A

dhiatto(fNs©

K

1

iti

pPipAi

0d

Cell Calcium 43 (2008) 344–355

NAADP+ is an agonist of the human P2Y11 purinergic receptor

Iliana Moreschi a, Santina Bruzzone a,∗, Nicoletta Bodrato a, Cesare Usai b, Lucrezia Guida a,Robert A. Nicholas c, Matthias U. Kassack d, Elena Zocchi a, Antonio De Flora a

a Department of Experimental Medicine, Section of Biochemistry, and Center of Excellence for Biomedical Research (CEBR),University of Genova, Viale Benedetto XV/1, 16132 Genova, Italy

b Institute of Biophysics, CNR, Via De Marini 6, 16149 Genova, Italyc Department of Pharmacology, University of North Carolina, Chapel Hill, NC, USA

d Institute of Pharmaceutical & Medicinal Chemistry, University of Duesseldorf, Building 26.23.01,Universitaetsstrasse 1, D-40225 Duesseldorf, Germany

Received 29 December 2006; received in revised form 8 May 2007; accepted 28 June 2007Available online 17 August 2007

bstract

Nicotinic acid adenine dinucleotide phosphate (NAADP+) is an intracellular second messenger releasing Ca2+ from intracellular stores inifferent cell types. In addition, it is also active in triggering [Ca2+]i increase when applied extracellularly and various underlying mechanismsave been proposed. Here, we used hP2Y11-transfected 1321N1 astrocytoma cells to unequivocally establish whether extracellular NAADP+

s an agonist of the P2Y11 receptor, as previously reported for �-NAD+ [I. Moreschi, S. Bruzzone, R.A. Nicholas, et al., Extracellular NAD+ isn agonist of the human P2Y11 purinergic receptor in human granulocytes, J. Biol. Chem. 281 (2006) 31419–31429]. Extracellular NAADP+

riggered a concentration-dependent two-step elevation of [Ca2+]i in 1321N1-hP2Y11 cells, but not in wild-type 1321N1 cells, secondary tohe intracellular production of IP3, cAMP and cyclic ADP-ribose (cADPR). Specifically, the transient [Ca2+]i rise proved to be related to IP3

verproduction and to consequent Ca2+ mobilization, while the sustained [Ca2+]i elevation was caused by the cAMP/ADP-ribosyl cyclaseADPRC)/cADPR signalling cascade and by influx of extracellular Ca2+. In human granulocytes, endogenous P2Y11 proved to be responsible

or the NAADP+-induced cell activation (as demonstrated by the use of NF157, a selective and potent inhibitor of P2Y11), unveiling a role ofAADP+ as a pro-inflammatory cytokine. In conclusion, we provide unequivocal evidence for the activation of a member of the P2Y receptor

ubfamily by NAADP+.2007 Elsevier Ltd. All rights reserved.

concen

Com

eywords: NAADP+; P2Y11 receptors; IP3; cADPR; Intracellular calcium

. Introduction

Nicotinic acid adenine dinucleotide phosphate (NAADP+)

s a potent Ca2+-releasing messenger in a wide variety of cellypes, from invertebrates to mammals [1–8]. In most systemsnvestigated, it has been demonstrated that NAADP+ triggers

Abbreviations: NAADP+, nicotinic acid adenine dinucleotide phos-hate; IP3, inositol-1,4,5-trisphosphate; ADPRC, ADP-ribosyl cyclase;PADS, pyridoxalphosphate-6-azophenyl-2′4′-disulfonic acid; [Ca2+]i,

ntracellular calcium concentration; cADPR, cyclic ADP-ribose; PLC, phos-holipase C; AC, adenylate cyclase; HBSS, Hank’s balanced salt solution;DPRP, 2′-phospho-ADP-ribose; PCA, perchloric acid; CI, chemotaxis

ndex∗ Corresponding author. Tel.: +39 010 3538158; fax: +39 010 354415.

E-mail address: santina.bruzzone@unige.it (S. Bruzzone).

[cdmt

qdbcc[

143-4160/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.oi:10.1016/j.ceca.2007.06.006

tration

a2+ release from intracellular stores through the activationf channels distinct from those sensitive to ryanodine and IP3,obilizing Ca2+ from thapsigargin-insensitive Ca2+ stores

1,9]. In sea urchin eggs, NAADP+ releases Ca2+ from acidicompartments, presumably lysosomes [10]. Other studies onifferent cell models, however, indicate that NAADP+ mightodulate the activity of ryanodine receptors, thus targeting

he endoplasmic reticulum [9,11–14].Beside the nature of the responsive stores, another open

uestion concerns the agonist-dependent endogenous pro-uction of NAADP+ in mammalian cells. NAADP+ has

een shown to be present in mammalian cells [15–17] andhanges in its intracellular content have been reported inholecystokinin-stimulated mouse pancreatic acinar cells16] and in human T cells upon stimulation of the T cell

l Calciu

rNg

ireCaa(mwoAiapNa

bercrcotnIiaddiatrd2(Ptbhiso

wtslwt

�neso[ewwtoat[lwnfleNl

2

2

Fsdfo

2

bfdPd

2

vMf

2

I. Moreschi et al. / Cel

eceptor/CD3 complex [17]. Therefore, in both cell systems,AADP+ fulfils the basic requirement of a second messen-er.

Although a CD38-unrelated and base-exchange-ndependent NAADP+-biosynthetic mechanism has beenecently reported in human myometrial cells [18], the base-xchange reaction catalyzed in mammals by the ectoenzymeD38 has been widely demonstrated to be the major mech-nism synthesizing NAADP+ from NADP+ and nicotiniccid [19]. CD38 is a member of the ADP-ribosyl cyclaseADPRC) family, which catalyzes the synthesis of variousolecules endowed with Ca2+-mobilizing activity [20,21],ith the first such reaction described being the formationf cyclic ADP-ribose (cADPR), formed from NAD+ [22].part from plasmamembrane localization, CD38 has been

dentified on various organelle membranes, thus probablyllowing the intracellular synthesis of its Ca2+-mobilizingroducts [20,23,24]. Nevertheless, presence of extracellularAADP+ is suggested by the ectocellular localization of thective site of CD38 and CD157 [20,24].

When exposed to extracellular NAADP+, both a ratasophilic and a neuroblastoma cell line actively transportxtracellular NAADP+ inside cells, with the consequentise in intracellular Ca2+ [25]. Murine cortical astrocytes,erebellar neurons, oligodendrocytes and microglial cellsespond with transient intracellular Ca2+ increases to extra-ellularly applied NAADP+ [26]. The underlying mechanismf intracellular Ca2+ increase apparently involves bothransport of NAADP+ through Connexin 43 hemichan-els (leading to Ca2+ release from lysosomes and fromP3-sensitive endoplasmic reticulum stores and to Ca2+

nflux through L-type Ca2+ channels), and activation ofdenosine receptors, likely resulting from extracellular degra-ation of NAADP+ to adenosine [26]. According to aifferent report, extracellularly applied NAADP+ promotesntracellular Ca2+ mobilization in rat astrocytes throughctivation of metabotropic P2Y purinergic receptors, sincehe Ca2+ increase was reduced or suppressed by the P2eceptor antagonist pyridoxalphosphate-6-azophenyl-2′4′-isulfonic acid (PPADS), by the IP3 receptor blocker-aminoethoxydiphenyl borate and by the phospholipase CPLC) inhibitor U73122 [27]. However, the specificity ofPADS is questionable and this reagent may act at sites dis-

inct from purinergic receptors [28–30]. Moreover, PPADSlocks the NAADP+-induced Ca2+-release in sea urchin eggomogenates and partly inhibits NAADP+ transport intontact NIH-3T3 cells [31]. Thus, PPADS lacks the requi-ite selectivity for unequivocal demonstration of activationf P2Y receptors by extracellular NAADP+.

To circumvent the problem and to conclusively establishhether P2Y receptors can be activated by NAADP+, we

ook advantage of transformed 1321N1 astrocytoma cells,

ince the wild-type cell line is not responsive to extracel-ular application of NAADP+ [26, this study]. Recently,e used 1321N1 cells expressing the human P2Y11 recep-

or (1321N1-hP2Y11 cells) to demonstrate that extracellularpw

m 43 (2008) 344–355 345

-NAD+ is an agonist of the P2Y11 receptor [32]. Sig-alling downstream of NAD+-induced P2Y11 stimulationvokes a two-step elevation of the [Ca2+]i: an initial tran-ient rise in [Ca2+]i is due to the production of IP3 (productf Gq-phospholipase C activation), and a second, sustainedCa2+]i increase is promoted by a cAMP-dependent gen-ration of cADPR. �-NAD+-induced increases in [Ca2+]iere observed only in 1321N1-hP2Y11 cells, but not inild-type 1321N1 cells [32]. Consistent with the aforemen-

ioned signalling mechanisms, the hP2Y11 receptor is thenly P2Y receptor that couples to both phospholipase C anddenylyl cyclase (AC) [33]. In the present study, we showhat extracellular NAADP+ triggers a two-step elevation ofCa2+]i in 1321N1-hP2Y11 cells, secondary to the intracel-ular production of IP3, cAMP and cADPR, similarly tohat was previously observed for �-NAD+. Since endoge-ous P2Y11 expressed on human granulocytes is responsibleor the NAD+-induced cell activation [32], we also chal-enged human granulocytes to migrate towards 0.1-10 �Mxtracellular NAADP+. The observed chemotaxis towardsAADP+ demonstrates a hitherto unknown role of extracel-

ular NAADP+ as a pro-inflammatory cytokine.

. Materials and methods

.1. Materials

The [3H]cyclic AMP and [3H]IP3 assay systems andicoll-Paque Plus were purchased from Amersham Bio-cience AB (Uppsala, Sweden). NF157 was synthesized asescribed [34]. Fluo-3AM and Fura-2AM were obtainedrom Calbiochem (Milan, Italy). All other chemicals werebtained from Sigma (Milan, Italy).

.2. Cell culture

1321N1 astrocytoma cell lines were cultured in Dul-ecco’s Modified Eagle’s Medium (DMEM) with 10%etal bovine serum and antibiotics (complete medium), asescribed before [32]. 1321N1 cells expressing the human2Y11 receptor were generated by retroviral infection asescribed previously [35].

.3. Isolation of human granulocytes

Buffy coats, prepared from freshly drawn blood of healthyolunteers, were provided by Galliera Hospital and by S.artino Hospital, Genova, Italy. Granulocytes were isolated

rom the buffy coats as described [36].

.4. Fluorimetric measurements of [Ca2+]i

1321N1 and 1321N1-hP2Y11 cells were seeded in 96 welllates at a density of 2 × 104 cells/well. After 24 h, cellsere loaded with 10 �M Fluo-3AM for 45 min at 37 ◦C in

3 l Calciu

caw4cGwslpe

ww594flcfp

2

pwPBv

2

s2Nit3ttBS

ast4I

2

s2

sp1otoeteA

apb(5Pca

2

s4Nas3ama

aosAo

2

(oaPntNsA3

46 I. Moreschi et al. / Cel

omplete medium and washed once with 0.2 ml Hank’s Bal-nced salt solution (HBSS, cat. no. H8264, Sigma). HBSSas then added (0.1 ml) and the fluorescence (excitation,85 nm; emission, 520 nm) measured every 3 s with a fluores-ence plate reader (Fluostar Optima; BMG LabtechnologiesmbH, Offenburg, Germany). The intensity of emitted lightas plotted as a function of time. Alternatively, cells were

eeded on 20-mm diameter coverslips (30 × 106 cells/dish),oaded with 10 �M Fura-2AM for 45 min at 37 ◦C in com-lete medium and washed once with 1 ml HBSS. [Ca2+]i wasvaluated as described previously [37].

Freshly prepared granulocytes (10 × 106/ml) were loadedith 10 �M Fluo-3AM for 45 min at 25 ◦C in RPMI medium,ashed with HBSS and resuspended in the same solution at× 106 cells/ml. [Ca2+]i measurements were performed in6 well plates (106 cells/well) and fluorescence (excitation,85 nm; emission, 520 nm) was measured every 3 s with theuorescence plate reader, as described above. For [Ca2+]ialibration, granulocytes were loaded with 10 �M Fura-2AMor 45 min at 25 ◦C in RPMI medium and measurements wereerformed as described before [36].

.5. Production of 2′-phospho-ADP-ribose (ADPRP)

We generated ADPRP by incubating 5 mM NADP+ in theresence of 0.06 U/ml NAD+-ase for 2.5 h at 37 ◦C, followinghich the enzyme was removed by filtration over Immobilon-membranes on a multiscreen vacuum manifold (Millipore,edford, MA). The complete degradation of NADP+ waserified on an aliquot of the reaction by HPLC analyses [38].

.6. Determination of intracellular IP3 levels

Control and hP2Y11-transfected 1321N1 cells wereeeded in 35 × 10 mm dishes (0.5 × 106 cells/dish). After4 h, the medium was replaced with 0.6 ml HBSS. NAADP+,AD+, NADP+ or buffer (control) were then added at the

ndicated concentrations, and at 0, 30 or 60 s after addition,he incubation was stopped by removal of HBSS, addition of00 �l ice-cold 0.6 M perchloric acid (PCA) and scraping ofhe cells. The IP3 content was measured on the supernatants ofhe neutralized cell extracts by radioimmunoassay ([3H]IP3iotrak Assay System, Amersham Bioscience AB, Uppsala,weden).

Granulocytes were resuspended in HBSS (40 × 106/ml)nd exposed to NAADP+. Aliquots (300 �l) of the suspen-ions were withdrawn at different times (0, 30 and 60 s) andhe reaction was stopped by adding 20 �l of 9 M PCA at◦C. PCA was removed as described [39]. The intracellular

P3 concentration was determined as described above.

.7. Determination of intracellular cAMP levels

Control and hP2Y11-transfected 1321N1 cells wereeeded in 35 × 10 mm dishes (0.5 × 106 cells/dish). After4 h, the medium was replaced with 0.6 ml HBSS. When

tc[a

m 43 (2008) 344–355

pecified, cells were pre-incubated for 10 min at 37 ◦C in theresence of the cAMP phosphodiesterase inhibitor Ro-20-724 (10 �M final concentration). NAADP+, NAD+, NADP+

r buffer (control) were then added at the indicated concentra-ions: incubations were stopped after 0 or 15 min by removalf HBSS and addition of 200 �l ice-cold PCA (0.6 M). Cellxtracts were collected and centrifuged to remove proteins;he cAMP content was measured on the neutralized cellxtracts by a radioimmunoassay ([3H]cAMP assay system,mersham Bioscience AB, Uppsala, Sweden).Granulocytes were resuspended in HBSS (40 × 106/ml)

nd exposed to NAADP+. When specified, granulocytes werere-incubated for 10 min at 25 ◦C with 10 �M Ro-20-1724efore exposure to NAADP+. Aliquots of the suspensions300 �l) were withdrawn at different times (0, 1, 2.5 andmin) and the reaction was stopped by adding 20 �l of 9 MCA at 4 ◦C. PCA was removed as described [39]. The intra-ellular cAMP concentration was determined as describedbove.

.8. Determination of intracellular cADPR levels

Control and hP2Y11-transfected 1321N1 cells wereeeded in 35 × 10 mm dishes (0.25 × 106 cells/dish). After8 h, the medium was replaced with HBSS (0.6 ml).AADP+, �-NAD+, NADP+ or buffer (control) were thendded at the indicated concentrations, and the incubation wastopped after 15 min by removal of HBSS and addition of00 �l ice-cold PCA (0.6 M). Cell extracts were collectednd centrifuged to remove proteins; the cADPR content waseasured on the neutralized perchloric acid cell extracts byhighly sensitive enzymatic cycling assay [39].

Granulocytes (40 × 106/ml) were incubated in HBSS for 0nd 15 min at 25 ◦C in the absence (control) or in the presencef NAADP+. At each time point, a 300 �l-aliquot of the celluspension was withdrawn and 20 �l of 9 M PCA were added.fter removal of proteins, the cADPR content was measuredn the neutralized extracts as described above.

.9. Assays of ADP-ribosyl cyclase activity

Intact control and hP2Y11-transfected 1321N1 cells4 × 106/ml) were incubated at 25 ◦C in the absence (control)r in the presence of 100 �M NAADP+ (for 10 min). Afterddition of 1:500 protease inhibitor cocktail (Sigma, cat. no.8340) and 1:100 phosphatase inhibitor cocktail (Sigma, cat.o. P2850), control and stimulated cells were lysed by sonica-ion in ice for 1 min at 3 W (Heat-System Ultrasonics, W380,ew York, NY, USA). ADP-ribosyl cyclase activity was mea-

ured at 37 ◦C on cell lysates by adding 0.1 mM �-NAD+.liquots (100 �l) were withdrawn at various times (0, 10,0 min) and the enzymatic reactions were stopped by addi-

ion of 220 �l of 0.9 M PCA to each aliquot. The cADPRoncentration was measured by the enzymatic cycling assay39]. Protein determination in each sample was performedccording to Bradford [40].

l Calciu

2

tCsU(tp

2

w

3

3P

vta(eAicda

tibtarorarwNs

ttwidNo1

i1F1Nohpt[<RFte1woAmpAHNvwtiFttatr

3Nc

(peuIscilWuh

I. Moreschi et al. / Cel

.10. Chemotaxis

Granulocytes were resuspended at 10 × 106/ml in chemo-axis buffer (HBSS, PBS and 5% albumin; 39:16:1 by vol.).hemotaxis assays were performed using 96-well ChemoTx

ystem microplates (Neuro Probe, Inc., Gaithersburg, MD,SA) with a 3-�m pore size polycarbonate filter. NAADP+

0.1–10 �M) was diluted in chemotaxis buffer and added inhe bottom wells. Cell migration was evaluated as describedreviously [36].

.11. Statistical analyses

All parameters were tested by paired t test. p-Values < 0.05ere considered significant.

. Results

.1. Extracellular NAADP+ increases the [Ca2+]i of2Y11

+ 1321N1 astrocytoma cells

In order to establish whether the P2Y11 receptor is acti-ated by extracellular NAADP+ (NAADP+

e ), we investigatedhe effect of this dinucleotide on [Ca2+]i in 1321N1 humanstrocytoma cells stably expressing the hP2Y11 receptor1321N1-hP2Y11 cells) [35]. To rule out potential interfer-nce by metabolic by-products of NAADP+

e (e.g., NAAD,MP, adenosine), we verified the stability of NAADP+

e dur-ng incubation with both native 1321N1 and 1321N1-hP2Y11ells. No significant degradation of NAADP+ (100 �M), asetermined by HPLC analysis [38], was observed over 60 mint 25 ◦C.

As shown in Fig. 1A and B, addition of 100 �M NAADP+e

o 1321N1-hP2Y11 cells resulted in a rapid and transientncrease of [Ca2+]i (from 65 ± 9 to 175 ± 20 nM), followedy a slow, sustained [Ca2+]i rise. Recording the [Ca2+]i for upo 30 min revealed that a plateau was reached 15 min after theddition of the dinucleotide (Fig. 1A). Conversely, as alreadyeported [26], NAADP+

e did not induce any Ca2+ responsen control 1321N1 cells (Fig. 1A), indicating that the P2Y11eceptor was responsible for NAADP+

e -promoted transientnd sustained increases of [Ca2+]i. As expected from aeceptor-mediated event, NAADP+

e -induced Ca2+ responsesere concentration-dependent. As shown in Fig. 1B, 1 �MAADP+

e was sufficient to trigger both the transient and theustained [Ca2+]i elevation.

To ascertain that the Ca2+ responses observed upon addi-ion of NAADP+

e to 1321N1-hP2Y11 cells were due solelyo NAADP+ and not to other contaminating nucleotides,e determined the levels of contaminating nucleotides

n commercial batches of NAADP+. HPLC analysis [38]

emonstrated that NAADP+ contains ≤3% ADPRP and ≤1%ADP+, ATP and ADP. The addition to 1321N1-hP2Y11 cellsf ADPRP at the concentration expected to be present in00 �M NAADP+

e (i.e. 3 �M, see Section 2), did not result

btt(

m 43 (2008) 344–355 347

n increases in [Ca2+]i. Likewise, addition of NADP+ up to00 �M also did not evoke any Ca2+ response (see below,ig. 2A). Thus, the possibility that the effects obtained with00 �M NAADP+

e were due to contaminating ADPRP orADP+ was ruled out. We also investigated whether ADPr ATP, at the concentrations expected in 100 �M NAADP+

e ,ad any effects on [Ca2+]i levels in 1321N1-hP2Y11 cells,articularly since these nucleotides have proven activity athe P2Y11 receptor [35]. ADP, at 1 �M, had no effect onCa2+]i, whereas 1 �M ATP increased Ca2+ only slightly, to10% of the level triggered by 100 �M NAADP+ (Fig. 1C).esponsiveness of 1321N1-hP2Y11 cells to ATP (inset ofig. 1C) was routinely checked and it was always similar

o that previously reported [32]. Although these data appar-ntly rule out the possibility that the effects obtained with00 �M NAADP+

e were due to contaminating ADP or ATP,e nevertheless incubated NAADP+ (10 mM) in the presencef 0.4 U/ml apyrase for 10 min at 25 ◦C to degrade ATP andDP, and then the enzyme was removed by filtration with aultiscreen vacuum manifold using Immobilon-P membrane

lates. The complete degradation of contaminating ADP andTP was verified on an aliquot of the incubation mixture byPLC analysis [38]. The addition of 100 �M apyrase-treatedAADP+ to 1321N1-hP2Y11 cells resulted in a two-step ele-ation of [Ca2+]i that was superimposable to the one obtainedith the same concentration of untreated NAADP+ (Fig. 1D),

hus conclusively excluding the possibility that contaminat-ng ADP or ATP were responsible for the effects on [Ca2+]i.inally, in other experiments, apyrase (0.1 U/ml, final concen-

ration) was added to cells immediately before their exposureo 100 �M NAADP+

e : as shown in Fig. 1E, the presence ofpyrase had no effect on NAADP+-promoted Ca2+ responses,hus ruling out the possibility that NAADP+ induced theelease of ADP or ATP from cells.

.2. Role of extracellular Ca2+, IP3 and cADPR inAADP+

e -promoted [Ca2+]i increase in 1321N1-hP2Y11

ells

Among the human G protein-coupled P2Y receptorhP2Y) subtypes cloned to date, only the hP2Y11 cou-les to both PLC and AC [33,35]. Addition of 100 �Mxtracellular NAD+ (NAD+

e ) to 1321N1-hP2Y11 cells stim-lated intracellular production of both IP3 and cAMP, withP3 mediating the transient Ca2+ release from intracellulartores [32]. Moreover, NAD+

e -promoted increases of intra-ellular cAMP was followed by a PKA-mediated increase inntracellular cADPR, which in turn was responsible for long-asting Ca2+ increases, due to extracellular Ca2+ influx [32].

e compared NAADP+e -promoted effects to those obtained

pon addition of extracellular NAD+ or NADP+ in 1321N1-P2Y11 cells. Fig. 2A shows the [Ca2+]i changes evoked

y each compound (100 �M). NAADP+ generated a greaterransient increase of [Ca2+]i than NAD+. In contrast, addi-ion of NADP+ did not evoke any transient [Ca2+]i increaseFig. 2A). NAADP+ was as effective as NAD+, with NADP+

348 I. Moreschi et al. / Cell Calcium 43 (2008) 344–355

Fig. 1. NAADP+e -induced [Ca2+]i increase in 1321N1-hP2Y11 cells. (A) Fluo-3AM-loaded control (P2Y11

−, grey) or hP2Y11-transfected (black) 1321N1cells were treated with 100 �M NAADP+; (B) Fura-2AM-loaded 1321N1-hP2Y11 cells were challenged with 1 (light grey), 10 (grey) or 100 �M NAADP+

(black). Fluo-3AM-loaded 1321N1-hP2Y11 cells were treated with: (C) 1 �M ATP (grey) or with 100 �M NAADP+ (black) or with 100 �M ATP (inset); (D)100 �M NAADP+ (black) or apyrase-treated NAADP+ (grey); (E) 100 �M NAADP+, in the presence (grey) or absence (black) of 0.1 U/ml apyrase in thewells. Recordings were carried out at 25 ◦C in HBSS. [Ca2+]i changes were measured in parallel using a fluorescence plate reader (A, C–E), or with the systemd of the nn the twoa

bv

paEi[iwtpiw

NttrCiwugl

escribed in Ref. [37] for [Ca2+]i calibration (B). Arrows indicate addition= 8; B, n = 3; C–E, n = 5) (S.D. < 15%). (A, C and D) The p-values betweent the maximum of the first peak and at 750 s: *p < 0.01, where indicated.

eing ineffective, for promotion of the sustained [Ca2+]i ele-ation (Fig. 2A).

In order to evaluate whether NAADP+e utilized a signalling

athway similar to that of NAD+e , 100 �M NAADP+ was

dded to 1321N1-hP2Y11 cells in the presence of eitherGTA to chelate extracellular Ca2+, or U73122, a PLC

nhibitor, or 8-Br-cADPR, a specific antagonist of cADPR41]. The rapid and transient NAADP+

e -promoted elevationn [Ca2+]i was due to release from intracellular stores, since itas not affected by extracellular EGTA (Fig. 2B). In contrast,

he sustained [Ca2+]i rise elicited by NAADP+e was com-

letely inhibited by extracellular EGTA, implicating Ca2+

nflux (Fig. 2B). Pre-incubation of 1321N1-hP2Y11 cellsith U73122 completely abrogated the rapid and transient

i1c

ucleotide. Traces are mean values recorded in different measurements (A,groups of Δ fluorescence values from each individual trace were calculated

AADP+e -promoted increase in [Ca2+]i, whereas the sus-

ained [Ca2+]i elevation was unaffected (Fig. 2C), suggestinghe involvement of a PLC activity downstream of the P2Y11eceptor in triggering the rapid [Ca2+]i rise. The sustaineda2+ rise, which was due to extracellular Ca2+ influx, was

nhibited (75%) by pre-incubation of 1321N1-hP2Y11 cellsith 8-Br-cADPR, whereas the first [Ca2+]i increase wasnaffected (Fig. 2D). Thus, the sustained [Ca2+]i rise is trig-ered by the P2Y11 receptor via a PLC-independent pathway,eading to a cADPR-mediated influx of extracellular Ca2+.

To further characterize the intracellular calcium storesnvolved in the NAADP+

e -promoted elevation of the [Ca2+]i in321N1-hP2Y11 cells, either the ER or the acidic organellesalcium stores were depleted, respectively, by addition of

I. Moreschi et al. / Cell Calciu

Fig. 2. Comparison of Ca2+ responses to NAADP+ and NAD+ in 1321N1-hP2Y11 cells: role of extracellular Ca2+, IP3 and cADPR. Fluo-3AM-loaded1321N1-hP2Y11 cells were: (A) treated with 100 �M NAADP+ (black),NAD+ (grey), or NADP+ (light grey) in HBSS; inset shows statistical analy-sis of the first peak (white bars) and of values at 750 s (black bars): *p < 0.01;(B) treated with 100 �M NAADP+ in Ca2+ buffer (HBSS) (black) or inCa2+-free HBSS (cat. no. H6648, Sigma), in the presence of 0.2 mM EGTA(grey); (C) pre-incubated (or not, black) for 10 min in HBSS in the presenceof 5 �M U73122 (grey) and then challenged with 100 �M NAADP+. (D)1321N1-hP2Y11 cells were pre-incubated (or not, black) for 90 min incomplete medium with 100 �M 8-Br-cADPR (grey), loaded with Fluo-3AMduring the last 45 min of pre-incubation and then treated with100 �MNAADP+. B-D, the p-values between the two groups of � fluorescencevalues from each individual trace were calculated at the maximal value of

tbAegspiwot[

ttiaC

3c

pitshcwNchtwncctwNswRri-i1a

tinn

m 43 (2008) 344–355 349

hapsigargin, an inhibitor of the SERCA pumps [42], or ofafilomycin A1, an inhibitor of V-type H+-ATPase [43–45].s shown in Fig. 3A, 1 �M thapsigargin induced a rapid

levation of [Ca2+]i and a second addition of 1 �M thapsi-argin was without further effect, indicating that the ER Ca2+

tores were depleted. Pre-treatment with thapsigargin com-letely abrogated the rapid and transient NAADP+

e -promotedncrease in [Ca2+]i, whereas the sustained [Ca2+]i elevationas unaffected (Fig. 3A). On the contrary, pre-incubationf cells with 100 nM bafilomycin A1 did not modify eitherhe transient or the sustained NAADP+

e -promoted increase inCa2+]i (Fig. 3B).

Taken together, these data demonstrate that the rapid andransient NAADP+

e -promoted elevation of the [Ca2+]i is dueo IP3-mediated release from ER Ca2+ stores. The sustainedncrease of the [Ca2+]i is independent of ER calcium storesnd is due to a cADPR-mediated influx of extracellulara2+.

.3. Extracellular NAADP+ increases intracellular IP3,AMP and cADPR levels in 1321N1-hP2Y11 cells

The results described above clearly indicate NAADP+e -

romoted activation of the hP2Y11 purinergic receptor, whichs coupled to both PLC and AC [33,35]. Thus, we measuredhe intracellular concentrations of IP3 and cAMP, the twoecond messengers generated following stimulation of theP2Y11 receptor in 1321N1-hP2Y11 cells. In native 1321N1ells, the basal value of [IP3]i (6.73 ± 0.21 pmol/106 cells)as not significantly increased by incubation with 100 �MAADP+

e . In contrast, addition of NAADP+e caused a rapid,

oncentration-dependent elevation of IP3 levels in 1321N1-P2Y11 cells, with 10 �M NAADP+

e approximately doublinghe basal value recorded in unstimulated cells (Table 1). Next,e determined whether NAADP+

e increased cAMP levels inative 1321N1 and 1321N1-hP2Y11 cells. In the transfectedells, concentrations of NAADP+

e as low as 1 �M NAADP+e

aused a significant increase of the intracellular cAMP con-ent (Table 1). cAMP levels measured in control 1321N1 cellsere not significantly increased in the presence of 100 �MAADP+

e . The basal amount of intracellular cAMP was notignificantly different in 1321N1-hP2Y11 cells pre-incubatedith or without the cAMP phosphodiesterase (PDE) inhibitoro-20-1724 (4.65 ± 0.81 versus 4.29 ± 0.72 pmol/106 cells,

espectively). Moreover, the percentage of increase of thentracellular cAMP content was similar in PDE-inhibited or

uninhibited conditions: in the presence of Ro-20-1724, thentracellular cAMP content was 6.05 ± 0.92, 7.16 ± 1.07 and4.64 ± 2.92 pmol/106 cells, as measured after 15 min fromddition of 1, 10 or 100 �M NAADP+

e , respectively (n = 9).

he first peak and at 750 s: *p < 0.01. [Ca2+]i changes were measured at 25 ◦Cn parallel using a fluorescence plate reader; arrows indicate addition of theucleotide. Traces are mean values recorded in different measurements (A,= 8; B, C and D, n = 6) (S.D. < 15%).

350 I. Moreschi et al. / Cell Calciu

Fig. 3. Effect of thapsigargin and bafilomycin A1 on NAADP+e -induced

[Ca2+]i increase in 1321N1-hP2Y11 cells. Fluo-3AM-loaded 1321N1-hP2Y11 cells were (A) treated (a and b) or not (c) with 1 �M thapsigargin(TG) and then challenged (a and c) or not (b) with 100 �M NAADP+; sta-tistical analysis between a and b: the p-values between the two groups ofvalues from each individual trace were calculated after 750 s from addition ofNAADP+; *p < 0.01. (B) 1321N1-hP2Y11 cells were pre-incubated (grey) ornot (black) for 15 min with 100 nM bafilomycin A1 and then challenged with100 �M NAADP+; the arrow indicates addition of the nucleotide; no statis-tical significance was observed between the two groups of Δ fluorescencevan

TR

be

i1tAaphii1

vftah1w

[NaNviitrwmi

cNii1pEc

3[h

m[wF[b

+

alues from each individual trace calculated at the maximum of the first peaknd at 750 s. Traces are mean values recorded in different measurements (A,= 6; B, n = 5) (S.D. < 15%).

he corresponding cAMP values measured in the absence of

o-20-1724 are shown in Table 1.

The inhibition of the sustained increase in [Ca2+]i affordedy 8-Br-cADPR (Fig. 2D) prompted us to measure theffects of NAADP+ on ADP-ribosyl cyclase (ADPRC) activ-

1oaf

m 43 (2008) 344–355

ty (responsible for cADPR generation) and cADPR levels in321N1-hP2Y11 cells. Cells were incubated for 10 min inhe presence or absence of 100 �M NAADP+

e and levels ofDPRC activity were measured in cell lysates using �-NAD+

s substrate. ADPRC activity increased from 0.11 ± 0.03mol cADPR/min/mg protein (in unstimulated 1321N1-P2Y11 cells) to 0.18 ± 0.04 pmol cADPR/min/mg proteinn NAADP+

e -treated cells (n = 5; p < 0.05). Conversely, noncrease of the ADPRC activity was recorded in wild-type321N1 cells stimulated with NAADP+

e (not shown).Likewise, the levels of cADPR increased to 160% of basal

alue (0.30 ± 0.04 pmol/106 cells) in 1321N1-hP2Y11 cellsollowing addition of 100 �M NAADP+

e (Table 1). The addi-ion of concentrations of NAADP+ as low as 1 �M caused

significant increase of intracellular cADPR in 1321N1-P2Y11 cells (Table 1), whereas cADPR levels in control321N1 cells (basal value 0.28 ± 0.04 pmol/106 cells, n = 5)ere not significantly increased by 100 �M NAADP+

e .We also comparatively measured increases in [IP3]i,

cAMP]i and [cADPR]i elicited by the addition of 100 �MAADP+, NAD+, or NADP+ to 1321N1-hP2Y11 cells. Ingreement with its effect on the transient [Ca2+]i increase,AADP+

e also induced the highest elevation (235% of basalalue in unstimulated cells after 1 min from the addition)n [IP3]i levels (Table 2). Notably, NAADP+

e -induced peakn [IP3]i showed a time-dependent pattern different fromhat promoted by NAD+

e : NAD+e -induced [IP3]i production

eached a maximum at 30 s after addition of the dinucleotide,hereas the [IP3]i production promoted by NAADP+

e wasaximal after 1 min (Table 2). NADP+

e did not evoke anyncrease of [IP3]i (Table 2).

Extracellular NAD+ (100 �M) triggered an increase inAMP levels which was comparable to that elicited byAADP+

e (Table 2). The intracellular production of cAMPnduced by NAD+

e and NAADP+e translated to comparable

ncreases (50% over basal value) in the [cADPR]i levels of321N1-hP2Y11 cells (Table 2), in agreement with the com-arable cADPR-induced sustained Ca2+ increases (Fig. 2A).xtracellular NADP+ did not evoke any increase of cAMP orADPR (Table 2).

.4. Extracellular NAADP+ increases [Ca2+]i, [IP3]i,cAMP]i and [cADPR]i and stimulates chemotaxis inuman granulocytes

Endogenous P2Y11 receptors have been demonstrated toediate NAD+

e -promoted activation of human granulocytes32]. We challenged freshly purified human granulocytesith increasing concentrations of NAADP+. As shown inig. 4A, NAADP+ promoted both a transient and a sustainedCa2+]i increase in a concentration-dependent manner, withoth responses occurring at [NAADP+] as low as 1 �M. At

00 �M, NAADP promoted a rapid and transient increasef [Ca2+]i (from 50 ± 7 to 200 ± 26 nM, n = 4), followed byslow, sustained [Ca2+]i rise (to 180 ± 21 nM after 10 min

rom the addition of the dinucleotide, n = 4). Prolonging the

I. Moreschi et al. / Cell Calcium 43 (2008) 344–355 351

Table 1NAADP+

e -induced intracellular IP3, cAMP and cADPR increases in 1321N1-hP2Y11 cells

NAADP+e (�M) IP3

a (pmol/106 cells) cAMPb (pmol/106 cells) cADPRc (pmol/106 cells)

0 27.94 ± 2.68 4.29 ± 0.72 0.30 ± 0.041 32.13 ± 2.91* 5.96 ± 0.83* 0.42 ± 0.03**

10 57.00 ± 6.84** 7.05 ± 1.11** 0.44 ± 0.04**

100 65.38 ± 10.02** 13.95 ± 2.37** 0.48 ± 0.02**

a IP3 levels were determined after 1 min from addition of NAADP+ (n = 6).b cAMP levels were determined after 15 min from addition of NAADP+ (n = 5).c cADPR levels were measured after 15 min from addition of NAADP+ (n = 6).* p < 0.05.

** p < 0.001.

Table 2Effect of NAADP+, NAD+ or NADP+ on intracellular IP3, cAMP and cADPR levels in 1321N1-hP2Y11 cells

Addition (100 �M) IP3a (pmol/106 cells) cAMPb (pmol/106 cells) cADPRc (pmol/106 cells)

– 28.12 ± 2.35 4.42 ± 0.96 0.31 ± 0.03NAADP+ 65.94 ± 9.89** 14.18 ± 3.65** 0.52 ± 0.04**

NAD+ 44.92 ± 5.38** 16.02 ± 2.96** 0.53 ± 0.03**

NADP+ 29.35 ± 3.02 4.25 ± 0.75 0.28 ± 0.04

a IP3 levels were determined after 0.5 and 1 min for NAD+ and NAADP+, respectively, from addition of the nucleotide (n = 8).ide (n =de (n =

ii(

[eiatfiN[oai

cp1s

sinp

cclTieario2

TN

N

1

b cAMP levels were determined after 15 min from addition of the nucleotc cADPR levels were measured after 15 min from addition of the nucleoti

** p < 0.001.

ncubation time (up to 30 min) resulted in a further slightncrease of the [Ca2+]i, suggesting that a plateau was reachednot shown).

In order to demonstrate that the NAADP+-promotedCa2+]i variations were due to activation of an endogenouslyxpressed P2Y11 receptor, human granulocytes were pre-ncubated with NF157, a selective and potent P2Y11 receptorntagonist [34], used previously to demonstrate activation ofhe P2Y11 receptor by NAD+

e [32]. We pre-emptively veri-ed that pre-incubation of 1321N1-hP2Y11 cells with 1 �MF157 strongly inhibited (by 90%) the NAADP+-induced

Ca2+]i rise (not shown). As shown in Fig. 4B, pre-incubationf granulocytes with 1 �M NF157 partially inhibited (bypproximately 60%) both the transient and the sustained Ca2+

ncrease induced by 100 �M NAADP+.Results comparable to those shown in 1321N1-hP2Y11

ells (Fig. 3), were also obtained on human granulocytes

re-incubated with thapsigargin or bafilomycin A1. Thus,�M thapsigargin completely abrogated the rapid and tran-

ient NAADP+e -promoted increase in [Ca2+]i, whereas the

mNwa

able 3AADP+

e -induced intracellular IP3, cAMP and cADPR increases in human granulo

AADP+e (�M) IP3

a (pmol/106 cells)

0 0.23 ± 0.031 0.35 ± 0.05*

10 0.46 ± 0.03**

00 0.52 ± 0.05**

a IP3 levels were determined after 1 min from addition of NAADP+ (n = 5).b cAMP levels were determined after 2.5 min from addition of NAADP+ (n = 6).c cADPR levels were measured after 15 min from addition of NAADP+ (n = 5).* p < 0.05.

** p < 0.001.

5).5).

ustained [Ca2+]i elevation was unaffected (not shown). Pre-ncubation of granulocytes with 100 nM bafilomycin A1 didot modify either the transient or the sustained NAADP+

e -romoted increase in [Ca2+]i (not shown).

As expected, application of NAADP+e increased intra-

ellular concentrations of IP3, cAMP and cADPR in aoncentration-dependent manner, and these increases corre-ated with the two [Ca2+]i responses to NAADP+

e (Table 3).he basal amount of intracellular cAMP was 10-fold higher

n granulocytes pre-incubated with the cAMP phosphodi-sterase (PDE) inhibitor Ro-20-1724, compared to cells in thebsence of the PDE inhibitor (1.60 ± 0.24 versus 0.15 ± 0.03,espectively). Nevertheless, the percentage of increase of thentracellular cAMP content was similar in PDE-inhibitedr -uninhibited cells: the intracellular cAMP content was.08 ± 0.25, 2.19 ± 1.12 and 2.44 ± 2.66 pmol/106 cells, as

easured after 2.5 min from addition of 1, 10 or 100 �MAADP+

e , respectively (n = 9) to granulocytes pre-incubatedith Ro-20-1724; the corresponding cAMP values in the

bsence of Ro-20-1724 are shown in Table 3.

cytes

cAMPb (pmol/106 cells) cADPRc (pmol/109 cells)

0.15 ± 0.03 25.5 ± 3.30.18 ± 0.02* 33.5 ± 5.2*

0.20 ± 0.04* 42.6 ± 6.5**

0.23 ± 0.03* 51.6 ± 6.9**

352 I. Moreschi et al. / Cell Calcium 43 (2008) 344–355

Fig. 4. NAADP+-induced [Ca2+]i increase in human granulocytes. (A) Fura-2AM-loaded granulocytes were challenged at 25 ◦C with 1 (light grey), 10(grey) or 100 �M NAADP+ (black); [Ca2+]i changes were measured asdescribed in Ref. [36]. (B) Granulocytes were pre-incubated (during the45 min loading with Fluo-3AM) in the absence (black) or in the presenceof 1 �M NF157 (grey) and then treated with 100 or �M NAADP+. [Ca2+]i

changes were measured at 25 ◦C in parallel using a fluorescence plate reader.Arrows indicate addition of the nucleotide. Traces are mean [Ca2+]i valuesrecorded in n = 4 (A) and n = 5 (B) different measurements (S.D. < 15%). (B)Tw

tNcN1TwNa11pi1rrw

Fig. 5. Chemotaxis of human granulocytes towards different concentrationsof NAADP+

e . Migration of granulocytes through 3 �m pore membranestowards a solution containing NAADP+ (white bars) at the concentrationsindicated, was measured in ChemoTx chambers and expressed as ChemotaxIndex (CI). Results shown are the mean ± S.D. from 7 experiments (p < 0.01for 0.1 �M; p < 0.0001 for 1 and 10 �M). Granulocytes pre-incubated with1(tg

m(s

4

ctttaorcecmlwaurraaPIarnNAADP+ transport in NIH-3T3 cells, thus indicating that

he p-values between the two groups of Δ values from each individual traceere calculated at the maximum of the first peak and at 750 s: *p < 0.01.

Finally, we measured the capacity of human granulocyteso migrate towards NAADP+. Different concentrations ofAADP+

e were placed in the bottom well of a chemotaxishamber: maximal migration was observed towards 10 �MAADP+ (Fig. 5), with concentrations of NAADP+ as low as00 nM sufficient to trigger granulocytes migration (Fig. 5).he chemotaxis index (CI) calculated for 10 �M NAADP+

as similar to that observed with the same concentration ofAD+ [36,32] and of ATP, a nucleotide known to behave aschemoattractant [46]. Interestingly, the CI calculated for�M NAADP+ was much higher than that obtained with�M NAD+ (4 versus 1.6, Ref. [36]). As shown in Fig. 5,re-incubation of cells with 1 �M NF157 partially inhib-ted (by 70%) chemotaxis of human granulocytes towards0 �M NAADP+, thus demonstrating that endogenous P2Y11

eceptor is responsible for NAADP+-induced chemotacticesponse. Moreover, pre-incubation of human granulocytesith 100 �M 8-Br-cADPR greatly reduced (by approxi-

Pir

�M NF157 for 45 min (grey bars) or with 100 �M 8-Br-cADPR for 90 minblack bars) were also challenged with 10 �M NAADP+. Results shown arehe mean ± S.D. from 5 experiments (p < 0.001 compared to CI values ofranulocytes not pre-treated with any inhibitor).

ately 65%) the migration induced by 10 �M NAADP+

Fig. 5), highlighting a major role of the cADPR/[Ca2+]iignalling cascade in the chemotactic response to NAADP+.

. Discussion

NAADP+ is the most potent Ca2+ mobilizer from intra-ellular stores in sea urchin homogenates [6]. In mammals,he catalytic domain of the known enzymes responsible forhe synthesis of NAADP+ (CD38 and CD157) localizes tohe outer surface of cells expressing ADP-ribosyl cyclasectivity. Thus, in addition to its role as an intracellular sec-nd messenger, recent studies have focused on a possibleole of extracellular NAADP+ in autocrine and paracrineell signalling processes, as previously demonstrated forxtracellular NAD+ and cADPR [24]. Specifically, murineortical astrocytes, cerebellar neurons, oligodendrocytes andicroglia [26], rat astrocytes [27] and a rat basophilic cell

ine [25] respond to extracellular application of NAADP+

ith a [Ca2+]i increase. As far as the underlying mechanismsre concerned, Heidemann et al. [26], by means of contin-ous application of ATP in order to desensitize purinergiceceptors, excluded the action of NAADP+ via purinergiceceptors in murine astrocytes. On the contrary, Singaravelund Deitmer [27], through the use of PPADS (a P2 receptorntagonist), 2-APB (an IP3 receptor blocker) and U73122, aLC inhibitor, suggested that NAADP+ recruits Ca2+ fromP3-sensitive Ca2+ stores in rat astrocytes, at least in part byctivating metabotropic P2Y receptors. However, a recenteport demonstrated that PPADS is a competitive antago-ist of the sea urchin NAADP+ receptor and that it inhibits

PADS is not a suitable molecule to investigate the potentialnvolvement of P2Y receptors in NAADP+-promoted Ca2+

esponses [31].

l Calciu

ecNtplNhamviNoAIciat[c[cofibiabC

1bmh1NarPT(oicitttPoP[

opgwnoami[Nb[[ihti

cma[

caets

A

AfSRFd

R

I. Moreschi et al. / Cel

Our study, exploiting a 1321N1 astrocytoma cell linexpressing the hP2Y11 receptor (1321N1-hP2Y11), con-lusively and unequivocally demonstrates that extracellularAADP+ activates the hP2Y11 purinergic receptor. Indeed,

he native cell line, which does not express endogenousurinergic receptors [47], showed no response to extracel-ular application of NAADP+ [26, this study]. In contrast,AADP+

e elicited a two-step [Ca2+]i elevation in 1321N1-P2Y11 cells. The possibility exists that NAADP+ can alsoctivate other members of the P2Y family. The signallingechanisms underlying NAADP+-induced [Ca2+]i increase

ia stimulation of the P2Y11 receptor are very similar, if notdentical, to those which follow binding of this receptor byAD+

e [32]. Thus, irrespective of activation with either NAD+

r NAADP+, the P2Y11 receptor couples to both PLC and AC.ctivation of PLC increases the intracellular production of

P3 and causes subsequent transient Ca2+ release from intra-ellular stores, while activation of AC results in an increase ofntracellular cAMP concentration, up-regulation of ADPRCctivity and accumulation of intracellular cADPR, whichriggers a 8-Br-cADPR-inhibitable extracellular Ca2+ influx32, this study]. cADPR, an intracellular Ca2+ mobilizer,ould stimulate Ca2+ influx by two possible mechanisms48,49]: (i) direct opening by intracellular cADPR of a Ca2+

hannel on the plasmamembrane; (ii) activation of store-perated calcium channels (SOCC) induced by Ca2+ releasedrom ryanodine receptors. The fact that the sustained [Ca2+]increase triggered by NAADP+ is insensitive to ER depletiony thapsigargin both in 1321N1-hP2Y11 cells (Fig. 3A) andn human granulocytes (see Section 3), seems to rule out thatcADPR-mediated Ca2+ release from ER stores is responsi-le for the generation of a subsequent influx of extracellulara2+.

The IP3-mediated, rapid and transient Ca2+ rise induced by00 �M NAADP+ was significantly higher than that inducedy 100 �M NAD+ (Fig. 2A). This was also confirmed byeasurements of the intracellular IP3 levels, which were

igher in cells stimulated with 100 �M NAADP+ than with00 �M NAD+ (Table 2). In contrast, 100 �M NAADP+ orAD+ induced a similar increase of the intracellular cAMPnd cADPR concentrations and a similar sustained [Ca2+]iise (Fig. 2A and Table 2). Neither the P2Y11/PLC nor the2Y11/AC coupling was induced by NADP+ (Fig. 2A andable 2). Thus, the simultaneous presence of nicotinic acidi.e. the negative charge on the 3-position of the pyridine ringf NAADP+) and of the 2′ phosphate group, as occurringn NAADP+, seems to play a significant role in the PLC-oupled P2Y11 activation. Interestingly, the phosphate group,n the presence of the nicotinamide moiety, apparently yieldshe opposite effect, rendering the NADP+ molecule unableo trigger either the PLC- or the AC-coupled P2Y11 activa-ion. The different activity of NAADP+ and NAD+, regarding

LC- and AC coupling with P2Y11 is not surprising: previ-us studies have demonstrated that P2Y11 couples to AC andLC with different efficiency depending on the agonist used33,35,32].

m 43 (2008) 344–355 353

Finally, although being aware of the fact that no physi-logical evidence is available so far for either extracellularresence or functions of NAADP+, we challenged humanranulocytes with NAADP+

e , at concentrations comparableith those used with other cell types [25–27], since endoge-ous P2Y11 is responsible for the NAD+-induced activationf these cells [32]. Interestingly, NAADP+ proved to beble to attract granulocytes along a concentration gradient,ainly via the P2Y11/cADPR pathway, as demonstrated by

nhibitory effect of NF157, a specific inhibitor of P2Y1134], and of 8-Br-cADPR (Fig. 5). Uptake of extracellularAADP+ by human granulocytes, as demonstrated in a ratasophilic cell line through a still unidentified transporter25], or in rat astrocytes through Connexin 43-hemichannels26], cannot be ruled out. The contribution of this hypothet-cal NAADP+ influx to the NAADP+

e -induced response inuman granulocytes would in any case be minimal, givenhe strong inhibition afforded by NF157 on the NAADP+

e -nduced Ca2+ increase and chemotactic response.

Failure of NADP+ to induce any changes in [Ca2+]i, IP3,AMP and cADPR levels in 1321N1-P2Y11 cells is in agree-ent with the previously reported failure of NADP+ to evokeCa2+ and a chemotactic response in human granulocytes

36].In conclusion, our data strengthen the view that NAADP+

an be considered, alongside a traditional second messenger,n autocrine/paracrine extracellular signal, acting via a hith-rto unidentified mechanism engaging P2Y11 receptors andargeting both the IP3/[Ca2+]i and the cAMP/cADPR/[Ca2+]iignalling pathways.

cknowledgements

This work was supported in part by grants fromssociazione Italiana per la Ricerca sul Cancro (AIRC),

rom the Italian Ministry of Education, University andcientific Research (MIUR-PRIN 2003, MIUR FIRBBAUO19A3C, MIUR FIRB RBNE01ERXR and MIURIRB RBLA039LSF), the University of Genova and Fon-azione Cassa di Risparmio di Genova e Imperia.

eferences

[1] H.C. Lee, R. Aarhus, A derivative of NADP mobilizes calcium storesinsensitive to inositol trisphosphate and cyclic ADP-ribose, J. Biol.Chem. 270 (1995) 2152–2157.

[2] E.N. Chini, K.W. Beers, T.P. Dousa, Nicotinate adenine dinucleotidephosphate (NAADP) triggers a specific calcium release system in seaurchin eggs, J. Biol. Chem. 270 (1995) 3216–3223.

[3] M. Albrieux, H.C. Lee, M. Villaz, Calcium signaling by cyclic ADP-ribose, NAADP, and inositol trisphosphate are involved in distinctfunctions in ascidian oocytes, J. Biol. Chem. 273 (1998) 14566–

14574.

[4] L. Santella, K. Kyozuka, A.A. Genazzani, L. De Riso, E. Carafoli,Nicotinic acid adenine dinucleotide phosphate-induced Ca(2+) release.Interactions among distinct Ca(2+) mobilizing mechanisms in starfishoocytes, J. Biol. Chem. 275 (2000) 8301–8306.

3 l Calciu

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

54 I. Moreschi et al. / Cel

[5] A.H. Guse, Cyclic ADP-ribose (cADPR) and nicotinic acid adeninedinucleotide phosphate (NAADP): novel regulators of Ca2+-signalingand cell function, Curr. Mol. Med. 2 (2002) 3216–3223.

[6] H.C. Lee, Nicotinic acid adenine dinucleotide phosphate (NAADP)-mediated calcium signaling, J. Biol. Chem. 280 (2005) 33693–33696.

[7] M. Yamasaki, G.C. Churchill, A. Galione, Calcium signalling by nico-tinic acid adenine dinucleotide phosphate (NAADP), FEBS J. 272(2005) 4598–4606.

[8] J.V. Gerasimenko, S.E. Flowerdew, S.G. Voronina, T.K. Sukhomlin,A.V. Tepikin, O.H. Petersen, O.V. Gerasimenko, Bile acids induce Ca2+

release from both the endoplasmic reticulum and acidic intracellularcalcium stores through activation of inositol trisphosphate receptorsand ryanodine receptors, J. Biol. Chem. 281 (2006) 40154–40163.

[9] A. Galione, O.H. Petersen, The NAADP receptor: new receptors or newregulation? Mol. Interv. 5 (2005) 73–79.

10] G.C. Churchill, Y. Okada, J.M. Thomas, A.A. Genazzani, S. Patel, A.Galione, NAADP mobilizes Ca(2+) from reserve granules, lysosome-related organelles, in sea urchin eggs, Cell 111 (2002) 703–708.

11] M. Hohenegger, I. Berg, L. Weigl, G.W. Mayr, B.V. Potter, A.H.Guse, Pharmacological activation of the ryanodine receptor in JurkatT-lymphocytes, Br. J. Pharmacol. 128 (1999) 1235–1240.

12] M. Hohenegger, J. Suko, R. Gscheidlinger, H. Drobny, A. Zidar, Nico-tinic acid-adenine dinucleotide phosphate activates the skeletal muscleryanodine receptor, Biochem. J. 367 (2002) 423–431.

13] J.V. Gerasimenko, Y. Maruyama, K. Yano, et al., NAADP mobilizesCa2+ from a thapsigargin-sensitive store in the nuclear envelope byactivating ryanodine receptors, J. Cell. Biol. 163 (2003) 271–282.

14] W. Dammermann, A.H. Guse, Functional ryanodine receptor expres-sion is required for NAADP-mediated local Ca2+ signaling inT-lymphocytes, J. Biol. Chem. 280 (2005) 21394–21399.

15] D. Churamani, E.A. Carrey, G.D. Dickinson, S. Patel, Determinationof cellular nicotinic acid-adenine dinucleotide phosphate (NAADP)levels, Biochem. J. 380 (2004) 449–454.

16] M. Yamasaki, J.M. Thomas, C.G. Churchill, et al., Role of NAADPand cADPR in the induction and maintenance of agonist-evoked Ca2+

spiking in mouse pancreatic acinar cells, Curr. Biol. 15 (2005) 874–878.17] A. Gasser, S. Bruhn, A.H. Guse, Second messenger function of nico-

tinic acid adenine dinucleotide phosphate revealed by an improvedenzymatic cycling assay, J. Biol. Chem. 281 (2006) 16906–16913.

18] S. Soares, M. Thompson, T. White, et al., NAADP as a second messen-ger: Neither CD38 nor the base-exchange reaction are necessary for thein vivo generation of the NAADP in myometrial cells, Am. J. Physiol.Cell Physiol. 292 (2006) C227–C239.

19] R. Aarhus, R.M. Graeff, D.M. Dickey, T.F. Walseth, H.C. Lee,ADP-ribosyl cyclase and CD38 catalyze the synthesis of a calcium-mobilizing metabolite from NADP, J. Biol. Chem. 270 (1995)30327–30333.

20] H.C. Lee, Cyclic ADP-ribose and NAADP: Structures, Metabolism andFunctions, Kluwer Acad. Publ., Norwell, MS, 2002.

21] G. Basile, O. Taglialatela-Scafati, G. Damonte, et al., ADP-ribosylcyclases generate two unusual adenine homodinucleotides with cyto-toxic activity on mammalian cells, Proc. Natl. Acad. Sci. U.S.A. 102(2005) 14509–14514.

22] H.C. Lee, T.F. Walseth, G.T. Bratt, R.N. Hayes, D.L. Clapper, Structuraldetermination of a cyclic metabolite of NAD+ with intracellular Ca2+-mobilizing activity, J. Biol. Chem. 264 (1989) 1608–1615.

23] L. Sun, O.A. Adebanjio, A. Koval, et al., A novel mechanism forcoupling cellular intermediary metabolism to cytosolic Ca2+ signalingvia CD38/ADP-ribosyl cyclase, a putative intracellular NAD+ sensor,FASEB J. 16 (2002) 302–314.

24] A. De Flora, E. Zocchi, L. Guida, L. Franco, S. Bruzzone, Autocrine

and paracrine calcium signaling by the CD38/NAD+/cyclic ADP-ribosesystem, Ann. N. Y. Acad. Sci. 1028 (2004) 176–191.

25] R.A. Billington, E.A. Bellomo, E.M. Floriddia, J. Erriquez, C. Distasi,A.A. Genazzani, A transport mechanism for NAADP in a rat basophiliccell line, FASEB J. 20 (2006) 521–523.

[

m 43 (2008) 344–355

26] A.C. Heidemann, C.G. Schipke, H. Kettenmann, Extracellular appli-cation of nicotinic acid adenine dinucleotide phosphate induces Ca2+

signaling in astrocytes in situ, J. Biol. Chem. 280 (2005) 35630–35640.27] K. Singaravelu, J.W. Deitmer, Calcium mobilization by nicotinic acid

adenine dinucleotide phosphate (NAADP) in rat astrocytes, Cell Cal-cium 39 (2006) 143–153.

28] P. Vigne, P. Pacaud, V. Urbach, E. Feolde, J.P. Breittmayer, C. Frelin,The effect of PPADS as an antagonist of inositol (1,4,5)trisphos-phate induced intracellular calcium mobilization, Br. J. Pharmacol. 119(1996) 360–364.

29] P. Vigne, P. Pacaud, G. Loirand, J.P. Breittmayer, C. Frelin, PPADSinhibits P2Y1 purinoceptors in rat brain capillary endothelial cells andin rat ileal myocytes by an indirect mechanism, Biochem. Biophys.Res. Commun. 244 (1998) 332–335.

30] D. Shehnaz, B. Torres, M.A. Balboa, P.A. Insel, Pyridoxal-phosphate-6-azophenyl-2′,4′-disulfonate (PPADS), a putative P2Y(1) receptorantagonist, blocks signaling at a site distal to the receptor in Madin-Darby canine kidney-D(1) cells, J. Pharmacol. Exp. Ther. 292 (2000)346–350.

31] R.A. Billington, A.A. Genazzani, PPADS is a reversible competitiveantagonist of the NAADP receptor, Cell Calcium 41 (2006) 505–511.

32] I. Moreschi, S. Bruzzone, R.A. Nicholas, et al., Extracellular NAD+ isan agonist of the human P2Y11 purinergic receptor in human granulo-cytes, J. Biol. Chem. 281 (2006) 31419–31429.

33] D. Communi, C. Govaerts, M. Parmentier, J.M. Boeynaems, Cloningof a human purinergic P2Y receptor coupled to phospholipase C andadenylyl cyclase, J. Biol. Chem. 272 (1997) 31969–31973.

34] H. Ullmann, S. Meis, D. Hongwiset, et al., Synthesis and structure-activity relationships of suramin-derived P2Y11 receptor antagonistswith nanomolar potency, J. Med. Chem. 48 (2005) 7040–7048.

35] A.-D. Qi, C. Kennedy, T. Harden, R.A. Nicholas, Differential cou-pling of the human P2Y(11) receptor to phospholipase C and adenylylcyclase, Br. J. Pharmacol. 132 (2001) 318–326.

36] S. Bruzzone, I. Moreschi, L. Guida, C. Usai, E. Zocchi, A. De Flora,Extracellular NAD+ regulates intracellular calcium levels and inducesactivation of human granulocytes, Biochem. J. 393 (2006) 697–704.

37] E. Zocchi, A. Daga, C. Usai, et al., Expression of CD38 increasesintracellular calcium concentration and reduces doubling time in HeLaand 3T3 cells, J. Biol. Chem. 273 (1998) 8017–8024.

38] L. Guida, L. Franco, E. Zocchi, A. De Flora, Structural role of disul-fide bridges in the cyclic ADP-ribose related bifunctional ectoenzymeCD38, FEBS Lett. 368 (1995) 481–484.

39] R. Graeff, H.C. Lee, A novel cycling assay for cellular cADP-ribosewith nanomolar sensitivity, Biochem. J. 367 (2002) 163–168.

40] M. Bradford, A rapid and sensitive method for the quantitation of micro-gram quantities of protein utilizing the principle of protein-dye binding,Anal. Biochem. 72 (1976) 248–252.

41] T.F. Walseth, H.C. Lee, Synthesis and characterization of antagonistsof cyclic-ADP-ribose-induced Ca2+ release, Biochim. Biophys. Acta1178 (1993) 235–242.

42] G. Inesi, Y. Sagara, Thapsigargin, a high affinity and global inhibitorof intracellular Ca2+ transport ATPases, Arch. Biochem. Biophys. 298(1992) 313–317.

43] E.J. Bowman, A. Siebers, K. Altendorf, Bafilomycins: a class ofinhibitors of membrane ATPases from microorganisms, animal cells,and plant cells, Proc. Natl. Acad. Sci. U.S.A. 85 (1988) 7972–7976.

44] M. Steen, T. Kirchberger, A.H. Guse, NAADP mobilizes calcium fromthe endoplasmic reticular Ca2+ store in T-lymphocytes, J. Biol. Chem.282 (2007) 18864–18871.

45] N.P. Kinnear, F.X. Boittin, J.M. Thomas, A. Galione, A.M. Evans,Lysosome-sarcoplasmic reticulum junctions. A trigger zone for cal-

cium signaling by nicotinic acid adenine dinucleotide phosphate andendothelin-1, J. Biol. Chem. 279 (2004) 54319–54326.

46] F. Di Virgilio, P. Chiozzi, D. Ferrari, et al., Nucleotide receptors: anemerging family of regulatory molecules in blood cells, Blood 97(2001) 587–600.

l Calciu

[

[

I. Moreschi et al. / Cel

47] T.M. Filtz, Q. Li, J.L. Boyer, R.A. Nicholas, T.K. Harden, Expression

of a cloned P2Y purinergic receptor that couples to phospholipase C,Mol. Pharmacol. 46 (1994) 8–14.

48] S. Partida-Sanchez, P. Iribarren, M.E. Moreno-Garcia, J.L. Gao, P.M.Murphy, N. Oppenheimer, J.M. Wang, F.E. Lund, Chemotaxis and cal-cium responses of phagocytes to formyl peptide receptor ligands is

[

m 43 (2008) 344–355 355

differentially regulated by cyclic ADP ribose, J. Immunol. 172 (2004)

1896–1906.

49] S.Y. Rah, K.H. Park, M.K. Han, M.J. Im, U.H. Kim, Activation ofCD38 by interleukin-8 signaling regulates intracellular Ca2+ level andmotility of lymphokine-activated killer cells, J. Biol. Chem. 280 (2005)2888–2895.