Nicotinic Acid Adenine Dinucleotide Phosphate (NAADP) Activates Global and Heterogeneous Local Ca2+ Signals from NAADP- and Ryanodine Receptor-gated Ca2+ Stores in Pulmonary Arterial Myocytes*

Background: NAADP activates Ca2+ release from endolysosomal organelles. Results: NAADP activates two-pore channels in pulmonary arterial smooth muscle cells to elicit global and heterogeneous subcellular Ca2+ signals from NAADP- and ryanodine-sensitive Ca2+ stores, which contribute to the agonist-induced response. Conclusion: NAADP mediates complex Ca2+ interactions between endolysosomes and the sarcoplasmic reticulum to regulate vascular reactivity and other cellular functions. Significance: The results improve our understanding of NAADP-dependent regulation of pulmonary vascular functions. Nicotinic acid adenine dinucleotide phosphate (NAADP) is the most potent Ca2+-mobilizing messenger that releases Ca2+ from endolysosomal organelles. Recent studies showed that NAADP-induced Ca2+ release is mediated by the two-pore channels (TPCs) TPC1 and TPC2. However, the expression of TPCs and the NAADP-induced local Ca2+ signals have not been examined in vascular smooth muscle. Here, we found that both TPC1 and TPC2 are expressed in rat pulmonary arterial smooth muscle cells (PASMCs), with TPC1 being the major subtype. Application of membrane-permeant NAADP acetoxymethyl ester to PASMCs elicited a biphasic increase in global [Ca2+]i, which was independent of extracellular Ca2+ and blocked by the NAADP antagonist Ned-19 or the vacuolar H+-ATPase inhibitor bafilomycin A1, indicating Ca2+ release from acidic endolysosomal Ca2+ stores. The Ca2+ response was unaffected by xestospongin C but was partially blocked by ryanodine or thapsigargin. NAADP triggered heterogeneous local Ca2+ signals, including a diffuse increase in cytosolic [Ca2+], Ca2+ sparks, Ca2+ bursts, and regenerative Ca2+ release. The diffuse Ca2+ increase and Ca2+ bursts were ryanodine-insensitive, presumably arising from different endolysosomal sources. Ca2+ sparks and regenerative Ca2+ release were inhibited by ryanodine, consistent with cross-activation of loosely coupled ryanodine receptors. Moreover, Ca2+ release stimulated by endothelin-1 was inhibited by Ned-19, ryanodine, or xestospongin C, suggesting that NAADP-mediated Ca2+ signals interact with both ryanodine and inositol 1,4,5-trisphosphate receptors during agonist stimulation. Our results show that NAADP mediates complex global and local Ca2+ signals. Depending on the physiological stimuli, these diverse Ca2+ signals may serve to regulate different cellular functions in PASMCs.

Nicotinic acid adenine dinucleotide phosphate (NAADP) is the most potent Ca 2؉ -mobilizing messenger that releases Ca 2؉ from endolysosomal organelles. Recent studies showed that NAADP-induced Ca 2؉ release is mediated by the two-pore channels (TPCs) TPC1 and TPC2. However, the expression of TPCs and the NAADP-induced local Ca 2؉ signals have not been examined in vascular smooth muscle. Here, we found that both TPC1 and TPC2 are expressed in rat pulmonary arterial smooth muscle cells (PASMCs), with TPC1 being the major subtype. Application of membrane-permeant NAADP acetoxymethyl ester to PASMCs elicited a biphasic increase in global [Ca 2؉ ] i , which was independent of extracellular Ca 2؉ and blocked by the NAADP antagonist Ned-19 or the vacuolar H ؉ -ATPase inhibitor bafilomycin A 1 , indicating Ca 2؉ release from acidic endolysosomal Ca 2؉ stores. The Ca 2؉ response was unaffected by xestospongin C but was partially blocked by ryanodine or thapsigargin. NAADP triggered heterogeneous local Ca 2؉ signals, including a diffuse increase in cytosolic [Ca 2؉ ], Ca 2؉ sparks, Ca 2؉ bursts, and regenerative Ca 2؉ release. The diffuse Ca 2؉ increase and Ca 2؉ bursts were ryanodine-insensitive, presumably arising from different endolysosomal sources. Ca 2؉ sparks and regenerative Ca 2؉ release were inhibited by ryanodine, consistent with cross-activation of loosely coupled ryanodine receptors. Moreover, Ca 2؉ release stimulated by endothelin-1 was inhibited by Ned-19, ryanodine, or xestospongin C, suggesting that NAADP-mediated Ca 2؉ signals interact with both ryanodine and inositol 1,4,5-trisphosphate receptors during agonist stimulation. Our results show that NAADP mediates complex global and local Ca 2؉ signals. Depending on the physiological stimuli, these diverse Ca 2؉ signals may serve to regulate different cellular functions in PASMCs.
Ca 2ϩ ion serves as a ubiquitous signal for numerous cellular functions ranging from muscle contraction to gene expression. Depending on the specific agonists and physiological stimuli, global and local Ca 2ϩ signals with unique spatiotemporal properties are generated by a multitude of extracellular Ca 2ϩ influx and intracellular Ca 2ϩ release pathways to precisely regulate the specific effectors in various subcellular compartments (1). There are three Ca 2ϩ -mobilizing messengers, namely inositol 1,4,5-trisphosphate (InsP 3 ), 3 cyclic ADP-ribose (cADPR), and nicotinic acid adenine dinucleotide phosphate (NAADP). NAADP is the most powerful Ca 2ϩ -mobilizing messenger of these three endogenous messengers and is capable of activating Ca 2ϩ release at low nanomolar concentrations (2), but its action mechanism is the least understood.
Increasing evidence suggests that NAADP plays important roles in vascular smooth muscle cell (VSMC) function, and NAADP-mediated Ca 2ϩ release is linked to agonist-induced vasoconstriction. For example, application of NAADP to microsomes of aortic smooth muscle cells elicited Ca 2ϩ release independent of InsP 3 and cADPR (15,16). Endothelin-1 (ET-1) caused an increase in NAADP production and activated the Ca 2ϩ response in coronary arterial myocytes (17). ET-1 and norepinephrine triggered the Ca 2ϩ response and vasoconstriction in renal afferent arterioles, and these responses were attenuated by the vacuolar H ϩ -ATPase inhibitors concanamycin A and bafilomycin A 1 and by the NAADP antagonist Ned-19 (18). In addition, a recent study showed that Fas ligand, an inducer of apoptosis, elicits NAADP-mediated lysosomal Ca 2ϩ release in mouse coronary arterial myocytes, suggesting that NAADP may involve in the inflammatory/apoptotic response in VSMCs (19).
In pulmonary arterial smooth muscle cells (PASMCs), intracellular dialysis of NAADP triggered "bursts" of spatially restricted Ca 2ϩ release and global Ca 2ϩ waves, which were blocked by depleting lysosomal Ca 2ϩ with bafilomycin A 1 or by inhibition of RyRs (20,21). It has been suggested that lysosomes and the RyR-gated SR are coupled to form specialized "trigger zones," at which NAADP-dependent Ca 2ϩ signals are amplified by RyRs through Ca 2ϩ -induced Ca 2ϩ release (21,22). We have previously found that integrin-specific ligands mobilize Ca 2ϩ in part through Ca 2ϩ release from the acidic lysosomal Ca 2ϩ stores in PASMCs (23), and the expression of integrins and their associated Ca 2ϩ responses are altered during the development of pulmonary hypertension (24). These studies suggest that NAADP-dependent Ca 2ϩ signals may be critically involved in the regulation of pulmonary circulation. However, the expression of NAADP channels and the properties of NAADP-dependent local Ca 2ϩ signals have not been examined in VSMCs.
In this study, we examined systematically the NAADP-dependent Ca 2ϩ signaling pathway in PASMCs by characterizing the expression of TPCs, identifying the associated Ca 2ϩ sources, quantifying the spatiotemporal properties of local Ca 2ϩ events activated by NAADP, and determining the contribution of NAADP in an agonist-induced Ca 2ϩ response. These experiments provide essential information for our understanding of NAADP-dependent Ca 2ϩ signaling in the pulmonary vasculature.

EXPERIMENTAL PROCEDURES
Isolation of Intralobar Pulmonary Arteries and Aortas-All animal procedures in this study were performed in accordance with the guidelines approved by The Johns Hopkins Animal Care and Use Committee. Pulmonary arteries (PAs) and aortas were isolated from male Wistar rats (150 -250 g) anesthetized with sodium pentobarbital (130 mg/kg intraperitoneally). Lungs and thoracic aortas were removed after exsanguination and transferred to a Petri dish filled with HEPES-buffered salt solution (HBSS) containing 130 mM NaCl, 5 mM KCl, 1.2 mM MgCl 2 , 1.5 mM CaCl 2 , 10 mM HEPES, and 10 mM glucose (pH 7.4 adjusted with NaOH). Intralobar large PAs (lPAs; ϳ300 -800 m), small PAs (sPAs; Ͻ300 m), and descending thoracic aortas were isolated and cleaned free of connective tissue. The endothelium was removed by gently rubbing the luminal surface with a cotton swab.
Isolation and Transient Culture of PASMCs-PASMCs were enzymatically isolated and transiently cultured as described previously (25). In brief, endothelium-denuded PAs were allowed to recover for 30 min in cold (4°C) HBSS, followed by 20 min in reduced Ca 2ϩ (20 M) HBSS at room temperature. The tissue was digested at 37°C for 20 min in 20 M Ca 2ϩ / HBSS containing collagenase (type I, 1750 units/ml), papain (9.5 units/ml), BSA (2 mg/ml), and dithiothreitol (1 mM). It was then washed with Ca 2ϩ -free HBSS to stop digestion, and PASMCs were dispersed gently by trituration with a small-bore pipette in Ca 2ϩ -free HBSS at room temperature. The dispersed PASMCs were placed on 25-mm glass coverslips and cultured transiently (16 -24 h) in Ham's F-12 medium (with L-glutamine) supplemented with 0.5% FCS, 100 units/ml streptomycin, and 0.1 mg/ml penicillin under 21% O 2 and 5% CO 2 before use.
RNA Preparation and RT-PCR-Endothelium-denuded intralobar PAs, sPAs, and aortas were frozen in liquid nitrogen and then mechanically pulverized and homogenized with a mortar and pestle kept on dry ice. Total RNA was extracted using the RNeasy mini kit (Qiagen) following standard procedures. Genomic DNA contamination was removed by TURBO DNA-free TM DNase (Ambion, Austin, TX). 1 g of total RNA were used for first-strand cDNA synthesis with random hexamer primers and Superscript III reverse transcriptase (Invitrogen) according to the manufacturer's protocol. The resulting first-strand cDNAs were directly used as templates for PCR amplification. Sense and antisense primers specific for TPC1 and TPC2 (listed in Table 1) were used. Reactions were carried out using PCR SuperMix (Invitrogen) with the following parameters: denaturation at 94°C for 30 s, annealing at 60°C for 45 s, and extension at 72°C for 90 s. A total of 35 cycles were performed. This was followed by a final extension at 72°C for 10 min and then storage at 4°C. PCR products were analyzed by 1.5% agarose gel electrophoresis and visualized by staining with ethidium bromide, and the sequences of PCR products were determined for verification. Parallel reactions were run for each RNA sample in the absence of Superscript III to access the degree of genomic DNA contamination.
Quantitative Real-time RT-PCR-Gene-specific real-time PCR primers for TPC1 and TPC2 were designed (Table 1).
PCRs were set up with iQ TM SYBR Green PCR Supermix (Bio-Rad) using 1 l of cDNA as the template in each 20-l reaction mixture. The PCR protocol consisted of an initial step at 95°C for 5 min, followed by 40 cycles at 95°C for 15 s, 60°C for 30 s, and 72°C for 1 min and was performed using an iQ5 multicolor real-time PCR detection system (Bio-Rad). Using the same protocol, we generated standard curves from serial dilutions of purified PCR products with known copy numbers measured by absorbance at 260 nm. The absolute copy number of the mRNA of interest was determined by interpolation of the standard curve with the threshold cycle value of each sample. To confirm the specificity of the PCR products, a melting curve was obtained at the end of each run. Standard gel electrophoresis was also performed to ensure the end product generated a single band with the predicted size (100 -150 bases). Data were normalized with the quantity of 18 S rRNA in individual samples to correct for sample variability.
Western Blotting-PAs frozen in liquid nitrogen were crushed and homogenized using a mortar and pestle and resuspended in ice-cold lysis buffer containing 50 mM Tris-Cl (pH 7.4), 150 mM NaCl, 1% deoxycholic acid, 0.1% SDS, 0.5% Nonidet P-40, and protease inhibitor mixture (Roche Applied Science). The homogenate was centrifuged at 1000 ϫ g for 5 min at 4°C, the supernatant was collected, and the protein concentration was estimated using the BCA assay. 20 g of protein sample was resolved on an 8% SDS-polyacrylamide gel and electrotransferred onto a PVDF membrane (Millipore). The membrane was blocked with 5% (w/v) nonfat dry milk in PBS containing 0.02% Tween 20 for 1 h at room temperature, followed by overnight incubation at 4°C with a specific primary antibody. The primary antibodies were polyclonal rabbit anti-TPC1 (1:500 dilution) from Abcam (Cambridge, MA) and anti-TPC2 (1:2500 dilution) from Alomone Labs (Jerusalem, Israel). The actin level was also determined and used as a loading control. The membrane was washed and incubated with peroxidase-conjugated goat anti-rabbit secondary antibody (Bio-Rad) at 1:2500 dilution at room temperature for 1 h. Protein bands were detected by enhanced chemiluminescence (Pierce) and imaged using a Gel Logic 200 image system (Kodak). Deglycosylation assays were performed on some samples to verify the double bands detected by the anti-TPC1 and anti-TPC2 antibodies. Protein samples (10 g) were incubated in the absence or presence of peptide:N-glycosidase F (New England Biolabs) according to the manufacturer's instruction. Protein denaturation was performed at 45°C for 10 min, and enzyme incubation was carried out at 37°C for 1 h. Protein samples were resolved and analyzed by immunoblotting as described above.

Measurement of Global [Ca 2ϩ ] i -[Ca 2ϩ
] i was monitored using the membrane-permeable Ca 2ϩ -sensitive fluorescent dye fluo 3-AM. PASMCs were loaded with 5-10 M fluo 3-AM (dissolved in Me 2 SO with 20% pluronic acid) for 45 min at room temperature (ϳ23°C) in normal Tyrode's solution containing 137 mM NaCl, 5.4 mM KCl, 2 mM CaCl 2 , 1 mM MgCl 2 , 10 mM HEPES, and 10 mM glucose (pH 7.4 adjusted with NaOH). Cells were then washed and rested for 15-30 min to allow complete de-esterification of the cytosolic dye. fluo 3 was excited at 488 nm, and emission light at Ͼ515 nm was detected from cells in a microscopic field using a Nikon Diaphot microscope (objective, Fluor 40ϫ/1.3 numerical aperture) equipped with a photomultiplier tube-based microfluorometer. Protocols were executed and data were collected on-line with a Digidata analog-to-digital interface and a pCLAMP software package (Axon Instruments, Inc., Foster City, CA). [Ca 2ϩ ] i was calibrated using the Measurement of Local Ca 2ϩ Events-Ca 2ϩ events were visualized using fluo 3-AM as described previously (25). Confocal images were acquired using a Zeiss LSM 510 inverted confocal microscope with a Zeiss Plan-Neofluar 40ϫ/1.3 oil immersion objective. The confocal pinhole was set to render a spatial resolution of 0.4 m in the x-y axis and 1.7 m in the z axis. fluo 3-AM was excited by the 488 nm light of an argon laser, and fluorescence was measured at Ͼ505 nm. Images were acquired in the line scan mode (digital zoom rendering a 38-m scan line), scanning at 0.075 m/pixel and 512 pixels/line at 2-ms intervals. Photobleaching and laser damage to the cells were minimized by attenuating the laser to ϳ1% of its maximum power (25 milliwatts) with an acousto-optical tunable filter, and each cell was imaged for Ͻ20 s. Images were processed and Ca 2ϩ sparks were analyzed by custom-written algorithms using the IDL software package (27) or the SparkMaster plug-in of ImageJ software (28).
Statistical Analysis-Data are expressed as means Ϯ S.E. Statistical significance (p Ͻ 0.05) of the changes was assessed by paired or unpaired Student's t tests, non-parametric Mann-Whitney U tests, or one-or two-way analysis of variance with Tukey's range test for post hoc analysis, wherever applicable.

RESULTS
Expression of TPC1 and TPC2 mRNAs and Proteins-To study the NAADP-dependent Ca 2ϩ response, the expression of the NAADP channels TPC1 and TPC2 was first characterized  APRIL 12, 2013 • VOLUME 288 • NUMBER 15

NAADP-induced Ca 2؉ Signaling in PASMCs
in sPAs, lPAs, and aortas using conventional RT-PCR. Fig. 1 (A and B) shows the amplified PCR products generated after 35 cycles from endothelium-denuded sPAs, lPAs, and aortas. TPC1 and TPC2 transcripts were detected in all three types of vascular tissues. The PCR-amplified products had sizes corresponding to the predicted values (309 bp for TPC1 and 262 bp for TPC2) and matched the predicted sequences. The relative expression of TPC1 and TPC2 was quantified by real-time RT-PCR. The TPC1 mRNA level was ϳ4 -5-fold higher than the TPC2 mRNA level in all three vascular tissues, with the values of individual samples normalized with 18 S rRNA. In addition, TPC1 and TPC2 mRNA expression in lPAs was the highest of the three vascular tissues, with the order lPAs Ͼ sPAs Ͼ aortas. TPC1 and TPC2 proteins in sPAs, lPAs, and aortas were detected by immunoblotting (Fig. 1, C and D). Specific anti-TPC1 antibodies detected two clear bands at ϳ100 and ϳ75 kDa; double bands were also detected at ϳ75 and ϳ60 kDa using anti-TPC2 antibody. The double bands were related to N-glycosylation of TPC1 and TPC2 proteins as previously described (6). Pretreatment of samples with peptide:N-glycosidase F to remove the N-glycan chains converted the blots to a single band of ϳ75 kDa for TPC1 and ϳ60 kDa for TPC2. The disappearance of the higher molecular mass bands after peptide:N-glycosidase F treatment indicated that they were the mature glycosylated proteins, whereas the lower bands were the core proteins. TPC1 and TPC2 protein levels were similar in sPAs, lPAs, and aortas (TPC1, n ϭ 7; and TPC2, n ϭ 7), with ␣-actin used as the internal standard for normalization. These results clearly show that the two types of NAADP-sensitive Ca 2ϩ channels are expressed in pulmonary arterial smooth muscle.
NAADP-induced Mobilization of Global Ca 2ϩ in PASMCs-The presence of functional NAADP-sensitive Ca 2ϩ channels in PASMCs was examined using the cell-permeant NAADP analog NAADP-AM (ISIS Innovation Ltd., Oxford, United Kingdom). Application of NAADP-AM activated a concentrationdependent increase in [Ca 2ϩ ] i (Fig. 2, A and B). NAADP-AM at 0.25 and 0.5 M elicited sustained increases in [Ca 2ϩ ] i , whereas 1 M activated a biphasic response with an initial transient rise, followed by a sustained increase in [Ca 2ϩ ] i . The 1 M NAADP-AM-induced response was unaffected by exchanging Ca 2ϩ -free solution (with 1 mM EGTA) 1 min prior to NAADP application (Fig. 2, C and D). The peak and sustained Ca 2ϩ responses were 216 Ϯ 13 and 91 Ϯ 6 nM (n ϭ 5), respectively, in the presence of Ca 2ϩ and 185 Ϯ 86 and 86 Ϯ 16 nM (n ϭ 5), respectively, in the absence of extracellular Ca 2ϩ . These results indicate that the NAADP-induced Ca 2ϩ response is solely dependent on Ca 2ϩ mobilization from intracellular Ca 2ϩ stores.
There is substantial evidence suggesting that NAADP-sensitive channels are expressed mainly in the acidic endolysosomal organelles (29,30). To examine the importance of endolysosomal Ca 2ϩ stores in the NAADP-activated Ca 2ϩ response, acidic Ca 2ϩ stores were depleted by inhibiting the vacuolar H ϩ -ATPase to disrupt the lysosomal H ϩ gradient for Ca 2ϩ entry via Ca 2ϩ /H ϩ exchange. Preincubation of PASMCs for 1 h with bafilomycin A 1 (3 M), a specific vacuolar H ϩ -ATPase inhibitor (31), significantly inhibited the peak and completely abolished the sustained phase of the Ca 2ϩ response activated by NAADP-AM (1 M) (Fig. 3A). The peak Ca 2ϩ response was 164 Ϯ 15 nM (n ϭ 5) in the control PASMCs and 50 Ϯ 11 nM (n ϭ 5; p Ͻ 0.05) in the bafilomycin A 1 -pretreated PASMCs (Fig. 3B). The specificity of the NAADP-AM-induced Ca 2ϩ responses was further verified using the selective NAADP receptor antagonist Ned-19 (Enzo Life Sciences, Ann Arbor, MI) (Fig. 3, C and D) (32). Pretreatment of PASMCs with Ned-19 (1 M) for 20 min eliminated the initial transient peak and significantly reduced the sustained phase of the NAADP-AM-activated Ca 2ϩ response (control, 97 Ϯ 12 nM (n ϭ 6), and Ned-19, 52 Ϯ 4 nM (n ϭ 6); p Ͻ 0.01). The NAADP-AM-activated Ca 2ϩ response was completely abolished by further increasing the concentration of Ned-19 to 100 M. The significant inhibition of the Ca 2ϩ response by bafilomycin A 1 and Ned-19 indicates that NAADP-AM mobilizes Ca 2ϩ mainly through the activation of specific NAADP receptors of the acidic endolysosomal organelles in PASMCs.
Previous studies in other cell types suggest that NAADPinduced Ca 2ϩ release is amplified by cross-activation of InsP 3 Rs and RyRs (20,33). To examine the possible interactions between NAADP-induced Ca 2ϩ signals and the InsP 3 R-and RyR-gated Ca 2ϩ stores, InsP 3 R-and RyR-dependent Ca 2ϩ release were either blocked separately using xestospongin C and ryanodine, respectively, or disabled simultaneously using the SERCA (sarco/endoplasmic reticulum Ca 2ϩ -ATPase) inhibitor thapsigargin. A 15-min pretreatment of PASMCs with 10 M xestospongin C had no significant effect on the peak and sustained Ca 2ϩ responses activated by NAADP-AM (Fig.  4A), suggesting that InsP 3 R does not contribute to NAADP-dependent Ca 2ϩ release in PASMCs. In contrast, inhibition of RyR with 50 M ryanodine caused a significant reduction in the initial transient Ca 2ϩ release (control, 244 Ϯ 31 nM (n ϭ 6), and ryanodine, 151 Ϯ 15 nM (n ϭ 7); p Ͻ 0.05) but did not affect the sustained phase of the Ca 2ϩ response (Fig. 4B). Similar to RyR inhibition, depletion of the SR Ca 2ϩ store with thapsigargin (10  M) also attenuated the peak Ca 2ϩ response (control, 278 Ϯ 33 nM (n ϭ 5), and ryanodine, 116 Ϯ 17 nM (n ϭ 6); p Ͻ 0.05), whereas the sustained Ca 2ϩ increase elicited by NAADP-AM was unaltered (Fig. 4C). These results suggest that the initial transient increase in [Ca 2ϩ ] i was mediated by the cross-activation of RyRs on the SR and that the sustained Ca 2ϩ response came from the NAADP-sensitive Ca 2ϩ stores independent of SR Ca 2ϩ release.
NAADP-induced Local Ca 2ϩ Events in PASMCs-Local Ca 2ϩ signals were further examined at the subcellular level using confocal Ca 2ϩ fluorescence microscopy in the line scan mode. Application of NAADP-AM to quiescent PASMCs activated robust local and global Ca 2ϩ events. The Ca 2ϩ response was heterogeneous, usually led by a diffuse increase in basal [Ca 2ϩ ] i , followed by an upsurge of Ca 2ϩ sparks, which fused to generate a global increase in [Ca 2ϩ ] i , where Ca 2ϩ sparks were no longer discernible (Fig. 5A, upper and middle panels). Repetitive local Ca 2ϩ events were observed in some subcellular sites, and large repetitive non-inactivating Ca 2ϩ bursts were also occasionally observed (Fig. 5A, lower panel). These Ca 2ϩ bursts had a higher amplitude, a larger spatial spread, and a much longer duration compared with Ca 2ϩ sparks (Fig. 6, A and B).    (Fig. 7), indicating that the RyR-gated Ca 2ϩ stores contributed significantly to both the local and global Ca 2ϩ signals. However, the diffuse increase in basal [Ca 2ϩ ] i and Ca 2ϩ bursts persisted in ryanodine-treated PASMCs (Fig. 7A, middle and lower pan-els), suggesting that they were Ca 2ϩ signals originating from the TPCs.
The spatiotemporal characteristics of local Ca 2ϩ events activated by NAADP-AM were further examined under steadystate conditions in a separate set of experiments in which Ca 2ϩ sparks were recorded in either the absence or continuous presence of 1 M NAADP-AM. The spark frequency was significantly higher in PASMCs continuously exposed to NAADP-AM compared with the control cells (control, 0.56 Ϯ 0.08 sparks/100 m/s (n ϭ 71 cells), and NAADP, 2.18 Ϯ 0.24 sparks/100 m/s (n ϭ 58 cells); p Ͻ 0.001). However, the spark amplitude (⌬F/F 0 ; control, 0.58 Ϯ 0.01 (n ϭ 436), and NAADP, 0.61 Ϯ 0.02 (n ϭ 438)), full duration at half-maximum (control, 59.9 Ϯ 4.3 ms (n ϭ 436), and NAADP, 46.0 Ϯ 2.68 ms (n ϭ 438)), and the spatial spread or full-width at half-maximum (control, 1.76 Ϯ 0.14 m (n ϭ 436), and NAADP, 1.5 Ϯ 0.04 m (n ϭ 438)) were not significantly different between the control and NAADP-AM-treated cells (Fig. 8). Hence, the spatiotemporal properties of local Ca 2ϩ events activated by NAADP were indistinguishable from spontaneous Ca 2ϩ sparks recorded under resting conditions. Our results therefore suggest that NAADP-dependent Ca 2ϩ signaling in PASMCs consists of heterogeneous Ca 2ϩ events, some of which are mediated by direct activation of NAADP receptors and others by cross-activation of RyRs.
NAADP-dependent Agonist-induced Ca 2ϩ Response in PASMCs-To further examine the contribution of NAADP to the agonist-induced response, we compared the effects of NAADP receptor, RyR, and InsP 3 R antagonists on the ET-1induced Ca 2ϩ response. ET-1 (10 nM) activated a biphasic Ca 2ϩ response in PASMCs (peak ⌬[Ca 2ϩ ] i ϭ 265 Ϯ 55 nM and sus-   inhibited the peak Ca 2ϩ response in a concentration-dependent manner without altering the sustained Ca 2ϩ response (Fig.  9, A and B), suggesting that the peak Ca 2ϩ response is mediated by a NAADP-dependent mechanism. Consistent with crossactivation of RyRs by Ca 2ϩ signals from NAADP-sensitive channels, the peak Ca 2ϩ response activated by ET-1 was significantly reduced by ryanodine (50 M), and the remaining peak Ca 2ϩ response was further inhibited by 1 M Ned-19 (Fig. 9, C  and D). It is well established in VSMCs that ET-1 binds to ET-A receptors, leading to activation of phospholipase C and production of InsP 3 to activate Ca 2ϩ release. Inhibition of InsP 3 R with 10 M xestospongin C almost completely abolished the peak Ca 2ϩ response of ET-1. The addition of Ned-19 did not further reduce the Ca 2ϩ signal (Fig. 9, E and F). These results suggest that both functional NAADP receptors and InsP 3 Rs are required for Ca 2ϩ release triggered by ET-1. In contrast, the sustained Ca 2ϩ response of ET-1 was mediated primarily by extracellular Ca 2ϩ influx, which was unaffected by Ned-19, ryanodine, or xestospongin C (Fig. 9, B, D, and F) but was completely abolished by removing extracellular Ca 2ϩ 100 s prior to the application of ET-1 (Fig. 10, A and B). Removal of extracellular Ca 2ϩ did not affect the ET-1-induced peak Ca 2ϩ response.

DISCUSSION
Recent studies have demonstrated that TPCs are the endolysosomal NAADP-sensitive channels, and they are widely expressed in major organs, including the brain, heart, kidney, liver, lung, intestine, spleen, thymus, ovary, and testis (6,8).
Here, we detected the mRNAs and proteins of TPC1 and TPC2 in endothelium-denuded rat sPAs, lPAs, and aortas. Preliminary screening also showed TPC1 and TPC2 expression in rat mesenteric, cerebral, and tail arteries (data not shown), suggesting that TPCs are expressed ubiquitously in VSMCs and may play essential roles in vascular functions. Quantitative RT-PCR data showed that the level of the TPC1 transcript is severalfold higher compared with TPC2 in PAs and aortas, indicating that TPC1 is the major endogenous NAADP channel in PASMCs. This is congruent with the observation that TPC1 is the predominant TPC subtype expressed in native human endothelial cells (35) and rat PC12 cells (7), as well as in the human cell lines SKBR3 and HEK293 (7,12). TPC1 also accounts for most of the NAADP-induced Ca 2ϩ response in SKBR3 and HEK293 cells (7,12). In this study, both N-glycosylated and non-glycosylated TPC1 and TPC2 were found in native PASMCs. This is similar to heterologously expressed TPCs in several cell lines (6,8,36,37). The different N-glycosylated forms of TPCs may reflect different stages of post-translational processing, but they could also be related to the regulation of TPC functions. It has been shown that the Nglycosylation sites of TPCs are located luminally, close to the pore-forming region in domain II (6,37). Removal of N-glycosylation residues of TPC1 had no effect on its subcellular localization but greatly enhanced the NAADP-induced Ca 2ϩ response (37). Because TPC1 is expressed in all stages of endolysosomes (including the recycling endosomes, early and late endosomes, and lysosomes) and TPC2 is confined more to the late endosomes and lysosomes (6 -10), the different glycosylated forms of TPCs in PASMCs may be related to specific types of endolysosomal organelles, which have different luminal environments, such as pH and [Ca 2ϩ ].
Cell-permeable NAADP-AM activates a robust biphasic global Ca 2ϩ response in PASMCs. Both the initial transient and sustained components of the NAADP-induced Ca 2ϩ response were independent of extracellular Ca 2ϩ but were inhibited by depleting Ca 2ϩ in acidic organelles with bafilomycin A 1 or using the specific NAADP antagonist Ned-19. This is consistent with endolysosomal Ca 2ϩ release via TPCs (11,17,18,21,32). The initial component of the Ca 2ϩ response is due to crossactivation of RyRs because it could be blocked by ryanodine and thapsigargin but not by xestospongin C. This supports the assertion that the NAADP-mediated Ca 2ϩ signal is amplified by Ca 2ϩ -induced Ca 2ϩ release in PASMCs (20 -22). However, the presence of a prominent sustained component of the thap- sigargin-insensitive but bafilomycin-sensitive Ca 2ϩ release clearly suggests that endolysosomes are capable of generating sizable Ca 2ϩ signals independent of SR Ca 2ϩ release. The NAADP-mediated sustained Ca 2ϩ release corresponds nicely with the sustained bafilomycin-sensitive Ca 2ϩ transient activated by the integrin ligand GRGDSP peptide reported previously in PASMCs (23).
The heterogeneous local Ca 2ϩ events activated by NAADP provide a close-up glimpse of the dynamic interactions of the NAADP-dependent Ca 2ϩ signaling pathways. At least four distinctive Ca 2ϩ events, namely the small diffuse rise in cytosolic [Ca 2ϩ ], the spatially discernible Ca 2ϩ sparks, the repetitive large localized Ca 2ϩ bursts, and the regenerative Ca 2ϩ releases, have been identified. The small diffuse increase in [Ca 2ϩ ] activated by NAADP is the initial Ca 2ϩ release directly from NAADP channels because it is insensitive to ryanodine and usually precedes the other Ca 2ϩ events. The diffuse nature of the Ca 2ϩ signal suggests that it arises from asynchronous TPCs and/or from Ca 2ϩ stores with a low Ca 2ϩ capacity. It has been shown that spatially discernible Ca 2ϩ events, such as Ca 2ϩ sparks, require concerted activation of multiple RyRs (38), whereas Ca 2ϩ signals from individual RyRs (Ca 2ϩ quarks) are submicroscopic (39,40). Because the single channel conductance of TPCs (ϳ15 picosiemens) (11) is significantly smaller than that of RyRs (ϳ120 picosiemens) (41) and the Ca 2ϩ content of endosomes and early lysosomes is lower than in the SR, Ca 2ϩ release via individual asynchronous TPCs from endosomes is unlikely to be resolved by confocal imaging. NAADP also activates discernible local Ca 2ϩ events, which are sensitive to ryanodine and have spatiotemporal properties indistinguishable from spontaneous Ca 2ϩ sparks (25,27,42,43). The gradual increase in spark frequency during the slow initial rise in cytosolic [Ca 2ϩ ] could be related to an increase in SR Ca 2ϩ loading as the result of continuous uptake of Ca 2ϩ released from the acidic stores. This scenario has been demonstrated in guinea pig ventricular and atrial myocytes (44,45).
Previous studies using conventional Ca 2ϩ fluorescence microscopy showed that intracellular dialysis of NAADP into PASMCs activated bursts of Ca 2ϩ release over a sizable region close to the perimeter of the cell (20 -22). These Ca 2ϩ bursts either stopped or eventually triggered global Ca 2ϩ waves that could be blocked by ryanodine or thapsigargin. It was later proposed that lysosomes localized around the nucleus are closely associated with the perinuclear SR, where NAADP-sensitive channels and RyR3 form a highly organized trigger zone for NAADP-mediated Ca 2ϩ signaling (21,22). In this study, we observed repetitive localized Ca 2ϩ bursts activated by NAADP. These Ca 2ϩ bursts were insensitive to ryanodine, thus unrelated to cross-activation of RyRs. However, the spatiotemporal profile of these Ca 2ϩ bursts is similar to the non-inactivation Ca 2ϩ events we reported previously in the perinuclear regions of PASMCs (42), where lysosomes are abundant (21)(22)(23). The robust Ca 2ϩ signal of these bursts suggests that a large number of NAADP channels are being activated simultaneously and that the Ca 2ϩ content of the store is high. Because the Ca 2ϩ content of mature lysosomes is the highest among endolysosomal organelles (46) and TPC2 is preferentially expressed in lysosomes (8), it is possible that Ca 2ϩ bursts are Ca 2ϩ signals coming primarily from TPC2 (and TPC1) of mature lysosomes. However, it is unclear how multiple TPCs are coordinated to generate repetitive Ca 2ϩ bursts. There is no information on Ca 2ϩ -induced activation of TPCs as in the case of RyRs, besides evidence for TPC regulation by voltage, luminal pH, and Ca 2ϩ (11,12). The diffuse increase in cytosolic [Ca 2ϩ ], Ca 2ϩ sparks, and Ca 2ϩ bursts activated by NAADP generally triggered regenerative global Ca 2ϩ release, which could be abolished by ryanodine. However, the occurrence of Ca 2ϩ bursts was not always associated with regenerative Ca 2ϩ release. In fact, solitary Ca 2ϩ bursts were frequently observed (Fig. 5A, middle and lower panels). This suggests that even though NAADP-induced Ca 2ϩ signals are amplified by RyRs, the coupling between lysosomes and SR Ca 2ϩ stores in PASMCs is loose, for example, compared with the coupling of L-type Ca 2ϩ channels and RyRs in cardiac myocytes (47,48) and the coupling of RyRs and Ca 2ϩactivated K ϩ channels in cerebral arteries (49,50).
Previous studies in systemic and pulmonary arteries suggested significant contributions of NAADP and lysosomal Ca 2ϩ stores to the agonist-induced Ca 2ϩ response (17,18,21,23). This notion is supported by our finding that Ned-19 inhibits dose-dependently the initial transient phase of the ET-1-induced Ca 2ϩ response. trans-Ned-19 is a highly specific NAADP antagonist. It inhibits NAADP-mediated Ca 2ϩ release with an IC 50 of 10 -70 nM (32,51) and completely antagonizes the single channel activity of TPC2 at 1 M (11), but it does not affect InsP 3 -or cADPR-induced Ca 2ϩ release at concentrations up to 100 M (32, 51). Ned-19 at 10 M also has no apparent effect on Ca 2ϩ influx through voltage-gated Ca 2ϩ channels activated by KCl, SOCE induced by thapsigargin, RyR-gated Ca 2ϩ release triggered by caffeine, or Ca 2ϩ release activated by Bt 3 -InsP 3 /AM in PASMCs. Complete inhibition of the transient Ca 2ϩ response by a low concentration of Ned-19 (1 M) therefore indicates that NAADP is a major mechanism for ET-1induced Ca 2ϩ release.
In contrast to the biphasic Ca 2ϩ response activated by NAADP-AM, endogenous NAADP generated by ET-1 stimulation contributes predominantly to the initial peak Ca 2ϩ response. The sustained phase of the ET-1-induced response is supported solely by extracellular Ca 2ϩ influx because it is insensitive to Ned-19, ryanodine, and xestospongin C but is completely abolished by the removal of extracellular Ca 2ϩ . The transient nature of NAADP-dependent Ca 2ϩ release could be related to the kinetics of ET-1-induced CD38 activation, production, and metabolism of NAADP, as well as desensitization of ET-1 receptors and inactivation of NAADP channels (52,53), such that endogenously produced NAADP is no longer available or effective after prolonged ET-1 exposure. It is interesting that the initial Ca 2ϩ release activated by ET-1 requires all three Ca 2ϩ stores. The interdependence of RyR-and NAADP-gated Ca 2ϩ stores is consistent with cross-activation of RyRs by Ca 2ϩ released from NAADP channels, as demonstrated by the earlier NAADP-AM experiments. However, the complete inhibition of the peak Ca 2ϩ response by xestospongin C suggests that InsP 3 R also plays a permissive role in ET-1-induced Ca 2ϩ release. Interactions between the three types of Ca 2ϩ stores are very complex. First, InsP 3 Rs are Ca 2ϩ -sensitive. They can be activated by Ca 2ϩ -induced Ca 2ϩ release, and this process is modulated by InsP 3 binding (54). It has been shown in other cells that NAADP-mediated Ca 2ϩ signals can be amplified by triggering further Ca 2ϩ release from InsP 3 Rs by Ca 2ϩ priming of the SR (8,10,33,55,56). We have demonstrated previously that Ca 2ϩ release from the InsP 3 R can cross-activate RyRs in PASMCs (27), in a manner similar to NAADP and RyRs observed in this study. Furthermore, recent evidence obtained with HEK cells shows that lysosomes are closely associated with the InsP 3 -gated SR, and they can selectively sequester released Ca 2ϩ from InsP 3 Rs (57). This process may facilitate lysosomal Ca 2ϩ loading for subsequent release. All of these mechanisms may be operating in PASMCs during agonist stimulation when InsP 3 Rs are sensitized by an increased level of InsP 3 and may allow InsP 3 Rs to play a much larger role in the integrated Ca 2ϩ release process compared with Ca 2ϩ release activated by NAADP-AM alone. However, it is unclear whether InsP 3 or NAADP is the primary trigger for the integrated Ca 2ϩ release event. The intricate interactions between these interdependent Ca 2ϩ stores during agonist stimulation require further investigations.
It is also interesting that exogenously applied NAADP-AM activates a sustained component of Ca 2ϩ release. The fact that the sustained response is not associated with RyR-mediated Ca 2ϩ release and is relatively insensitive to Ned-19 suggests that these NAADP-sensitive stores are not coupled to RyRs. They are perhaps gated by NAADP channels with properties different from those activated by ET-1. This group of NAADPsensitive stores could be endo/lysosomes gated by TPC1 or maybe even gated by channels other than TPCs. It has been suggested that the transient receptor potential channel TRPML1 is a NAADP-sensitive lysosomal Ca 2ϩ release channel (58 -60), but its role as a NAADP-sensitive channel is currently under dispute (61). Nevertheless, this group of NAADPsensitive stores may participate in other signaling pathways serving different cellular functions. In fact, NAADP-dependent signaling is involved in many endolysosomal functions, such as regulation of lysosomal pH, endocytosis, lipid transport and storage, and autophagy (46,56,62), in addition to contributions to vascular reactivity.
In conclusion, we have characterized the expression of TPCs and the global and local Ca 2ϩ signals mediated by NAADP in PASMCs. We found that TPC1, which is widely expressed in endosomes and lysosomes, is the major NAADP channel in PASMCs. Moreover, NAADP-induced subcellular Ca 2ϩ signals are heterogeneous, reflecting Ca 2ϩ release from different endolysosomal organelles cross-activating loosely coupled RyRs of the SR. NAADP also plays a crucial role in the agoniststimulated Ca 2ϩ release response through complex interactions with RyRs and InsP 3 Rs. Depending on the physiological stimuli and conditions, these heterogeneous NAADP-mediated Ca 2ϩ signals serve to regulate different endolysosomal functions in PASMCs.