Nitric Oxide-induced Mobilization of Intracellular Calcium via the Cyclic ADP-ribose Signaling Pathway*

Cyclic adenosine diphosphate ribose (cADPR) is a po- tent endogenous calcium-mobilizing agent synthesized from (cid:98) -NAD (cid:49) by ADP-ribosyl cyclases in sea urchin eggs and in several mammalian cells A., and White, A. (1994) Trends Cell Biol. 4, 431–436). Pharmacological studies suggest that cADPR is an endogenous modulator of Ca 2 (cid:49) -induced Ca 2 (cid:49) release mediated by ryanodine-sensitive Ca 2 (cid:49) release channels. An unresolved question is whether cADPR can act as a Ca 2 (cid:49) -mobilizing intracellular messenger. We show that exogenous application of nitric oxide (NO) mobilizes Ca 2 (cid:49) from intracellular stores in intact sea urchin eggs and that it releases Ca 2 (cid:49) and elevates cADPR levels in egg homogenates. 8-Ami- no-cADPR, a selective competitive antagonist of cADPR-mediated Ca 2 (cid:49) release, and nicotinamide, an inhibitor of ADP-ribosyl cyclase, inhibit the Ca 2 (cid:49) -mobilizing actions of NO, while, heparin, a competitive antagonist of the inositol 1,4,5-trisphosphate receptor, did not affect NO-induced Ca 2 (cid:49) release. Since the Ca 2 (cid:49) -mobilizing effects of NO can be mimicked by cGMP, are inhibited by the cGMP-dependent-protein kinase inhibitor, R p -8-pCPT- cGMPS, and in egg homogenates show a requirement for the guanylyl cyclase substrate, GTP, we suggest a novel action of NO in mobilizing intracellular calcium from microsomal stores via a signaling pathway involving cGMP and cADPR. These results suggest that cADPR has the capacity to act as a Ca 2 (cid:49) -mobilizing intracellular messenger.

Nitric oxide (NO) 1 is now recognized as a signaling molecule in many mammalian tissues where it has diverse functions as a neurotransmitter as well as an agent mediating apoptosis (1)(2)(3)(4)(5)(6)(7). Although NO was first discovered as a mediator of vascular smooth muscle relaxation, where it leads to a decrease in intracellular free calcium [Ca 2ϩ ] i (8), recent reports in interstitial cells in the mammalian gut (9), a macrophage line (10), and pancreatic ␤ cells (11) demonstrate that treatments with NO and NO donors elicit increases in [Ca 2ϩ ] i . These effects persist in the absence of extracellular calcium and can be blocked by pretreatment with ryanodine (9,11), suggesting that NO may activate a signal transduction cascade, which activates ryanodine-sensitive calcium release channels (RyRs). We have studied this novel aspect of NO action in the sea urchin egg, since Ca 2ϩ release mechanisms have been extensively studied in this system (12) and where multiple calcium mobilization pathways have been shown and are amenable to detailed analysis. In the sea urchin egg one Ca 2ϩ release mechanism is gated by the established second messenger, inositol 1,4,5-trisphosphate (IP 3 ), which is produced in response to the interaction of many extracellular stimuli with cell surface receptors (13). Another involves the activation of ryanodine-sensitive calcium release channels (14). RyRs are present on intracellular calcium stores in a wide range of cell types including sea urchin eggs (15). Here ryanodine receptors have been shown to be regulated by cADPR (16), a novel calcium-mobilizing metabolite that is synthesized from ␤-NAD ϩ by ADP-ribosyl cyclases (12). Accumulating evidence suggests that cADPR is a widespread modulator of ryanodine receptor-mediated calcium release in many different types of mammalian cells (17)(18)(19)(20)(21)(22)(23)(24)(25)(26) as well as in plants (27).
A second messenger role for cADPR requires that it mediates the intracellular actions of hormones or neurotransmitters. We show that NO mobilizes calcium from intracellular stores in the sea urchin egg via a pathway in part involving cGMP and leading to the activation of the cADPR-sensitive calcium release mechanism.

EXPERIMENTAL PROCEDURES
Collection of Sea Urchin Eggs-Eggs were obtained by stimulating ovulation of female Lytechinus pictus (Marinus, Inc., Long Beach, CA) with an intracoelomic injection of 0.5 M KCl solution. These were then washed twice in artificial seawater (435 mM NaCl, 40 mM MgCl 2 , 15 mM MgSO 4 , 11 mM CaCl 2 , 10 mM KCl, 2.5 mM NaHCO 3 , 1 mM EDTA at pH 8.0), and jelly was removed by filtration through 85-m Nitex mesh.
Imaging of Intracellular [Ca 2ϩ ] i in Eggs-Eggs were transferred to poly-L-lysine (10 mg/ml)-coated glass coverslips, allowed to adhere, and microinjected with fura-2, pentapotassium salt (10 mM in the pipette), in buffer consisting of 0.5 M KCl, 20 mM Pipes at pH 6.7 to a final cellular concentration of approximately 10 M. Injection volumes did not exceed 1% of cell volume. All experiments were performed at 22°C. Free cytosolic Ca 2ϩ concentration was determined by ratioing fluorescence intensities at excitation wavelengths of 340 and 380 nm, using an emission wavelength of 510 nm. Ratio images were obtained using a fluorimetric imaging system and Ionvision software supplied by Improvision Ltd., University of Warwick Science Park, Coventry, UK. Standard CaCl 2 solutions were used to calibrate the system, and viscosity corrections were made (28).
NO Applications-NO-containing solutions in respective buffers were prepared by bubbling NO gas (Aldrich) at 4°C and under oxygenfree conditions to reduce oxidation. The NO concentration was measured with a NO-sensitive electrode (ISO-NO meter, 2-mm diameter electrode; range of 1 nM to 20 M, World Precision Instruments, Stevenage, UK). The electrode was calibrated in accordance with the manu-facturer's recommended methodology using a chemical titration method. Known concentrations of KNO 2 are converted to NO in the presence of reducing agents KI and H 2 SO 4 . Our stock solutions were ϳ350 M with respect to gaseous NO. Exposure of eggs to NO was achieved by adding 50 l of the solution into the imaging chamber (volume, ϳ 500 l) or by the addition of 5-20 l of the stock solution to the cuvette containing 500 l of 5% sea urchin egg homogenate in homogenate experiments.
Measurement of cGMP-Cyclic GMP was radioimmunoassayed according to Doshi et al. (29), using a kit supplied by Amersham International plc. Protein was measured by the method of Lowry et al. (30), as modified by Miller (31).
Ca 2ϩ Release Assays-Homogenates (5%) of unfertilized L. pictus eggs (Marinus, Inc.) were prepared as described previously (32). 5% microsomes were purified from homogenates by Percoll centrifugation (33). Ca 2ϩ loading was achieved by incubation at room temperature for 3 h in an intracellular medium (IM) consisting of 250 mM potassium gluconate, 250 mM N-methylglucamine, 20 mM Hepes (pH 7.2), 1 mM MgCl 2 , 1 mM ATP, 10 mM phosphocreatine, 10 units/ml creatine phosphokinase, 1 g/ml oligomycin, 1 g/ml antimycin, and 3 M fluo-3. Fluorimetry was performed at 17°C using 500 l of 5% homogenate, continuously stirred, in a Perkin-Elmer LS-50B fluorimeter. Free Ca 2ϩ concentration was measured by monitoring fluorescence intensity at excitation and emission wavelengths of 490 and 535 nm, respectively. Additions were made in 1-5-l volumes, and all chemicals were added in IM containing 10 M EGTA. Basal concentrations of Ca 2ϩ were typically between 100 and 150 nM. Sequestered Ca 2ϩ was determined by monitoring the decrease in fluo-3 fluorescence during microsomal loading and by measuring Ca 2ϩ release in response to ionomycin (5 M) and was constant between experiments.
cADPR Determinations in Egg Homogenate Treated with NO-containing Solutions-Fluorescent increases obtained in homogenates treated with NO aliquots were translated into cADPR levels from a standard curve of fluorescence versus cADPR concentrations obtained from homogenates to which known amounts of authentic cADPR had been added. The specificity of the bioassay for cADPR (34) was demonstrated by the complete inhibition of NO-induced increases in fluo-3 fluorescence by the prior addition of a desensitizing concentration of cADPR (1 M) (32) or treatment with the cADPR antagonist, 8-amino-cADPR (35).
Materials-cADPR and 8-amino-cADPR were synthesized as described previously. Ryanodine, fluo-3, and fura-2 were purchased from Calbiochem; R p -8-pCPT-cGMPS was from Biolog Life Science Institute, Bremen, Germany. All other chemicals were from Sigma. Fig. 1 shows that in single sea urchin eggs microinjected with the Ca 2ϩ indicator fura-2, application of exogenous NO dis-solved in seawater (approximate final concentration of 32 M) caused an increase in [Ca 2ϩ ] i . There was a latency of 17 Ϯ 3 s (n ϭ 15; S.E.) before the initiation of the [Ca 2ϩ ] i signal, which occurred at a discrete locus and then spread across the egg as a rapid but short-lived Ca 2ϩ wave. The magnitude of NOinduced calcium transients (800 Ϯ 30 nM, n ϭ 12; S.E.) was generally smaller than those elicited at fertilization (1 Ϯ 0. Intracellular free Ca 2ϩ was calibrated as described previously (38). There was a latency of at least 14 s before a calcium rise was detected, and the calcium often increased in one region of the egg before spreading across the cell (see columns 1, 2, and 5). The increase in calcium was transient with a return to baseline, often occurring within 10 s of its initiation. The approximate wave velocity was of the order of 15 Ϯ 3 m/s (n ϭ 12; S.E.). cantly affected by the removal of extracellular calcium (Fig. 1, 5th column), indicating that it was produced predominantly by release from intracellular Ca 2ϩ stores.

RESULTS
To confirm that NO was mobilizing Ca 2ϩ from intracellular stores we tested the effects of NO on Ca 2ϩ release in sea urchin egg homogenates. Fig. 2 shows the simultaneous measurement of NO concentration changes and transient Ca 2ϩ release in sea urchin egg homogenates stimulated with a bolus of NO-containing IM solution. There was a rapid increase in NO concentration in the homogenate, which reached a peak of approximately 5 M as measured with a NO electrode that declined over 150 s. The Ca 2ϩ release elicited by this stimulus occurred only after a latency of around 120 s. Fig. 3A shows the effect of varying the concentration of NO (3-10 M) in the presence of ␤-NAD ϩ (50 M) and GTP (250 M). The magnitude of response increased with increasing NO concentrations, whereas the latency was inversely dependent and was as long as 180 s at lower NO concentrations. Using the sea urchin egg microsomes as a bioassay for cADPR, we obtained a concentration-response relationship for NO-induced cADPR production in egg homogenates (Fig. 3A, inset).
The mechanism of NO on Ca 2ϩ release was indirect since it was unable to mobilize Ca 2ϩ from purified microsomes (data not shown), suggesting the requirement for cytosolic factors present in crude homogenate. In addition it also required the presence of ␤-NAD ϩ and GTP. The dependence of both ␤-NAD ϩ and GTP for the Ca 2ϩ -mobilizing effect of NO is shown in Fig.  3B. Addition of NO (9 M) in the absence of either ␤-NAD ϩ or GTP to egg homogenates alone caused no Ca 2ϩ release. However, Ca 2ϩ release by NO (9 M) could be reconstituted in the presence of both ␤-NAD ϩ (50 M) and GTP (250 M) (Fig. 3B). NO-induced release displayed a latency of ϳ100 s. Increasing the GTP concentration to 500 M shortened the latency and increased the magnitude of response (Fig. 3B). A rapid Ca 2ϩ release by subsequent addition of cADPR (100 nM) with no apparent delay could be achieved under all four conditions; however, the magnitude of the cADPR response was diminished in proportion to release obtained with NO. The effects of NO-induced Ca 2ϩ release on the magnitude of subsequent release by either cADPR is shown in Fig. 3C. The more Ca 2ϩ released by NO reduces that triggered by cADPR (100 nM). The pharmacology of NO-induced Ca 2ϩ release is shown in Fig. 3D. NO-induced Ca 2ϩ release was antagonized by 8-amino-cADPR (400 nM), which also blocked Ca 2ϩ release by a subsequent addition of cADPR (200 nM) (Fig. 3D), suggesting that the effect of NO was mediated by cADPR. Consistent with this result was that nicotinamide, which inhibits ␤-NAD ϩ conversion to cADPR catalyzed by ADP-ribosyl cyclases, 2 abolished NO but not cADPR-induced Ca 2ϩ release (Fig. 3D). The cGMP-dependent protein kinase inhibitor, R p -8-pCPT-cGMPS (200 M) (36), also blocked Ca 2ϩ release by NO but not by cADPR (Fig. 3D). These data suggest that NO-induced calcium release requires the participation of cGMP-dependent protein kinases and ADPribosyl cyclases, which may explain the requirement for cytosol as well as GTP and ␤-NAD ϩ . Since NO required the cGMP precursor GTP for Ca 2ϩ release and a cGMP-dependent protein kinase inhibitor blocked the effects, we examined the effects of cGMP on Ca 2ϩ release in egg homogenates. Previous studies have indicated that cGMP mobilizes Ca 2ϩ in sea urchin eggs (37) and in egg homogenates (38). cGMP alone does not have direct Ca 2ϩ mobilizing activity but has been reported to enhance the synthesis of cADPR from ␤-NAD ϩ (38). Fig. 4 shows that in microsomal fractions derived from sea urchin eggs in the presence of 25% supernatant (32,33), treatment with cGMP leads to the release of calcium from microsomes after a variable latency of a number of seconds. The amplitude of [Ca 2ϩ ] i release with cGMP was dose-dependent, and the latency in the range of 300 -600 s was inversely dependent on cGMP concentration (Fig. 4A). Mobilization of calcium was absolutely dependent on the presence of ␤-NAD ϩ (38) and abolished (Fig. 4B) by the competitive cADPR antagonist, 8amino-cADPR (34). R p -8-pCPT-cGMPS (200 M) also com-pletely abolished Ca 2ϩ release by cGMP (data not shown). Heparin (0.2 mg/ml), which blocks IP 3 receptors in sea urchin eggs (32,38) and other tissues, had no inhibitory effect on Ca 2ϩ release in response to cGMP, although it blocked release by IP 3 (1 M) (Fig. 4C).
We investigated whether the Ca 2ϩ -mobilizing actions of NO in sea urchin eggs were mediated by cGMP, since NO is well characterized as an activator of soluble guanylate cyclase (7). We measured intracellular levels of cGMP in eggs and egg homogenates treated with NO. In NO-treated eggs there was an approximate doubling in the intracellular levels of cGMP (Fig. 5A). In egg homogenates NO treatments lead to a concentration-dependent increase in the cGMP by over 3-fold (Fig. 5B).
To investigate the mechanism of NO-induced Ca 2ϩ mobilization from intracellular stores in intact sea urchin eggs, eggs were treated with pharmacological agents that inhibit NO or cGMP effects or inhibit cADPR-induced Ca 2ϩ release mechanisms (Fig. 6). Eggs microinjected with 8-amino-cADPR to a final concentration of 1 M (Fig. 6, column 2) showed substantially reduced NO-induced calcium increases in the egg (approximately 90%) compared with the control (Fig. 6, column 1). In experiments in Ca 2ϩ -free medium the response in 8-amino-CADPR-injected eggs was completely abolished (data not shown). R p -8-pCPT-cGMPS (25 M) also reduced the NO-induced Ca 2ϩ transient substantially (Fig. 6, column 3), suggesting a role of cGMP and cGMP-dependent protein kinase in mediating Ca 2ϩ mobilization by NO. Hemoglobin that scavenges NO and blocks NO-mediated effects in mammalian systems (2) also blocked the NO-induced Ca 2ϩ signal in sea urchin 2 J. Sethi and A. Galione, unpublished observations.

FIG. 4. cGMP-induced Ca 2؉ release from sea urchin egg microsomes and homogenates.
A, comparison between cADPR-and cGMPinduced Ca 2ϩ release from sea urchin egg microsomes (5%) in the presence of 25% supernatant (v/v). cADPR caused an immediate Ca 2ϩ release in a dose-dependent manner. cGMP-induced Ca 2ϩ release was dose-dependent and only occurred in the presence of ␤-NAD ϩ and showed a latency whose duration was inversely dependent on cGMP concentration. B, the cADPR receptor antagonist, 8-amino-cADPR (80 nM), completely abolished cGMP-induced Ca 2ϩ release. C, IP 3 -induced Ca 2ϩ release is immediate and dose-dependent. Heparin (Hep, 0.2 mg/ml) abolished Ca 2ϩ release by IP 3 (1 M) but had no effect on cGMP-induced Ca 2ϩ release. All figures are representative of at least three experiments. eggs (column 4). However, intracellular injection of heparin (0.4 mg/ml, final concentration) did not reduce the calcium transient in response to NO (column 5). There was a slight enhancement of the response (n ϭ 5 eggs). One possibility is that heparin weakly activates the RyR as has been reported for RyRs in lipid bilayers (39), and heparin has been shown to enhance ryanodine-induced Ca 2ϩ release in sea urchin eggs (40), although it did not augment Ca 2ϩ release by cGMP in egg homogenates (Fig. 4C). DISCUSSION cADPR has been identified as a potent Ca 2ϩ -releasing agent through a Ca 2ϩ release mechanism that is distinct from that regulated by IP 3 (32). In many systems, including sea urchin eggs, cADPR appears to act as modulator of CICR through RyRs (41). Although the number of cell types in which cADPR is an effective Ca 2ϩ -releasing agent continues to grow, little is known about possible receptor mechanisms that may be coupled to intracellular cADPR production.
In this investigation we identified NO as an agonist that can mobilize intracellular Ca 2ϩ by selectively activating a Ca 2ϩ signaling pathway involving cADPR while having no effect on the IP 3 receptor pathway. We have previously shown that cGMP can enhance cADPR synthesis in sea urchin eggs and homogenates (38) and that this may underlie the Ca 2ϩ -mobilizing action of cGMP in this cell (37,42). We have investigated whether the guanylyl cyclase activator NO can also release Ca 2ϩ from intracellular stores by activating the cADPR signaling pathway. Surprisingly, for an agent that was first discovered as a relaxant of smooth muscle (8), NO was found to elicit a large Ca 2ϩ transient in intact sea urchin eggs loaded with the intracellular Ca 2ϩ reporter fura-2 (Fig. 1). The sea urchin egg is rapidly becoming a useful system in which to investigate Ca 2ϩ mobilization since Ca 2ϩ stores in these eggs express multiple Ca 2ϩ release channels that participate in the fertilization Ca 2ϩ wave (43,44) and the regulation of these channels can be directly investigated in egg homogenates or microsomal preparations (16). The Ca 2ϩ -mobilizing action of NO could be reconstituted in the egg homogenate system, greatly facilitating the analysis of its mechanism of action. From homogenate experiments we have shown that the NO-induced Ca 2ϩ mobilization operates predominantly via the cADPR rather than the IP 3 -sensitive Ca 2ϩ -release mechanism, since the effect of NO is abolished by the cADPR antagonist, 8-amino-cADPR (34). The mechanism of NO action was indirect since it required additional factors such as GTP and ␤-NAD ϩ , the precursor for cADPR synthesis. Since the NO effects are reduced by cGMPdependent protein kinase inhibitors and cGMP has been reported to stimulate ␤-NAD ϩ metabolism to cADPR and ADPribose (38), a possible pathway for the Ca 2ϩ -mobilizing effects of NO is that NO activates a soluble guanylate cyclase, and the resulting cGMP elevation (Fig. 5) activates a cGMP-dependent protein kinase, which then phosphorylates ADP-ribosyl cyclase or a regulator of this enzyme, resulting in an increase in cADPR levels. cADPR then binds to its receptor (45), leading to the opening of a RyR-like Ca 2ϩ channel in the endoplasmic reticulum (16,42) resulting in a rise in [Ca 2ϩ ] i (Fig. 7). However, since the increases in cGMP in NO-stimulated eggs are modest (Fig. 5) compared with the concentrations of cGMP required to mimic the effects of NO, we cannot exclude other actions of NO upon cADPR synthesis such as direct ADPribosylation (46), although the requirement for the cGMP pre- FIG. 5. NO increases cGMP levels in unfertilized sea urchin eggs and egg homogenates. cGMP levels were measured using a radioimmunoassay protocol (see "Experimental Procedures"). A, nitric oxide was bubbled into degassed artificial seawater, yielding a solution of ϳ350 M. 50 l of this solution was applied to sea urchin eggs (500 l) suspended in artificial seawater at t ϭ 0 s. Incubations were then continued at 22°C. Levels of cGMP were measured at the times indicated. Results are expressed as the mean of four separate estimations Ϯ S.D. B, cGMP levels were also measured in 5% egg homogenates before and after NO additions (350 nM to 105 M) to homogenates. The initial concentrations of NO in the homogenate were measured with a NO electrode. cGMP levels were measured in unstimulated homogenates and 10 s after the addition of NO, the time for peak Ca 2ϩ release. Results are expressed as the mean of six separate estimations Ϯ S.D. for each NO concentration.
cursor GTP for NO-induced Ca 2ϩ release in homogenates and the ability of G-kinase inhibitors to block NO effects in both intact eggs and homogenates may favor a cGMP-dependent mechanism. The other possibility that NO/cGMP sensitizes Ca 2ϩ release through RyRs by endogenous cADPR is unlikely since R p -8-pCPT-cGMPS or nicotinamide blocks NO-induced Ca 2ϩ release but does not inhibit Ca 2ϩ by exogenously added cADPR (Fig. 3D).
Since NO synthesis by constitutive NO synthases is often calcium-dependent (1), a NO-induced rise in [Ca 2ϩ ] i may serve to amplify NO production as previously seen (9) and could also give rise to regenerative Ca 2ϩ waves seen in many single cells and tissues (47). Whether nitric oxide has a role in calcium signaling at fertilization in the sea urchin egg remains to be determined. Since the magnitude of the Ca 2ϩ wave elicited by the high concentrations of NO required to induce Ca 2ϩ release is insufficient to activate sea urchin eggs, if such a mechanism is employed at fertilization it is likely to be modulatory. One possible role of the NO-activated pathway being investigated is that NO could be locally produced at the site of sperm-egg fusion, which would rapidly diffuse across the entire cell. This could lead to a global rise in cADPR, which could facilitate a wave of CICR across the egg to activate it by sensitizing the egg's CICR mechanism to activation by increases in [Ca 2ϩ ] i .
The NO-stimulated Ca 2ϩ mobilization pathway involving cADPR/RyRs might augment the recently described effects of NO and cGMP in regulating receptor-mediated Ca 2ϩ influx across the plasma membrane in other cells (48), contribute to RyR-based subsarcolemmal Ca 2ϩ sparks, which have recently been implicated in regulating relaxation of vascular smooth muscle (49), and be important in NO-induced changes in neuronal plasticity (50). Whether the recently described stimulation of ADP-ribosyl cyclases in longitudinal smooth muscle by cholecystokinin (51) involves either NO or cGMP as intermediates remains to be determined.