Fertilization and Nicotinic Acid Adenine Dinucleotide Phosphate Induce pH Changes in Acidic Ca2+ Stores in Sea Urchin Eggs*

The second messenger nicotinic acid adenine dinucleotide phosphate (NAADP) releases Ca2+ from the acidic Ca2+ stores of many organisms, including those of the sea urchin egg. We investigated whether the pH within the lumen of these acidic organelles changes in response to stimuli. Fertilization activates the egg by Ca2+ release dependent upon NAADP, and accordingly, we report that fertilization also alters organellar pH in a spatio-temporally complex manner. Upon sperm fusion, vesicles deep in the egg center slowly acidify, whereas cortical vesicles undergo a rapid alkalinization. The cortical vesicle alkalinization is independent of exocytosis and cytosolic pH but coincides with the NAADP-dependent fertilization Ca2+ wave. Microinjection of NAADP mimicked the fertilization cortical response, suggesting that it occurred within NAADP-sensitive acidic Ca2+ stores. Our data show that NAADP and physiological stimuli alter the pH within intracellular organelles and suggest that NAADP signals through pH as well as Ca2+.

The sea urchin egg is an invaluable cell model for studying Ca 2ϩ signaling since the discovery of new second messengers and new Ca 2ϩ stores in this system has subsequently proven to have a major impact on mammalian biology (1). The mechanism of activation of the egg upon fertilization encompasses the rapid increases in the cytosolic pH and intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) important for fertilization envelope formation, DNA and protein synthesis, and embryological development (2). The pattern of [Ca 2ϩ ] i changes within the egg are highly organized spatio-temporally and reflect the precisely timed and placed recruitment of different families of Ca 2ϩ channels resident upon either the plasma membrane or intracellular organelles (2,3). After the initial transient "cortical flash" caused by Ca 2ϩ influx across the plasma membrane (2)(3)(4)(5), Ca 2ϩ release from intracellular stores is manifest as a Ca 2ϩ wave that propagates across the entire egg initiating from the point of sperm entry (2). These phasic elevations in [Ca 2ϩ ] i are driven by a complex interplay between the second messen-gers inositol 1,4,5-trisphosphate, cyclic adenosine diphosphoribose, and nicotinic acid adenine dinucleotide phosphate (NAADP) 2 (1). Importantly, the relative contribution of each messenger to each phase of the Ca 2ϩ signal is different, e.g. NAADP is the only messenger implicated in the cortical flash (4,5), whereas cyclic adenosine diphosphoribose probably plays a later role in prolonging the main Ca 2ϩ spike (6).
NAADP not only evokes a cortical flash but, as first revealed in sea urchin egg, also releases Ca 2ϩ from acidic Ca 2ϩ stores that are lysosome-like organelles (possibly yolk platelets in eggs), whereas inositol 1,4,5-trisphosphate and cyclic adenosine diphosphoribose release Ca 2ϩ from the neutral endoplasmic reticulum (1,7). Recently in sea urchin egg homogenate, we revealed that the luminal pH (pH L ) of these acidic stores is dynamic and increases upon NAADP-induced Ca 2ϩ release, which is a direct effect of NAADP (and not cytosolic Ca 2ϩ ) via changes in luminal proton buffering/transport (8). We have now confirmed that the pH L increases observed in homogenate are physiologically relevant in the intact egg upon fertilization. Unexpectedly, these changes are confined to the NAADP-sensitive stores in the egg cortex during a novel, early fertilization event, which may have ramifications for the pH L of NAADPsensitive stores in mammalian cells.
40ϫ and 63ϫ oil immersion objectives (NA 0.3, 1.3, and 1.4, respectively) and maintained at room temperature in ASW. Simultaneous monitoring of several fluorescent dyes was carried out using the Multitrack mode of the microscope, which rapidly alternates channel collection and thereby greatly reduces bleed-through. For UV excitation, the 364-nm line of a Coherent Enterprise laser was used. The green channel was excited using the 488-nm line of an argon laser (emission, 505-530 nm), whereas the red channel used a 543-nm HeNe laser (emission, Ͼ560 nm). In the experiments using an Alexa Fluor 647 Dextran, this third fluorophore was excited with a 633-nm HeNe laser and emission collected at 645-719 nm using the Meta head.
Microinjection-Micropipettes were pulled from capillary glass with an internal filament and backfilled. The pipettes were then mounted in the electrode holder of an Injectman pressure injection system (Eppendorf) used at typical pressures of 200 hPa for 2 s, which produces ϳ1% injection volumes. Injectate compounds were usually prepared in 0.5 M KCl at 100 times the final required concentration, and insoluble debris was removed by centrifugation through a Centricon filter (100,000 molecular weight cutoff). Rhod Dextran was first prepared as 2 mM aliquots in 100 mM Tris-HCl, pH 8.0, to assist solubility and then diluted 1:1 with 1 M KCl. The injectate concentration of Alexa Fluor 647 Dextran was 50 -200 M.
pH L Measurements-We measured pH L with two different approaches. First, eggs were loaded with 2-10 M Lysosensor Yellow/Blue DND-160 for 15 min with qualitatively similar results at all dye concentrations. Lysosensor Yellow/Blue was imaged ratiometrically on the confocal microscope with excitation at 364 nm and dual emission recorded at 385-470 nm (channel 1) and Ͼ505 nm (channel 2). The results are expressed as the ratio of channel 1/channel 2, which is directly proportional to pH L . In the majority of experiments, however, acridine orange was used because of its superior signal-to-noise and its documented use to monitor pH L in both sea urchin egg homogenate (8) and intact eggs (10). However, we modified the acridine orange protocol to measure pH L ratiometrically to circumvent movement and cell volume artifacts. Therefore, intact eggs were simultaneously loaded with 10 M acridine orange and 1 M Lysotracker Red DND-99 for 15-20 min at room temperature, at which time the fluorescence reaches an equilibrium; the dyes were present throughout the rest of the experiment. Acridine orange responds rapidly and profoundly to changes in pH L , whereas Lysotracker Red responds only poorly (in agreement with Invitrogen) and remains essentially unchanged throughout the experiment until the addition of NH 4 Cl. The results are expressed as the ratios of the acridine orange/Lysotracker Red signals such that an increase in the ratio reflects an increase in pH L . At higher magnification (see Fig. 6), Lysotracker Red exhibited some photobleaching that resulted in an artifactual rise in the basal ratio in some eggs. This fall in fluorescence was corrected by using a fifth order polynomial curve fit.
In experiments investigating the effect of sodium removal, the eggs were dye-loaded in normal ASW prior to washes in Na ϩ -free ASW (also containing the pH L dyes) immediately before recording. Na ϩ -free ASW was prepared by replacing NaCl with 435 mM N-methyl-D-glucamine and replacing NaHCO 3 with 2.5 mM KHCO 3 .
Measurement of [Ca 2ϩ ] i -The eggs were microinjected with the Ca 2ϩ -sensitive dye fluo-4 dextran (10 kDa, injectate concentration 1 mM, plus 250 mM KCl, Tris pH 8). After a 15-min recovery period, a second micropipette containing 50 -200 M Alexa Fluor 647 Dextran with or without 100 M NAADP was used for injection.
Simultaneous Measurement of pH L and [Ca 2ϩ ] i -The eggs were first microinjected with Ca 2ϩ -sensitive Rhod Dextran (K d 0.7 M, 1 mM injectate) and then loaded with acridine orange. However, because Rhod Dextran is a much dimmer dye under basal conditions than Lysotracker Red, acridine orange was used at the lower concentration of 1 M to reduce bleedthrough into the necessarily more sensitive red channel. Qualitatively similar results were seen at both 1 and 10 M acridine orange when eggs were fertilized, but the fluorescence change was smaller with 1 M.
Simultaneous Measurement of pH L and Exocytosis-Exocytosis was assessed with two different fluorescent techniques. First, by monitoring the disappearance of cortical granules; acridine orange monomers aggregate in acidic vesicles as a function of pH L , and consequently their fluorescence undergoes a red shift as a function of aggregation (10,11). When excited with the red laser, a selective staining of a narrow (1-2 m) peripheral ring is observed that corresponds to the acidic cortical granules docked at the plasma membrane (12). The eggs were therefore labeled as before with 10 M acridine orange for 15-20 min, but this time they were simultaneously viewed in the green and red channels using the Multitrack mode of the microscope (emission/excitation, 488 nm/505-530 nm (green); 543 nm/Ͼ560 nm (red)). Note that in the previous ratiometric experiments, the Lysotracker Red signal swamped the dimmer acridine orange red fluorescence.
We used another approach to validate the loss of red acridine orange fluorescence; the extracellular space was also fluorescently labeled with a long wavelength, cell-impermeant dye (10 M Alexa Fluor 647 Dextran), whose signal was simultaneously collected with the red (cortical granule) and green (pH L ) channels as indicated above. At high magnification (63ϫ objective, 2ϫ digital zoom), the exocytosis of individual cortical granules observed with the red acridine orange fluorescence was confirmed when the space vacated by the granule filled with extracellular dye (data not shown), as originally developed by Terasaki (13). The data were analyzed by using Student's t test, paired where appropriate, or by creating a 2 ϫ 2 contingency table analyzed with Fisher's exact test. The significance was set at p Ͻ 0.05.

RESULTS
Fertilization-induced pH L Changes-Sperm activate eggs and evoke Ca 2ϩ release from acidic stores via NAADP (5). Measuring pH L changes within these acidic Ca 2ϩ stores revealed that fertilization altered the pH L after a short lag (Fig.  1). The acridine orange signal changed substantially more than did Lysotracker Red (which is less pH L -sensitive, see "Experimental Procedures"), consistent with an alteration in pH L rather than a morphological artifact (Fig. 1, B and C). Fig. 1D shows the calculated ratio, and unexpectedly, this was not a spatially homogeneous response; around the periphery (cortex) of the egg, pH L promptly increased in a narrow band, whereas in the center of the egg pH L decreased, albeit with slower kinetics ( Fig. 1, D and F). The transient and spatially restricted nature of the pH L increase was highly reminiscent of the [Ca 2ϩ ] i cortical flash (4,5), for which reason we have termed this novel phenomenon a "pH L ASH" (pronounced "flash"). This peripheral vesicular alkalinization did not reflect an apparent unresponsiveness of the egg center because NH 4 Cl evoked a profound alkalinization across the entire cell ( Fig.  1, A, C, and F). The results were confirmed using the chem-ically and spectrally dissimilar ratiometric dye, Lysosensor Yellow/Blue DND-160, albeit with an inferior signal-to-noise (see supplemental Fig. S1). Fertilization therefore induces changes in organellar pH that are highly organized in time and space.
In cells fortuitously inseminated in the right orientation, this pH L ASH proceeded as an alkalinization "wave" around the cell cortex (Fig. 1E). It is worth noting that in many eggs, the exocytotic lifting of the fertilization envelope results in such a dramatic change in cell shape/focus that a parallel fall in the fluorescence of both dyes occurs that is unrelated to changes in pH L and is depicted as a broken line in the ratio plot ( Fig. 1D). (Note that this also accounts for the artifactual differences in the ratiometric NH 4 Cl response in the periphery versus the center; Fig. 1F.) When measuring pH L alone, cellto-cell asynchrony (Fig. 1A) made it difficult to gauge how this pH L ASH related temporally with the fertilization Ca 2ϩ response, i.e. would the pH L ASH coincide with the Ca 2ϩ cortical flash (Ca 2ϩ influx) or the main Ca 2ϩ wave (intracellular Ca 2ϩ release)? Therefore, we simultaneously measured pH L and [Ca 2ϩ ] i using acridine orange and the Ca 2ϩ dye Rhod Dextran. In these dual labeling experiments, the first detectable insemination response was the Ca 2ϩ cortical flash ( Fig. 2A, left panels), followed after a delay by the main Ca 2ϩ wave (Fig. 2).
Temporally, the pH L ASH overlapped with the main Ca 2ϩ wave (Fig. 2, A and B), with the pH L upstroke occurring ostensibly with or just after the sharp [Ca 2ϩ ] i upstroke; the bar graph in Fig. 2C shows that the lag between sperm addition and the cortical flash is Ͻ30 s, whereas the main Ca 2ϩ wave and pH L ASH both peak at ϳ65 s. In other words, despite their spatial overlap, the Ca 2ϩ cortical flash and pH L ASH are temporally divorced; the latter coincides more with the main wave. This timing is consistent with the pH L ASH being associated with the intracellular Ca 2ϩ release phase.
Spatially, the Ca 2ϩ wave and pH L ASH wave initiated from the same pole in 7/10 cells (Fisher's exact test, p Ͻ 0.02) and propagated through the egg periphery to the antipode (Fig. 2B). Despite a heterogeneous pH L response, the main Ca 2ϩ wave FIGURE 1. Fertilization-induced changes in pH L in intact sea urchin eggs. The eggs were co-loaded 10 M acridine orange (AO) and 1 M Lysotracker Red (LTR) and fertilized with 0.1% (v/v) ejaculate where indicated. All of the images shown were captured using a 10ϫ objective. A, pseudocolor images of the ratio of acridine orange/Lysotracker Red fluorescence where cool colors represent low pH L , and warmer colors represent higher pH L . The time in seconds is shown in the upper left corner of each image. B, raw fluorescence of acridine orange (solid traces) or Lysotracker Red (broken traces) taken from the egg periphery in accordance with the inset schematic of a single egg. Sperm added when indicated followed by 10 mM NH 4 Cl (C) as for B except that the raw data is taken from the egg center. D, peripheral (red) and central (green) ratiometric (acridine orange/ Lysotracker Red) pH L changes derived from the traces plotted in B and C. The broken portion of the red trace corresponds to the artifactual ratio because of the loss of fluorescence from the peripheral ROI as the fertilized egg apparently changes shape. E, the ratiometric pH L response proceeds as a wave around the egg cortex. The traces correspond to their color-matched ROIs in the inset for the same egg in B and C. F, bar graph showing the change in ratio in the egg periphery (Peri) and center (means Ϯ S.E. of 77 eggs). amplitude was the same in the periphery and center (data not shown). Because the Ca 2ϩ wave initiates at the point of sperm entry (2), we conclude that the pH L wave also starts at the point of sperm fusion.
pH L Events and Exocytosis-However, it was possible that the pH L change is secondary to Ca 2ϩ -stimulated exocytosis (and/or endocytosis), e.g. upon vesicle fusion with the plasma membrane, the alkaline extracellular medium exchanges with the acidic lumen of cortical granules, the primary egg exocytotic vesicles (12,14). Therefore, we simultaneously recorded pH L changes and exocytosis to see whether there was any spatial or temporal overlap. Exocytosis was monitored as the disappearance of cortical granule labeling (12).
Measuring the fluorescence at the periphery, fertilization induced a prompt fall in the red fluorescence, which corresponded to the loss of cortical granules via exocytosis (Fig. 3,  A and B). By altering the focus, we verified that this was a true loss of staining rather than egg movement in the focal plane (data not shown). The initiation of the pH L ASH and the loss of cortical granules were, on average, simultaneous (Fig. 3A) with a lag of only 5 Ϯ 3 s, which was not significant (n ϭ 34, p Ͼ 0.05). However, this obscured the fact that the pH L ASH and exocytosis were seldom coincident on an individual egg basis, and plotting the initiation times revealed that there was no significant correlation between these two parameters (Fig. 3B, r 2 ϭ 0.003, p Ͼ 0.7). These data indicate that the pH L ASH and exocytosis can be divorced kinetically.
In addition to their different kinetics, the pH L ASH and exocytosis could be spatially distinguished. Fig. 3C shows a profile plot across the cell as indicated in Fig. 3A (upper left) and shows the mean fluorescence changes in the green and red channels after fertilization (Note that the red signal has been inverted for clarity); whereas the red changes were limited to the narrow outer cortical granule layer, the pH L ASH occurred in a broader band (up to 25 m away from the plasma membrane at low magnification).
The final piece of evidence to discount exocytosis as a causal factor comes from experiments with the wasp venom mastoparan. Known to stimulate fertilization envelope formation in sea urchin eggs (15), mastoparan is thought to do so independently of changes in [Ca 2ϩ ] i (16), presumably by interacting with the exocytotic protein machinery itself. Indeed, although mastoparan readily evoked cortical granule loss (76 Ϯ 3% of sperm control, n ϭ 37), it did so without any detectable pH L ASH (O4 Ϯ 6% sperm, p Ͻ 0.001; Fig. 3D). Moreover, the addition of sperm after mastoparan elicited no response compared with parallel fertilization controls (O8 Ϯ 1%, p Ͻ 0.001). This is consistent with a physical barrier (i.e. the fertilization envelope) being induced by the venom.
The fact that exocytosis and pH L ASH can be divorced under a variety of circumstances strongly supports an independent mechanism for the latter. Because it is well known that endocytosis takes place many minutes after the exocytotic burst (12,17), we conclude that neither of these events is responsible for the pH L ASH.
Unitary pH L Events-Not only did the higher resolution images confirm that the pH L ASH occurred well away from the fusion with the plasma membrane, but pH L ASH was revealed as a summation of smaller events (which we term "pH L ARES," pronounced "flares"). When focused at the egg equator, pH L ARES were seen to propagate as a wave along a 15-mwide cortical band (Fig. 4A). When plotted as a function of distance from the plasma membrane, both the rate of rise and amplitude of these events were well maintained up to ϳ15 m  DECEMBER 28, 2007 • VOLUME 282 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 37733 from the cortical granules (Fig. 4A, red to yellow ROIs), but both parameters fell off dramatically at deeper loci.

Ca 2؉ and pH L at Fertilization
Discrete pH L ARES were more clearly observed when focused along the bottom of the egg in the plane of the egg cortex. Viewed en face (Fig. 4B), individual pH L ARES with a ϳ2-m diameter were more clearly resolved (Fig. 4B, white arrows) presumably because of reduced scatter and turbidity at the eggsubstratum interface. Unlike the equatorial slice, the bottom of the egg shows pH L ARES firing across the entire field of view that were invariant with distance from the cell edge (Fig. 4B, cf. red and black ROIs). Together, the data indicate that pH L changes can be resolved as small events that are spatially divorced from exocytosis. pH L and Cytosolic pH-In a previous egg population study, the slow vesicular acidification upon fertilization was secondary to changes in cytosolic pH driven by the plasma membrane sodiumproton exchanger (NHE) (10). Therefore, we tested whether the NHE was also involved in the subplasmalemmal pH L ASH by inhibiting the NHE with Na ϩ -free medium. However, this had no effect upon the pH L ASH amplitude (Fig. 5, p Ͼ 0.1), pH L ASH kinetics (time to peak in s: control, 67 Ϯ 2; Na ϩ -free, 69 Ϯ 2; p Ͼ 0.4), or fertilization envelope formation (data not shown). Nonetheless, it entirely blocked the acidification of granules deeper in the egg (Fig. 5, B and C, p Ͻ 0.001) in agreement with previous work (10). We conclude that the cortical pH L response was not secondary to NHE-driven changes in cytosolic pH.
NAADP-induced pH L ash-It was unexpected that fertilization could evoke vesicular alkalinization in such a restricted subcellular locus when acidic vesicles are found throughout the egg (7,10). Given that fertilization is NAADPdependent, and NAADP is the only Ca 2ϩ -mobilizing second messenger that evokes an increase in pH L (8), we microinjected eggs with NAADP to map the spatial distribution of NAADP-sensitive acidic Ca 2ϩ stores that respond with a change in pH L .
In control eggs microinjected with a fluorescent injection marker alone, no substantial change of the pH L (Fig. 6, A and C) or Ca 2ϩ (Fig. 6, D and F) was observed in any part of the egg. By contrast, when NAADP was co-injected into the center of the egg (as judged by the fluorescent marker, left hand images), a rapid and profound alkalinization was evoked in the characteristic peripheral pattern of the pH L ASH (Fig. 6, B and  C). This was not due to a heterogeneous Ca 2ϩ response because NAADP evoked a global increase (Fig. 6, E and F). In other words, the pH L ASH response seen with sperm is entirely consistent with the intrinsic nature of a subpopulation of NAADPsensitive stores rather than upstream signal compartmentation.
Closer examination revealed that the pH L response to NAADP exhibited aspects that were both similar to as well as different from those induced by sperm. The spatial aspects of the pH L ASH were clearly well preserved and sometimes pro-  ceeded in a quasi-wave-like manner akin to fertilization (data not shown). The quantitative differences we observed between NAADP-and sperm-induced responses suggest that NAADP generation is limiting during fertilization; the onset of the NAADP-induced response occurred after much a shorter delay compared with that to sperm (5.0 Ϯ 0.5 s after NAADP injection versus ϳ35 s delay after insemination; Fig. 2C) and was of a much faster upstroke (time to peak, seconds: sperm, 31 Ϯ 1; NAADP, 7.6 Ϯ 0.5). The slower fertilization pH L ASH kinetics are entirely consistent with the time taken for the egg to generate NAADP. In summary, NAADP alone is sufficient to generate a pH L ASH because of the resident cellular architecture.

DISCUSSION
The release of Ca 2ϩ from internal stores has long been recognized as the primary drive for the fertilization-induced Ca 2ϩ signal (2) as well as other mammalian processes (1), but the involvement of non-endoplasmic reticulum Ca 2ϩ stores was underscored when NAADP was shown to evoke Ca 2ϩ release from acidic, lysosome-related organelles in both sea urchin egg and mammalian cells (1,7). In sea urchin egg, NAADP-induced Ca 2ϩ release seems to contribute to all phases of the fertilization Ca 2ϩ spike from the initial cortical flash through to the falling phase of the main Ca 2ϩ transient (5). Recently we demonstrated that the pH L of these acidic Ca 2ϩ stores is increased by NAADP in sea urchin egg homogenate by a mechanism coupled to Ca 2ϩ release via the NAADP receptor on acidic Ca 2ϩ stores (8). Interestingly, this pH L change is not secondary to an increase in [Ca 2ϩ ] at the cytosolic face, nor can it be mimicked by Ca 2ϩ release from the endoplasmic reticulum but rather is a direct effect of NAADP upon the vesicles themselves (8). However, this is not at the level of the H ϩ -ATPase or the vesicle membrane potential but may be due to changes in luminal H ϩ buffering (8). In view of this, we have investigated NAADP-dependent pH L events under physiological conditions, i.e. at fertilization.
The pH L of acidic vesicles in unfertilized eggs has been reported to be approximately pH 5.5 for both cortical granules (14) and yolk platelets (18), and it has been proposed that the pH within these vesicles serves to regulate luminal enzyme activity (14, 19 -21) and/or vesicular integrity (18). That the pH L within egg acidic vesicles can slowly acidify has been suggested during oogenesis (11,20,21) and post-fertilization (10), but with the exception of the rapid alkalinization of cortical granules upon fusion and exocytosis (12,14), rapid increases in the pH L of acidic vesicles have not been reported.
Corroborating our egg homogenate studies (8), the luminal pH of acidic organelles in intact eggs is coupled to both fertilization and NAADP-induced Ca 2ϩ release with a prompt alkalinization. We would argue that our assays of pH L are reliable because similar results were ratiometrically recorded with both acridine orange, which has been well characterized in this system (8,10), as well as with Lysosensor Yellow/Blue DND-160. Importantly, pH L changes were clearly independent of exocytosis/endocytosis and cytosolic pH changes. Moreover, acridine orange is unlikely to be responding to the oxidative burst at fertilization because this initiates much later, 2-3 min after insemination (19,(22)(23)(24)(25), and direct addition of H 2 O 2 (10 M to 1 mM) to eggs did not induce a pH L ASH (data not shown). Taken together, our data strongly support an alkalinization of cortical vesicles.
Spatial Characterization of the pH L Response-Fertilization evoked a spatially heterogeneous change in organellar pH, with a rapid, cortical alkalinization followed by a central, slow acidification. Only the latter was previously detected in asynchronous populations of eggs (10), hardly surprising when the transient pH L ASH only extends 15-25 m into the cortex (in an egg ϳ120 m in diameter). Moreover, alkalinization propagated as a wave around the egg cortex at ϳ5 m/s, a velocity similar to the main [Ca 2ϩ ] i wave and the exocytotic lifting of the fertilization envelope (26,27). At higher magnification, the alkalinization wave was manifest as the summation of smaller events (pH L ARES, 2-3 m in diameter) that appear to represent alkalinization at the individual vesicle level.
From which vesicles these pH L ARES emanate is presently unclear. From electron micrographs the 1-3-m diameter yolk platelets (20,28,29) and other candidate "acidic vesicles" (28) correlate well with the size of a pH L ARES, particularly if one allows for diffusion of acridine orange lost from the vesicle upon alkalinization. Deeper exocytotic cortical vesicle classes have been described (30) and the so-called "basal laminar" and "apical" vesicles are both in the right place to host the pH L ARES. Defining the vesicle population will be crucial to understanding the ramifications of this novel phenomenon.
Why the fertilization pH L ASH is restricted to the egg cortex is not entirely clear but might reflect either localized messenger accumulation or a unique property of the acidic stores themselves. Our data are more consistent with the latter because global NAADP microinjection evoked a local pH L ASH, and it is known that the cortical Ca 2ϩ stores are the most sensitive to NAADP (31,32). Nonetheless, we cannot formally exclude local messenger production as a contributory factor at fertilization.
Temporal Characterization of the pH L ash-Despite the spatial similarity of the pH L ASH and Ca 2ϩ cortical flash, there was no temporal overlap, and the pH L ASH occurred with or just after the main Ca 2ϩ wave, ϳ35 s later. However, this timing makes sense because it coincides with NAADP-induced Ca 2ϩ release from intracellular stores (5). Indeed, both the pH L ASH and main Ca 2ϩ response propagate as waves, which is additional evidence for their being interrelated phenomena. Note that the difference in the upstroke kinetics of the Ca 2ϩ and pH L changes in intact eggs is consistent with those observed in homogenate (8).
We were concerned that the pH L ASH may be related to Ca 2ϩstimulated exocytosis, which occurs  Similarly, in the egg schematic a rainbow scheme reflects the concentration of Ca 2ϩ . Cortical granules and acidic vesicles are red when acidic, and green when alkalinized during the pH L ASH. The red arrows correspond to Ca 2ϩ influx across the plasma membrane (the cortical flash) or release from acidic stores through NAADP receptors (NAADP R).
6 -8 s after the main [Ca 2ϩ ] i rise in urchin eggs (13,33) and which alkalinizes cortical granules upon plasmalemmal fusion (12,14). Our conclusion is that the pH L ASH is not due to exocytosis based on the following evidence: (a) Deeper pH L events are too remote to be due to fusion with the plasma membrane, although it might be argued that these deeper pH L ARES artifactually represents the "tail" of exocytotic events emanating from out of the plane of focus, recording from thin optical slices (2-3% of the egg diameter) at the egg equator means that the curvature of the plasma membrane is minimal, and out-of-focus cortical granules will not overlap this in-focus region. In addition, the amplitude and upstroke kinetics of the pH L ASH are well maintained over 15-25 m, which is inconsistent with the tail of out-of-focus events, which would be blunted; (b) Exocytosis and pH L changes can be divorced, at the level of individual eggs, as well as by mastoparan, which stimulated cortical granule exocytosis but no pH L ASH. Therefore, the approximate post-fertilization sequence of events is Ca 2ϩ cortical flash Ͼ Ͼ main Ca 2ϩ wave Ϸ pH L ASH Ͼ exocytosis (Fig. 7).
Not only do we describe a novel fertilization pH L event, but we also offer a mechanism whereby sperm elevate NAADP levels in the egg (5), which in turn raise pH L via activation of the Ca 2ϩ -mobilizing NAADP receptor on acidic vesicles (8) (Fig.  7). Moreover, we reveal that the unique spatial profile of the pH L ASH reflects the intrinsic nature of the cortical stores themselves rather than some upstream mechanism. We also conclude that vesicle alkalinization is independent of changes in cytosolic Ca 2ϩ , vesicular membrane potential, H ϩ -ATPase activity (8), cytosolic pH, and exocytosis and appears to be at the level of NAADP-induced changes in H ϩ buffering (8).
The importance of the egg cortex for egg function is underscored by many cell biological aspects (reviewed in Ref. 34). In addition to supporting layers of exocytotic vesicle populations (30), it also hosts the [Ca 2ϩ ] i cortical flash (4, 5), a unique tubular endoplasmic reticulum morphology with an exclusive complement of luminal Ca 2ϩ -binding proteins (35,36), as well as a greater density of ryanodine receptors (37). To this and other properties, we add the fertilization-dependent alkalinization of NAADP-sensitive acidic Ca 2ϩ stores. The regulation of organelle pH by activation of the NAADP receptor on the vesicle membrane poses exciting questions relating not just to sea urchin biology but also to mammalian biology. Certainly, key enzymes in the lumen of egg vesicles are regulated by pH changes at fertilization (14,38), but in the acidic organelles of mammalian cells, luminal pH influences processes as diverse as secretory vesicle fusion (39), autophagy (40), and proteolysis (41). Our results may have far-reaching implications for NAADP regulating various cellular pathways via pH and not just by Ca 2ϩ .