Effects of Adenophostin-A and Inositol-1,4,5-trisphosphate on Cl− Currents in Xenopus laevis Oocytes

  1. H. Criss Hartzell,
  2. Khaled Machaca and
  3. Yoshiyuki Hirayama
  1. Department of Anatomy and Cell Biology, Emory University School of Medicine, Atlanta, Georgia 30322-3030
     
    Next Section

    Abstract

    Adenophostin-A, a novel compound isolated from cultures ofPenicillium brevicompactum, has been shown to stimulate Ca2+ release from inositol-1,4,5-trisphosphate (IP3)-sensitive Ca2+ stores in microsomal preparations, permeabilized cells, and lipid vesicles containing purified IP3 receptor. The purpose of the current study was to compare the effects of adenophostin-A and IP3 on Ca2+ release from stores and Ca2+ influx in intact Xenopus laevis oocytes. Ca2+ influx though store-operated Ca2+ channels and Ca2+release from stores were monitored by measuring two Ca2+-activated Cl− currents that can be used as real-time indicators of Ca2+ release and Ca2+ influx (ICl-1 and ICl-2, respectively). We find that high concentrations (final intraoocyte concentrations of 5–10 μm) of adenophostin-A and IP3 stimulate a large Ca2+ release from stores (as measured by ICl-1) followed by Ca2+ influx (as measured by ICl-2). Low concentrations (∼50 nm) of IP3 stimulate oscillations in Ca2+ release without stimulating Ca2+ influx. In contrast, low concentrations of adenophostin-A can stimulate Ca2+ influx without stimulating a large Ca2+ release. However, Ca2+ influx did not occur in the complete absence of Ca2+ release. Therefore, it is unlikely that adenophostin-A directly stimulates store-operated Ca2+ channels. We hypothesize that adenophostin-A releases Ca2+ from a subpopulation of stores that is tightly coupled to store-operated Ca2+ channels.

    The concentration of cytosolic free Ca2+ regulates many physiological processes as diverse as fertilization and programmed cell death. One of the key pathways that controls the level of cytosolic free Ca2+ involves G protein-coupled and tyrosine kinase-coupled receptor stimulation of phospholipase C, production of IP3, and the release of Ca2+ from internal stores (1-4). Release of Ca2+ from internal stores is often followed by a sustained influx of extracellular Ca2+ (5-8). This influx [capacitative Ca2+ entry (9)] is mediated by SOCCs in the plasmalemma that are apparently controlled by the level of Ca2+ in the internal store.

    Although IP3 is a very potent stimulator of Ca2+ release from internal stores (ED50 ∼ 200 nm), it has recently been reported that a structurally different compound, adenophostin-A, is ∼100-fold more potent than IP3 in releasing Ca2+ from internal stores (10). Adenophostin-A, isolated from the broth of cultures ofPenicillium brevicompactum, is 2′-AMP linked through its 3′-hydroxyl to glucose-3,4-diphosphate. It has been proposed that the 3- and 4-phosphates on the glucose ring of adenophostin-A assume the same role as the 4- and 5-phosphates in IP3 (10). Consistent with this idea is the finding that 2-hydroxyethyl-α-d-glucopyranoside-2,3′,4′-trisphosphate is also capable of binding to the IP3 receptor and releasing Ca2+ from internal stores, although with ∼1000-fold lower potency than adenophostin-A (11). Although the effects of adenophostin-A on Ca2+ release have been demonstrated in microsomal preparations (10), permeabilized cells (10), and purified reconstituted IP3 receptors (12), the effects of adenophostin-A in intact cells have not been investigated. In the current study, we examined the effects of adenophostin-A injected intoXenopus laevis oocytes and compared its effects with those of IP3.

    X. laevis oocytes are a very useful model system for studying Ca2+ signaling, in part because they express Ca2+-activated Cl− channels that can be used as real-time indicators of cytosolic Ca2+ concentration (13) and in part because their large size facilitates the study of spatial and temporal changes in cytosolic Ca2+concentrations (14). We have recently described two distinct Ca2+-activated Cl− currents in X. laevis oocytes whose activation depends on the source of Ca2+: ICl-2 is activated only by Ca2+ influx through SOCCs, and ICl-1 can be activated both by Ca2+ influx and by Ca2+release from internal stores, depending on the voltage protocol used (15). The purpose of the current study was to use these two currents to compare the effects of intracellular injection of IP3 and adenophostin-A on Ca2+ release from internal stores and the subsequent capacitative Ca2+ influx through SOCCs.

    Previous SectionNext Section

    Materials and Methods

    Electrophysiological methods.

    X. laevis oocytes (stage V–VI) were isolated according to the method of Dascal (13) and voltage-clamped with two microelectrodes filled with 3 mKCl (1–2 MΩ) as described previously (15). Typically, the membrane was held at −35 mV, and voltage steps were applied as described in the text. Stimulation and data acquisition were controlled by pCLAMP 6.01 (Axon Instruments, Burlingame, CA) via a Digidata 1200 A/D-D/A converter (Axon Instruments) and a Gateway P5–90 computer (Intel Pentium, 90 MHz). During recording, the oocyte was superfused with normal Ringer’s solution at a rate of 2 ml/min (∼300-μl chamber). Experiments were performed at room temperature (22–26°).

    Microinjection.

    Oocytes were injected with IP3or adenophostin-A using a Drummond Nanoject Automatic Oocyte Injector (Broomall, PA). The injection pipette was pulled from glass capillary tubing in a manner similar to the recording electrodes and then broken so that it had a beveled tip with an inside diameter of 10–20 μm. The final concentrations ([X]calc) of injected solutions in the oocyte were calculated assuming an oocyte volume of 1 μl and uniform distribution of the solute in the oocyte. The figures show the pipette concentrations and volumes of solutions injected.

    Solutions.

    Normal Ringer’s solution consisted of 123 mm NaCl, 2.5 mm KCl, 1.8 mmCaCl2, 1.8 mm MgCl2, and 10 mm HEPES, pH 7.4. Zero-Ca2+ Ringer’s was the same except CaCl2 was omitted, MgCl2 was increased to 5 mm, and 0.1 mm EGTA was added. Stock solutions of IP3 and adenophostin-A were made at 10 mm in H2O, stored at −20°, and diluted in water to the final concentrations indicated for injection. In all cases, injection of the same volume of water had no effect on the Cl− currents. Adenophostin-A was the generous gift of Drs. M. Takahashi, S. Takahashi, and K. Tanzawa (Sankyo Co., Ltd., Tokyo, Japan).

    Previous SectionNext Section

    Results

    Effects of IP3 and adenophostin-A at a constant membrane potential of −100 mV.

    Initially, we examined the effects of IP3 and adenophostin-A on the membrane current recorded at a constant holding potential of −100 mV (Fig. 1). When 13 nl of a 1 mm solution of IP3 was injected into an X. laevis oocyte, an inward current developed. The current was biphasic: the initial component peaked in ∼30 sec and was followed by a slowly developing current that took ∼10 min to develop fully (Fig. 1A). A similar result was obtained when the oocyte was injected with 10 nl of a 1 mm solution of adenophostin-A (Fig. 1B). These concentrations of drug produced calculated drug concentrations in the oocyte of 5–13 μm(assuming an oocyte volume of 1 μl), which would be expected to be supramaximal for stimulating Ca2+ release from IP3-sensitive stores (4, 10). In contrast, when the oocyte was injected with 200-fold less adenophostin-A (intraoocyte concentration ∼50 nm), the transient phase was not detectable and only the slow phase was present (Fig. 1C). The slow phase of the currents evoked by IP3 or adenophostin-A were abolished by removal of extracellular Ca2+ (not shown). This confirms the results of other studies (15-20) that have shown that the transient phase of the inward current produced by IP3 injection corresponds to a Cl− current that is activated by Ca2+ released from internal stores and the slowly developing phase is a Cl− current that requires Ca2+ influx from the extracellular space. The data of Fig.1C, therefore, suggest that low concentrations of adenophostin-A might stimulate Ca2+ influx without first causing a substantial release of Ca2+ from internal stores.

    Figure 1
    View larger version:
    Figure 1

    Effects of injection of IP3 and adenophostin in X. laevis oocytes at −100 mV. Oocytes were held at −100 mV and (A) 13 nl of 1 mm IP3(calculated IP3 concentration = 13 μm), (B) 10 nl of 1 mm adenophostin-A ([adenophostin-A]CALC = 10 μm), or (C) 4.6 nl of 10 μm adenophostin-A ([adenophostin-A]CALC = 46 nm) was injected at the time indicated (arrows).

    Although the voltage protocol used in Fig. 1 has been a standard approach for investigating Ca2+ release and influx inX. laevis oocytes, it is not very sensitive for detecting Ca2+ release from stores because the transient current (ICl-1) that is stimulated by Ca2+ released from stores is not significantly activated at negative potentials (15). At −100 mV, ICl-1 is a rather insensitive indicator of Ca2+ release from stores (15).

    Effects of IP3 on ICl-1 and ICl-2

    To examine this possibility more closely, we compared the effects of IP3 and adenophostin-A on ICl-1 and ICl-2 as we have previously described (15). Fig. 2 shows the effect of injection of large concentrations of IP3 into an oocyte. The potential of the oocyte was repetitively stepped from a holding potential of −35 mV to +40 mV for 1 sec and then to −120 mV for 1 sec. The current at the end of the +40 mV pulse was taken as a measure of ICl-1, and the current at the end of the −120 mV pulse was taken as a measure of ICl-2. We have previously shown that under these conditions, ICl-1 is an indicator of Ca2+released from stores and ICl-2 is an indicator of Ca2+ influx (15). Injection of 4.6 nl of 10 mmIP3 into an oocyte caused an immediate but transient increase in ICl-1 (Fig. 2, A and B). This increase peaked in ∼1 min and declined back to base-line in ∼2 min. As ICl-1 at +40 mV declined, ICl-2 at −120 mV began to increase and reached a maximum in ∼10 min. The time courses of ICl-1 and ICl-2 paralleled the time courses of the transient and slow phases, respectively, of the current recorded in Fig. 1. A qualitatively similar result was obtained when a 100-fold lower amount of IP3 was injected (Fig. 2, C and D).

    Figure 2
    View larger version:
    Figure 2

    Effects of high concentrations of IP3on two Cl− currents. Oocytes were voltage-clamped and repeatedly stepped from −35 mV to +40 mV for 1 sec and then to −120 mV for 1 sec (D, top trace). The current at the end of the +40-mV pulse was taken as ICl-1, and the current at the end of the −120-mV pulse was taken as ICl-2. A and C, Plot of the change in (○) ICl-1 and (▵) ICl-2 in response to (A) 4.6 nl of 1 mm IP3 (calculated IP3 concentration = 4.6 μm) or (C) 4.6 nl of 0.1 mm IP3 (calculated IP3concentration = 460 nm). B and D, Selected traces corresponding to the plots in A and C, respectively.a–c, Times in A–C at which the traces were selected.

    In contrast, when very low concentrations of IP3 were injected into the oocyte, a different result was obtained (Fig.3). In the experiment illustrated, the oocyte was injected three times with 10 μm IP3. The first injection of 10 nl of 10 μm IP3produced a small ICl-1 and no ICl-2. The second injection of the same amount produced a somewhat larger ICl-1 that oscillated in amplitude for several minutes before it declined to base-line. No ICl-2 was detected. The third injection of 23 nl produced an even larger ICl-1. ICl-1 reached a maximum after ∼4 min but then began to oscillate and declined to base-line after ∼8 min. In response to this injection, ICl-2 was stimulated. ICl-2 began to develop ∼1 min after ICl-1 began to increase and reached a peak and declined to zero with approximately the same time course as ICl-1. The decline of ICl-2 coincided with the onset of ICl-1 oscillation. The observation that the amplitudes of ICl-1 and ICl-2 were not linearly related suggested that release of Ca2+ from internal stores needed to reach a threshold level before Ca2+ influx was stimulated (21). The first two injections of IP3 failed to release sufficient Ca2+ from the store to activate influx, whereas the last injection produced sufficient release to stimulate influx. We believe that the influx was transient because release was terminated as the injected IP3 was metabolically inactivated and the stores were refilled by Ca2+ influx. The length of time ICl-2 remained elevated correlated with the dose of IP3 injected. With large injections of IP3 (50 μm intraoocyte concentration), as in Fig. 2A, Ca2+ influx usually remained elevated for >1 hr, but with lower concentrations (0.5 μm) as in Fig. 2B, ICl-2 declined to base-line in ∼30 min.

    Figure 3
    View larger version:
    Figure 3

    Effects of low concentrations of IP3 on Cl− currents measured as described in legend to Fig. 2. A, Plot of (○) ICl-1 and (▵) ICl-2 as a function of time. IP3 (10 μm in the pipette) was injected at the three times indicated (arrows). B,Traces (a–c) corresponding to the times indicated in A.

    In an attempt to simplify the interpretation of these experiments, we repeated them using the metabolically stable derivatives of IP3, 2,3-dideoxy IP3 and 2-deoxy-3-fluoro IP3 (21). Injection of high concentrations of these analogs (4.6 nl of a 1 mm solution) had the same effect as injections of similarly high concentrations of native IP3(as in Fig. 2, but not shown). However, injection of 4.6 nl of 10 μm 2,3-dideoxy-IP3 invariably produced an increase in ICl-1 that then oscillated in amplitude for >20 min (Fig. 4). Under these conditions, when ICl-1 was oscillating, ICl-2 was usually not activated or was activated only transiently. Presumably, when ICl-1 was oscillating, the level of Ca2+ in the stores did not reach a sufficiently low level for a sufficiently long time to initiate the signal required to activate Ca2+influx.

    Figure 4
    View larger version:
    Figure 4

    Effects of low concentration of 2,3-dideoxy-IP3 on Cl− currents. A, Plot of (○) ICl-1 and (▵) ICl-2 as a function of time. Dideoxy-IP3 (4.6 nl, 10 μm) was injected at the time indicated (arrow) (calculated dideoxy-IP3 concentration = 46 nm). B,Traces (a–c) corresponding to the times indicated in A.

    Effects of adenophostin-A on ICl-1 and ICl-2

    Injection of high concentrations of adenophostin-A produced a similar response to that seen with injection of high concentrations of IP3 (Fig.5). It seemed that ICl-1 declined more slowly in response to adenophostin-A than in response to IP3 (compare Figs. 2 and 5), but this was not analyzed quantitatively. In contrast, injection of a low concentration of adenophostin-A produced a different response than injection of a low concentration of IP3 (Fig. 6). Fig. 6 shows an example typical of >50 oocytes, in which the injection of 4.6 nl of a 1 μm solution of adenophostin-A stimulated ICl-1 only a little, whereas ICl-2 was strongly stimulated. Compare this result (Fig. 6) with that shown in Fig. 3, in which IP3 stimulated a larger ICl-1 but no ICl-2.

    Figure 5
    View larger version:
    Figure 5

    Effect of a high concentration of adenophostin-A on Cl− currents. A, Plot of (○) ICl-1 and (▵) ICl-2 as a function of time. Adenophostin-A (4.6 nl of 1 mm) was injected at the time indicated (arrow) ([adenophostin-A]CALC = 4.6 μm). B, Traces (a–c) corresponding to the times indicated in A.

    Figure 6
    View larger version:
    Figure 6

    Effect of a low concentration of adenophostin-A on Cl− currents. A, Plot of (○) ICl-1 and (▵) ICl-2 as a function of time. Adenophostin-A (4.6 nl of 1 μm) was injected at the time indicated (arrow) ([adenophostin-A]CALC = 4.6 nm). Recording was interrupted between 10 and 14 min while other voltage protocols were being run. B, Traces(a–c) corresponding to the times indicated in A.

    To verify that the currents stimulated by adenophostin-A had the same properties as those stimulated by IP3, we characterized the adenophostin-A evoked currents in more detail. In Fig.7, we tested the dependence of ICl-1 and ICl-2 on extracellular Ca2+ . In Fig. 7A,10 nl of 100 μm adenophostin was injected at 40 sec. At ∼30 sec after ICl-1 became maximally stimulated, we switched to zero-Ca2+ Ringer’s solution. Removal of extracellular Ca2+ had little effect on ICl-1, but ICl-2 did not develop until Ca2+ was added back to the extracellular solution at ∼9 min. This shows that ICl-1 is not dependent on Ca2+ influx, whereas ICl-2 is dependent on Ca2+ influx. This confirms that the currents stimulated by adenophostin-A have the same dependence on store-released Ca2+ and influxed Ca2+ as the currents stimulated by IP3 (15). Furthermore, the adenophostin-A-stimulated currents have the same voltage-dependent activation and current-voltage relationships as the IP3-stimulated currents (data not shown).

    Figure 7
    View larger version:
    Figure 7

    Effect of zero external Ca2+ on currents evoked by adenophostin-A. A, Traces(a–e) corresponding to the times indicated in B. B, Plot of (○) ICl-1 and (▵) ICl-2 as a function of time. Adenophostin-A (10 nl of 100 μm) was injected at the time indicated (arrow) ([adenophostin-A]CALC = 1 μm). The solution was normal Ringer’s except for the period during which the solution was switched to zero-Ca2+ Ringer’s (0 Ca) (see Methods).

    To compare the responses to adenophostin-A and IP3quantitatively, we measured the maximal amplitude of ICl-1and the amplitude of ICl-2 10 min after injection of adenophostin-A or 2,3-dideoxy IP3. For these experiments, we wanted to compare concentrations of adenophostin-A and 2,3-dideoxy IP3 that produced the minimal possible stimulation of ICl-1. We determined that the minimal [adenophostin-A]CALC required to produce an effect on ICl-2 varied from oocyte to oocyte but was ∼5 nm. The minimal calculated dideoxy-IP3concentration required to produce a response was ∼25 nm. To compare the effects of low concentrations of these drugs, we began by injecting 5–10 nl of a 0.5–2 μm solution of adenophostin-A or 5–10 nl of a 1–10 μm solution of 2,3-dideoxy IP3. If the first injection did not produce a response in 2 min, a second injection was given and the final concentration of drug was calculated as the sum of the two injections. For the data shown in Fig. 8, the average calculated intraoocyte concentration of 2,3-dideoxy IP3 was 60 ± 20 nm, and the average adenophostin-A concentration was 18 ± 4 nm. The principal difference between the responses to adenophostin-A and IP3 was that adenophostin-A was much less effective than IP3 in stimulating ICl-1 but was more effective than IP3 in stimulating ICl-2. Adenophostin-A produced a ∼4-fold smaller ICl-1 (237 ± 39 nA, 25 cells, for adenophostin-A; 904 ± 184 nA, 19 cells, for 2,3-dideoxy IP3),but stimulated ICl-2 to a greater extent (−483 ± 71 nA for adenophostin-A; −298 ± 101 nA for 2,3-dideoxy IP3). Thus, the ratio of ICl-2 to ICl-1 was 5-fold greater for adenophostin-A than for IP3 (Fig. 8B).

    Figure 8
    View larger version:
    Figure 8

    Summary of the effect of low concentrations of 2,3-dideoxy IP3 and adenophostin on Cl−currents. A, The maximal amplitude of ICl-1 and the amplitude of ICl-2 at 10 min after injection of drug were measured. Error bars, standard errors. B, The ratio of ICl-2 at 10 min to the maximal ICl-1 observed in that oocyte was calculated for each individual oocyte in A.

    These results, and those of Fig. 1, suggest that adenophostin-A is capable of stimulating Ca2+ influx without depleting the Ca2+ stores to the same extent as IP3. Indeed, Fig. 1 might suggest that Ca2+ release from stores is not at all necessary to stimulate Ca2+ influx, but as we pointed out, the voltage protocol used in Fig. 1 was very insensitive for measuring ICl-1. Using the more sensitive protocol of Fig. 8, we have never seen development of ICl-2 without some stimulation of ICl-1. Injection of 5–10 nl of 0.5–1 μm adenophostin-A either had no effect on ICl-1 or ICl-2 or produced an increase in ICl-2 that was preceded by an increase in ICl-1, albeit sometimes the increase was very small (for example, Fig. 9B). These concentrations of adenophostin-A are in the range of concentrations that are effective in releasing Ca2+ via IP3 receptors (10, 12). In the range below 1 nm, adenophostin-A had no discernible effect on the Ca2+-activated Cl− currents. We wondered whether the stimulation of ICl-1 by low concentrations of adenophostin might be explained by the higher concentration of adenophostin near the tip of the injection pipette on injection. To test this possibility, we injected larger volumes (≤46 nl) of lower pipette concentration adenophostin solutions (≤50 nm). Although in a few of these injections increases in ICl-1 were not detectable at +40 mV, they were detectable at +80 mV. Furthermore, under these conditions, the development of ICl-2 was extremely slow: it often took >2 hr for ICl-2 to develop fully. This observation suggested that at low [adenophostin-A]CALC, release of Ca2+from stores was very slow but was eventually sufficient to trigger Ca2+ influx. This conclusion has recently been supported by confocal imaging of Ca2+-green-1 fluorescence.1 On injection of 9.2 nl of 0.2 μm adenophostin-A, we observed release of Ca2+ from stores. Obviously, this does not prove that Ca2+ must be released from stores for ICl-2 to be stimulated, but we have so far been unable to dissociate the two processes.

    Figure 9
    View larger version:
    Figure 9

    The effect of IP3 on Cl−currents after injection of adenophostin-A. A, Plot of (○) ICl-1 and (▵) ICl-2 as a function of time. Adenophostin-A (10 nl of 1 μm) was injected at the time indicated (first arrow) ([adenophostin-A]CALC = 10 nm). At ∼30 min later (second arrow), 4.6 nl of 1 mmIP3 was injected (calculated IP3concentration = 4.6 μm). B, Another oocyte showing a different result. Inset, On an expanded scale, the increase in ICl-1 after adenophostin-A injection.

    Effect of IP3 after adenophostin-A.

    To determine whether adenophostin-A could stimulate influx without depleting Ca2+ stores, we investigated whether Ca2+ could be released from stores by IP3 after ICl-2 had developed fully in response to a previous adenophostin-A injection. In Fig. 9A, injection of 10 nl of 1 μm adenophostin produced a ∼400 nA ICl-1 and a ∼4500 nA ICl-2, which developed over ∼35 min. After ICl-2 had plateaued, 4.6 nl of 1 mm IP3 was injected. This produced only a very small increase in ICl-1 and ICl-2. These results suggested that adenophostin-A under these conditions had either emptied the IP3-sensitive stores or had somehow inactivated the IP3 receptor. In other cells, however, a different result was obtained, as shown in Fig. 9B. In this experiment, a slightly smaller amount of adenophostin (4.6 nl of 0.5 μmadenophostin) produced a ∼10-fold smaller increase in ICl-1 (∼70 nA, inset) and a smaller ICl-2 (∼1000 nA). After ICl-2 had plateaued, 10 nl of 1 mm IP3 was injected. In this cell, the IP3 injection produced a large ICl-1 and rapidly stimulated ICl-2. Thus, in this cell, the IP3-sensitive stores were clearly not completely depleted of Ca, even though a large ICl-2 current was present in response to adenophostin-A injection. Similar results were obtained in seven other cells injected with adenophostin to give ∼5–25 nm calculated final concentration. In contrast, after a first injection of IP3 activated ICl-2, a second injection of IP3 never evoked ICl-1 as long as ICl-2 was present.

    Previous SectionNext Section

    Discussion

    These experiments demonstrate a difference in the ability of adenophostin-A and IP3 to activate two Cl−currents in X. laevis oocytes. We have previously shown that one of these Cl− currents (ICl-1) is activated by Ca2+ released from internal stores because it is absent unless the oocyte is injected with IP3 and that its activation on stepping from −35 mV to positive potentials is independent of extracellular Ca2+ but is blocked by intracellular 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (15). The other current (ICl-2) is activated by Ca2+ influx and is abolished by removal of extracellular Ca2+ (15). These currents are most likely due to different channels because ICl-1 has a linear instantaneous current-voltage relationship with an activation range at positive potentials and ICl-2 has a strongly outwardly rectifying current-voltage relationship with activation at negative potentials.

    Does adenophostin activate SOCCs without depleting stores?.

    If we assume that ICl-1 is a reliable indicator of Ca2+ release from stores (see below) and that ICl-2 is a reliable indicator of Ca2+ influx, these data suggest the possibility that adenophostin-A may be capable of activating Ca2+ influx through SOCCs without depleting Ca2+ stores. If this is true, this is very exciting because it suggests that the signal transmitted from Ca2+ stores to SOCCs may be an adenophostin-A-like compound. At the present time, the nature of the signaling pathway between stores and SOCCs remains completely unknown (8). One hypothesis for the coupling mechanism states that when store Ca2+ falls, the store releases a diffusible messenger. However, experiments supporting the existence of a diffusible messenger remain controversial (22-25). The alternative hypothesis, proposed by Berridge et al. (8, 26) hypothesizes that a “Ca2+-sensor” in the membrane of the Ca2+ store directly couples to the SOCC and opens it. This conformational coupling hypothesis is attractive because it is analogous to the dihydropyridine receptor-ryanodine receptor coupling that occurs in skeletal muscle excitation-contraction coupling, but there remains little direct evidence either in support of or against this hypothesis (8).

    The idea that adenophostin-A may activate SOCCs without depleting Ca2+ stores is supported by the finding that adenophostin-A stimulates a larger ICl-2, whereas stimulation of ICl-1 is >20-fold less compared with IP3(compare Figs. 9B and 2C). If ICl-1 is an accurate indicator of Ca2+ released from stores, then the data suggest that adenophostin-A can activate SOCCs without depleting Ca2+ stores to the extent that is necessary for IP3 to activate SOCCs. Furthermore, in cells in which ICl-2 is activated but ICl-1 is stimulated only marginally by adenophostin-A, IP3 is capable of stimulating a large ICl-1 by releasing Ca2+ from stores. Thus, it is clear that Ca2+ stores are not completely depleted of Ca2+ even though ICl-2 has developed significantly. In contrast, a second injection of IP3 at any time after ICl-2 has begun to develop in response to an initial IP3 injection is ineffective in stimulating additional Ca2+ release (as measured by ICl-1). Thus, it seems that IP3-sensitive stores must be more fully emptied of Ca2+ in response to IP3 than in response to adenophostin-A to initiate Ca2+ influx.

    ICl-1 may be an imperfect indicator of store-released Ca2+.

    There are two possible explanations of these results. The first explantation, as suggested above, is that adenophostin-A has a direct effect on SOCCs. This interpretation should be accepted with caution, however, because we have never observed the development of ICl-2 in the complete absence of stimulation of ICl-1. This observation could simply be explained if adenophostin-A has two sites of action, the IP3 receptor and the SOCC, and the dose-response curves for the two sites overlap partially. However, another possibility is that ICl-1 is an imperfect indicator of Ca2+ release from stores. For example, the response of the ICl-1 channel may depend on the rapidity with which Ca2+ is released from the store (different rates of release resulting in different local concentrations of cytosolic Ca2+) or possibly the temporal pattern of Ca2+ release from the store. Thus, if low concentrations of adenophostin-A release Ca2+ from stores slowly, it might not be revealed as an increase in ICl-1. Alternatively, because adenophostin-A binds to the IP3receptor with very high cooperativity (10, 12), it is possible that adenophostin-A stimulates a very rapid release of Ca2+ from stores that is too fast to be detected by the ICl-1channel. In support of the idea that ICl-1 may be an imperfect indicator of steady state cytosolic Ca2+concentration is the finding by Parker and Yao (27) that Cl− current amplitude correlated better with the rate of rise in the Ca2+ transient measured by Fluo-3 fluoresence than with the steady state cytosolic Ca2+ level.

    Functionally different IP3-sensitive stores.

    An alternative explanation for the differential ability of adenophostin-A and IP3 to stimulate ICl-1 and ICl-2 is that there are functionally different IP3-sensitive Ca2+ stores and that only a subset are tightly coupled to SOCCs. There is evidence in the literature that not all of the stores must be completely depleted of Ca2+ to stimulate capacitative Ca2+ entry through SOCCs. Montero et al. (28) suggested that there is a linear relationship between Ca2+ influx and the amount of Ca2+ in the store. However, there are suggestions that the relationship between “store Ca2+” and Ca2+influx may be more complex. Mathes and Thompson (29) have shown that a ∼60% reduction in store Ca2+ is sufficient to maximally activate Ca2+ influx in neuroblastoma cells. Furthermore, in X. laevis oocytes, it seems that there is no direct relationship between the level of store Ca2+ depletion and Ca2+ influx. Lupu-Meiri et al. (20) have shown that although acetylcholine or incubation of oocytes in zero-Ca2+ solution produce comparable reductions in cell Ca2+, only acetylcholine produces significant Ca2+ influx. Likewise, different inositol phosphates have differential ability to activate Ca2+ release from stores and Ca2+ influx (30). These and other data (31) suggest the possibility that there are discrete Ca2+ stores in the cell and that only one subset of the stores is coupled to Ca2+influx. If this interpretation is true, it might suggest that low concentrations of adenophostin-A are capable of stimulating Ca2+ release from a discrete Ca2+ store that is more closely associated with SOCCs, whereas IP3 may indiscriminately release Ca2+ from all IP3-sensitive stores in X. laevis oocytes.

    One attractive hypothesis is that the type-3 IP3 receptor is more tightly coupled to stores than the type-1 IP3receptor and that adenophostin-A has a higher affinity for the type-3 receptor than for the type-1 receptor. In support of this suggestion is the observation that overexpression of the type-3 IP3receptor in X. laevis oocytes increases capacitative Ca2+ entry and that this receptor is preferentially localized near the plasma membrane (32).

    Another important difference that we have observed between adenophostin-A and IP3 is that low concentrations of IP3 and IP3 analogs produce oscillations in ICl-1, whereas adenophostin does not. Oscillations in Cl− currents produced by IP3 injection have been described by other investigators (33-36). The oscillations of ICl-1 parallel the Ca2+ waves produced by injection of oocytes with low concentrations of IP3 analogs (37, 38) and are probably related to the bell-shaped Ca2+dependence of the IP3 receptor (39-41). Thus, as IP3 releases Ca2+ from the store, the high concentration of cytosolic Ca2+ inhibits the action of IP3. As the cytosolic Ca2+ is lowered by uptake into the endoplasmic reticulum by Ca2+-ATPases, the inhibition is relieved and IP3 can act again to stimulate release (42). The lack of oscillations with low concentrations of adenophostin suggests that Ca2+ may not modulate the action of adenophostin-A as it modulates the effect of IP3. An attractive hypothesis is that adenophostin-A may act on a subtype of IP3 receptor that is modulated in a different way by Ca2+ than is the type-1 receptor. In this regard, Mikoshibaet al. (43) recently showed that the type-1 and type-3 IP3 receptors are regulated differently by Ca2+. Alternatively, the inactivation of the type-1 IP3receptor may depend on the agonist that activates it. If this is true, it suggests that inactivation may not be solely due to the increase by IP3 of the access of Ca2+ to an inhibitory binding site on the cytoplasmic surface of the IP3 receptor (44).

    Other interpretations.

    Another difference between adenophostin and IP3 that should be considered is that adenophostin may be much more metabolically stable than IP3. Although the metabolic pathways that use adenophostin as substrate and the products of adenophostin metabolism have not been characterized, it is known that IP3 is relatively rapidly metabolized into other inositol phosphates that may have their own biological actions. Thus, the products of IP3 metabolism could have complex effects that could explain the differences in responsiveness of ICl-1 and ICl-2 to adenophostin and IP3. For example, it has been shown that inositol-3,4,5,6-tetrakisphosphate is capable of inhibiting Ca2+-activated Cl− channels in colonic and intestinal epithelial cells (45, 46). If inositol-3,4,5,6-tetrakisphosphate or other inositol phosphates were to have effects not shared by adenophostin-A on Cl− channels, this could confound the interpretation of the results described here. One possible scenario would be that a metabolite of IP3selectively inhibits the ICl-2 channel because of the time required for metabolic conversion of the injected IP3. The strongest argument against such a mechanism is shown in Fig. 9. This hypothesis would predict that IP3 injection after ICl-2 had developed in response to adenophostin injection should produce an inhibition of the current. However, although the IP3 injection does sometimes produce a very small and transient inhibition of ICl-2, this negative effect is too small and too short-lived to explain the differences we have described in the responses to IP3 and adenophostin-A. Furthermore, we see a similar difference between adenophostin-A and two slowly metabolized analogs of IP3, 2,3-dideoxy IP3 and 2-deoxy-3-fluoro IP3.

    Another consequence of the difference in the metabolic stability of adenophostin-A and IP3 might be spatial differences in the spread of the drugs through the cell after injection. For example, IP3 may not diffuse throughout the cell before it is metabolized, whereas adenophostin-A might be able to diffuse a longer distance before it is inactivated. If so, compared with IP3, adenophostin-A might deplete a larger fraction of the stores, which would activate more SOCCs, and activate more ICl-2. Although this could theoretically explain why adenophostin-A activates less ICl-1 and more ICl-2 than does IP3, it is does not explain why IP3 releases Ca2+ from stores that activate ICl-1 and adenophostin-A does not (unless one assumes that adenophostin-A-sensitive stores are located farther from the membrane than IP3-sensitive stores).

    Summary.

    The results provide interesting suggestions regarding the possible mechanisms of action of adenophostin-A. However, distinguishing between the possibilities discussed above will require imaging changes in store Ca2+ and cytosolic Ca2+ using fluorescent or luminescent Ca2+probes and correlating these changes with the currents we have recorded here. These studies are currently in progress.

    Previous SectionNext Section

    Acknowledgments

    We would like to thank Drs. M. Takahashi, S. Takahashi, and K. Tanzawa (Sankyo, Tokyo, Japan) for their generous gift of adenophostin-A. We would also like to thank Dr. Lynne Quarmby for helpful discussions and comments on the manuscript and Amber Rinderknecht for technical assistance.

    Previous SectionNext Section

    Footnotes

    • Send reprint requests to: Dr. H. Criss Hartzell, Department of Anatomy and Cell Biology, Room 333, 1648 Pierce Drive, Emory University School of Medicine, Atlanta, GA 30322-3030. E-mail:criss{at}anatomy.emory.edu

    • ↵1 K. Machaca and H. C. Hartzell. Manuscript in preparation.

    • This work was supported by National Institutes of Health Grants HL54074 and GM55276.

    • Abbreviations:
      IP3
      inositol-1,4,5-trisphosphate
      SOCC
      store-operated Ca2+channel
      EGTA
      ethylene glycol bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid
      HEPES
      4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
      [adenophostin-A]CALC
      calculated adenophostin-A concentration
      ICl-1 and ICl-2
      Ca2+-activated Cl− currents
      • Received October 23, 1996.
      • Accepted December 20, 1996.
    Previous Section
     

    References

    1. ↵
      1. Putney J. W.
      1. Rhee S. G.,
      2. Choi K. D.
      (1992) Multiple forms of phopholipase C isozymes and their activation mechanisms. in Advances in Second Messenger and Phosphoprotein Research, ed Putney J. W. (Raven Press, New York), pp 35–61.
    2. ↵
      1. Berridge M. J.,
      2. Irvine R. F.
      (1989) Inositol phosphates and cell signaling. Nature (Lond.) 341:197–205.
      CrossRefMedline
    3. ↵
      1. Rana R. S.,
      2. Hokin L. E.
      (1990) Role of phosphoinositides in transmembrane signaling. Physiol. Rev. 70:115–164.
      FREE Full Text
    4. ↵
      1. Berridge M. J.
      (1993) Inositol trisphosphate and calcium signaling. Nature (Lond.) 361:315–325.
      CrossRefMedline
    5. ↵
      1. Fasolato C.,
      2. Innocenti B.,
      3. Pozzan T.
      (1994) Receptor-activated Ca2+ influx: how many mechanisms for how many channels? Trends. Pharmacol. Sci. 15:77–83.
      CrossRefMedline
    6. ↵
      1. Meldolesi J.,
      2. Clementi E.,
      3. Fasolato C.,
      4. Zacchetti D.,
      5. Pozzan T.
      (1991) Ca2+ influx following receptor activation. Trends. Pharmacol. Sci. 12:289–292.
      CrossRefMedline
    7. ↵
      1. Putney J., Jr.
      (1992) Inositol phosphates and calcium entry. Adv. Second Messenger Phosphoprotein Res. 26:143–160.
      Medline
    8. ↵
      1. Berridge M. J.
      (1995) Capacitative calcium entry. Biochem. J. 312:1–11.
    9. ↵
      1. Putney J. W.
      (1990) Capacitative calcium entry revisited. Cell Calcium 11:611–624.
      CrossRefMedlineWeb of Science
    10. ↵
      1. Takahashi M.,
      2. Tanzawa K.,
      3. Takahashi S.
      (1994) Adenophostins, newly discovered metabolites of Penicillium brevicompactum, act as potent agonists of the inositol 1,4,5-trisphosphate receptor. J. Biol. Chem. 269:369–372.
      Abstract/FREE Full Text
    11. ↵
      1. Wilcox R. A.,
      2. Erneux C.,
      3. Primrose W. U.,
      4. Gigg R.,
      5. Nahorski S. R.
      (1995) 2-Hydroxyethyl-a-d-glucopyranoside-2,3′,4′-trisphosphate, a novel, metabolically resistant, adenophostin A and myo-inositol-1,4,5-trisphosphate analogue, potently interacts with the myo-inositol-1,4,5-trisphosphate receptor. Mol. Pharmacol. 47:1204–1211.
      Abstract
    12. ↵
      1. Hirota J.,
      2. Michikawa T.,
      3. Miyawaki A.,
      4. Takahashi M.,
      5. Tanzawa K.,
      6. Okura I.,
      7. Furuichi T.,
      8. Mikoshiba K.
      (1995) Adenophostin-mediated quantal Ca2+ release in the purified and reconstituted inositol 1,4,5-trisphosphate receptor type 1. FEBS Lett. 368:248–252.
      CrossRefMedlineWeb of Science
    13. ↵
      1. Dascal N.
      (1987) The use of Xenopus oocytes for the study of ion channels. CRC Crit. Rev. Biochem. 22:317–387.
      MedlineWeb of Science
    14. ↵
      1. Lechleiter J. D.,
      2. Clapham D. E.
      (1992) Molecular mechanisms of intracellular calcium excitability in X. laevis oocytes. Cell 69:283–294.
      CrossRefMedlineWeb of Science
    15. ↵
      1. Hartzell H. C.
      (1996) Activation of different Cl currents in Xenopus oocytes by Ca liberated from stores and by capacitative Ca influx. J. Gen. Physiol. 108:157–175.
      Abstract/FREE Full Text
    16. ↵
      1. Snyder P. M.,
      2. Krause K.,
      3. Welsh M. J.
      (1988) Inositol triphosphate isomers, but not inositol 1,3,4,5-tetrakisphosphate, induce calcium influx in Xenopus laevis oocytes. J. Biol. Chem. 263:11048–11051.
      Abstract/FREE Full Text
    17. ↵
      1. Parker I.,
      2. Miledi R.
      (1987) Inositol trisphosphate activates a voltage-dependent calcium influx in Xenopus oocytes. Proc. R. Soc. Lond. 231:27–36.
      Medline
    18. ↵
      DeLisle, S., D. Pittet, B. V. L. Potter, P. D. Lew, and M. J. Welsh. InsP3 and Ins(1,3,4,5)P4 act in synergy to stimulate influx of extracellular Ca2+ in Xenopus oocytes.Am. J. Physiol. C1456–C1463, 1992..
    19. ↵
      1. Petersen C. C. H.,
      2. Berridge M. J.
      (1994) The regulation of capacitative calcium entry by calcium and protein kinase C in Xenopus oocytes. J. Biol. Chem. 269:32246–32253.
      Abstract/FREE Full Text
    20. ↵
      1. Lupu-Meiri M.,
      2. Beit-Or A.,
      3. Christensen S. B.,
      4. Oron Y.
      (1993) Calcium entry in Xenopus oocytes: effects of inositol trisphosphate, thapsigargin and DMSO. Cell Calcium 14:101–110.
      CrossRefMedline
    21. ↵
      1. Parker I.,
      2. Miledi R.
      (1989) Nonlinearity and facilitation in phosphoinositide signaling studied by the use of caged inositol trisphosphate in Xenopus oocytes. J. Neurosci. 9:4068–4077.
      Abstract
    22. ↵
      1. Parekh A. B.,
      2. Terlau H.,
      3. Stuhmer W.
      (1993) Depletion of InsP3 stores activates a Ca and K current by means of a phosphatase and a diffusible messenger. Nature (Lond.) 364:814–818.
      CrossRefMedline
    23. ↵
      1. Randriamampita C.,
      2. Tsien R. Y.
      (1993) Emptying of intracellular Ca stores releases a novel small messenger that stimulates Ca influx. Nature (Lond.) 364:809–814.
      CrossRefMedline
    24. ↵
      1. Thomas D.,
      2. Kim H. Y.,
      3. Hanley M. R.
      (1996) Regulation of inositol trisphosphate-induced membrane currents in Xenopus oocytes by a Jurkat cell calcium influx factor. Biochem. J. 318:649–656.
    25. ↵
      1. Gilon P.,
      2. Bird G. S. J.,
      3. Bian X.,
      4. Yakel J. L.,
      5. Putney J. W.
      (1995) The Ca mobilizing actions of Jurkat extracts on mammalian cells and Xenopus laevis oocytes. J. Biol. Chem. 270:8050–8055.
      Abstract/FREE Full Text
    26. ↵
      1. Petersen C. C. H.,
      2. Berridge M. J.
      (1996) Capacitative calcium entry is colocalised with calcium release in Xenopus oocytes: evidence against a highly diffusible calcium influx factor. Pflueg. Arch. Eur. J. Physiol. 431:286–292.
    27. ↵
      1. Parker I.,
      2. Yao Y.
      (1994) Relation between intracellular Ca2+ signals and Ca2+-activated Cl− current in Xenopus oocytes. Cell Calcium 15:276–288.
      CrossRefMedlineWeb of Science
    28. ↵
      1. Montero M.,
      2. Alvarez J.,
      3. Garcia-Sancho J.
      (1991) Agonist-induced Ca2+ influx in human neutrophils is secondary to the emptying of intracellular calcium stores. Biochem. J. 277:73–79.
    29. ↵
      1. Mathes C.,
      2. Thompson S. T.
      (1995) The relationship between depletion of intracellular Ca stores and activation of Ca current by muscarinic receptors in neuroblastoma cells. J. Gen. Physiol. 106:975–993.
      Abstract/FREE Full Text
    30. ↵
      1. Lupu-Meiri M.,
      2. Lipinsky D.,
      3. Ozaki S.,
      4. Watanabe Y.,
      5. Oron Y.
      (1994) Independent external calcium entry and cellular calcium mobilization in Xenopus oocytes. Cell Calcium 16:20–24.
      Medline
    31. ↵
      1. Shapira H.,
      2. Lupu-Meiri M.,
      3. Lipinisky D.,
      4. Oron Y.
      (1996) Agonist-evoked calcium efflux from a functionally discrete compartment in Xenopus oocytes. Cell Calcium 19:201–210.
      CrossRefMedlineWeb of Science
    32. ↵
      1. DeLisle S.,
      2. Blondel O.,
      3. Longo F. J.,
      4. Schnabel W. E.,
      5. Bell G. I.,
      6. Welsh M. J.
      (1996) Expression of inositol 1,4,5-trisphosphate receptors changes the Ca2+ signal of Xenopus oocytes. Am. J. Physiol. 270:C1255–C1261.
      Abstract/FREE Full Text
    33. ↵
      1. Berridge M. J.
      (1988) Inositol triphosphate-induced membrane potential oscillations in Xenopus oocytes. J. Physiol. 403:589–599.
      Abstract/FREE Full Text
    34. ↵
      1. Parker I.,
      2. Miledi R.
      (1986) Changes in intracellular calcium and in membrane currents evoked by injection of inositol trisphosphate into Xenopus oocytes. Proc. R. Soc. Lond. 228:307–315.
      Medline
    35. ↵
      1. Oron Y.,
      2. Dascal N.,
      3. Nadler E.,
      4. Lupu M.
      (1985) Inositol 1,4,5-trisphosphate mimics muscarinic response in Xenopus oocytes. Nature (Lond.) 313:141–143.
      CrossRefMedline
    36. ↵
      1. DeLisle S.,
      2. Mayr G. W.,
      3. Welsh M. J.
      (1995) Inositol phosphate structural requisites for Ca2+ influx. Am. J. Physiol. 268:C1485–C1491.
      Abstract/FREE Full Text
    37. ↵
      1. Lechleiter J.,
      2. Girard S.,
      3. Peralta E.,
      4. Clapham D.
      (1991) Sprial calcium wave propagation and annihilation in Xenopus laevis oocytes. Science (Washington D. C.) 252:123–126.
      Abstract/FREE Full Text
    38. ↵
      1. Parker I.,
      2. Yao Y.
      (1991) Regenerative release of calcium from functionally discrete subcellular stores by inositol trisphosphate. Proc. R. Soc. Lond. 246:269–274.
      Medline
    39. ↵
      1. Iino M.
      (1990) Biphasic Ca2+-dependence of inositol 1,4,5-trisphosphate-induced Ca release in smooth muscle cells of the guinea pig Taenia caeci. J. Gen. Physiol. 95:1103–1122.
      Abstract/FREE Full Text
    40. ↵
      1. Bezprozvanny I.,
      2. Watras J.,
      3. Ehrlich B. E.
      (1991) Bell-shaped calcium-response curves of Ins(1,4,5)P3 and calcium-gated channels from endoplasmic reticulum of cerebellum. Nature (Lond.) 351:751–754.
      CrossRefMedline
    41. ↵
      1. Finch E. A.,
      2. Turner T. J.,
      3. Goldin S. M.
      (1991) Calcium as a co-agonist of inositol 1,4,5-trisphosphate-induced calcium release. Science (Washington D. C.) 252:443–446.
      Abstract/FREE Full Text
    42. ↵
      1. Atri A.,
      2. Amundson J.,
      3. Clapham D.,
      4. Sneyd J.
      (1993) A single-pool model for intracellular calcium oscillations and waves in the Xenopus oocyte. Biophys. J. 65:1727–1739.
      MedlineWeb of Science
    43. ↵
      Yoneshima, H., Miyawaki, A., Michikawa, T., Furuichi, T., and K. Mikoshiba. Ca2+ differentially regulates the ligand affinity states of type-1 and type-3 inositol 1,4,5-trisphosphate receptors. Biophys. J., in press..
    44. ↵
      1. Ehrlich B. E.
      (1995) Functional properties of intracellular calcium-release channels. Curr. Opin. Neurobiol. 5:304–309.
      CrossRefMedlineWeb of Science
    45. ↵
      1. Xie W.,
      2. Kaetzel M. A.,
      3. Bruzik K. S.,
      4. Dedman J. R.,
      5. Shears S. B.,
      6. Nelson D. J.
      (1996) Inositol 3,4,5,6-tetrakisphosphate inhibits the calmodulin-dependent protein kinase II-activated chloride conductance in T84 colonic epithelial cells. J. Biol. Chem. 271:14092–14097.
      Abstract/FREE Full Text
    46. ↵
      1. Ismailov I. I.,
      2. Fuller C. M.,
      3. Berdiev B. K.,
      4. Shlyonsky V. G.,
      5. Benos D. J.,
      6. Barrett K. E.
      (1996) A biologic function for an ‘orphan’ messenger: d-myo-inositol 3,4,5,6-tetrakisphosphate selectively blocks epithelial calcium-activated chloride channels. Proc. Natl. Acad. Sci. USA 93:10505–10509.
      Abstract/FREE Full Text
    « Previous | Next Article »Table of Contents

    This Article

    1. Molecular Pharmacology April 1, 1997 vol. 51 no. 4 683-692
    1. AbstractFree
    2. » Full Text
    3. Full Text (PDF)

    Classifications

    Responses

    1. Submit a response
    2. No responses published

    Navigate This Article

    Current Issue

    1. November 2009, 76 (5)
    1. Current Issue
    1. Alert me to new issues of Molecular Pharmacology