- Email: [email protected]
Identifying the Ca++ signalling sources activating chloride currents in Xenopus oocytes using ionomycin and thapsigargin
Cellular Signalling 12 (2000) 629–635 http://www.elsevier.com/locate/cellsig
Identifying the Ca⫹⫹ signalling sources activating chloride currents in Xenopus oocytes using ionomycin and thapsigargin Carl L. Thurmana,*, Jon S. Burnsa, Roger G. O’Neilb b
a Department of Biology, University of Northern Iowa, Cedar Falls, IA 50614-0421, USA Department of Integrative Biology and Pharmacology, University of Texas Medical School, Houston, TX 77030, USA
Received 16 April 2000; accepted 28 June 2000
Abstract The calcium ionophore, ionomycin (IM), and the sarcoplasmic/endoplasmic reticulum (SER) calcium pump inhibitor, thapsigargin (TG), were used to study the roles of Ca⫹⫹ from different sources in regulating Ca⫹⫹-dependent Cl⫺ currents in Xenopus oocytes. The Ca⫹⫹-dependent Cl⫺ currents, Ic, were measured in voltage-clamped oocytes (Vc ⫽ ⫺60 mV). In the presence of extracellular Ca⫹⫹, both TG (0.1 to 10 M) and IM (0.1 to 10 M) induce release of Ca⫹⫹ from SER and activated capacitative Ca⫹⫹ entry (CCE) across the plasma membrane leading to activation of both “fast” and “slow” Cl⫺ currents. The fast Ic was produced by Ca⫹⫹ release from SER while Ca⫹⫹ entry across the plasma membrane activated the slow Ic. Intracellular application of the calcium buffer, BAPTA, blocked activation of the slow Ic due to Ca⫹⫹ entry via CCE pathways, but not via IM-mediated movement across the plasma membrane. It is concluded that predominantly Ca⫹⫹ release from stores regulates a fast Ic while Ca⫹⫹ entry through CCE pathways regulates a slow Ic. Further, the CCE and slow Ic pathways must be located in spatially separated compartments since BAPTA can effectively abolish the effects of Ca⫹⫹ entry via the CCE pathway, but not by the IM-mediated entry pathway.
1. Introduction Ion currents through the membrane of the unfertilized Xenopus laevis oocyte are mediated primarily by Cl⫺ channels [1–3]. Because the activity of the oocyte Cl⫺ channels is dependent upon cytoplasmic Ca⫹⫹ [4–6], these channels appear to provide a continuous and reliable indication of the submembranous Ca⫹⫹ level [7]. The source of Ca⫹⫹ is from either extracellular solutions or cytoplasmic stores [8,9]. Pathways of Ca⫹⫹ metabolism in the Xenopus oocyte are illustrated in Figure 1. One pathway for Ca⫹⫹ entry across the plasma membrane is capacitative calcium entry (CCE). Calcium entry via CCE is regulated indirectly [10,11]. As Ca⫹⫹ stores become depleted, plasma memAbbreviations: Vm, open-circuit membrane voltage; Vc, voltageclamped; Ic, net electrical current under Vc; CCE, capacitive calcium entry; IP3, inositol 2,4,5-triphosphate; SER, sarcoplasmic/endoplasmic reticulum; TG, thapsigargin; IM, ionomycin; HEPES, N-[2-hydroxyethyl]piperazine-N⬘-[ethanesulfonic acid]; BAPTA,1,2-bis (2-aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid); BAPTA-AM, 1,2-bis (2-aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid-tetra (acetoxymethyl)ester. * Corresponding author. Tel.: 319-273-2276; Fax: 319-273-7125. E-mail address: [email protected] (C.L. Thurman)
brane Ca⫹⫹ transport via CCE pathways is activated by an unknown mechanism. Once in the cytosol, Ca⫹⫹ is sequestered in the sarcoplasmic/endoplasmic reticulum (SER) by a membrane-bound Ca⫹⫹ pump (SERCA), which is blocked by thapsigargin (TG), specifically [12]. Another route for Ca⫹⫹ entry into the cytoplasm is through an IP3-mediate pathway in the SER membrane. IP3 produced enzymatically from membrane phospholipids can bind to a receptor on the SER leading to the release of Ca⫹⫹ from internal stores [13–16]. Increasing cytosolic Ca⫹⫹ levels within the oocyte leads to positive feedback on the IP3 system to enhance Ca⫹⫹ release producing a phenomenon known as calcium induced calcium release (CICR) [9]. Ca⫹⫹ released from the internal stores appears to regulate predominantly the “fast” membrane Cl current (ICl-1) activity while Ca⫹⫹ entry via CCE appears to control the “slower” Cl current activity (ICl-2) [1]. Ionomycin was used by both Yoshida and Plant [8] and Morgan and Jacob [17] to partition the fast and slow Cl currents into distinct, observable entities. In the presence of extracellular Ca⫹⫹, both currents are active while the fast current predominates under conditions of low extracellular Ca⫹⫹. Chelation of Ca⫹⫹ from the cytoplasm also eliminates the fast current.
0898-6568/00/$ – see front matter 2000 Elsevier Science Inc. All rights reserved. PII: S0898-6568(00)00106-6
630
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
ml/min. Occasionally, perfusion was interrupted to introduce a reagent. Glass micropipettes were fashioned with a P-97 Programmable Pipette Puller (Sutter Instruments Co.; Novato, CA). Capillaries (WPI; Sarasota, FL; 1B120F-4) were drawn, filled with 3.0 M KCl and used only if the tip potential was less than ⫾3 mV and resistance was 3.0 (⫾1.0) M⍀. Estimates of membrane voltage (Vm) and current (Ic), using a dual-electrode oocyte clamping system (Warner Instrument Corp.; Model OC725C), were transcribed on a Kipp and Zonen (Holland) two-channel chart recorder. The Ic was the electrical current (in nA) applied to sustain the oocyte Vm at ⫺60 ⫾ 0.5 mV. 2.3. Solutions Ca⫹⫹-rich Xenopus Ringer’s consisted of (in mM): 100 NaCl, 2 KCl, 1.8 CaCl2•H2O, 1.0 MgCl2•H2O and 5 HEPES (pH 7.5). Ca⫹⫹-free Ringer’s consisted of (in mM): 82.5 NaCl, 2.5 KCl, 1.0 MgCl2•H2O and 5 HEPES (pH 7.5). Fig. 1. Dynamic cellular model for submembrane Ca⫹⫹ distribution and homeostasis in the Xenopus oocyte. SER—endoplasmic reticulum; PM—plasma membrane; IM—ionomycin; IP3—inositol trisphosphate; TG—thapsigargin; ATP—adenosine triphosphate; CIF—signal for CCE (specific nature unknown); Cl⫺ both fast and slow Ca⫹⫹dependent Cl⫺ channels.
In the present article, we describe the use of a Ca⫹⫹ ionophore, ionomycin (IM), and an inhibitor of the SERCA pump, thapsigargin (TG), at differing concentrations to control the release/entry of Ca⫹⫹ from various reservoirs available to the Xenopus oocyte. Here, we demonstrate that by using IM and TG in the presence of BAPTA-AM, a cytoplasmic Ca⫹⫹ buffer, we can differentially control the Ca⫹⫹ source regulating Ca⫹⫹dependent Cl⫺ currents in the oocyte. Our results indicate a spatial separation of the Ca⫹⫹ entry pathway from the chloride channels. 2. Materials and methods 2.1. Preparation of oocytes Oocytes were obtained from healthy, mature Xenopus laevis as described by Hartzell [1–3]. After anesthesia by submersion in 0.17% MS-222 for 15 min, oocytes in stage V–VI were surgically harvested from the frogs. Postsurgical recovery for the frogs was in isolation. Oocytes were defollicated with collagenase and kept in frog Ringer’s solution (18⬚C) with continuous stirring. The bath was changed twice per day. 2.2. Electrophysiology Single oocytes were placed in a small Plexiglas chamber (Warner Instruments Corp.; Hamden, CT; #RC26Z) and bathed constantly at a perfusion rate of 1–2
2.4. Chemicals The following reagents were purchased from Sigma Chemical Co. (St. Louis, MO): HEPES (N-[2-hydroxyethyl]piperazine-N⬘-[ethanesulfonic acid]), ionomycinCa⫹⫹salt, thapsigargin, BAPTA [1,2-bis (2-Aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid)]. BAPTA-AM [1,2-bis (2-Aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid-tetra (acetoxymethyl)ester] was obtained from Calbiochem (San Diego, CA). 2.5. Statistical treatment Data are reported as averages with SEM. Statistical comparisons were made using a “one-tailed” Student’s paired t-test. p-Values indicate the probability of accepting the null hypothesis (Ho). 3. Results 3.1. Ionomycin dose response Eight oocytes from Xenopus, bathed in Ca⫹⫹-free media, were voltage clamped to ⫺60 mV. The resting, open circuit membrane potentials (Vm) averaged ⫺42.8 ⫾ 2.9 mV. Upon voltage clamping (Vc) to ⫺60 mV, the membrane current, Ic, averaged ⫺25.0 ⫾ 8.9 nA. To determine the effect of IM on Ca⫹⫹ stores and the Cl⫺ current, net membrane current was monitored under voltage clamping in the absence of extracellular Ca⫹⫹ as IM was administered (Table 1). The Ic response to 10 M IM is biphasic (Fig. 2). Application of IM produced a fast transient spike followed by an elevated plateau of the Ic (Fig. 2A). This can be attributed to Ic contributions from different kinds of Ca⫹⫹-activated Cl⫺ channels. On the other hand, increasing doses of IM, between 0.1 and 10.0 M, increased only the plateau
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
631
Table 1 Ic (nA) response of oocytes in Ca-freea solutions to Ionomycin (IM) Ic (nA) at each IM concentrationb
Mean SEM N a b
Vm (mV)
0 M
0.1 M
0.3 M
1.0 M
5.0 M
10 M
⫺42.8 2.9 8
⫺25.0 8.9
⫺45.0 18.3
⫺69.5 33.1
⫺102.0 48.9
⫺161.0 61.6
⫺173.4 61.3
Mean baseline Ic 440 ⫾ 20 nA after 10 M IM (n ⫽ 3) in Ca⫹⫹-rich media. Values taken from “slow” plateau Ic in Ca⫹⫹-free media.
component of the slow Ic. It rose from ⫺25.0 ⫾ 8.9 to ⫺173.4 ⫾ 61.3 nA (p ⬍ 0.05). If 10 M IM is applied in the presence of Ca⫹⫹, the Ic response is also biphasic. It is similar to the response observed in the absence of Ca⫹⫹. However, the elevation of the slow Ic plateau was much greater in Ca⫹⫹ rich medium. The Ic increased from ⫺95 ⫾ 45 to ⫺440 ⫾ 20 nA (n ⫽ 3) (p ⬍ 0.01). The plateau remains elevated in the presence of extracellular Ca⫹⫹ (Fig. 2B) and is significantly higher than the plateau Ic in the absence of Ca⫹⫹ (p ⬍ 0.05). 3.2. Thapsigargin dose response Four Xenopus oocytes were used to study the effects of TG on Ic in Ca⫹⫹-free medium (Table 2A). A typical response of the ooctye to TG is shown in Figure 3A. In the absence of TG, the Ic averaged ⫺72.5 ⫾ 21.7 nA. When oocytes were treated with TG, the Ic began to increase slowly. Increasing TG concentration from 0.1 to 5 M, the Ic increased from a baseline of ⫺72.5 ⫹ 21.7 to ⫺155.0 ⫾ 56.8 nA (0.05 ⭐ p ⭐ 0.1. If, however, oocytes ere treated with a single 10 M concentration of TG in the presence of Ca⫹⫹ (Figure 3B), the Ic increased from ⫺50.0 ⫾ 14.1 to ⫺586.7 ⫾ 210 nA (n ⫽ 3; p ⬍ 0.05), reflecting an enhancement of the slow current. If 10 M IM was subsequently administered following the TG treatment, the IM-induced fast Ic com-
Fig. 2. Ca⫹⫹ modulation of IM response in Xenopus oocytes. (A) IM (10 M) in Ca⫹⫹-free solution, initial Ic 80 nA. (B) IM (10 M) in complete Ringer’s solution, initial Ic 50 nA. Box indicates period of IM application. Scale shown.
ponent was still apparent, producing an Ic transient with a peak of ⫺2062.5 ⫾ 736.6 nA (Table 2A). Because TG is known to selectively inhibit SER Ca⫹⫹ uptake, 10 M IM was used to to test for release of sequestered Ca⫹⫹ when the oocyte was bathed with a Ca⫹⫹-free medium. After voltage clamping, six oocytes were treated with 10 M IM. The usual biphasic Ic response was observed (not shown). Peak Ic after IM averaged ⫺791.7 nA while the sustained Ic, however, averaged only ⫺86.7 ⫾ 13.9 nA (Table 2B). The response to increasing TG concentrations under these conditions was still evident as indicated by the Ic changing from ⫺111.7 to ⫺150 nA (Table 2B; p ⬎ 0.1). Subsequently, a second challenge with IM produced no Ic response (⫺158 ⫾ 48 nA). (Compare the last column 10 M IM in Table 2A to Table 2B.) This indicates that in the presence of TG following IM, the Ca⫹⫹ stores are completely depleted. A second challenge with IM reveals the deprived state of the stores. 3.3. Mediation of Ic by BAPTA and BAPTA-AM In an effort to buffer cytoplasmic Ca⫹⫹ fluctuations, either BAPTA or BAPTA-AM was added to the oocyte bathing solution. When BAPTA (10 mM) in a Ca⫹⫹-free medium was applied to oocytes to chelate extracellular Ca⫹⫹, unstable membrane properties were observed. Four control oocytes having membrane potentials of ⫺23.7 ⫾ 2.4 mV, had initial Ics of ⫺222.5 ⫾ 52.7 nA. Upon exposure to extracellular BAPTA for only 5 min, spontaneous extrusion of the cytoplasm (blebbing) and sharp increases in Ic were observed. Application of 10 M BAPTA in Ca⫹⫹ -free media for 2 h produced oocytes with Ics of ⫺4106.4 ⫾ 1343.6 nA. The combined effect of BAPTA and Ca⫹⫹ -free media appear to be deleterious to the egg. Because BAPTA is a charged moiety, it is unlikely to gain access to the cytoplasm. Hence, its effects are probably due to chelation of extracellular Ca⫹⫹. This appears to destabilize the membrane and increase Ic [18]. On the other hand, application of BAPTA-AM (10 M) resulted in oocytes with stable electrical properties (Table 3). BAPTA-AM is nonpolar, and permeates the cellular membrane. Once in the cell, it is deesterfied to BAPTA and chelates intracellular Ca⫹⫹. Oocytes incubated in Ca⫹⫹-rich medium containing 10 M
632
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
Table 2 Effect of thapsigargina (TG) on Ic (nA) in oocytes depending upon pretreatment with ionomycin A: Effect of IM after TG dosagea Ic (nA) at each TG concentration
Mean SEM N a b
Vm (mV)
0 M
0.1 M TG
⫺16.3 1.5 4
⫺72.5 21.7
⫺97.5 22.5
0.3 M TG
1.0 M TG
5.0 M TG
10 Mb TG
⫺112.5 28.7
⫺130.0 36.7
⫺155.0 56.8
⫺2062.5 736.1
In Ca⫹⫹ solutions, Ic after 10 M TG, 586 ⫾ 56.0 nA (n ⫽ 3). Value taken at peak of fast spike.
B: Effect of TG after 10 M IM Ic (nA) at each IM/TG concentration
Mean SEM N a b
Vm (mV)
0 M
10 Ma IM
0.1 M TG
0.3 M TG
1.0 M TG
5.0 M TG
10 Mb TG
⫺31.9 0.3 6
⫺70.0 19.8
⫺86.7 13.9
⫺111.7 30.3
⫺144.2 42.4
⫺144.0 50.5
⫺150.0 48.6
⫺158.0 48.6
Value taken from “slow” plateau Ic in Ca⫹⫹-free media No “fast” current observed.
BAPTA-AM for 5 to 6 h had average Ics of ⫺96.7 ⫾ 26.7 nA. Subsequent application of IM to oocytes in a Ca⫹⫹-free medium increased the Ic only moderately. As IM was increased from 0.1 to 10 M, the Ic of the BAPTA-AM treated oocytes increased from ⫺96.7 ⫾ 23.3 to ⫺163.3 ⫾ 53.3 nA (Table 3; p ⬍ 0.05). Most notedly, the fast transient spike induced by 10 M IM was eliminated and only the slow Ic was observed when BAPTA-AM was presented in Ca⫹⫹-free media. Conse-
quently, it appears that IM increases the slow component of the Ic by changing Ca⫹⫹ concentration near the site of the Cl⫺ current (e.g., Ca⫹⫹-dependent Cl⫺ channels). This Ca⫹⫹ influx is localized to regions near the membrane, and appears to occur too rapidly to be buffered immediately by cytoplasmic BAPTA. 3.4. Synergistic effects of BAPTA-AM and TG To understand the cytoplasmic effect of Ca⫹⫹ from different reservoirs on Ic, the combined effects of BAPTA-AM and TG were examined (Table 4; Figure 4). Oocytes were incubated at room temperatures with 10 M BAPTA-AM in a Ca⫹⫹-rich medium for 5 h, then transferred to a Ca⫹⫹ -free Ringer’s solution. Under this condition, the mean of initial Vm was ⫺20.7 ⫾ 4.3 mV, while Ic averaged ⫺100.0 ⫾ 15.3 nA. Superfusing the voltage-clamped oocytes with 10 M TG modestly increased the Ic to ⫺113.3 ⫾ 18.6 nA. If the oocyte was then treated with 10 M IM, the Ic did not increase significantly (p ⬎ 0.2) in the absence of extracellular Ca⫹⫹ (Table 4A; Figure 4A). In an additional set of experiments, adding Ca⫹⫹ to the extracellular medium Table 3 Effect of 10 M BAPTA-AMa on oocyte response to IM Ic (nA) response at each IM concentrations (with BAPTA)b
Fig. 3. Ca⫹⫹ modulation of TG-induced Ic in Xenopus oocytes. (A) TG (10 M) in Ca⫹⫹-free solution, initial Ic 80 nA. (B) TG (10 M) in Ca⫹⫹-rich solution, initial Ic 40 nA. Arrows indicates time of TG application. Discontinuity of tracing for 720 s show in (A). Scale shown.
Mean SEM N a b
Vm (mv)
0 M
0.1 M
0.3 M
1.0 M
10 M
⫺18.4 1.7 3
⫺96.7 26.7
⫺96.7 23.3
⫺100.0 20.8
⫺150.0 60.3
⫺163.0 53.3
Cell bathed for 5 h in BAPTA-AM. Values of Ic in Ca⫹⫹ rich media.
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
633
cannot sufficiently chelate all IM-induce Ca⫹⫹ entry across the plasma membrane to eliminate activation of a Cl⫺ current.
Table 4 Effect of IM on BAPTA-AM/TG-treated oocytea A: Ca⫹⫹-free solution Ic (nA) Vm (mV) Mean SEM N B:
⫺20.7 4.3 3
Ca⫹⫹-rich
Initial Ic
TG 10 M
—
IM 10 M
⫺100.0 15.3
⫺113.3 18.6
— —
⫺123.3 14.5
solution Ic (nA)
Vm(mV) Mean SEM N a
⫺27.6 5.5 5
Initial Ic
TG 10 M
Ca⫹⫹ 1.8 mM
IM 10 M
⫺142.0 25.5
⫺100.0 14.7
⫺96.0 14.5
⫺260.0 60.4
Oocytes treated for 5 h with 10 M BAPTA-AM.
following TG treatment did not influence the Ic despite having the stores discharged and CCE pathways thereby activated (Table 4B). However, with subsequent administration of IM in a Ca⫹⫹-rich medium, Ic increased to ⫺260.0 ⫾ 60.4 nA (p ⬍ 0.05; Figure4B). As with previous observations in BAPTA-AM–treated cells, the fast transient spike in the Ic response was not evident. The Ic change with IM was thus considered to mimic a slow current increase demonstrating that IM-induced Ca⫹⫹ influx via the ionophore action at the plasma membrane can activate Ic when TG induced Ca⫹⫹ release due to activation of CCE could not. On the other hand, if Ca⫹⫹ were not present in the bath (Table 4A; Figure 4A), the IM-induced slow current transient was eliminated. Thus, BAPTA-AM can effectively minimize the contributions of cytoplasmic calcium stores in regulating Ic by either direct blockade of Ca⫹⫹ release or signalling for Ca⫹⫹ entry via CCE. However, cytoplasmic BAPTA
Fig. 4. Ca⫹⫹ modulation of TG and IM response in BAPTA-AM treated Xenopus oocytes. (A) TG (10 M) and IM in Ca⫹⫹-free solution, initial Ic 80 nA. (B) TG (10 M) and IM in Ca⫹⫹-rich solution, initial Ic 110 nA. Arrows indicate addition of TG and IM. Scale shown.
4. Discussion 4.1. Intracellular Ca⫹⫹ and regulation of Cl⫺ currents Xenopus oocytes have long been known to display Ca⫹⫹-dependent Cl⫺ currents. Elevation of intracellular Ca⫹⫹ levels activates one or more different Cl⫺ currents [1,5 ], as confirmed in the present study. The regulation by Ca⫹⫹ of the underlying Cl⫺ channels and their associated currents has been shown to be complex, in part due to the complexity of the Ca⫹⫹ transport pathways regulating intracellular Ca⫹⫹ levels. Indeed, it appears that the method of altering intracellular Ca⫹⫹ levels, such as that shown for TG or IM in the present study, can produce widely varying Cl⫺ currents. This complexity may reside in the Ca⫹⫹ sensitivity of the various Cl⫺ channels and their spatial relations to the Ca⫹⫹ transport pathways [7]. As heretofore noted, the regulation of intracellular Ca⫹⫹ levels is a highly integrated process that reflects a dynamic balance between Ca⫹⫹ influx and efflux pathways [16]. As illustrated in Figure 1, Ca⫹⫹ influx can arises from Ca⫹⫹ in both the extracellular space and intracellular storage compartments via passive influx through Ca⫹⫹ channels; Ca⫹⫹ efflux from the cytosol occurs via active extrusion through Ca⫹⫹ pumps in the plasma membrane and intracellular membranes of Ca⫹⫹ storage sites. The generation of Ca⫹⫹ signals, reflected as a transient rise in intracellular Ca⫹⫹, typically arises from a transient activation of Ca⫹⫹ influx via activation of one or more Ca⫹⫹ channels, followed by activation of the Ca⫹⫹ efflux mechanisms to restore Ca⫹⫹ levels to near normal levels. Consequently, the transition from rest to active state and back again to a resting state is marked by redistribution of Ca⫹⫹ among the cytosol, the intracellular storage sites and the extracellular medium. The dynamics of Ca⫹⫹ movement among these compartments is rapidly becoming increasingly important to the understanding of cellular homeostasis [14,15]. In Xenopus oocytes, and other nonexcitable cells, the SER is now recognized as the main compartment of the intracellular Ca⫹⫹ stores [16]. The SER Ca⫹⫹ stores work in a coordinated manner with the plasma membrane Ca⫹⫹ channels to provide appropriate Ca⫹⫹ for different physiological processes. IP3 receptors on the SER are themselves regulated Ca⫹⫹ channels that control the release of Ca⫹⫹ from the SER storage site. Various hormonal receptors on the plasma membrane are coupled through the phosphatidyloinostol signal transduction pathways controlling the generation of intracellular IP3 [15]. Stimulation of IP3 production will lead to binding of IP3 to the SER IP3 receptor, leading
634
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
to activation of Ca⫹⫹ efflux from the SER to bring about a rise in intracellular Ca⫹⫹ levels. The release of Ca⫹⫹ from the SER stores can, in turn, activate Ca⫹⫹ channels in the plasma membrane. In the oocyte and many other cells, as the SER becomes depleted, Ca⫹⫹ channels in the plasma membrane are activated, leading to enhanced Ca entry. This process, first described by Putney and coworkers [19,20] has been called CCE or, more recently “store-operated calcium entry.” The exact nature of the mechanism coordinating Ca⫹⫹-channel activity between the SER and plasma membrane is unclear [15,16]. Originally, it was believed that a calcium influx factor (CIF) was released from the SER. Recent studies point to a potential role for phosphorylated glycopyranosides as a possible signal in oocytes [21]. In some instances, the IP3 receptor is directly coupled to CCE, and may control Ca⫹⫹ entry [22]. While release of Ca⫹⫹ from stores and the entry of Ca⫹⫹ via CCE pathways and other pathways appear to play critical roles in regulating Ca⫹⫹-dependent Cl⫺ channels, the two pathways of Ca⫹⫹ entry can appear to display widely differential control of one or more Cl⫺ channels (see below). 4.2. Ca⫹⫹-dependent Cl⫺ channels in oocytes It has long been known that Xenopus oocyte expresses endogenous outward Cl⫺ currents that are Ca⫹⫹ dependent and regulated by intracellular Ca⫹⫹ levels [4]. The present study also verified the presence of Ca⫹⫹ dependent Cl⫺ currents showing that both release of Ca⫹⫹ from internal stores and activation of Ca⫹⫹ entry across the plasma membrane can differentially regulate plasma membrane Ca⫹⫹-dependent Cl⫺ currents. Studies by Boton and coworkers [5] further demonstrated that there may be at least two distinct classes of Ca⫹⫹ dependent Cl currents that could be unmasked by the Ca⫹⫹ ionophore, A23187. In the presence of extracellular Ca⫹⫹, a “slow” Cl⫺ channel is activated. In the presence of low or normal extracellular Ca⫹⫹, a “fast” Cl⫺ channel is activated. Others [8,17] have noted that ionomycin, a Ca⫹⫹ ionophore that may also activate release of Ca⫹⫹ from SER stores, similarly activated a fast and slow Cl⫺ current. However, if the cells were treated with BAPTA to buffer intracellular Ca⫹⫹ levels to low levels, both Cl⫺ currents were suppressed. In contrast, in the absence of BAPTA, but where extracellular Ca⫹⫹ was removed, ionomycin now only activated the fast Cl⫺ current, demonstrating that the fast Cl⫺ current may be predominantly controlled by Ca⫹⫹ release from internal stores and the slow Cl⫺ current controlled by Ca⫹⫹ entry across the plasma membrane. The concept of at least two separate Ca⫹⫹-dependent Cl⫺ currents in oocytes controlled by separate sources of Ca⫹⫹ was most poignantly demonstrated by Hartzell and coworkers [1–3]. Indeed, these workers clearly demonstrated the presence of at least two Cl⫺ currents, one being primarily controlled by release of Ca⫹⫹ from the SER
and the other typically controlled by Ca⫹⫹ entry through endogenous, CCE channels. The present study confirms and extends the concept of two or more Ca⫹⫹-dependent Cl⫺ channels in Xenopus oocytes. Indeed, in the presence of extracellular Ca⫹⫹ ionomycin was demonstrated to induce both a “fast” spiking Cl⫺ current and a second, “slow” plateau Cl current (Figure 2). The slow current was closely controlled by Ca⫹⫹ entry through CCE pathways as unloading of stores via TG treatment was observed to activate the slow Cl⫺ current in the presence of extracellular Ca⫹⫹, but not in the absence of extracellular Ca⫹⫹ (Figure 4). Most important, however, was the observation that the CCE pathway and the slow Cl⫺ current appear to be spatially separated on the oocyte surface. If cells are treated with BAPTA to buffer intracellular Ca⫹⫹ to low levels, activation of Ca⫹⫹ entry through CCE pathways (TG treatment) did not activate the Cl⫺ current, implying that the CCE pathways and Cl⫺ channels are sufficiently spatially separated to allow time for Ca⫹⫹ in the immediate vicinity of the Cl⫺ channel to be fully buffered by BAPTA. In contrast, if Ca⫹⫹ entry were induced by treatment with ionomycin in the presence of extracellular Ca⫹⫹, which should induce Ca⫹⫹ entry with a uniform spatial distribution across the cell surface, the slow Cl⫺ current was now activated even in the presence of BAPTA. This can only occur if BAPTA does not sufficiently buffer Ca⫹⫹ changes in the vicinity of the Cl⫺ channels so that the Cl⫺ channels are activated. It, therefore, seems reasonably to conclude that the Cl⫺ channels and the CCE pathways may be, at least in part, spatially separated. Hartzell and coworkers have previously shown that the slow Cl⫺ channels may be widely distributed throughout both the animal and vegetable poles of the oocyte [1–3,7]. If such is the case, then it is highly probable that the CCE pathways have a more restricted distribution, perhaps only in one pole of the oocyte. Alternatively, if sufficiently spatially separated from the slow Cl⫺ channels, Ca⫹⫹ entry via CCE pathways should be effectively buffered by intracelluar BAPTA. Regardless, the exact distribution of both the Ca⫹⫹ and Cl⫺ channels will require a more detailed assessment of channel localization when suitable probes become available. Then, perhaps, the impact of spatial separation on the interaction of Ca⫹⫹ and Cl⫺ channels in the oocyte can be clearly delineated. Finally, the present study provides further confirmation that ionomycin can act both as a plasma membrane and a SER membrane Ca⫹⫹ ionophore. Previous studies in oocytes [8] and in umbilical-vein endothelial cells [17] have shown that application of ionomycin in the absence of extracellular Ca⫹⫹ can induce release of Ca⫹⫹ from internal stores. The present study likewise demonstrated that in the absence of extracellular Ca⫹⫹, ionomycin treatment induced a rapid release of Ca⫹⫹ from stores. This effect most probably reflects the ionophoretic ac-
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
635
tion of the antibiotic on the SER membrane. Ionomycin is incorporated into the lipid bilayer where it operates as a Ca⫹⫹/H⫹ exchanger [23]. Furthermore, it was shown that in the presence of extracellular Ca⫹⫹, ionomycin induces a rapid influx of Ca⫹⫹ across the plasma membrane as expected, due to its actions as a Ca⫹⫹ ionophore. Hence, the actions of ionomycin include effects on both plasma membranes and intracellular membranes. Interpreting the actions of ionomycin must, therefore, be viewed with caution, as it has effects at several cellular sites. In summary, the present study demonstrates that the regulation of Ca⫹⫹-dependent Cl⫺ channel in Xenopus oocytes is a complex process being controlled by both release of Ca⫹⫹ from internal stores and influx of Ca⫹⫹ across the plasma membrane. Part of this complexity reflects not only the presence of multiple Ca⫹⫹-dependent Cl⫺ channels, but a spatial separation of plasma membrane Cl⫺ channels from the dominant Ca⫹⫹ influx pathway, the CCE pathway. The spatial dependence of the Cl⫺ channels on other Ca⫹⫹ entry pathways and the precise spatial interdependence of the Cl⫺ and Ca⫹⫹ channels is currently not known, but remains to be determined in future studies.
J.S.B.) was provided by the UT-Houston Summer Undergraduate Research Program. This work was completed during the 1999–2000 professional development leave of C.L.T. from UNI. Matt Friez and Eugene M. Barnes, at Baylor College of Medicine, Houston, generously provided oocytes for this research.
Acknowledgments
[19] [20] [21]
This work was supported, in part, by NIDDK 2RO1DK40545 (to R.G.O.) and NSF DUE-9750754 (to C.L.T.). Summer salary (to C.L.T.) during 1999 was provided by fellowships from UNI Graduate College and Applied Technologies funds. Summer salary (to
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]
[22]
[23]
Hartzell HC. J Gen Physiol 1986;108:157–75. Machaca K, Hartzell HC. Biophys J 1998;74:1286–95. Kimura A, Hartzell HC. Am J Physiol 1999;276:C161–75. Miledi R, Parker J. J Physiol (Lond) 1984;357:173–83. Boton R, Dascal N, Gillo B, Lass Y. J Physiol 1989;408:511–34. Parekh AB. Pflugers Arch Eur J Physiol 1995;430:954–63. Gomez-Hernandez J-M, Stuhmer W, Parekh, AB. J Physiol 1997;502:569–74. Yoshida S, Plant S. J Physiol 1992;458:307–18. Petersen CCH, Dupont G. Cell Calcium 1994;16:391–403. Petersen CCH, Berridge MJ. J Biol Chem 1994;51:32246–53. Gillo B, Sealfon SC, Minke B. J Photochem Photobiol 1996;35B:77–82. Goudeau H, Goudeau M. Dev Biol 1998;193:225–38. Parker I, Miledi R. Proc R Soc Lond.1987;231:27–36. Pozzan T, Pizzuto R, Volpe P, Meldolesi J. Physiol Rev 1994;74:595–636. Clapham D. Cell 1995;80:259–68. Alvarez J, Montero M, Garcia-Sancho J. NIPS 1999;14:161–8. Morgan A, Jacob R. Biochem J 1994;300:665–72. Weber W-M, Liebold KM, Reifarth FW, Uhr U, Clauss W. Pflugers Arch Eur J Physiol 1995;429:820–4. Putney JW. Cell Calcium 1986;7:1–12. Putney JW. Cell Calcium 1990;11:611–24. Hartzell HC, Machaca K, Hirayama Y. Mol Pharmacol 1997;51:683–92. Boulay G, Brown MB, Qin N, Jiang M, Dietrich A, Zhu MX, Chen Z, Birnbaumer M, Mikoshiba K, Birnbaumer L. Proc Natl Acad Sci USA 1999;96:14955–60. Wheeler JJ, Veiro JA, Cullis PR. Mol Membr Biol 1994;11:151–7.
Identifying the Ca⫹⫹ signalling sources activating chloride currents in Xenopus oocytes using ionomycin and thapsigargin Carl L. Thurmana,*, Jon S. Burnsa, Roger G. O’Neilb b
a Department of Biology, University of Northern Iowa, Cedar Falls, IA 50614-0421, USA Department of Integrative Biology and Pharmacology, University of Texas Medical School, Houston, TX 77030, USA
Received 16 April 2000; accepted 28 June 2000
Abstract The calcium ionophore, ionomycin (IM), and the sarcoplasmic/endoplasmic reticulum (SER) calcium pump inhibitor, thapsigargin (TG), were used to study the roles of Ca⫹⫹ from different sources in regulating Ca⫹⫹-dependent Cl⫺ currents in Xenopus oocytes. The Ca⫹⫹-dependent Cl⫺ currents, Ic, were measured in voltage-clamped oocytes (Vc ⫽ ⫺60 mV). In the presence of extracellular Ca⫹⫹, both TG (0.1 to 10 M) and IM (0.1 to 10 M) induce release of Ca⫹⫹ from SER and activated capacitative Ca⫹⫹ entry (CCE) across the plasma membrane leading to activation of both “fast” and “slow” Cl⫺ currents. The fast Ic was produced by Ca⫹⫹ release from SER while Ca⫹⫹ entry across the plasma membrane activated the slow Ic. Intracellular application of the calcium buffer, BAPTA, blocked activation of the slow Ic due to Ca⫹⫹ entry via CCE pathways, but not via IM-mediated movement across the plasma membrane. It is concluded that predominantly Ca⫹⫹ release from stores regulates a fast Ic while Ca⫹⫹ entry through CCE pathways regulates a slow Ic. Further, the CCE and slow Ic pathways must be located in spatially separated compartments since BAPTA can effectively abolish the effects of Ca⫹⫹ entry via the CCE pathway, but not by the IM-mediated entry pathway.
1. Introduction Ion currents through the membrane of the unfertilized Xenopus laevis oocyte are mediated primarily by Cl⫺ channels [1–3]. Because the activity of the oocyte Cl⫺ channels is dependent upon cytoplasmic Ca⫹⫹ [4–6], these channels appear to provide a continuous and reliable indication of the submembranous Ca⫹⫹ level [7]. The source of Ca⫹⫹ is from either extracellular solutions or cytoplasmic stores [8,9]. Pathways of Ca⫹⫹ metabolism in the Xenopus oocyte are illustrated in Figure 1. One pathway for Ca⫹⫹ entry across the plasma membrane is capacitative calcium entry (CCE). Calcium entry via CCE is regulated indirectly [10,11]. As Ca⫹⫹ stores become depleted, plasma memAbbreviations: Vm, open-circuit membrane voltage; Vc, voltageclamped; Ic, net electrical current under Vc; CCE, capacitive calcium entry; IP3, inositol 2,4,5-triphosphate; SER, sarcoplasmic/endoplasmic reticulum; TG, thapsigargin; IM, ionomycin; HEPES, N-[2-hydroxyethyl]piperazine-N⬘-[ethanesulfonic acid]; BAPTA,1,2-bis (2-aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid); BAPTA-AM, 1,2-bis (2-aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid-tetra (acetoxymethyl)ester. * Corresponding author. Tel.: 319-273-2276; Fax: 319-273-7125. E-mail address: [email protected] (C.L. Thurman)
brane Ca⫹⫹ transport via CCE pathways is activated by an unknown mechanism. Once in the cytosol, Ca⫹⫹ is sequestered in the sarcoplasmic/endoplasmic reticulum (SER) by a membrane-bound Ca⫹⫹ pump (SERCA), which is blocked by thapsigargin (TG), specifically [12]. Another route for Ca⫹⫹ entry into the cytoplasm is through an IP3-mediate pathway in the SER membrane. IP3 produced enzymatically from membrane phospholipids can bind to a receptor on the SER leading to the release of Ca⫹⫹ from internal stores [13–16]. Increasing cytosolic Ca⫹⫹ levels within the oocyte leads to positive feedback on the IP3 system to enhance Ca⫹⫹ release producing a phenomenon known as calcium induced calcium release (CICR) [9]. Ca⫹⫹ released from the internal stores appears to regulate predominantly the “fast” membrane Cl current (ICl-1) activity while Ca⫹⫹ entry via CCE appears to control the “slower” Cl current activity (ICl-2) [1]. Ionomycin was used by both Yoshida and Plant [8] and Morgan and Jacob [17] to partition the fast and slow Cl currents into distinct, observable entities. In the presence of extracellular Ca⫹⫹, both currents are active while the fast current predominates under conditions of low extracellular Ca⫹⫹. Chelation of Ca⫹⫹ from the cytoplasm also eliminates the fast current.
0898-6568/00/$ – see front matter 2000 Elsevier Science Inc. All rights reserved. PII: S0898-6568(00)00106-6
630
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
ml/min. Occasionally, perfusion was interrupted to introduce a reagent. Glass micropipettes were fashioned with a P-97 Programmable Pipette Puller (Sutter Instruments Co.; Novato, CA). Capillaries (WPI; Sarasota, FL; 1B120F-4) were drawn, filled with 3.0 M KCl and used only if the tip potential was less than ⫾3 mV and resistance was 3.0 (⫾1.0) M⍀. Estimates of membrane voltage (Vm) and current (Ic), using a dual-electrode oocyte clamping system (Warner Instrument Corp.; Model OC725C), were transcribed on a Kipp and Zonen (Holland) two-channel chart recorder. The Ic was the electrical current (in nA) applied to sustain the oocyte Vm at ⫺60 ⫾ 0.5 mV. 2.3. Solutions Ca⫹⫹-rich Xenopus Ringer’s consisted of (in mM): 100 NaCl, 2 KCl, 1.8 CaCl2•H2O, 1.0 MgCl2•H2O and 5 HEPES (pH 7.5). Ca⫹⫹-free Ringer’s consisted of (in mM): 82.5 NaCl, 2.5 KCl, 1.0 MgCl2•H2O and 5 HEPES (pH 7.5). Fig. 1. Dynamic cellular model for submembrane Ca⫹⫹ distribution and homeostasis in the Xenopus oocyte. SER—endoplasmic reticulum; PM—plasma membrane; IM—ionomycin; IP3—inositol trisphosphate; TG—thapsigargin; ATP—adenosine triphosphate; CIF—signal for CCE (specific nature unknown); Cl⫺ both fast and slow Ca⫹⫹dependent Cl⫺ channels.
In the present article, we describe the use of a Ca⫹⫹ ionophore, ionomycin (IM), and an inhibitor of the SERCA pump, thapsigargin (TG), at differing concentrations to control the release/entry of Ca⫹⫹ from various reservoirs available to the Xenopus oocyte. Here, we demonstrate that by using IM and TG in the presence of BAPTA-AM, a cytoplasmic Ca⫹⫹ buffer, we can differentially control the Ca⫹⫹ source regulating Ca⫹⫹dependent Cl⫺ currents in the oocyte. Our results indicate a spatial separation of the Ca⫹⫹ entry pathway from the chloride channels. 2. Materials and methods 2.1. Preparation of oocytes Oocytes were obtained from healthy, mature Xenopus laevis as described by Hartzell [1–3]. After anesthesia by submersion in 0.17% MS-222 for 15 min, oocytes in stage V–VI were surgically harvested from the frogs. Postsurgical recovery for the frogs was in isolation. Oocytes were defollicated with collagenase and kept in frog Ringer’s solution (18⬚C) with continuous stirring. The bath was changed twice per day. 2.2. Electrophysiology Single oocytes were placed in a small Plexiglas chamber (Warner Instruments Corp.; Hamden, CT; #RC26Z) and bathed constantly at a perfusion rate of 1–2
2.4. Chemicals The following reagents were purchased from Sigma Chemical Co. (St. Louis, MO): HEPES (N-[2-hydroxyethyl]piperazine-N⬘-[ethanesulfonic acid]), ionomycinCa⫹⫹salt, thapsigargin, BAPTA [1,2-bis (2-Aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid)]. BAPTA-AM [1,2-bis (2-Aminophenoxy)ethane-N,N,N⬘,N⬘-tetraactic acid-tetra (acetoxymethyl)ester] was obtained from Calbiochem (San Diego, CA). 2.5. Statistical treatment Data are reported as averages with SEM. Statistical comparisons were made using a “one-tailed” Student’s paired t-test. p-Values indicate the probability of accepting the null hypothesis (Ho). 3. Results 3.1. Ionomycin dose response Eight oocytes from Xenopus, bathed in Ca⫹⫹-free media, were voltage clamped to ⫺60 mV. The resting, open circuit membrane potentials (Vm) averaged ⫺42.8 ⫾ 2.9 mV. Upon voltage clamping (Vc) to ⫺60 mV, the membrane current, Ic, averaged ⫺25.0 ⫾ 8.9 nA. To determine the effect of IM on Ca⫹⫹ stores and the Cl⫺ current, net membrane current was monitored under voltage clamping in the absence of extracellular Ca⫹⫹ as IM was administered (Table 1). The Ic response to 10 M IM is biphasic (Fig. 2). Application of IM produced a fast transient spike followed by an elevated plateau of the Ic (Fig. 2A). This can be attributed to Ic contributions from different kinds of Ca⫹⫹-activated Cl⫺ channels. On the other hand, increasing doses of IM, between 0.1 and 10.0 M, increased only the plateau
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
631
Table 1 Ic (nA) response of oocytes in Ca-freea solutions to Ionomycin (IM) Ic (nA) at each IM concentrationb
Mean SEM N a b
Vm (mV)
0 M
0.1 M
0.3 M
1.0 M
5.0 M
10 M
⫺42.8 2.9 8
⫺25.0 8.9
⫺45.0 18.3
⫺69.5 33.1
⫺102.0 48.9
⫺161.0 61.6
⫺173.4 61.3
Mean baseline Ic 440 ⫾ 20 nA after 10 M IM (n ⫽ 3) in Ca⫹⫹-rich media. Values taken from “slow” plateau Ic in Ca⫹⫹-free media.
component of the slow Ic. It rose from ⫺25.0 ⫾ 8.9 to ⫺173.4 ⫾ 61.3 nA (p ⬍ 0.05). If 10 M IM is applied in the presence of Ca⫹⫹, the Ic response is also biphasic. It is similar to the response observed in the absence of Ca⫹⫹. However, the elevation of the slow Ic plateau was much greater in Ca⫹⫹ rich medium. The Ic increased from ⫺95 ⫾ 45 to ⫺440 ⫾ 20 nA (n ⫽ 3) (p ⬍ 0.01). The plateau remains elevated in the presence of extracellular Ca⫹⫹ (Fig. 2B) and is significantly higher than the plateau Ic in the absence of Ca⫹⫹ (p ⬍ 0.05). 3.2. Thapsigargin dose response Four Xenopus oocytes were used to study the effects of TG on Ic in Ca⫹⫹-free medium (Table 2A). A typical response of the ooctye to TG is shown in Figure 3A. In the absence of TG, the Ic averaged ⫺72.5 ⫾ 21.7 nA. When oocytes were treated with TG, the Ic began to increase slowly. Increasing TG concentration from 0.1 to 5 M, the Ic increased from a baseline of ⫺72.5 ⫹ 21.7 to ⫺155.0 ⫾ 56.8 nA (0.05 ⭐ p ⭐ 0.1. If, however, oocytes ere treated with a single 10 M concentration of TG in the presence of Ca⫹⫹ (Figure 3B), the Ic increased from ⫺50.0 ⫾ 14.1 to ⫺586.7 ⫾ 210 nA (n ⫽ 3; p ⬍ 0.05), reflecting an enhancement of the slow current. If 10 M IM was subsequently administered following the TG treatment, the IM-induced fast Ic com-
Fig. 2. Ca⫹⫹ modulation of IM response in Xenopus oocytes. (A) IM (10 M) in Ca⫹⫹-free solution, initial Ic 80 nA. (B) IM (10 M) in complete Ringer’s solution, initial Ic 50 nA. Box indicates period of IM application. Scale shown.
ponent was still apparent, producing an Ic transient with a peak of ⫺2062.5 ⫾ 736.6 nA (Table 2A). Because TG is known to selectively inhibit SER Ca⫹⫹ uptake, 10 M IM was used to to test for release of sequestered Ca⫹⫹ when the oocyte was bathed with a Ca⫹⫹-free medium. After voltage clamping, six oocytes were treated with 10 M IM. The usual biphasic Ic response was observed (not shown). Peak Ic after IM averaged ⫺791.7 nA while the sustained Ic, however, averaged only ⫺86.7 ⫾ 13.9 nA (Table 2B). The response to increasing TG concentrations under these conditions was still evident as indicated by the Ic changing from ⫺111.7 to ⫺150 nA (Table 2B; p ⬎ 0.1). Subsequently, a second challenge with IM produced no Ic response (⫺158 ⫾ 48 nA). (Compare the last column 10 M IM in Table 2A to Table 2B.) This indicates that in the presence of TG following IM, the Ca⫹⫹ stores are completely depleted. A second challenge with IM reveals the deprived state of the stores. 3.3. Mediation of Ic by BAPTA and BAPTA-AM In an effort to buffer cytoplasmic Ca⫹⫹ fluctuations, either BAPTA or BAPTA-AM was added to the oocyte bathing solution. When BAPTA (10 mM) in a Ca⫹⫹-free medium was applied to oocytes to chelate extracellular Ca⫹⫹, unstable membrane properties were observed. Four control oocytes having membrane potentials of ⫺23.7 ⫾ 2.4 mV, had initial Ics of ⫺222.5 ⫾ 52.7 nA. Upon exposure to extracellular BAPTA for only 5 min, spontaneous extrusion of the cytoplasm (blebbing) and sharp increases in Ic were observed. Application of 10 M BAPTA in Ca⫹⫹ -free media for 2 h produced oocytes with Ics of ⫺4106.4 ⫾ 1343.6 nA. The combined effect of BAPTA and Ca⫹⫹ -free media appear to be deleterious to the egg. Because BAPTA is a charged moiety, it is unlikely to gain access to the cytoplasm. Hence, its effects are probably due to chelation of extracellular Ca⫹⫹. This appears to destabilize the membrane and increase Ic [18]. On the other hand, application of BAPTA-AM (10 M) resulted in oocytes with stable electrical properties (Table 3). BAPTA-AM is nonpolar, and permeates the cellular membrane. Once in the cell, it is deesterfied to BAPTA and chelates intracellular Ca⫹⫹. Oocytes incubated in Ca⫹⫹-rich medium containing 10 M
632
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
Table 2 Effect of thapsigargina (TG) on Ic (nA) in oocytes depending upon pretreatment with ionomycin A: Effect of IM after TG dosagea Ic (nA) at each TG concentration
Mean SEM N a b
Vm (mV)
0 M
0.1 M TG
⫺16.3 1.5 4
⫺72.5 21.7
⫺97.5 22.5
0.3 M TG
1.0 M TG
5.0 M TG
10 Mb TG
⫺112.5 28.7
⫺130.0 36.7
⫺155.0 56.8
⫺2062.5 736.1
In Ca⫹⫹ solutions, Ic after 10 M TG, 586 ⫾ 56.0 nA (n ⫽ 3). Value taken at peak of fast spike.
B: Effect of TG after 10 M IM Ic (nA) at each IM/TG concentration
Mean SEM N a b
Vm (mV)
0 M
10 Ma IM
0.1 M TG
0.3 M TG
1.0 M TG
5.0 M TG
10 Mb TG
⫺31.9 0.3 6
⫺70.0 19.8
⫺86.7 13.9
⫺111.7 30.3
⫺144.2 42.4
⫺144.0 50.5
⫺150.0 48.6
⫺158.0 48.6
Value taken from “slow” plateau Ic in Ca⫹⫹-free media No “fast” current observed.
BAPTA-AM for 5 to 6 h had average Ics of ⫺96.7 ⫾ 26.7 nA. Subsequent application of IM to oocytes in a Ca⫹⫹-free medium increased the Ic only moderately. As IM was increased from 0.1 to 10 M, the Ic of the BAPTA-AM treated oocytes increased from ⫺96.7 ⫾ 23.3 to ⫺163.3 ⫾ 53.3 nA (Table 3; p ⬍ 0.05). Most notedly, the fast transient spike induced by 10 M IM was eliminated and only the slow Ic was observed when BAPTA-AM was presented in Ca⫹⫹-free media. Conse-
quently, it appears that IM increases the slow component of the Ic by changing Ca⫹⫹ concentration near the site of the Cl⫺ current (e.g., Ca⫹⫹-dependent Cl⫺ channels). This Ca⫹⫹ influx is localized to regions near the membrane, and appears to occur too rapidly to be buffered immediately by cytoplasmic BAPTA. 3.4. Synergistic effects of BAPTA-AM and TG To understand the cytoplasmic effect of Ca⫹⫹ from different reservoirs on Ic, the combined effects of BAPTA-AM and TG were examined (Table 4; Figure 4). Oocytes were incubated at room temperatures with 10 M BAPTA-AM in a Ca⫹⫹-rich medium for 5 h, then transferred to a Ca⫹⫹ -free Ringer’s solution. Under this condition, the mean of initial Vm was ⫺20.7 ⫾ 4.3 mV, while Ic averaged ⫺100.0 ⫾ 15.3 nA. Superfusing the voltage-clamped oocytes with 10 M TG modestly increased the Ic to ⫺113.3 ⫾ 18.6 nA. If the oocyte was then treated with 10 M IM, the Ic did not increase significantly (p ⬎ 0.2) in the absence of extracellular Ca⫹⫹ (Table 4A; Figure 4A). In an additional set of experiments, adding Ca⫹⫹ to the extracellular medium Table 3 Effect of 10 M BAPTA-AMa on oocyte response to IM Ic (nA) response at each IM concentrations (with BAPTA)b
Fig. 3. Ca⫹⫹ modulation of TG-induced Ic in Xenopus oocytes. (A) TG (10 M) in Ca⫹⫹-free solution, initial Ic 80 nA. (B) TG (10 M) in Ca⫹⫹-rich solution, initial Ic 40 nA. Arrows indicates time of TG application. Discontinuity of tracing for 720 s show in (A). Scale shown.
Mean SEM N a b
Vm (mv)
0 M
0.1 M
0.3 M
1.0 M
10 M
⫺18.4 1.7 3
⫺96.7 26.7
⫺96.7 23.3
⫺100.0 20.8
⫺150.0 60.3
⫺163.0 53.3
Cell bathed for 5 h in BAPTA-AM. Values of Ic in Ca⫹⫹ rich media.
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
633
cannot sufficiently chelate all IM-induce Ca⫹⫹ entry across the plasma membrane to eliminate activation of a Cl⫺ current.
Table 4 Effect of IM on BAPTA-AM/TG-treated oocytea A: Ca⫹⫹-free solution Ic (nA) Vm (mV) Mean SEM N B:
⫺20.7 4.3 3
Ca⫹⫹-rich
Initial Ic
TG 10 M
—
IM 10 M
⫺100.0 15.3
⫺113.3 18.6
— —
⫺123.3 14.5
solution Ic (nA)
Vm(mV) Mean SEM N a
⫺27.6 5.5 5
Initial Ic
TG 10 M
Ca⫹⫹ 1.8 mM
IM 10 M
⫺142.0 25.5
⫺100.0 14.7
⫺96.0 14.5
⫺260.0 60.4
Oocytes treated for 5 h with 10 M BAPTA-AM.
following TG treatment did not influence the Ic despite having the stores discharged and CCE pathways thereby activated (Table 4B). However, with subsequent administration of IM in a Ca⫹⫹-rich medium, Ic increased to ⫺260.0 ⫾ 60.4 nA (p ⬍ 0.05; Figure4B). As with previous observations in BAPTA-AM–treated cells, the fast transient spike in the Ic response was not evident. The Ic change with IM was thus considered to mimic a slow current increase demonstrating that IM-induced Ca⫹⫹ influx via the ionophore action at the plasma membrane can activate Ic when TG induced Ca⫹⫹ release due to activation of CCE could not. On the other hand, if Ca⫹⫹ were not present in the bath (Table 4A; Figure 4A), the IM-induced slow current transient was eliminated. Thus, BAPTA-AM can effectively minimize the contributions of cytoplasmic calcium stores in regulating Ic by either direct blockade of Ca⫹⫹ release or signalling for Ca⫹⫹ entry via CCE. However, cytoplasmic BAPTA
Fig. 4. Ca⫹⫹ modulation of TG and IM response in BAPTA-AM treated Xenopus oocytes. (A) TG (10 M) and IM in Ca⫹⫹-free solution, initial Ic 80 nA. (B) TG (10 M) and IM in Ca⫹⫹-rich solution, initial Ic 110 nA. Arrows indicate addition of TG and IM. Scale shown.
4. Discussion 4.1. Intracellular Ca⫹⫹ and regulation of Cl⫺ currents Xenopus oocytes have long been known to display Ca⫹⫹-dependent Cl⫺ currents. Elevation of intracellular Ca⫹⫹ levels activates one or more different Cl⫺ currents [1,5 ], as confirmed in the present study. The regulation by Ca⫹⫹ of the underlying Cl⫺ channels and their associated currents has been shown to be complex, in part due to the complexity of the Ca⫹⫹ transport pathways regulating intracellular Ca⫹⫹ levels. Indeed, it appears that the method of altering intracellular Ca⫹⫹ levels, such as that shown for TG or IM in the present study, can produce widely varying Cl⫺ currents. This complexity may reside in the Ca⫹⫹ sensitivity of the various Cl⫺ channels and their spatial relations to the Ca⫹⫹ transport pathways [7]. As heretofore noted, the regulation of intracellular Ca⫹⫹ levels is a highly integrated process that reflects a dynamic balance between Ca⫹⫹ influx and efflux pathways [16]. As illustrated in Figure 1, Ca⫹⫹ influx can arises from Ca⫹⫹ in both the extracellular space and intracellular storage compartments via passive influx through Ca⫹⫹ channels; Ca⫹⫹ efflux from the cytosol occurs via active extrusion through Ca⫹⫹ pumps in the plasma membrane and intracellular membranes of Ca⫹⫹ storage sites. The generation of Ca⫹⫹ signals, reflected as a transient rise in intracellular Ca⫹⫹, typically arises from a transient activation of Ca⫹⫹ influx via activation of one or more Ca⫹⫹ channels, followed by activation of the Ca⫹⫹ efflux mechanisms to restore Ca⫹⫹ levels to near normal levels. Consequently, the transition from rest to active state and back again to a resting state is marked by redistribution of Ca⫹⫹ among the cytosol, the intracellular storage sites and the extracellular medium. The dynamics of Ca⫹⫹ movement among these compartments is rapidly becoming increasingly important to the understanding of cellular homeostasis [14,15]. In Xenopus oocytes, and other nonexcitable cells, the SER is now recognized as the main compartment of the intracellular Ca⫹⫹ stores [16]. The SER Ca⫹⫹ stores work in a coordinated manner with the plasma membrane Ca⫹⫹ channels to provide appropriate Ca⫹⫹ for different physiological processes. IP3 receptors on the SER are themselves regulated Ca⫹⫹ channels that control the release of Ca⫹⫹ from the SER storage site. Various hormonal receptors on the plasma membrane are coupled through the phosphatidyloinostol signal transduction pathways controlling the generation of intracellular IP3 [15]. Stimulation of IP3 production will lead to binding of IP3 to the SER IP3 receptor, leading
634
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
to activation of Ca⫹⫹ efflux from the SER to bring about a rise in intracellular Ca⫹⫹ levels. The release of Ca⫹⫹ from the SER stores can, in turn, activate Ca⫹⫹ channels in the plasma membrane. In the oocyte and many other cells, as the SER becomes depleted, Ca⫹⫹ channels in the plasma membrane are activated, leading to enhanced Ca entry. This process, first described by Putney and coworkers [19,20] has been called CCE or, more recently “store-operated calcium entry.” The exact nature of the mechanism coordinating Ca⫹⫹-channel activity between the SER and plasma membrane is unclear [15,16]. Originally, it was believed that a calcium influx factor (CIF) was released from the SER. Recent studies point to a potential role for phosphorylated glycopyranosides as a possible signal in oocytes [21]. In some instances, the IP3 receptor is directly coupled to CCE, and may control Ca⫹⫹ entry [22]. While release of Ca⫹⫹ from stores and the entry of Ca⫹⫹ via CCE pathways and other pathways appear to play critical roles in regulating Ca⫹⫹-dependent Cl⫺ channels, the two pathways of Ca⫹⫹ entry can appear to display widely differential control of one or more Cl⫺ channels (see below). 4.2. Ca⫹⫹-dependent Cl⫺ channels in oocytes It has long been known that Xenopus oocyte expresses endogenous outward Cl⫺ currents that are Ca⫹⫹ dependent and regulated by intracellular Ca⫹⫹ levels [4]. The present study also verified the presence of Ca⫹⫹ dependent Cl⫺ currents showing that both release of Ca⫹⫹ from internal stores and activation of Ca⫹⫹ entry across the plasma membrane can differentially regulate plasma membrane Ca⫹⫹-dependent Cl⫺ currents. Studies by Boton and coworkers [5] further demonstrated that there may be at least two distinct classes of Ca⫹⫹ dependent Cl currents that could be unmasked by the Ca⫹⫹ ionophore, A23187. In the presence of extracellular Ca⫹⫹, a “slow” Cl⫺ channel is activated. In the presence of low or normal extracellular Ca⫹⫹, a “fast” Cl⫺ channel is activated. Others [8,17] have noted that ionomycin, a Ca⫹⫹ ionophore that may also activate release of Ca⫹⫹ from SER stores, similarly activated a fast and slow Cl⫺ current. However, if the cells were treated with BAPTA to buffer intracellular Ca⫹⫹ levels to low levels, both Cl⫺ currents were suppressed. In contrast, in the absence of BAPTA, but where extracellular Ca⫹⫹ was removed, ionomycin now only activated the fast Cl⫺ current, demonstrating that the fast Cl⫺ current may be predominantly controlled by Ca⫹⫹ release from internal stores and the slow Cl⫺ current controlled by Ca⫹⫹ entry across the plasma membrane. The concept of at least two separate Ca⫹⫹-dependent Cl⫺ currents in oocytes controlled by separate sources of Ca⫹⫹ was most poignantly demonstrated by Hartzell and coworkers [1–3]. Indeed, these workers clearly demonstrated the presence of at least two Cl⫺ currents, one being primarily controlled by release of Ca⫹⫹ from the SER
and the other typically controlled by Ca⫹⫹ entry through endogenous, CCE channels. The present study confirms and extends the concept of two or more Ca⫹⫹-dependent Cl⫺ channels in Xenopus oocytes. Indeed, in the presence of extracellular Ca⫹⫹ ionomycin was demonstrated to induce both a “fast” spiking Cl⫺ current and a second, “slow” plateau Cl current (Figure 2). The slow current was closely controlled by Ca⫹⫹ entry through CCE pathways as unloading of stores via TG treatment was observed to activate the slow Cl⫺ current in the presence of extracellular Ca⫹⫹, but not in the absence of extracellular Ca⫹⫹ (Figure 4). Most important, however, was the observation that the CCE pathway and the slow Cl⫺ current appear to be spatially separated on the oocyte surface. If cells are treated with BAPTA to buffer intracellular Ca⫹⫹ to low levels, activation of Ca⫹⫹ entry through CCE pathways (TG treatment) did not activate the Cl⫺ current, implying that the CCE pathways and Cl⫺ channels are sufficiently spatially separated to allow time for Ca⫹⫹ in the immediate vicinity of the Cl⫺ channel to be fully buffered by BAPTA. In contrast, if Ca⫹⫹ entry were induced by treatment with ionomycin in the presence of extracellular Ca⫹⫹, which should induce Ca⫹⫹ entry with a uniform spatial distribution across the cell surface, the slow Cl⫺ current was now activated even in the presence of BAPTA. This can only occur if BAPTA does not sufficiently buffer Ca⫹⫹ changes in the vicinity of the Cl⫺ channels so that the Cl⫺ channels are activated. It, therefore, seems reasonably to conclude that the Cl⫺ channels and the CCE pathways may be, at least in part, spatially separated. Hartzell and coworkers have previously shown that the slow Cl⫺ channels may be widely distributed throughout both the animal and vegetable poles of the oocyte [1–3,7]. If such is the case, then it is highly probable that the CCE pathways have a more restricted distribution, perhaps only in one pole of the oocyte. Alternatively, if sufficiently spatially separated from the slow Cl⫺ channels, Ca⫹⫹ entry via CCE pathways should be effectively buffered by intracelluar BAPTA. Regardless, the exact distribution of both the Ca⫹⫹ and Cl⫺ channels will require a more detailed assessment of channel localization when suitable probes become available. Then, perhaps, the impact of spatial separation on the interaction of Ca⫹⫹ and Cl⫺ channels in the oocyte can be clearly delineated. Finally, the present study provides further confirmation that ionomycin can act both as a plasma membrane and a SER membrane Ca⫹⫹ ionophore. Previous studies in oocytes [8] and in umbilical-vein endothelial cells [17] have shown that application of ionomycin in the absence of extracellular Ca⫹⫹ can induce release of Ca⫹⫹ from internal stores. The present study likewise demonstrated that in the absence of extracellular Ca⫹⫹, ionomycin treatment induced a rapid release of Ca⫹⫹ from stores. This effect most probably reflects the ionophoretic ac-
C.L. Thurman et al. / Cellular Signalling 12 (2000) 629–635
635
tion of the antibiotic on the SER membrane. Ionomycin is incorporated into the lipid bilayer where it operates as a Ca⫹⫹/H⫹ exchanger [23]. Furthermore, it was shown that in the presence of extracellular Ca⫹⫹, ionomycin induces a rapid influx of Ca⫹⫹ across the plasma membrane as expected, due to its actions as a Ca⫹⫹ ionophore. Hence, the actions of ionomycin include effects on both plasma membranes and intracellular membranes. Interpreting the actions of ionomycin must, therefore, be viewed with caution, as it has effects at several cellular sites. In summary, the present study demonstrates that the regulation of Ca⫹⫹-dependent Cl⫺ channel in Xenopus oocytes is a complex process being controlled by both release of Ca⫹⫹ from internal stores and influx of Ca⫹⫹ across the plasma membrane. Part of this complexity reflects not only the presence of multiple Ca⫹⫹-dependent Cl⫺ channels, but a spatial separation of plasma membrane Cl⫺ channels from the dominant Ca⫹⫹ influx pathway, the CCE pathway. The spatial dependence of the Cl⫺ channels on other Ca⫹⫹ entry pathways and the precise spatial interdependence of the Cl⫺ and Ca⫹⫹ channels is currently not known, but remains to be determined in future studies.
J.S.B.) was provided by the UT-Houston Summer Undergraduate Research Program. This work was completed during the 1999–2000 professional development leave of C.L.T. from UNI. Matt Friez and Eugene M. Barnes, at Baylor College of Medicine, Houston, generously provided oocytes for this research.
Acknowledgments
[19] [20] [21]
This work was supported, in part, by NIDDK 2RO1DK40545 (to R.G.O.) and NSF DUE-9750754 (to C.L.T.). Summer salary (to C.L.T.) during 1999 was provided by fellowships from UNI Graduate College and Applied Technologies funds. Summer salary (to
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]
[22]
[23]
Hartzell HC. J Gen Physiol 1986;108:157–75. Machaca K, Hartzell HC. Biophys J 1998;74:1286–95. Kimura A, Hartzell HC. Am J Physiol 1999;276:C161–75. Miledi R, Parker J. J Physiol (Lond) 1984;357:173–83. Boton R, Dascal N, Gillo B, Lass Y. J Physiol 1989;408:511–34. Parekh AB. Pflugers Arch Eur J Physiol 1995;430:954–63. Gomez-Hernandez J-M, Stuhmer W, Parekh, AB. J Physiol 1997;502:569–74. Yoshida S, Plant S. J Physiol 1992;458:307–18. Petersen CCH, Dupont G. Cell Calcium 1994;16:391–403. Petersen CCH, Berridge MJ. J Biol Chem 1994;51:32246–53. Gillo B, Sealfon SC, Minke B. J Photochem Photobiol 1996;35B:77–82. Goudeau H, Goudeau M. Dev Biol 1998;193:225–38. Parker I, Miledi R. Proc R Soc Lond.1987;231:27–36. Pozzan T, Pizzuto R, Volpe P, Meldolesi J. Physiol Rev 1994;74:595–636. Clapham D. Cell 1995;80:259–68. Alvarez J, Montero M, Garcia-Sancho J. NIPS 1999;14:161–8. Morgan A, Jacob R. Biochem J 1994;300:665–72. Weber W-M, Liebold KM, Reifarth FW, Uhr U, Clauss W. Pflugers Arch Eur J Physiol 1995;429:820–4. Putney JW. Cell Calcium 1986;7:1–12. Putney JW. Cell Calcium 1990;11:611–24. Hartzell HC, Machaca K, Hirayama Y. Mol Pharmacol 1997;51:683–92. Boulay G, Brown MB, Qin N, Jiang M, Dietrich A, Zhu MX, Chen Z, Birnbaumer M, Mikoshiba K, Birnbaumer L. Proc Natl Acad Sci USA 1999;96:14955–60. Wheeler JJ, Veiro JA, Cullis PR. Mol Membr Biol 1994;11:151–7.