Ultrashort nanosecond electric pulses evoke heterogeneous patterns of Ca2+ release from the endoplasmic reticulum of adrenal chromaffin cells

Graphical Abstract

1. Introduction

Nanosecond-duration electric pulses (NEPs) have the ability to penetrate into the cell interior and porate membranes of intracellular Ca²⁺-storing organelles such as the endoplasmic reticulum (ER), causing a rapid release of Ca²⁺ into the cytoplasm that increases intracellular Ca²⁺ levels ([Ca²⁺]_i) [1,2]. Experimentally, Ca²⁺ mobilization from the ER has been demonstrated in a variety of cell types exposed to NEPs varying from 4 to 600 ns in duration. Cell types include non-excitable cells such as Jurkat T-lymphocytes [3,4], CHO cells [5,6], HL-60 cells [7,8] and blood platelets [9], and excitable cells such as cardiomyocytes [10] and neuroblastoma cells [11].

The main strategy that was used to identify the ER as the source of Ca²⁺ was to deplete the ER of Ca²⁺ prior to NEP exposure, typically by pretreating cells with a sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor such as thapsigargin (TG). The role of SERCA is to pump Ca²⁺ from the cytoplasm into the lumen of the ER [12]. Thus, blocking SERCA prevents Ca²⁺ from accumulating both in inositol-1,4,5-triphosphate (IP₃)- and caffeine-sensitive ER Ca²⁺ stores [13]. SERCA inhibition also causes Ca²⁺ stored in the ER to leak into the cytoplasm. When this approach was used to deplete the ER of Ca²⁺, NEP-evoked Ca²⁺ mobilization was found to be either totally abolished [5,11] or significantly reduced [3], thereby establishing that Ca²⁺ release occurs from the ER. Moreover, two studies reported that NEP-evoked Ca²⁺ release from intracellular stores co-localized with the ER [4,11].

In addition to Ca²⁺ mobilization caused by ER membrane permeabilization [3–7], another effect of NEPs involves release of Ca²⁺ from the ER via IP₃ receptors (IP₃Rs). Semenov et al. [5] reported that CHO cells exposed to a single 60 ns pulse at an electric (E)-field amplitude that caused [Ca²⁺]_i to reach a critical level (200 – 300 nM) resulted in Ca²⁺-induced Ca²⁺-release (CICR) mediated by activation of IP₃Rs. This Ca²⁺-triggered response was attributed both to Ca²⁺ release into the cytoplasm due to direct ER membrane permeabilization, and to Ca²⁺ influx as a result of direct electropermeabilization of the plasma membrane. The occurrence of IP₃R-mediated Ca²⁺ release from the ER by NEPs was similarly reported in CHO cells by Tolstykh et al. [14]. They found that a 600 ns pulse caused activation of plasma membrane lipid signaling pathways that resulted in IP₃ accumulation in the cytoplasm, which in turn triggered Ca²⁺ release from the ER. Thus, NEPs can cause release of Ca²⁺ from ER stores both directly as a result of electropermeabilization, and indirectly via the generation of second messengers.

In isolated bovine adrenal chromaffin cells, which serve as a non-transformed cell model of neurosecretion, Ca²⁺ mobilization from internal stores has not been detected in cells exposed to 5 ns pulses applied at an E-field of 5–6 MV/m [15–17]. At this E-field intensity, a single pulse applied to the cells in the presence of extracellular Ca²⁺ causes an increase in [Ca²⁺]_i mediated solely by Ca²⁺ influx via voltage-gated Ca²⁺ channels (VGCCs) [15,17,18]. In contrast, when a single pulse or a pulse train is applied at 5 MV/m in the absence of extracellular Ca²⁺, [Ca²⁺]_i remains unchanged at resting basal levels. Recently we determined that when the E-field was raised to 8 MV/m, a 4% increase in Ca²⁺ fluorescence was detected in 50% of the cells under Ca²⁺-free conditions [19]. We also found that by increasing the E-field amplitude further, both the number of cells in which Ca²⁺ mobilization was observed as well as the magnitude of the rise in [Ca²⁺]_i increased [19]. Based on cell modeling, the ER rather than other Ca²⁺ storing organelles in these cells, in particular secretory granules [20–22], was tentatively identified as the intracellular target of the NEP [19,23].

The goal of the present study was to determine experimentally whether the ER is the intracellular source from which Ca²⁺ is released in adrenal chromaffin cells exposed to a single 5 ns, 17 MV/m pulse. Our approach was to monitor [Ca²⁺]_i by fluorescence imaging, using strategies that included blocking SERCA as well as ones that allowed us to distinguish between Ca²⁺ mobilization from IP₃- and caffeine/ryanodine-sensitive Ca²⁺ stores. In chromaffin cells, each ER store plays a distinct role in Ca²⁺ signaling. Whereas IP₃-sensitive stores mediate Ca²⁺ release from the ER via IP₃Rs following stimulation of G-protein-coupled receptors that leads to IP₃ generation [24–26], caffeine/ryanodine-sensitive Ca²⁺ stores mediate Ca²⁺ efflux from the ER via ryanodine receptors (RyRs) that are activated by Ca²⁺ that enters the cells via VGCCs [25,27–29].

2. Materials and methods

2.1. Chromaffin cell culturing and preparation

Bovine chromaffin cells were isolated from the medulla of fresh adrenal glands by collagenase digestion and maintained in suspension culture in Ham’s F-12 medium supplemented with 10% bovine calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.25 μg/ml fungizone, and 6 μg/ml cytosine arabinoside at 36.5°C under a humidified atmosphere of 5% CO₂ as previously described [30]. Cells were used up until 3 weeks in culture. For all experiments, the large aggregates of cells that form in suspension culture were dissociated into single isolated cells with the protease dispase [31], and the cells then plated onto fibronectin-coated 35 mm glass bottom culture dishes. Cells were used for a period not exceeding 2 days after attachment.

2.2. Fluorescence imaging of intracellular Ca²⁺

Attached cells were incubated with the cell permeant Ca²⁺ indicator Calcium Green-1-AM (1 μM; 480_Ex/535_Em nm) for 45 min at 37°C in a balanced salt solution (BSS) with the following composition: 145 mM NaCl, 5 mM KCl, 1.2 mM NaH₂PO₄, 2 mM CaCl₂, 1.3 mM MgCl₂, 10 mM glucose, 0.1% bovine serum albumin (BSA) and 15 mM HEPES, pH 7.4. After incubation, cells were washed twice with dye-free BSS lacking BSA and for experiments, placed in Ca²⁺-free BSS that contained 1 mM ethylene glycol tetraacetic acid (EGTA). Cells were placed on the stage of a Nikon TE2000 epifluorescence microscope equipped with a 100X objective. Fluorescence images of the cells were captured by an iXonEM + DU-897 EMCCD camera (Andor Technology Ltd., Belfast, UK) using the open source microscopy software Micro-Manager (version 1.4, Vale Lab, UCSF, San Francisco, CA). The exposure time of the camera was set to 100 ms and images captured at a rate of ~8 frames/s. Baseline Ca²⁺ fluorescence of the cells was monitored 10 s prior to NEP application and continued for 30 s after the pulse. Sequences were analyzed using the public domain image-processing program ImageJ (https://imagej.nih.gov/ij/). Background fluorescence was subtracted from the fluorescence of the cells, and all fluorescence data were normalized to the intensity value measured just before the stimulus was applied (F/F₀). Bright field images of each cell were obtained at the start and end of an experiment. All experiments were performed at ambient room temperature.

2.3. ER Ca²⁺ store depletion

Prior to NEP exposure, cells were incubated for at least 15 min at room temperature with one or more of the following agents: caffeine, 50 mM; ryanodine, 1 μM; cyclopiazonic acid (CPA), 30 μM and 2-aminoethyl diphenylborinate (2-APB), 100 μM. Caffeine was prepared in Ca²⁺-free BSS (containing 1 mM EGTA) in which an equivalent amount of NaCl (25 mM) was removed to maintain isotonicity. Stock solutions of 2-APB, CPA and ryanodine were prepared in dimethyl sulfoxide (DMSO) and added to the Ca²⁺-free, EGTA BSS immediately before use such that the final concentration of DMSO did not exceed 0.1%. All agents were delivered to the cells via perfusion. As a control for IP₃-sensitive store depletion, the cholinergic receptor agonist carbachol, dissolved in Ca²⁺-free, EGTA BSS, was applied directly to a single cell by a pressure ejection pipette.

2.4. NEP exposure

A 5 ns pulse (Fig. 1A) was delivered to a cell via a pair of electrodes consisting of two cylindrical gold-plated tungsten rods (127 μm diameter) spaced 100 μm apart. The tips of the electrodes were immersed in the BSS bathing the cells and placed 40 μm above the bottom of the dish using a motorized MP-225 micromanipulator (Sutter Instruments, Novato, CA), with the target cell located at the center of the gap between the electrode tips (Fig. 1B). Single pulses that produced E-fields of 17 MV/m at the location of the cell were generated by a custom-fabricated nanosecond pulse generator (Transient Plasma Systems, Torrance, CA). The E-field distribution in the vicinity and at the location of the target cell (Fig. 1C) was computed using the commercially available Finite-Difference Time-Domain (FDTD) software package SEMCAD X (version 14.8.5, SPEAG, Zurich, Switzerland). Delivery of pulses was triggered externally by a program written in LabVIEW. A cell was exposed to the E-field only once.

Fig. 1.

NEP exposure. (A) Representative waveform of the 5 ns, 3.5 kV electric pulse that was delivered to the cells, resulting in an E-field of 17 MV/m at the location of the cell. The pulse duration corresponds to the width at half maximum. (B) Photomicrograph (60X) of a chromaffin cell located between the electrode tips. (C) Computed E-field distribution in the vicinity and at the location of a cell (black square between the electrodes) exposed to a 5 ns pulse.

2.5. Statistical Analysis

Experiments were repeated at least once using cells from different days in culture and different cell preparations. The normalized Ca²⁺ responses of the cells are represented as the mean ± standard error (s.e.). An unpaired Student’s t test was used to determine the statistical significance between two means. p-values < 0.05 were considered statistically significant.

2.6. Reagents

Calcium Green-1-AM was purchased from Molecular Probes (Eugene, OR). Ham’s F-12 medium, the antibiotic-antimycotic and dispase II were obtained from Gibco Laboratories (Grand Island, NY). Bovine calf serum was purchased from Gemini Bio-products (West Sacramento, CA), collagenase B was obtained from Roche Diagnostics (Indianapolis, IN), carbachol, caffeine, 2-APB and CPA were purchased from Sigma-Aldrich (St. Louis, MO), and ryanodine was obtained from AG Scientific Inc. (San Diego, CA). All other chemicals and reagents were reagent grade and purchased from standard commercial sources.

3. Results and Discussion

3.1. NEP exposure caused two patterns of Ca²⁺ release from internal stores

As previously reported, an E-field amplitude of at least 8 MV/m is required to cause detectable (> 4% increase) mobilization of Ca²⁺ from internal stores in chromaffin cells exposed to a 5 ns electric pulse [19]. To best study this intracellular effect, chromaffin cells were exposed to a single pulse applied at an E-field of 17 MV/m, which is close to the highest E-field amplitude achievable with our exposure setup. We found that in the absence of extracellular Ca²⁺, 94% of exposed cells underwent an instantaneous rise in [Ca²⁺]_i that varied in magnitude from 1.08 to 1.35 (peak rise: 1.18 ± 0.01, n = 46). Thus, even at this high E-field intensity, a few cells still did not show evidence of Ca²⁺ mobilization when exposed to a pulse (Fig. 2A). In addition, the magnitude of the changes in [Ca²⁺]_i was smaller than that evoked by agonists that cause Ca²⁺ release from IP₃- and caffeine-sensitive stores (average peak rise 1.35 versus 1.88 and 2.12, respectively; see Section 3.4) demonstrating that the pulse is not very efficient at mobilizing Ca²⁺ from internal stores.

Fig. 2.

Ca²⁺ responses in chromaffin cells exposed to a single 5 ns, 17 MV/m pulse in the absence of extracellular Ca²⁺. (A) Averaged responses ± s.e. (n = 3) of cells that did not respond to the pulse. (B) Averaged cell responses ± s.e. (n = 24) of cells exhibiting single Ca²⁺ transients. The arrow indicates when the pulse was delivered to the cells. Here and in all subsequent figures, F/F₀ was calculated as described in Methods. Error bars not shown are within the thickness of the line. Fluorescence photomicrographs of chromaffin cells exposed to a single 5 ns pulse showing (C) global versus (D) more localized Ca²⁺ responses before and after pulse application. The red circles represent the areas with increased fluorescence.

Notably, the characteristics of NEP-evoked Ca²⁺ responses exhibited one of two distinct patterns. As shown in Fig. 2B, the pattern observed in half of the cells was a single Ca²⁺ transient (peak rise: 1.17 ± 0.01, n = 24). The transient nature of the Ca²⁺ response resembled that reported in other cell types [4,5,7] and tended to occur heterogeneously inside the cell. Specifically, in some cases the increase in fluorescence was global (Fig. 2C) while in others it was more restricted in area (Fig. 2D), spreading across only one to two thirds of the area of the cell rather than over the entire cell. The latter was not further investigated because of the relatively poor spatial resolution of standard epifluorescence imaging. In the remaining half of exposed cells, the Ca²⁺ response was characterized by an instantaneous transient rise in [Ca²⁺]_i that was followed by multiple, short-lived Ca²⁺ spikes (Fig. 3A). Of note was the observation that Ca²⁺ spiking activity was triggered by E-field amplitudes greater than 15 MV/m, and was absent when the cells were not subjected to a pulse (Fig. 3B). To our knowledge, this behavior has not been observed in any other cell type. However, it is possible that such behavior was missed either because Ca²⁺ responses were averaged [5] or that the length of time that Ca²⁺ responses were being recorded was too short to allow for such events to be captured [11].

Fig. 3.

Ca²⁺ responses in chromaffin cells exposed to a single 5 ns, 17 MV/m pulse. (A) Representative traces of Ca²⁺ spikes (n = 22) in chromaffin cells exposed to a single 5 ns pulse in the absence of extracellular Ca²⁺. The arrow indicates when the pulse was delivered to the cells. (B) Averaged responses ± s.e. (n = 12) in control cells that were not subjected to the pulse displaying no Ca²⁺ activity in the absence of external Ca²⁺.

3.2. Inhibiting SERCA abolished Ca²⁺ mobilization in the majority of cells

Our approach to determine whether the ER was the intracellular store from which Ca²⁺ was released by NEPs was the same as that employed in other laboratories, which was to block SERCA [3,5,11]. We used the SERCA inhibitor CPA at 30 μM [32]. As shown in Fig. 4A, perfusing cells with CPA caused a gradual release of Ca²⁺ from the ER into the cytoplasm that peaked around 80 s once the inhibitor reached the cells (peak rise: 1.38 ± 0.03, n = 5). Ca²⁺ release continued for at least 5 min. Based on this time course, cells were preincubated with the blocker for a minimum of 15 min prior to NEP exposure.

Fig. 4.

NEP-evoked Ca²⁺ responses in chromaffin cells treated with CPA. (A) Representative trace of Ca²⁺ released from the ER in response to CPA (30 μM) applied by perfusion (gray bar). The shutter was closed at 300 s and opened 70 s later to minimize photobleaching. (B) and (C) Representative traces of immediate (n = 2) and delayed (n = 6) Ca²⁺ responses, respectively, observed in CPA-treated cells exposed to a 5 ns pulse (arrow) in the continued presence of CPA (gray bar).

CPA treatment decreased the number of cells undergoing an increase in [Ca²⁺]_i in response to a 5 ns, 17 MV/m pulse. Out of 33 cells, 25 cells (76%) did not show a Ca²⁺ response. Similar results (data not shown) were obtained using the SERCA blocker TG at 1 μM [32]. These results indicate that in the majority of chromaffin cells, the ER was the source of the Ca²⁺ that was mobilized by the NEP. However, that some cells still showed Ca²⁺ responses contrasted with results reported for other cell types where SERCA inhibitors were completely effective for abolishing the Ca²⁺-mobilizing effects of an NEP [5,11]. This meant either that the source of released Ca²⁺ in some chromaffin cells might have involved another Ca²⁺-storing organelle, such as secretory granules [20–22], or that ER Ca²⁺ stores might not have been fully depleted. The latter possibility is addressed in Section 3.4.

3.3. SERCA inhibition uncovered additional patterns of Ca²⁺ mobilization

In the 8 out of 33 SERCA-inhibited cells in which an NEP evoked a Ca²⁺ response, the increase in [Ca²⁺]_i observed in some cells exhibited patterns that differed from the two previously described. Whereas one cell exhibited an immediate release of Ca²⁺ that was characterized by a single, small magnitude Ca²⁺ transient (Fig. 4B; blue trace, peak rise: 1.04), similar to that shown in Fig. 2B, another cell (Fig. 4B; red trace) exhibited an instantaneous small magnitude Ca²⁺ transient that was followed by multiple Ca²⁺ transients in which Ca²⁺ levels did not return to baseline, leading to a sustained increase in [Ca²⁺]_i. The mechanism underlying the long-lived nature of the increase in [Ca²⁺]_i is unknown. The other 6 cells exhibited yet another type of Ca²⁺ response in which Ca²⁺ release occurred at various times after the pulse was applied. As shown in Fig. 4C, these delayed Ca²⁺ responses consisted of either a single Ca²⁺ transient (peak rise: 1.14 ± 0.04, n = 4) or multiple Ca²⁺ transients (n = 2) similar to those shown in Fig. 4B wherein [Ca²⁺]_i remained elevated. In all probability, they represent a secondary Ca²⁺ mobilizing effect of the pulse since spontaneous Ca²⁺ release from internal stores was never observed in cells monitored in Ca²⁺-free BSS (n = 12; Fig. 3B).

3.4. Residual Ca²⁺ mobilization in SERCA-inhibited cells was attributed to incomplete depletion of caffeine-sensitive stores

As indicated in the previous two sections, 24% of CPA-treated chromaffin cells still showed intracellular Ca²⁺ release following NEP exposure, possibly because ER Ca²⁺ stores may not have been completely emptied. To assess the effectiveness of SERCA blockers for depleting ER Ca²⁺ stores under our experimental conditions, we determined first the extent to which IP₃-sensitive stores retained Ca²⁺ after cells were treated with CPA. For the assessment, CPA-treated cells were stimulated with the cholinergic receptor against carbachol (500 μM), which in the absence of extracellular Ca²⁺ causes the cells to undergo an increase in [Ca²⁺]_i that is due solely to activation of G-protein-coupled muscarinic receptors and subsequent IP₃ generation [33]. As shown in Fig. 5A for control (non-CPA treated) cells, carbachol evoked a transient rise in [Ca²⁺]_i that ranged in amplitude from 1.51 to 2.42 (1.88 ± 0.1, n = 9). The magnitude of the response was much greater than that evoked by a 5 ns pulse (Fig. 2B). Note also that Ca²⁺ was released from the ER with a delay of at least 2 s. A delayed Ca²⁺ response is consistent with observations obtained previously by us [19] and other groups [24,26] and is explained by the time needed for IP₃ generation to reach a threshold level for eliciting Ca²⁺ release from intracellular stores via IP₃Rs. In contrast to control cells, carbachol failed to elicit a rise in [Ca²⁺]_i in CPA-treated cells (Fig. 5B; 24 out of 24 cells), indicating that IP₃-sensitve stores were fully depleted of Ca²⁺ by CPA.

Fig. 5.

Differential effect of CPA on depleting IP₃-sensitive versus caffeine-sensitive Ca²⁺ stores. (A) Averaged Ca²⁺ responses ± s.e. (n = 6) of chromaffin cells to carbachol (500 μM) delivered to the cells via a pressure ejection pipette (arrow). Error bars not shown are within the thickness of the line. (B) Averaged fluorescence traces ± s.e. (n = 24) showing the lack of a Ca²⁺ response to carbachol (arrow) in the presence of CPA (gray bar). Error bars not shown are within the thickness of the line. (C) Representative trace of Ca²⁺ released from the ER in CPA-treated cells (upper solid line) in response to caffeine (50 mM) applied by perfusion (lower solid line). In both B and C, cells were pretreated with 30 μM CPA for 15 min prior to the experiment, which was carried out in the continued presence of CPA. (D) Representative traces of Ca²⁺ released from the ER in non-CPA-treated cells in response to caffeine (50 mM) applied by perfusion (solid line).

For determining if caffeine-sensitive ER stores may also have retained Ca²⁺, CPA-treated cells were exposed to a caffeine concentration (50 mM) that was previously shown to be necessary to fully deplete the caffeine-sensitive stores in chromaffin cells [27,28]. As shown in Fig. 5C, caffeine caused significant release of Ca²⁺ from the ER in CPA-treated cells that resembled that in non-CPA-treated cells (Fig. 5D). In both cases, Ca²⁺ release was characterized by a rapid, global rise in [Ca²⁺]_i that reached an amplitude of 1.84- to 2.55-fold over baseline (2.15 ± 0.12, n = 5 CPA-treated cells), and 1.34- to 2.51-fold over baseline (2.12 ± 0.13, n = 9 non-CPA-treated cells) and which continued for 60 s. Thus, caffeine-sensitive ER stores still retained significant Ca²⁺ under conditions in which SERCA was inhibited, a finding that has been reported by some laboratories [32,34]. The differential efficacy of SERCA inhibitors to fully deplete IP₃-sensitive but not caffeine-sensitive ER stores in chromaffin cells could be attributed to the presence of two Ca²⁺-ATPases on the ER, SERCA 2b and SERCA 3 isoenzymes of Ca²⁺ pumps that differ both in their sensitivities to SERCA inhibitors and in their distribution on each ER Ca²⁺ store [13]. SERCA 2b, a 140K Ca²⁺-ATPase-like protein, has been associated with IP₃-sensitive stores whereas SERCA 3, a 100K Ca²⁺-ATPase-like protein, has been associated with caffeine-sensitive stores [13,35]. Our results showing full depletion of IP₃-sensitive stores by CPA is consistent with only the 140K Ca²⁺-ATPase SERCA 2b being effectively blocked. Moreover, that caffeine-sensitive ER stores were not fully depleted could explain why NEP-evoked Ca²⁺ mobilization still occurred in some CPA-treated cells, which we next investigated.

3.5. Ca²⁺ mobilization was totally abolished when ER stores were fully depleted of Ca²⁺

To fully deplete the ER of Ca²⁺, we used a strategy in which SERCA was blocked with CPA and RyRs activated with two agonists, caffeine and ryanodine [27,28,36–38]. Although caffeine alone is very effective for triggering Ca²⁺ release from the ER of chromaffin cells via RyRs (Fig. 5C and [27–29]), ryanodine, when used at a low concentration (i.e., 1–10 μM), will have the added effect of locking irreversibly RyRs in an open sub-conductance state [27,28,37,38], further ensuring that caffeine-sensitive ER stores are fully depleted of Ca²⁺ [28,37,39]. Wang et al. [10] used a similar strategy to show that the sarcoplasmic reticulum was the source of Ca²⁺ mobilized in cardiac myocytes exposed to 4 ns pulses.

The sequence in which chromaffin cells were treated with these agents is shown in the schematic diagram in Fig. 6A. Cells were first perfused for 15 min with Ca²⁺-free BSS containing CPA (30 M) and ryanodine (1 μM). The latter was included with CPA to allow sufficient time for the agonist to bind to the inactive RyR channel (t_1/2 ≈ 25 min) [40]. Caffeine (50 mM) was then added to the CPA/ryanodine-containing Ca²⁺-free BSS and the cells perfused with all three agents for 15 min. Similar to the results shown in Fig. 5C, the addition of caffeine caused the rapid release of Ca²⁺ from the ER that lasted for 45 to 130 s (Fig. 6B). To verify that ER caffeine-sensitive Ca²⁺ stores were now depleted, the cells were placed in Ca²⁺-free BSS and after 2 min, CPA, ryanodine and caffeine were reapplied to the cells. In 9 out of 9 cells tested, there was no additional release of Ca²⁺ from the ER (data not shown), indicating that full ER depletion of Ca²⁺ was achieved.

Fig. 6.

Lack of effect of a 5 ns pulse on [Ca²⁺]_i in chromaffin cells treated with CPA, caffeine and ryanodine. (A) Schematic diagram showing the order in which cells were treated with CPA, ryanodine and caffeine to fully deplete the ER of Ca²⁺. (B) Representative traces of release of Ca²⁺ from the ER by caffeine (lower gray bar) in the continued presence of CPA and ryanodine (upper gray bar). (C) Averaged fluorescence traces ± s.e. (n = 46) showing the lack of effect of a 5 ns, 17 MV/m pulse (arrow) applied to cells treated according to the strategy depicted in (A) in the continued presence of CPA, ryanodine and caffeine (gray bar). Error bars not shown are within the thickness of the line.

After chromaffin cells were subjected to this ER Ca²⁺-depleting strategy, exposing the cells to a 5 ns pulse did not lead to a rise in [Ca²⁺]_i (Fig. 6C) in any cell examined (46 out of 46 cells). These results not only indicated that the ER was the sole source of the Ca²⁺ that was mobilized by the NEP but that both IP₃- and caffeine-sensitive ER Ca²⁺ stores were involved.

3.6. Ca²⁺ spikes correlated with IP3R activation

Because NEP-evoked Ca²⁺ responses included both a single Ca²⁺ transient and multiple Ca²⁺ spikes, it was possible that each response was mediated by a particular ER Ca²⁺ store. We first assessed the involvement of IP₃-sensitive Ca²⁺ stores by blocking IP₃Rs with 2-APB (100 μM) [41] prior to NEP exposure. As a control for verifying IP₃R inhibition, carbachol (500 μM) was applied to cells treated with 2-APB. Carbachol failed to elicit a Ca²⁺ rise in 15 out of 15 cells under these conditions (Fig. 7A), confirming that IP₃Rs were completely blocked.

Fig. 7.

Effect of 2-APB on Ca²⁺ responses evoked by carbachol or a 5 ns pulse. In (A-C), cells were pretreated with 100 μM 2-APB for 15 min prior to the experiment, which was carried out in the continued presence of 2-APB (gray bar). (A) Averaged fluorescence traces ± s.e. (n = 15) showing the lack of a Ca²⁺ response to carbachol (500 μM), which was applied by a pressure ejection pipette (arrow). (B) Averaged Ca²⁺ responses ± s.e. (n = 24) evoked by a 5 ns pulse (arrow). Error bars not shown are within the thickness of the line. (C) Fluorescence traces of the two cells that showed Ca²⁺ spikes evoked by the pulse (arrow).

When chromaffin cells were exposed to a single 5 ns pulse in the presence of 2-APB, there was an instantaneous release of Ca²⁺ from ER stores in 77% of the cells tested. In the majority of the cells (24 of 26 cells), the response consisted of a single Ca²⁺ transient (Fig. 7B) resembling that observed in the absence of 2-APB (Fig. 2B). We noted, however, that the magnitude of the Ca²⁺ transient (peak rise: 1.11 ± 0.02, n = 24) was less than that obtained when IP₃Rs were not blocked (peak rise: 1.18 ± 0.01), a difference that reached statistical significance (p-value < 0.05, n = 24). We attributed this difference as well as the fewer number of cells responding to the pulse to possible non-specific actions of 2-APB which, as shown in one cell type, can include an inhibitory effect on SERCA [42]. It is also known that 2-APB exerts inhibitory activity on non-selective cation channels, in particular those involved in store-operated Ca²⁺ entry [43]. However, this was not a concern in our conditions since all our experiments were carried out using Ca²⁺-free BSS.

Although IP₃R activation did not appear to be responsible for the initial Ca²⁺ transient, it was on the other hand associated with the appearance of Ca²⁺ spikes. In cells in which IP₃Rs were blocked, the percentage of cells exhibiting Ca²⁺ spiking behavior was substantially reduced. Out of 26 cells that were treated with 2-APB and that underwent a rise in [Ca²⁺]_i, only two cells (8%) exhibited Ca²⁺ spikes (Fig. 7C) versus 48% of the cells not treated with 2-APB (Fig. 3A). Given that NEP exposure has been reported to perturb the plasma membrane, resulting in IP₃ generation, IP₃Rs activation and Ca²⁺ release from IP₃-sensitive stores [14], this signaling cascade could presumably underlie the Ca²⁺ spikes triggered by a 5 ns, 17 MV/m pulse. In this regard, Ca²⁺ spikes were only detected when the E-field amplitude reached a critical value of 15 MV/m, suggesting that the E-field must be of sufficient magnitude to cause plasma membrane perturbations leading to the generation of IP₃.

Other considerations regarding the involvement of IP₃ stores in Ca²⁺ spiking activity are the following. First, Ca²⁺ spikes occurred with a delay, which is consistent with Ca²⁺ release from the ER via a mechanism that is mediated by a second messenger such as IP₃. Second, Ca²⁺ spikes originated from discrete areas within a cell, which is consistent with IP₃-generating agonists causing localized increases in [Ca²⁺]_i rather than increases that are global in nature [35]. In fact, it has been shown that IP₃-generating agonists evoke increases in [Ca²⁺]_i that co-localize with IP₃-sensitive Ca²⁺ stores and the 140K Ca²⁺-ATPase [24,13,44]. Due to the small amplitude and transient nature of Ca²⁺ spikes coupled with the limitations of our imaging setup to monitor with sufficient resolution where the changes in Ca²⁺ occur inside the cell, we were unable to obtain spatial information regarding the origin of Ca²⁺ spikes that would allow us to confirm co-localization of these events with IP₃-sensitive stores. Nevertheless, it should be noted that Ca²⁺ responses to carbachol tended to be spatially delimited rather than global (results not shown).

In the two cells that still exhibited Ca²⁺ spikes when IP₃Rs were blocked, a possible scenario that could explain such activity is a CICR mechanism involving Ca²⁺ release from caffeine-sensitive stores [28]. Because these ER Ca²⁺ stores would be fairly intact in the presence of 2-APB, an NEP could cause Ca²⁺ to be released via ER membrane poration where Ca²⁺ could act on RyRs of closely apposed caffeine-sensitive stores, resulting in subsequent Ca²⁺ spikes.

3.7. Instantaneous increases in [Ca²⁺]i were due to Ca²⁺ release from caffeine-sensitive stores

The initial Ca²⁺ transient evoked by a 5 ns pulse was not abolished when IP₃Rs were blocked, suggesting that Ca²⁺ mobilization most likely originated from caffeine-sensitive stores. This possibility was investigated by depleting this ER Ca²⁺ store prior to NEP exposure, using the strategy for Ca²⁺ depletion shown in Fig. 8A in which cells were perfused for 15 min with ryanodine, then with ryanodine plus caffeine for another 15 minutes. As shown previously, caffeine caused a rapid release of Ca²⁺ from the ER (Fig. 8B; peak rise: 2.26 ± 0.12, n = 8) that lasted 45 to 130 s.

Fig. 8.

Effect of depleting caffeine-sensitive stores on Ca²⁺ responses evoked by a 5 ns pulse. (A) Schematic diagram showing the order in which ryanodine and caffeine were applied to the cells. (B) Representative traces showing release of Ca²⁺ from the ER by caffeine in cells pretreated with ryanodine for 15 min. (C) Representative Ca²⁺ responses (n = 11) evoked by a 5 ns pulse (arrow) in cells treated with ryanodine and caffeine. (D) Averaged fluorescence traces ± s.e. (n = 14) showing the lack of effect of a 5 ns pulse (arrow) on [Ca²⁺]_i in cells treated with ryanodine, caffeine and 2-APB (100 μM). Error bars not shown are within the thickness of the line.

When cells were exposed to a single 5 ns pulse under these conditions, only 11 out of 34 cells underwent an increase in [Ca²⁺]_i. Importantly, the rise in [Ca²⁺]_i was not instantaneous but instead occurred with a variable delay following the application of the pulse. Responses varied from a single (n = 4) to multiple (n = 7) Ca²⁺ transients (Fig. 8C). These results support the conclusion that the initial rise in [Ca²⁺]_i was mediated by Ca²⁺ release from caffeine-sensitive stores. In addition, since IP₃-sensitive stores were intact, the appearance of delayed Ca²⁺ spikes when caffeine-sensitive stores were depleted is consistent with Ca²⁺ release from the ER taking place via IP₃Rs, as previously discussed. This latter observation was verified by treating the cells with a combination of ryanodine, caffeine and 2-APB. Under these conditions, we found that in 14 out of 14 cells exposed to a 5 ns pulse, none of the cells underwent an increase in [Ca²⁺]_i, as exemplified by the trace shown in Fig. 8D. These results not only provided additional evidence that the ER was the source of Ca²⁺ mobilized by a 5 ns, 17 MV/m pulse, but also allowed us to propose a role for each specific ER Ca²⁺ store in NEP-evoked Ca²⁺ responses. Ca²⁺ spikes originated from IP₃-sensitive Ca²⁺ stores, resulting in localized Ca²⁺ release [35,44] that was delayed in onset, most likely reflecting the time course required to generate IP₃. On the other hand, the instantaneous increase in [Ca²⁺]_i following NEP exposure correlated with release of Ca²⁺ from caffeine-sensitive Ca²⁺ stores, causing a rise in [Ca²⁺]_i throughout the cytoplasm in most cells and thus mimicking global Ca²⁺ release triggered by caffeine observed by us and others [35]. We attributed the immediate Ca²⁺ mobilizing effect of the NEP to ER membrane poration as reported by others [3–5] and previously suggested by our cell modeling studies [19].

4. Conclusion

Neuroendocrine adrenal chromaffin cells possess both IP₃-sensitive and caffeine-sensitive ER Ca²⁺ stores. The present study shows that when chromaffin cells are exposed to a single 5 ns, high intensity (17 MV/m) NEP, Ca²⁺ is released from each store in a manner that differs with respect to the mechanism involved, as well as to the pattern and time course of released Ca²⁺. As reported for other types of cells, we found that NEP exposure causes chromaffin cells to undergo an instantaneous, transient rise in [Ca²⁺]_i that can be attributed to permeabilization of the ER membrane. For reasons not yet understood, the permeabilizing effect of the NEP appears to be confined mainly to the ER membrane of caffeine-sensitive Ca²⁺ stores. We also found that NEP exposure evokes, in about half of the cells, multiple Ca²⁺ spikes that are delayed in onset. Ca²⁺ spiking is attributed to the generation of IP₃ and hence the release of Ca²⁺ from IP₃-sensitive stores via IP₃Rs. These results are the first to demonstrate that a single NEP can differentially affect ER Ca²⁺ stores, resulting in heterogeneous Ca²⁺ responses. Depending on the cell type, the occurrence of two different Ca²⁺-mobilizing responses can have functional significance. In chromaffin cells, for example, Ca²⁺ released from the caffeine-sensitive pool could enhance Ca²⁺-dependent catecholamine release by increasing the concentration of Ca²⁺ at sites on the plasma membrane where exocytosis occurs. If IP₃-mediated Ca²⁺ release also occurred, catecholamine release could be enhanced further. Whether a differential effect of NEPs on the ER occurs in other types of cells possessing both types of ER Ca²⁺ stores remains to be determined.

Highlights.

• 5 ns pulses cause diverse patterns of Ca²⁺ release from the ER of chromaffin cells

• Patterns include a single Ca²⁺ transient or multiple Ca²⁺ spikes

• IP₃- and caffeine-sensitive ER Ca²⁺ stores are differentially involved

• Ca²⁺ release due to membrane poration occurs from caffeine-sensitive stores

• Ca²⁺ spikes are associated with Ca²⁺ release mediated by IP₃ receptors