Methods for Studying Store-Operated Calcium Entry

Abstract

Activation of surface membrane receptors coupled to phospholipase C results in the generation of cytoplasmic Ca²⁺ signals comprised of both intracellular Ca²⁺ release, and enhanced entry of Ca²⁺ across the plasma membrane. A primary mechanism for this Ca²⁺ entry process is attributed to store-operated Ca²⁺ entry, a process that is activated by depletion of Ca²⁺ ions from an intracellular store by inositol 1,4,5-trisphosphate. Our understanding of the mechanisms underlying both Ca²⁺ release and store-operated Ca²⁺ entry have evolved from experimental approaches that include the use of fluorescent Ca²⁺ indicators and electrophysiological techniques. Pharmacological manipulation of this Ca²⁺ signaling process has been somewhat limited; but recent identification of key molecular players, STIM and Orai family proteins, has provided new approaches. Here we describe practical methods involving fluorescent Ca²⁺ indicators and electrophysiological approaches for dissecting the observed intracellular Ca²⁺ signal to reveal characteristics of store-operated Ca²⁺ entry, highlighting the advantages, and limitations, of these approaches.

I. Introduction

In many cell types, activation of hormone, neurotransmitter, or growth factor receptors coupled to phospholipase-C results in the breakdown of phosphatidylinositol 4,5-bisphoshate, resulting in production of inositol 1,4,5-trisphosphate (IP₃) which stimulates a Ca²⁺ signaling process that is biphasic [1,2]. This biphasic response involves the release of Ca²⁺ ions from an intracellular organelle, the endoplasmic reticulum (ER) or a specialized component of the ER, followed by the entry of Ca²⁺ ions across the plasma membrane. Much is known about the first phase of intracellular Ca²⁺ release from an intracellular organelle, an effect mediated by IP₃ acting on its own receptor, the IP₃ receptor [3]. Until recently, however, the mechanisms regulating the Ca²⁺ entry process have been less well understood, although there is a basic and well established concept for this second phase of Ca²⁺ entry. That is, the degree of emptying of the Ca²⁺-storage organelle, generated by intracellular Ca²⁺ release, initiates a retrograde signaling process that regulates the rate of Ca²⁺ entry across the plasma membrane. This process is known as capacitative Ca²⁺ entry or store–operated Ca²⁺ entry (SOCE) [4]. The signaling processes underlying SOCE have been the subject of intense study for more than 20 years, yet only recently have the key molecular components been identified. Stim family proteins (Stim1 and 2), appear to function as Ca²⁺ sensors within the ER, and Orai family proteins (Orai1, 2 and 3, also known as CRACM1, 2 and 3) function as SOC channels in the plasma membrane (for recent reviews, see [5,6,7,8]).

Identification of the molecular makeup of the SOCE pathway has been facilitated by optical techniques utilizing fluorescent Ca²⁺ indicators [9,10]. Although the primary focus of this volume is optical techniques, for the study of Ca²⁺ influx mechanisms it is almost always advisable to combine the use of fluorescent indicators with electrophysiological techniques, and these will be discussed in this review as well. In developing these methodological approaches it is important to discriminate SOCE from other pathways that may influence [Ca²⁺]_i. For example, additional mechanisms initiated by phospholipase C activation that can regulate Ca²⁺ entry not related to SOCE have been reported in non-excitable cells [11].

II. Fluorescence-based Measurements of SOCE

Fluorescence-based measurements of [Ca²⁺]_i have provided a robust and widely used technique for monitoring Ca²⁺ signaling processes, including SOCE. These fluorescence based techniques have been productive due to the availability of a broad range of Ca²⁺ indicators that can be easily introduced into intact cells, and an extensive range of ‘turnkey’ equipment for measuring Ca²⁺ with good temporal and spatial resolution.

A. Fluorescent Ca²⁺ Indicator Selection

The choice of fluorescent Ca²⁺ indicator is the foundation for a successful study of Ca²⁺ signaling. It can influence the spatial and temporal information that one can collect, and the choices one has for analyzing a response. This selection will also be influenced by the available equipment for measuring fluorescence, such as excitation wavelength selection.

Single Wavelength and Ratiometric Ca²⁺ Indicators

The working core of a Ca²⁺ indicator is centered around its ability to reversibly bind Ca²⁺ ions, with affinities that lie within the physiological range of cytoplasmic [Ca²⁺]_i. With most Ca²⁺ indicators, such as fluo dyes, the level of [Ca²⁺]_i can be directly monitored as a change in fluorescence intensity of the indicator where, usually, elevation of [Ca²⁺]_i leads to higher intensity of fluorescence at a single wavelength (Fig.1) . Alternatively, there are some Ca²⁺ indicators that exhibit a spectral shift upon binding Ca²⁺ ions [12] such as the Indo and fura dyes (Fig.1). Using spectral shift indicators has the enormous advantage of providing ratiometric [Ca²⁺]_i measurements that are independent of the concentration of the Ca²⁺ indicator present in the cells.

Figure. 1. Spectral Characteristic of Ratiometric and Single Wavelength Ca²⁺Indicators.

(A) Emission spectra of fura-5F (free-acid form) recorded at 530nm while scanning excitation wavelengths from 320-400nm. With 10 μM fura-5F dissolved in buffer containing 100 mM KCl, 20 mM HEPES, pH 7.2, switching between ‘low Ca²⁺‘ (buffer +200 μM BAPTA) and ‘high Ca²⁺’ (buffer + 1 mM CaCl₂) conditions demonstrates the spectral shift characteristics of fura-5F. This [Ca²⁺] change can be quantified by ratioing the emission fluorescence measured at 340nm and 380nm excitation wavelengths, as shown in (B).

(C) Spectra for fluo-4 (free-acid form) was recorded with excitation at 485nm while scanning emission wavelengths from 500-600nm. Using the same buffers as in (A), switching between ‘low Ca²⁺‘(buffer +200 μM BAPTA) and ‘high Ca²⁺’ (buffer + 1 mM CaCl₂) demonstrates the fluorescence intensity change in the spectra. This [Ca²⁺] change can be quantified by selecting a single emission wavelength at which the intensity change is maximal, as shown in (D) (measured at 520nm).

Ca²⁺ Binding Affinity and Ion Selectivity

Different Ca²⁺ indicator dyes bind Ca²⁺ with different affinities, and Ca²⁺ affinity is therefore an important consideration when choosing a dye. For example, for most experimental situations fura-5F has proven a useful, if not better, choice than fura-2 for measuring cytoplasmic [Ca²⁺]_i. A derivative of fura-2, fura-5F is handled in exactly the same way in terms of cell loading and recording of fluorescence signals. However, fura-5F has a higher K_D than fura-2 (400 nM vs. 140 nM, respectively), and given that most Ca²⁺ studies examine [Ca²⁺]_i in the 100 nM – 1 μM range, the advantages of the higher K_D indicator are obvious. The characteristics of fura-5F are shown in Fig.1 and tables 1 and 2, as are those of another commonly used single wavelength Ca²⁺ indicator, fluo-4.

Table 1.

General Characteristics of Fura-5F and Fluo4

	Ratiometric	Ba2+-entry	Mn2+-quench	KD (nM)
Fura-5F	Yes	Yes	Yes	400
Fluo-4	No	No	No	390

Table 2.

Spectral and biological effects of Ca²⁺ and other ions on Ca²⁺ indicators and Ca²⁺ signaling processes.

	Ca2+	Ba2+	La3+	Gd3+	Mn2+
Fura Signal	Yes	Yes	Yes	Yes	No
Fura Quench	No	No	No	No	Yes
Fluo-4 Signal	Yes	Poor	Yes	Yes	Poor
Fluo-4 Quench	No	No	No	No	No
Block SOCE	No	No	Yes	Yes	No
Substrate for PMCA/SERCA	Yes	No	?	?	No
Block PMCA	No	No	Yes	Yes	?

While it may seem intuitive that a Ca²⁺ indicator should be selected based on its selectivity for Ca²⁺ ions over other cations, the ability of these indicators to interact with other cations can provide experimental flexibility for studying Ca²⁺ signaling and SOCE. This has been particularly useful with cations such as Mn²⁺ and Ba²⁺ which, in addition to their ability to interact with Ca²⁺ indicators, can also substitute for Ca²⁺ ions in a number of biological processes, or pharmacologically interfere with these same processes (Table 2). Thus, experimental approaches can be designed that involve substituting alternative cations for Ca²⁺ ions and these approaches may then reveal specific aspects of the underlying Ca²⁺ signaling process. This has been a particularly useful approach with the fura-based Ca²⁺ indicators and substituting Mn²⁺ [13] or Ba²⁺ [14,15] for Ca²⁺ ions (see below).

Introduction of Ca²⁺ Indicators into Cells

Since Ca²⁺ indicators are charged molecules and do not freely cross biological membranes, a critical factor for ensuring their successful application to biological systems was developing a way to load the indicator into the cytoplasm of intact cells. Fortunately, an innovative approach was developed whereby charged residues were chemically masked with acetoxy methyl ester groups (AM groups) [16]. The uncharged products cross the plasma membrane into the cytoplasm where non-specific esterases cleave the AM groups leaving the charged and Ca²⁺-sensitive form of the indicator to accumulate in that cell compartment.

One complicating problem associated with AM derivatives of Ca²⁺ indicators is that their accumulation may not be restricted to the cell cytoplasm. That is, the AM derivative may cross other membranes and accumulate in compartments such as the ER and mitochondria. The extent of this compartmentalization can vary depending on the type of cell being used. For example, no such problem is apparent with HEK 293 cells, whereas it presents a major obstacle in primary rat hepatocytes [17]. Fortunately, there are ways to diagnose and minimize this problem. Diagnosis can include: (i) A non-uniform and punctate spatial distribution of the Ca²⁺ indicator; (ii) Calibrated Ca²⁺ indicator values indicative of high resting cytoplasmic [Ca²⁺]_i; (iii) Poor or lack of agonist-induced [Ca²⁺]_i response; (iv) A combination of agonist and a sarco-endoplasmic reticulum Ca²⁺-ATPase (SERCA)-pump inhibitor (thapsigargin) leads to a drop in resting cytoplasmic [Ca²⁺] levels (See [17,18]). Minimizing compartmentalization can usually be achieved by manipulating indicator concentration, duration of incubation with AM-indicator, and the temperature of the incubation. In the event that compartmentalization cannot be resolved, one has the option of introducing free acid forms of Ca²⁺ indicators directly into the cytoplasm by techniques such as microinjection or electroporation [17,19].

Leakage of Ca²⁺ indicators out of the intact cell is an additional complication with contributing factors including the specific Ca²⁺ indicator choice and/or the cell type being used. In some instances, the problem appears due to the Ca²⁺ indicator being a substrate for organic anion transporters, an effect that can be minimized by including anion transport inhibitors such as probenicid and sulfinpyrazone in all experimental solutions [20].

Data Representation

Due to the non-linear relationship between Ca²⁺ indicator fluorescence changes and [Ca²⁺]_i, calibration of the fluorescence signal [12] is advisable for instances such as looking for subtle quantitative effects on [Ca²⁺]_i signaling, or as a means to compare experimental data collected from different instrumentation. However, in many instances a general understanding of [Ca²⁺]_i signaling mechanisms such as SOCE can be gleaned from fluorescence intensities or fluorescence ratios. If uncalibrated ratios are used it is minimally recommended that you (i) ensure the Ca²⁺ indicator used has an appropriate affinity to maximize the dynamic range of the measurement and avoid the possibility of saturating the indicator, and (ii) correct data for fluorescence signals not related to Ca²⁺-sensitive fluorescence changes (auto-fluorescence), which minimizes cell-to-cell variation as well as variability between experiments.

In the case of ratiometric fura dyes, auto-fluorescence is readily estimated by quenching fura dyes with Mn²⁺ ions (treat cells with ionomycin and MnCl₂, 10 μM and 20 mM, respectively, in a nominally Ca²⁺-free bathing solution). Single wavelength indicators like fluo-4 and Calcium Green-1 are not completely quenched by Mn²⁺, thus non-indicator loaded cells may provide an estimation of autofluorescence.

B. Quantitative Assessment of SOCE

Activating SOCE: SERCA Inhibitors/Ca²⁺-Ionophores

As described above, depletion of intracellular Ca²⁺-stores located within the ER plays a key role in activating SOCE and is primarily achieved through IP₃-induced Ca²⁺-mobilization during agonist activation. The ER also serves as a critical Ca²⁺-buffer, with Ca²⁺-ATPases (SERCA pumps) that can rapidly sequester Ca²⁺ ions from the cell cytoplasm to prevent untoward changes in [Ca²⁺]_i and replenish ER Ca²⁺-stores following agonist activation. Even in unstimulated cells, Ca²⁺ ions are continually cycling across the ER membrane with the actions of SERCA pumps sequestering Ca²⁺ balanced against a poorly defined ‘Ca²⁺ leak’ process out of the ER (but see [21,22]). Fortunately, there are a number of pharmacological reagents that target and inhibit the function of the SERCA Ca²⁺-pumps and, with the ‘Ca²⁺ leak’ process in effect, results in depletion of ER Ca²⁺-stores and full activation SOCE (Fig.2B).

Figure. 2. ‘Ca²⁺re-addition’ Protocol and Biphasic Ca²⁺Signaling.

HEK 293 cells were loaded with fura-5F and intracellular [Ca²⁺]_i measured as described in [57]. In (A) and (B), HEK 293 cells were subjected to the ‘Ca²⁺ re-addition’ protocol by switching between buffer solutions with extracellular Ca²⁺ present or absent. In unstimulated HEK 293 cells (A), the ‘Ca²⁺ re-addition’ protocol elicits no detectable change in [Ca²⁺]_i. In (B), the ‘Ca²⁺ re-addition’ protocol was combined with treatment of cells with a SERCA pump inhibitor (2 μM thapsigargin) to demonstrate a biphasic Ca²⁺ response. In the absence of extracellular [Ca²⁺], the transient first phase of intracellular Ca²⁺ release is observed. On ‘Ca²⁺ re-addition’, the seconds phase of SOCE is observed. As shown in (C), the extent of SOCE activity can be quantified either as a rate of Ca²⁺ entry, or appeal response.

Discovery of SERCA inhibitors, namely thapsigargin, cyclopiazonic acid (CPA) and tBHQ [23] provided an important validation of the SOCE model of Ca²⁺ entry, especially as these reagents made it possible to deplete the IP₃-sensitive Ca²⁺ stores and activate SOCE without formation of any inositol phosphates associated with agonist activation [24]. SERCA pump inhibitors are membrane permeant, and provide a non-invasive technique for manipulating ER Ca²⁺-pools and SOCE activation in intact cells. Activating SOCE with thapsigargin is, however, a difficult process to reverse. In contrast, CPA is more water soluble than thapsigargin, and appears more amenable to washing out of cells, and offers the possibility of achieving partial depletion of intracellular Ca²⁺ stores with a range of CPA concentrations, resulting in partial activation of SOCE [25].

An alternative to SERCA pump inhibition is to deplete ER Ca²⁺-pools using a Ca²⁺-ionophore such as ionomycin [26]. A carboxylic acid antibiotic, ionomycin is a mobile ion carrier that has proven effective in manipulating the movement of Ca²⁺ ions and intracellular Ca²⁺-pools in intact cells. Low concentrations of ionomycin (<1μM) appear to selectively release intracellular Ca²⁺-stores (ED₅₀ = 50 nM) without directly increasing the permeability of the plasma membrane to extracellular Ca²⁺ [27,28]. At higher concentrations (1-10 μM), the ionophore increases the permeability of the plasma membrane, ER and mitochondria to Ca²⁺ ions.

In general, ionomycin is preferred over A23187 since, with fluorescent Ca²⁺ indicators excited in the UV range, A23187 can contribute significant autofluorescence to the measured signal (a non-fluorescent derivative, 4Br-A23187, is available [29].

‘Ca²⁺ Re-Addition’ Protocol: Dissecting out SOCE

In unstimulated cells, the concentration of Ca²⁺ ions in the cytoplasm (~10^-7 M) reflects a steady state balance of homeostatic mechanisms involving Ca²⁺ pumps and channels. Cells utilize ATP-dependent Ca²⁺ pumps to maintain this level against a background of enormous Ca²⁺ gradients (extracellular milieu ~10^-3 M; ER ~10^-4 M). In addition in most, but not all cell types, plasma membrane permeability to Ca²⁺ is very low unless altered by a physiological or pathological mechanism. Figure 2A illustrates a principle method for disrupting steady state [Ca²⁺]_i conditions by simply modifying extracellular Ca²⁺ conditions ([Ca²⁺]_o). In this ‘Ca²⁺ re-addition’ protocol, the salt solution bathing cells is switched alternatively from one containing normal [Ca²⁺]_o (1-2 mM) to a salt solution that is depleted or nominally Ca²⁺-free, and then back again to normal [Ca²⁺]_o conditions. Depending on the water source and purity of salts used, nominally Ca²⁺-free buffers can contain [Ca²⁺] in the range of a few μM. Alternatively, one can ensure the solutions are Ca²⁺-free by supplementing them with a Ca²⁺-chelator such as EGTA or BAPTA (100-500 μM). As shown in Fig. 2A for an unstimulated HEK 293 cell loaded with fura-5F, the Ca²⁺ re-addition protocol does not result in any detectable change in [Ca²⁺]_i (this is not always the case, for example see [30]). Characterizing this response in unstimulated cells is an important control to characterize the cell type being used and the full extent of SOCE activation measured in stimulated cells. In Fig.2B, the ‘Ca²⁺ re-addition’ protocol is now used in the same fura-5F-loaded HEK 293 cells, but this time treated with thapsigargin to empty Ca²⁺ stores and thus fully activate SOCE. In the absence of extracellular Ca²⁺, thapsigargin treatment results in a [Ca²⁺]_i response that is transient, representing the release of Ca²⁺ ions from intracellular Ca²⁺ pools. On adding back extracellular Ca²⁺ to the bathing medium, the second phase of SOCE is revealed resulting in a rise in [Ca²⁺]_i to an elevated and sustained phase. The cells will remain at this elevated [Ca²⁺]_i level until the Ca²⁺-store depletion stimulus is removed, or [Ca²⁺]_o is modified.

In summary, the basic approach described in the Ca²⁺ re-addition protocol is to separate the two phases of Ca²⁺ mobilization, Ca²⁺ release and SOCE. At the point of restoring [Ca²⁺]_o, manipulations of the bathing solution, including cation substitutions, can enhance or distinguish Ca²⁺-entry pathways on the basis of their pharmacological properties (see below). An interesting variation of this protocol is a ‘Ca²⁺-overshoot protocol’. In this method, phospholipase C-coupled receptors are transiently activated in the absence of extracellular Ca²⁺, with Ca²⁺ re-addition occurring after the stimulus is removed. With this protocol, and despite the removal of the cell stimulus, intracellular Ca²⁺ pools remain depleted, and SOCE remains activated until extracellular Ca²⁺ is restored and Ca²⁺ pools refilled. In some cell types, this SOCE activity can be observed as a transient rise in [Ca²⁺]_i, or‘Ca²⁺-overshoot’, following Ca²⁺ re-addition [31].

Quantifying SOCE: Peaks and Rates

Three general methods have been used to quantitatively analyze Ca²⁺ influx from experiments such as in Fig. 2C: initial rate of [Ca²⁺]_i rise, peak [Ca²⁺]_i level, and area under the [Ca²⁺]_i curve. In theory at least, initial rates should give the cleanest indication of Ca²⁺ influx, although it is clear that cellular Ca²⁺ buffering occurs so rapidly that even this measure can be influenced by many factors. These factors can sometimes be minimized by the use of Ca²⁺ surrogates such as Ba²⁺ or Mn²⁺, which do not activate negative feedbacks and are poor substrates for cellular Ca²⁺ buffers. This is discussed in more detail below. Peak responses are probably more commonly reported, but these have the disadvantage that [Ca²⁺]_i levels are often limited in non-linear ways by cellular feedback mechanisms and Ca²⁺ buffers. Least desirable is the use of integrated areas under the [Ca²⁺]_i curve; these are generally sustained responses and thus the areas will depend upon arbitrarily selected time intervals, and will to varying degrees be even more affected by secondarily activated pumps and inactivation mechanisms.

Monitoring SOCE with Ca²⁺ Surrogates

The complexity of homeostatic mechanisms that regulate [Ca²⁺]_i can lead to a number of potential artifacts when using Ca²⁺ ions to monitor SOCE in fluorescence-based experiments. Any mechanism that limits the extent of a [Ca²⁺]_i rise can result in non-linear behaviors that belie appropriate quantitative analysis. These mechanisms can include activation of plasma membrane pumps to extrude Ca²⁺ ions and/or feedback inhibition of SOC channels. For example, a treatment that inhibits intracellular Ca²⁺-ATPases (such at thapsigargin) at the same time as activating SOCE may have a quantitatively different effect on elevating [Ca²⁺]_i compared to SOCE activation by other means. These problems can sometimes by minimized by substituting for Ca²⁺ cations that will permeate through SOC channels but are poor substrates for other Ca²⁺-dependent enzymes such as Ca²⁺-pumps. Mn²⁺ or Ba²⁺ have proven very useful substitutes for [Ca²⁺]_o for SOCE measurements, particularly when using the Ca²⁺ re-addition protocol. However, it should be noted that not all Ca²⁺-indicators behave in the same way to these Ca²⁺ surrogates (see Table 2).

Mn²⁺ ions appear to traverse most Ca²⁺-permeable channels [32] and substitute for Ca²⁺ in SOCE [33], despite the fact that CRAC channels are less permeable to Mn²⁺ than to Ca²⁺ [34]. In fluorescence-based [Ca²⁺]_i measurements, fura-indicators are ideally suited to monitor Mn²⁺-entry as they irreversibly bind Mn²⁺ (due to high affinity for the cation), which quenches the fura fluorescence. Quantifying Mn²⁺-entry can easily be done by monitoring the Mn²⁺-quench process at fura-2’s isobestic wavelength (~360 nm), or by using an algebraic summation of fluorescence values obtained from excitation at two different wavelengths [35]. Thus, activation of SOCE activity can easily be quantified by measuring initial rates of Mn²⁺ quench in fura-loaded cells [36]. In quantifying SOCE activity, it is important to measure basal Mn²⁺-quench rates in unstimulated cells, since most cells exhibit a basal, constitutive permeability to Mn²⁺. A useful facet of the Mn²⁺-quench technique is that you can monitor Ca²⁺-release events and Mn²⁺-quench (SOCE) simultaneously [35,37].

There are some limitations with using the Mn²⁺-quench technique. One is that it is not applicable to all Ca²⁺-indicators, for example the single wavelength indicator Fluo-4 is poorly responsive to Mn²⁺ ions, and Calcium Green-1 fluorescence intensity is actually increased by Mn²⁺, not quenched. Another surrounds an artifact when applying Mn²⁺-quench to cells with significant compartmentalization of fura-indicators in the ER. While Mn²⁺ is not a substrate for intracellular SERCA pumps, it can easily pass through activated IP₃ receptors in a retrograde manner [17]. This presents a problem when monitoring SOCE with agonist stimulation since the observed Mn²⁺-quench may reflect IP₃ mediated movement into the ER rather than flux across the plasma membrane.

Monitoring Ba²⁺ entry by substituting Ba²⁺ for Ca²⁺ ions in the Ca²⁺ re-addition protocol has provided a useful measure of SOCE [15]. Fura-based indicators are amenable for this technique since Ba²⁺ can substitute for Ca²⁺ and render an excitation spectrum similar to that for Ca²⁺. Importantly, Ba²⁺ is a poor substrate for Ca²⁺-pumping ATPases and does not enter ER Ca²⁺-pools. Thus, monitoring Ba²⁺-entry during cell activation provides a measure of SOCE activity without complications due to buffering by Ca²⁺-pumps and Ca²⁺-pool refilling. As for Mn²⁺-quench, quantifying initial rates of Ba²⁺ entry is recommended, as well as taking into account basal rates of entry in the absence of cell stimulation. One potential artifact that could be encountered when using Ba²⁺ ions involves altering membrane potential [38], which can influence SOCE activity (see below).

III. Electrophysiological Measurement of SOC Currents

A. Biophysical Characteristics of Store-operated Currents

An understanding of SOCE and its regulation has been facilitated by the electrophysiological characterization of plasma membrane Ca²⁺ currents associated with SOCE. The most extensively studied and most thoroughly characterized store-operated current was first identified in mast cells as Ca²⁺ release-activated Ca²⁺ current, or I_crac [39,34]. I_crac is a small current, often below the level of reliable detection. It is largest in hematopoetic cells where it can be up to 2 pA/pF, under optimal recording conditions. But for many cell types, the measurement of store-operated entry by electrophysiological means has not always been a practical approach since the currents apparently fall near or below the limits of detection. For instance, in HEK293 cells the store-operated Ca²⁺ current is maximally only about 0.5 pA/pF. This limitation can be circumvented to some extent by carrying out experiments under conditions where cells are maintained in an extracellular solution that is divalent cation-free (DVF, see below). Studies on I_crac, indicate that the underlying channels have a small pore diameter [40], low permeability to Cs⁺, and an extremely small unitary conductance for Ca²⁺ and Na⁺ [34,41,40,42,43]. While no single channel events have been detected for I_crac, fluctuation analysis predicts a monovalent unitary conductance of less than 1 pS [40].

B. Biophysical Assessment of Store-operated Currents

Electrophysiological measurement of I_crac is usually carried out in whole-cell patch clamp mode (Fig.3A). This mode offers a number of experimental advantages, in particular the ability to define and control the conditions bathing the intracellular and extracellular milieu, and to precisely control membrane potential. With fluorescence-based [Ca²⁺]_i measurements on SOCE, the inability to control membrane potential in intact cells is a problem that can potentially give rise to experimental artifacts (see below for discussion).

Figure 3. Biophysical Assessment of Store-Operated Currents.

(A) Diagram showing the whole-cell configuration of the patch-clamp technique. As depicted, the use of this technique allows for access to the inside of the cell through the internal pipette solution (See Table 3 for example solutions), allowing for depletion of internal Ca²⁺ stores passively with BAPTA alone, or in combination with activating reagents such as IP₃.

(B) Schematic of a voltage ramp protocol used to assess store-operated Ca²⁺ currents. The protocol is repetitively applied over time, with 1 to 2 second intervals at 0 mV, to reveal currents develop during Ca²⁺ store depletion.

(C) An example I_CRAC in an RBL cell recorded using the protocol shown in panel B, and which was repetitively applied every two seconds. Internal Ca²⁺ stores were actively depleted by including 20 μM IP₃ and 10 mM BAPTA in the patch pipette solution. 10 mM Ca²⁺ was present in the external bathing solution. Following break-in with the patch pipette, a small (2 pA/pF) inward current develops at -100 mV, while no change is observed at +100 mV). Upon removal of all external divalent cations (divalent-free, DVF) in the extracellular media, Na⁺ ions now permeate the store-operated channel to reveal Na⁺-I_CRAC. Over time, Na⁺-I_CRAC depotentiates, a process which is the reciprocal of a process known as Ca²⁺ dependent potentiation (CDP).

(D) Current-voltage relationship of Ca²⁺-I_CRAC and Na⁺-I_CRAC are inwardly rectifying Ca²⁺- and Na⁺-I_CRAC (data taken from recordings shown in panel C).

(E) In this whole-cell recording, DVF conditions were used to reveal I_CRAC in HEK 293 cells where, normally, Ca²⁺- I_CRAC is difficult to detect. Following break-in, the protocol entails switching the external bathing solutions between Ca²⁺ containing (10mM) and DVF bathing solutions. Using optimal conditions for passive depletion of Ca²⁺ stores (10 mM BAPTA in patch pipette solution), no I_CRAC is observed with 10mM Ca²⁺ in the external bathing solution. By switching to DVF conditions the developing Na⁺-I_CRAC is revealed (black line trace). In contrast, supplementing the internal pipette solution with enough Ca²⁺ to “clamp” free Ca²⁺ at 100 nM (calculated using Maxchelator software; www.stanford.edu/~cpatton/maxc) prevents the passive depletion of Ca²⁺ stores upon break in, and no I_CRAC current develops (grey line trace) under all external bathing solution conditions.

General protocols to record I_crac using the whole-cell patch-clamp technique are reviewed in [4]. This technique involves the establishment of a tight seal between a recording pipette electrode and the surface membrane of a cell, followed by rupture of the membrane under the pipette opening to establish electrical and chemical continuity between the pipette and the cytoplasm (Fig 3A). This then allows the cell membrane potential to be clamped (usually at a depolarized potential, 0 mV) until, every 1 to 2 seconds, a voltage ramp is applied from -100 mV up to +100 mV (see Fig. 3B). With this protocol, the measured currents can reveal how I_crac develops over time (Fig. 3C) as well as its current-voltage relationship which, for I_crac, is inwardly rectifying (Fig. 3D). After completing each voltage ramp, it is routine to return to the depolarized potential (0 mV). This is an important step in the protocol as it reduces the driving force for Ca²⁺ ions entering the cell, and thus lessens Ca²⁺ dependent feedback pathways that inactivate CRAC channels.

Figure 3C illustrates a typical experiment utilizing RBL cells in the whole-cell configuration to measure I_crac activation. After break-in and establishing a whole cell patch, the cell was held at 0 mV followed by applying a voltage ramp from -100 mV to +100 mV every 2 seconds. The internal pipette solution (described in Table 3) contained IP₃ (20 μM) and BAPTA (10 mM), optimal conditions to rapidly deplete ER Ca²⁺ stores and activate SOCE. The extracellular bathing solution (described in Table 3) contained 10 mM Ca²⁺. Over a period of seconds after break-in, a small inward current can be observed developing at -100 mV, while little to no change is detected at +100 mV. Plotting the current-voltage relationship of this data (Fig. 3D; please note that these figures have been leak subtracted) clearly demonstrates that I_crac, activated by Ca²⁺ store depletion, is strongly inwardly rectifying.

Table 3.

Basic solutions for I_crac measurements

Extracellular Solutions (in mM)
	Ca2+-Icrac	Na+-Icrac
NaCl	145	145
KCl	3	3
MgCl2	2	0
Glucose	10	10
CsCl2	10	10
EGTA	10	0
HEPES (pH 7.4)	0	0.1
Internal Pipette Solutions (in mM)
	Passive Depletion	No Depletion Control
Cs Methansulfonate	145	145
MgCl2	8	8
BAPTA	10	10
CaCl2	0	3.5 (free [Ca2+] ~100nM)
HEPES (pH 7.2 with CsOH)	10	10

Note: High [Mg²⁺] required to block MIC currents [43].

Free [Ca²⁺] calculated using MaxChelator (www.stanford.edu/~cpatton/maxc)

An important characteristic of the CRAC channel is that it completely loses its Ca²⁺ selectivity when all extracellular divalent cations are removed, and Na⁺ ions are now allowed to permeate the channel. So under extracellular divalent free (DVF) conditions (see table 3), CRAC channel activity can now be measured as a function of current carried by Na⁺ ions. Interestingly, Na⁺ currents carried through CRAC channels (Na⁺-I_crac) run down over time by a process known as depotentiation, which is the reciprocal of a process known as Ca²⁺ dependent potentiation (see below).

Critically, the Na⁺-I_crac current densities measured under DVF conditions are large, and has provided an invaluable technique for characterizing SOC channels in cells where there is little to no detectable Ca²⁺-I_crac (I_crac measured with divalent cations present in extracellular solutions) (Fig. 3E, and examples in [44,45]). However, it is advisable that the Na⁺ currents recorded be characterized pharmacologically to ensure that they represent the activity of CRAC channels (see below).

C. Activating Store-operated Currents

The single initial signal for the activation of SOCE and I_crac is the depletion of intracellular Ca²⁺ stores located in the ER. As mentioned above, I_crac activation is measured using the whole-cell patch, allowing the intracellular milieu to be modified directly with the internal patch pipette solution. Thus, rather than relying of the actions of SERCA pump inhibitors like thapsigargin or ionomycin (both added to extracellular solution) to activate SOC channels, intracellular Ca²⁺ store depletion can be achieved by additions to the internal pipette solution.

The simplest mode of I_crac activation is through ‘passive depletion’ of intracellular Ca²⁺ stores. By simply breaking into cells with a patch pipette solution containing a high concentration of BAPTA or EGTA and no added Ca²⁺ (see table 3), the intracellular Ca²⁺ stores are gradually lost. This technique is at least in principle equivalent to the use of SERCA inhibitors. The passive process is slow, and I_crac activation may take in the order of several minutes to fully develop (Fig.4A). Alternatively, one can use the whole-cell patch clamp technique to introduce membrane impermeant reagents to actively and rapidly deplete intracellular Ca²⁺ stores and activate I_crac. The most efficient way is to include metabolizable or non-metabolizable analogs of IP₃ (in conjunction with EGTA or BAPTA) [46]. Under these conditions, the time course of I_crac activation is significantly faster than with passive depletion alone (Fig.4A).

Figure 4. Methods for activating I_CRAC.

(A) Whole –cell currents recorded in HEK 293 cells transfected with Stim1 and Orai1. From a holding potential of 0 mV, developing store-operated currents were recorded during voltage ramps (-100 mV to +100 mV) applied every two seconds using. This data shows a time course of Ca²⁺-I_CRAC development recorded -100 mV. With coexpression of Stim1 and Orai1, “monster” Ca²⁺-I_CRAC currents develop, and without the need to use DVF conditions. In this experiments, we compare the time course of Ca²⁺-I_CRAC activation when Ca²⁺ stores are depleted passively (20mM BAPTA alone in the internal pipette solution) or actively (20mM BAPTA plus 20 μM IP₃ in the pipette). In contrast, “clamping” free [Ca²⁺] in the internal pipette solution to ~100 nM prevents Ca²⁺ store depletion, and no “monster” Ca²⁺-I_CRAC is observed. In all recordings, the external [Ca²⁺] was 10 mM.

(B) Current-voltage relationships for each of the “monster” Ca²⁺-I_CRAC currents recorded in panel A are inwardly rectifying akin to endogenous I_CRAC.

Ca²⁺-dependent feedback regulation of I_crac

I_crac is both positively and negatively regulated by Ca²⁺ in a complex manner. First, Ca²⁺ regulates SOC channels in a positive manner through a process known as Ca²⁺ dependent potentiation (CDP) [47,48], and appears to result from Ca²⁺ ions interacting with an extracellular binding site. While this binding site has not been identified as yet, it is possible that it is the same as the Ca²⁺ binding site in the selectivity filter. Nonetheless, the presence of Ca²⁺ at this putative CDP site potentiates I_crac, and can be detected electrophysiologically when switching between external DVF solution and Ca²⁺ containing (10 mM) solutions. As mentioned above, the reverse of CDP can be detected when switching from a Ca²⁺ containing extracellular solution to DVF solutions to detect Na⁺I_crac, the effect being that the Na⁺ current depotentiates (Figure 3C and 3E).

Under conditions whereby [Ca²⁺]_i levels are high in close proximity to the CRAC channels, CRAC channels can experience two types of Ca²⁺-dependent feedback regulation described as either fast or slow inactivation [34,49,50,51]. Fast inactivation of I_crac can be observed electrophysiologically by using brief hyperpolarizing steps from depolarized holding potentials (for protocol, see [34,49,51]), and is believed to result from Ca²⁺ entering through the channels binding directly to the channels at or near the channel mouth. This effect occurs in the millisecond time range, and differences in the rates of fast inactivation can be seen by comparing the effects of the slower Ca²⁺ chelator EGTA with the faster chelator, BAPTA [34,49,51] in the internal pipette solution.

In contrast, slow inactivation of I_crac occurs over a period of tens of seconds [52,50,53]. While the mechanism underlying slow inactivation is not fully understood, slow inactivation might occur due to refilling of intracellular Ca²⁺ stores and could occur if cytoplasmic [Ca²⁺]_i levels were significantly elevated in spatially restricted areas and thus locally overwhelmed the Ca²⁺ chelator included in the internal pipette solution. Alternatively, slow inactivation might be due to a regulatory feedback mechanism on the SOCE signaling process, for example it has been suggested that PKC regulates I_crac activity [52].

In summary, feedback regulation of I_crac by Ca²⁺ ions is a complex multitude of processes that are yet to be fully defined. While these Ca²⁺ signaling pathways may play crucial roles in the physiology of many cells types, awareness of these processes is crucial for studying SOC channels electrophysiologically and for understanding how they can be controlled experimentally.

IV. Manipulating SOCE

Cytoplasmic [Ca²⁺]_i reflects a homeostatic balance of Ca²⁺ pumps and Ca²⁺ channel activities. Any disturbance in these processes, through receptor-dependent activation of Ca²⁺ channel activity for example, can lead to the rapid movement of Ca²⁺ down concentration gradients until a new steady state is achieved in concert with Ca²⁺ pump and feedback mechanisms. Experimental manipulations and pharmacological agents that interfere with any of these Ca²⁺ homeostatic processes can have a profound effect on the cytoplasmic [Ca²⁺]_i and plasma membrane currents. Thus, an intimate understanding of the underlying Ca²⁺ homeostatic processes and the various ways they can be manipulated will allow the observed fluorescence-based [Ca²⁺]_i signals and membrane currents to be interpreted in meaningful ways that speak to the underlying Ca²⁺ signaling process. Some of these approaches are outlined below.

A. Pharmacological Manipulation of SOCE

Pharmacological manipulation of SERCA pumps with inhibitors such as thapsigargin was crucial for developing our understanding of SOCE (see above). Unfortunately, our ability to manipulate SOC channels directly and inhibit them with any degree of specificity has been rather limited. Amongst the arsenal of tools proposed as SOC channel inhibitors [23,4], lanthanides and 2-aminoethyldiphenyl borate (2-APB) have proven most useful in manipulating SOC entry, with lanthanides appearing to be the most selective (see below).

Due to their physical chemistry [54], lanthanides were exploited as tools to block both Ca²⁺ entry and efflux processes [55]. Fortunately, SOCE is much more sensitive to inhibition by lanthanides than Ca²⁺-efflux across the plasma membrane. The lanthanide Gd³⁺ is generally used in this respect and appears specific for SOCE over non-SOCE pathways when used in the appropriately low (≤5 μM) concentration range [56], and only begins to block PMCA activity above 100 μM [57].

When using this approach, care should be taken to use freshly prepared lanthanide salt solutions in buffers devoid of divalent anions. Also the presence of extracellular BAPTA or EGTA will bind lanthanides with high affinity [56] and adversely affect their concentration in solution. Another potential problem stems from the ability of lanthanides to render an excitation spectrum with fura-based Ca²⁺-indicators. Usually, lanthanides poorly traverse the plasma membrane of intact cells, if at all [15]. However, should bathing an unstimulated cell with a lanthanide increase the fura-signal, this could be indicative of ‘leaky’ or unhealthy cells.

2-APB has proven a somewhat selective inhibitor for SOCE [58,59,60,61,62], an effect that appears to be extracellular at the plasma membrane [59,61,63,64], and independently of IP₃ receptor inhibition [65]. However, a careful evaluation of 2-APB effects on Ca²⁺-signaling processes is advised. Originally shown to be a membrane permeable inhibitor of IP₃ receptors [66], the effects of 2-APB are proving complex and this drug may lack the desired specificity. These effects can include inhibiting a magnesium-inhibitable cation channel [59,43], or activating distinct ion channel activity depending on the concentration of 2-APB [67,68,69].

B. Membrane Potential

Contrary to voltage-activated Ca²⁺ channels, membrane depolarization does not play a role in SOCE activation. However, since Ca²⁺ entry through SOC channels is an electrogenic process driven by chemical (concentration gradient) and electrical (membrane potential) forces, plasma membrane potential can regulate SOCE such that less Ca²⁺ enters upon depolarization and membrane hyperpolarization promotes Ca²⁺ influx [70,71].

Membrane potential is well controlled when using voltage-clamp techniques to measure SOC channel activity (see below). However, when measuring SOCE with fluorescent Ca²⁺-indicators alone, changes in membrane potential are usually not controlled, but it is possible to carry out fluorescence measurements of [Ca²⁺]_i under voltage clamped conditions [72]. Lack of control of membrane potential must be considered when characterizing pharmacological agents that affect SOCE [73], since modulation of the Ca²⁺-entry signal could result from depolarization/hyperpolarization rather than direct effects on the SOC channel itself.

Monitoring membrane potential and characterizing the effects of candidate pharmacological agents for SOCE is one way to control for this. Alternatively, these complications can be avoided by combining voltage-clamp techniques with fluorescence Ca²⁺ measurements. If this option is not available, one can manipulate the bathing solution to contain high [KCl] and create a “poor man’s voltage-clamp” and carry out fluorescence experiments with the plasma membrane fully depolarized [74,75,76].

C. Manipulating the Molecular Players for SOCE

As described above, the main limitation to developing approaches that specifically manipulate SOCE has been our poor understanding of the molecular players involved in this process. Fortunately, the recent identification of STIM and Orai (or CRACM) family proteins has facilitated our understanding of the target plasma membrane channel mediating SOCE, as well as the mechanism by which depletion of intracellular Ca²⁺ stores is communicated to these SOC channels at the plasma membrane [77].

These discoveries have opened new experimental approaches to dissect and manipulate SOCE through techniques such as RNAi [78,79,80,81,82,83], protein modification and expression [84], and the generation of animal models [45]. These approaches have enabled structure function studies on Orai family proteins to identify residues involved in pore formation and the selectivity filter [85,86,87]. These studies may very well lay the foundation to develop new pharmacological strategies to manipulate SOCE [88].

V. Concluding Remarks

The purpose of this review is to delineate useful strategies for dissecting Ca²⁺ signaling processes and studying SOCE using experimental approaches that employ fluorescent Ca²⁺ indicators and electrophysiological techniques. The ability to use these diverse experimental approaches to monitor [Ca²⁺]i and SOC currents in real time, with varying degrees of spatial and temporal resolution, has significantly contributed to our current understanding of the basic processes of regulated SOCE.

It remains a challenge to continually refine protocols and experimental conditions to optimally discriminate SOCE from other possible routes of Ca²⁺-entry across the plasma membrane. This is especially important when identifying a role for SOCE under conditions of physiological activation of phospholipase C-linked pathways and SOCE, or in considering a role for SOCE in excitable cells. The identification of the molecular players for SOCE, STIM and Orai family proteins, has also opened up exciting possibilities to derive new strategies to specifically manipulate SOCE, as well as delineate all the steps involved in this Ca²⁺ signaling pathway.