Receptor-Activated Calcium Entry Channels—Who Does What, and When?

See allHide authors and affiliations

Science's STKE  27 Jul 2004:
Vol. 2004, Issue 243, pp. pe40
DOI: 10.1126/stke.2432004pe40


The recent Science STKE E-Conference on Defining Calcium Entry Signals highlighted many of the outstanding problems and questions regarding the nature and regulation of the receptor-activated entry of Ca2+, particularly as it relates to Ca2+ signaling in nonexcitable cells. Frequently, these stem from the current lack of any clear candidates for the molecular identity of many of the major conductances involved. Moreover, there is considerable confusion in the field, largely as a result of the use of sometimes inappropriate or imprecise methodologies and inconsistent terminology. Nevertheless, much useful information is beginning to be revealed about the biophysical characterization of the fundamental properties of the channels involved and, at least in some cases, the specific conditions under which they are active. As a result, it is becoming clear that cells often contain various Ca2+ entry channels in addition to the ubiquitous store-operated, or capacitative, channels. These different channels are activated in distinct ways and operate under different conditions of stimulation.

The receptor-activated entry of Ca2+ is a key component in cellular Ca2+ signaling, playing a particularly important role in nonexcitable cells. Despite intensive study over the past 20 years, our understanding of the nature of the pathways responsible and the mechanisms that control their activity is far from clear. Moreover, in the past few years it has been revealed that entry pathways in addition to the ubiquitous store-operated or capacitative pathway exist and apparently play a unique role under specific, possibly physiologically relevant, conditions. The recent E-Conference on Defining Calcium Entry Signals (1) has served to highlight many of the outstanding problems in the field, not least of which is the current lack of any clear candidates for the molecular identity of many of the major conductances involved. However, the accumulation of data that define the specific properties of the channels involved, their regulation, and identification of the particular conditions under which they are active should help in this search. At a minimum, this will provide a physiological and biophysical profile against which any potential molecular candidate can be compared. In this Perspective, I will focus on two key aspects of this endeavor—the methodologies for characterizing the channels responsible for Ca2+ entry, and which channels are active under which conditions and what do they do?

Characterizing the Channels Responsible for Ca2+ Entry

Debate regarding the best approach to characterizing the channels responsible for Ca2+ entry has generally revolved around the relative merits of fluorescence versus electrophysiological measurements. As is often the case, there are advantages and limitations to both techniques. Fluorescence measurements involve minimal interference with normal cell integrity and regulation, but it is difficult to isolate Ca2+ entry per se with this approach. Use of Mn2+, Ba2+, or Sr2+ as surrogates for Ca2+ goes some way to overcoming this limitation, but the use of such surrogate ions is not without its own problems. Moreover, other factors that are inadequately controlled (for example, membrane potential) can influence the measured rate of influx. More importantly, fluorescence measurements provide very little information on the actual characteristics of the channels involved. Realistically, such information can only be obtained by using patch-clamp techniques to develop a detailed biophysical "fingerprint" for the relevant conductances. In this way, the magnitude of the macroscopic current, the current-voltage profile, and the Ca2+ selectivity (particularly the Ca2+:Na+ permeability ratios) of the conductance can be obtained. These, along with various other biophysical parameters, help construct the channel fingerprint. Of course, such measurements necessarily involve a substantial disturbance of the normal cellular environment, but concerns about the implications of such disruption can be largely overcome by confirming key findings using the perforated patch technique. An additional problem is that the specific conductances involved are often relatively small, placing technical demands on the experimenter. This is particularly true in the case of freshly dissociated primary cells, where it is often difficult to obtain seals with high enough resistance to reliably obtain good current recordings.

To date, we have only begun to scratch the surface of such biophysical characterization of receptor-operated Ca2+ channels. The number of different cell types studied, and the different conditions under which they have been examined, is limited. Nevertheless, I think a broad classification of different channel types is beginning to become apparent. First, the various conductances measured seem to fall into two distinct groups based on their relative Ca2+ selectivity. The first group includes channels that are highly selective for Ca2+, whereas the second group contains channels that are relatively nonselective. Even here, there is some confusion in the field. To some, the term "Ca2+ selectivity" refers to the ability of the entry pathway to pass Ca2+ relative to other divalent cations (usually Ba2+ or Sr2+ or both). However, I think it is preferable to use this term to define the selectivity of the channel to Ca2+ over other ions to which it is exposed under normal conditions—most notably Na+. This definition of selectivity is, at least, more consistent with the channel field in general. For example, voltage-gated channels often show a high permeability to Ba2+ (2), yet no one in the field would suggest that, physiologically, such channels are anything other than highly Ca2+ selective. More importantly, this definition has far greater relevance to the actual function of the channel in vivo (see below).

Based on this definition, existing evidence indicates that the conductances activated by agonists are often highly Ca2+ selective. Obvious examples would include the Ca2+ release–activated Ca2+ (CRAC) channels and their close relatives (36), and the arachidonate-regulated Ca2+ (ARC) channels (7, 8). Under normal circumstances, both of these conductances have a selectivity for Ca2+ over Na+ that is at least 1000-fold, their current-voltage relationships show a very positive reversal potential, and substitution of external Na+ with the impermeant N-methyl-d-glucamine (NMDG+) has a negligible effect on the measured macroscopic current. As such, these conductances have been identified in various different cell types, including primary cells, and clearly play a major role in the generation of Ca2+ entry signals. In contrast to these channels, there are several examples in which agonist-activated conductances are essentially nonselective (911). For these conductances, the observed current-voltage relationship is more or less linear, and any selectivity for Ca2+ over Na+ is modest at best (generally less than 10-fold). This distinction has relevance to the debate over any potential role of members of the canonical transient receptor potential channel (TRPC) family of proteins (1214) in receptor-activated Ca2+ entry. Despite extensive (often highly contradictory) evidence based on either overexpression or knockdown studies, I would argue that any claim that these channels represent candidates for any Ca2+-selective (as opposed to nonselective) conductance must remain moot—at least until some heteromeric combination of these proteins can be demonstrated to possess the appropriate biophysical properties.

Another classification system is based on the mechanism by which the channels are activated—namely, store-operated (capacitative), or independent of the status of intracellular Ca2+ stores (noncapacitative). Although this classification has a much longer history in the literature (15, 16), it remains far more problematic than the basic biophysical property of Ca2+ selectivity. Both Penner and Nilius have spelled out some of the key criteria for defining a conductance as being "store-operated" (1). These criteria include its nonadditive activation following depletion of the intracellular Ca2+ stores by several independent means involving both endogenous signaling molecules [such as inositol 1,4,5-trisphosphate (InsP3)], as well as pharmacological approaches (such as thapsigargin). However, a major hindrance to defining these conductances remains in that we simply do not have any consistent picture as to how the depletion of intracellular stores results in the activation of these channels (17).

What about the "noncapacitative" conductances? To be of any relevance, such conductances must be activated following activation of the appropriate cell surface receptors. However, because such activation may induce store depletion (detectable or not), how can such conductances conclusively be described as noncapacitative? The best approach would seem to be to demonstrate the activation of such a conductance in an additive manner following complete depletion of the stores by any, or all, of the methods discussed above. As pointed out by Bolotina and by Byron, among others (1), such a conductance should also demonstrate unique features that permit its distinction from the coexisting store-operated conductances in the same cell. Ideally, these distinguishing features should include unique biophysical properties that can be unequivocally assigned to the channel itself. Unless demonstrated to be genuinely specific, distinctions based on pharmacological approaches are questionable. Most current pharmacological tools fail to meet this criterion.

Which Channels Are Active Under Which Conditions, and What Do They Do?

I think there can be little doubt that, at high agonist concentrations, store-operated channels, whether Ca2+ selective or nonselective, are active. Under these conditions, evidence indicates that the intracellular Ca2+ stores are largely depleted and remain so in the continued presence of the agonist. The questions arise when low agonist concentrations are used. Here, the typically oscillatory Ca2+ signals that are generated are associated with only a transient and partial depletion of the overall Ca2+ store (18).

In contrast, current evidence, at least for the store-operated CRAC channels, indicates that a rather profound and sustained depletion of the stores is required for their effective activation (19). How, then, can such channels be operating under these circumstances? Yet there can be no doubt that enhanced Ca2+ entry occurs during such responses and has the important function of modulating the frequency of the oscillations (20, 21). This question has recently been discussed at length, and some intriguing models have been proposed (1). For example, Penner and Berridge have both suggested that activation of CRAC at low agonist concentrations may be achieved as a result of high, spatially confined, InsP3 concentrations close to the site of phospholipase C (PLC) activity—either through depletion of a local Ca2+ store or through direct conformational coupling between the InsP3 receptor and the entry channel (1). Although such mechanisms are entirely possible, the means to test the validity of these models are not immediately clear, and the fact remains that direct measurement of CRAC channel activity at these concentrations has yet to be demonstrated.

There is, however, extensive evidence for the activation of conductances under these conditions that are entirely distinct from those conductances activated, in the same cells, by store depletion. Although various nonselective cation conductances can be placed in this category, the most thoroughly studied are the highly Ca2+-selective ARC channels (7). These channels are biophysically distinct from, although superficially similar to, the coexisting CRAC-like store-operated conductances in the same cell (7). They are specifically activated by low agonist concentrations (22) in a manner that is entirely dependent on the generation of arachidonic acid, and independent of store depletion. In the case of human embryonic kidney (HEK) 293 cells, this activation is even independent of PLC activity and InsP3 generation (22, 23). Moreover, macroscopic ARC currents are additive to maximally activated store-operated currents in the same cell. Activation of these channels can be detected at agonist concentrations that produce just minimal Ca2+ signals, and becomes maximal at concentrations associated with the beginning of the transition from oscillatory responses to sustained elevations in cytosolic Ca2+ concentration and the initiation of store-operated Ca2+ (SOC) channel activity (22). Interestingly, the sustained elevation in cytosolic Ca2+ concentration induces the inhibition of the ARC channels through a calcineurin-dependent dephosphorylation process (24), resulting in what we have termed the "reciprocal regulation" of Ca2+ entry (22) (Fig. 1).

Fig. 1.

Diagram illustrating the "reciprocal regulation" of Ca2+ entry through ARC channels and store-operated Ca2+ (SOC) channels as agonist concentrations increase, and their relationship with the associated cytosolic Ca2+ signals generated. Whereas the ARC channels play a key role in determining the frequency of Ca2+ oscillations, the SOC channels principally regulate the amplitude of sustained Ca2+ signals. See (22, 24) for details.

Based on the above categorizations, we can begin to isolate the specific functions that these various conductances may have in overall Ca2+ signaling. For the nonselective conductances, whether store-operated or noncapacitative, it seems likely that their main action will be to depolarize the cell membrane potential. This would seem to impose some limits on their direct contribution to any Ca2+ entry. However, in cells that possess voltage-gated Ca2+ channels (such as smooth muscle cells, endocrine cells, etc.), this depolarization could activate these channels, resulting in substantial overall Ca2+ entry. Perhaps this is the main role of nonselective conductances in Ca2+ signaling in such cells. In contrast, the highly Ca2+-selective conductances (such as CRAC and ARC channels) presumably have a much more direct role in Ca2+ signaling. Consistent with the specific activation of ARC channels at low agonist concentrations, current evidence from various cell types (for instance, HEK293, HeLa, and mouse parotid acinar cells) indicates that Ca2+ entry through these channels is critical in modulating the frequency of Ca2+ oscillations (25, 26). Similarly, Ca2+ entry through the CRAC channels (and other similar store-operated channels) has been shown to be responsible for the amplitude of the sustained Ca2+ signals seen at high agonist concentrations, and for the refilling of the stores on termination of the signal. The question has been raised as to whether this constitutes a "physiologically relevant" function. Certainly, the maintenance of an appropriate Ca2+ concentration in the endoplasmic reticulum is critical for various functions [for instance, appropriate folding and processing of cellular proteins (27, 28)], and this could certainly represent a major function for store-operated channels. Whether these same channels can also play any role in modulating oscillatory Ca2+ signals will depend on whether their activation can be demonstrated at the relevant agonist concentrations.

So, where do we go from here? I think we have little alternative but to continue trying to characterize the relevant conductances involved in Ca2+ entry processes, and the specific conditions under which the different channel types are activated. This, at least, will provide the basic specifications to which any subsequently identified candidate channel proteins would have to conform. The problems involved are obvious, not least because it will often require detailed measurements of small conductances operating under conditions of low levels of stimulation. In this endeavor, I think we should be aware that so-called PLC-coupled receptors often also activate additional signaling pathways, so we need to consider mechanisms beyond the straightforward, essentially linear, receptor-PLC-InsP3-InsP3 receptor sequence. Moreover, although store-operated pathways have had a long-standing place in this field, we should not be surprised by the demonstration of additional distinct pathways for Ca2+ entry. We need look no further than the voltage-gated Ca2+ channel field to appreciate that a variety of different conductances have been tailored to provide unique signals under specific conditions, often within the same cell.

Editor's Note: This Perspective is part of a series related to an E-Conference held at Science's STKE. Invited participants were asked to provide Perspectives that summarized the ideas, conclusions, areas of controversy, and challenges for future work that emerged in the online discussion at;14.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
View Abstract

Stay Connected to Science Signaling

Navigate This Article