PerspectiveCell Biology

The IP3 Receptor as a Hub for Bcl-2 Family Proteins in Cell Death Control and Beyond

See allHide authors and affiliations

Science Signaling  11 Feb 2014:
Vol. 7, Issue 312, pp. pe4
DOI: 10.1126/scisignal.2005093


Bcl-2 family proteins generally have a proapoptotic or antiapoptotic role. In addition to their known roles at the mitochondria, Bcl-2 family proteins also influence signaling at the endoplasmic reticulum by binding to the inositol 1,4,5-trisphosphate receptor (IP3R), the most ubiquitous intracellular Ca2+ channel. Surprisingly, multiple Bcl-2 family proteins target different sites on the IP3R to elicit various effects on IP3R function, which suggests that the IP3R is an important hub for Bcl-2 family proteins in various physiological settings.

The inositol 1,4,5-trisphosphate (IP3) receptors (IP3Rs) are intracellular Ca2+-release channels present in all tissues of the body. Each of the three isoforms of IP3Rs (IP3R1, IP3R2, and IP3R3) can be divided in three main parts: the IP3-binding region, the regulatory region, and the C-terminal region containing the transmembrane and channel-forming domains. They form large homo- or heterotetrameric complexes whereby each monomer is approximately 2700 amino acids. IP3Rs play a central role in the generation and propagation of complex spatiotemporal Ca2+ signals that coordinate a plethora of cellular functions, including cell death (1).

The Bcl-2 family of proteins comprises about 20 homologs characterized by one or more Bcl-2 homology (BH) domains that mediate protein-protein interactions. Bcl-2 family proteins primarily regulate apoptosis by interaction with mitochondrial proteins but may also participate in other cellular functions (2). The interaction of Bcl-2 family proteins with IP3Rs to regulate IP3-induced Ca2+ release (IICR) participates in the control of programmed cell death as well as other cellular processes that depend on IP3R function. In this issue, in the context of Ca2+ signaling during early zebrafish development, Bonneau et al. (3) demonstrate a previously unknown interaction between the IP3R and the zebrafish homolog of mammalian Nr2 (also known as Bcl-2–like 10), (3), an antiapoptotic protein of the Bcl-2 family. Nrz binds the IP3-binding domain of the IP3R, which is now the fourth documented site on the IP3R that directly interacts with Bcl-2 family proteins (Fig. 1). Here, I describe the structural details and functional ramifications of these various interactions.

Fig. 1

Model indicating how the various Bcl-2 family proteins interact with and regulate the IP3R. IP3R1 is represented in the model, as most information is available for this isoform. Two of the four subunits are shown as line diagrams. From the N to the C terminus, IP3R1 consists of the N-terminal IP3-binding region (amino acids 1 to 578), the regulatory region (amino acids 579 to 2275), and the C-terminal region (amino acids 2276 to 2749) consisting of the channel domain with six transmembrane helices, a pore region, and a short cytosolic tail. IP3 binding induces a conformational change in the IP3R, leading to channel opening and Ca2+ release (white arrow). There is evidence for four different binding sites for Bcl-2 family members. The possible interactions are (A) Bcl-2, Bcl-xL, and Mcl-1, (B) Bcl-2, (C) Bok, and (D) Nrz. Proteins involved in the mechanism of action of the Bcl-2 family proteins are depicted: K-Ras4B, calcineurin (CaN), DARPP-32, protein phosphatase 1 (PP1), protein kinase A (PKA), and caspase 3 (Casp3). Phosphorylation sites on IP3R1 and on DARPP-32 are indicated by a P. ER, endoplasmic reticulum; GTP, guanosine triphosphate.


The C-terminal regions of IP3Rs (Fig. 1A) interact with Bcl-xL (4), Bcl-2, and Mcl-1 (5), Bcl-2 family proteins that are known to have antiapoptotic functions. The specific residues involved in these interactions are not known, but the sixth transmembrane domain of the IP3R is required (5), which suggests a hydrophobic interaction. Recruitment of Bcl-2 family proteins to the C terminus of the IP3R promotes Ca2+ signaling associated with cell survival at low concentrations of IP3—for example, by promoting mitochondrial bioenergetics. However, when Bcl-xL is bound to the C terminus of the IP3R in the presence of active K-Ras4B phosphorylated by protein kinase C, they form a trimolecular complex that antagonizes the Bcl-xL–dependent prosurvival Ca2+ signaling and thus promotes cell death (6).

In addition to the C terminus, Bcl-2 binds the middle of the regulatory region of the IP3R (7) (Fig. 1B). This site (amino acids 1389 to 1408 in IP3R1) is conserved among IP3R isoforms and binds the BH4 domain of Bcl-2 but not of Bcl-xL, because of the substitution of a critical lysine (Lys17) in Bcl-2 by an aspartate (Asp11) in Bcl-xL (8). The interaction between Bcl-2 and the regulatory region of IP3R protects cells from death by preventing large Ca2+ signals that stimulate apoptosis (7, 8). Mechanistically, the potent inhibition of the IP3R by Bcl-2 is not fully understood, but it may be at least partly due to the interaction of Bcl-2 with calcineurin and DARPP-32, which in the presence of Ca2+ promotes protein phosphatase 1–mediated dephosphorylation of the IP3R (9).

The multidomain proapoptotic protein Bok also interacts with the regulatory region of IP3R but elicits a different response than Bcl-2. Unlike Bcl-2, the BH4 domain of Bok binds a different site in the regulatory region (amino acids 1895 to 1903 in IP3R1) and interacts with IP3R1 and IP3R2 but not IP3R3 (10) (Fig. 1C). The binding affinity of Bok to the IP3R is higher than that of Bcl-2 or Bcl-xL, but Bok does not affect IP3R function directly. Rather, the binding of Bok protects IP3R1 against caspase 3–mediated proteolysis at amino acids 1888 to 1891, potentially by steric hindrance. Given that caspase 3 is active during apoptosis (11), Bok is likely to stabilize the IP3R during this process.

Similar to Bcl-2 and Bok, the BH4 domain of Nrz interacts with IP3R1, but it binds the IP3-binding domain (amino acids 226 to 581) rather than the regulatory region (3) (Fig. 1D). Whereas the BH4 domain of Bcl-2 is sufficient for both binding the IP3R and inhibiting IICR (8), and even though the BH4 domain of Nrz is sufficient to bind IP3R1, a 94–amino acid region of the N terminus of Nrz including the BH4, BH3, and BH1 domains is required to inhibit IICR (3). The interaction between Nrz and IP3R1 appears to be regulated because phospho-mimetic mutations in Nrz antagonize its interaction with IP3R1, which correlates with the increased phosphorylation of Nrz observed during epiboly. Thus, Nrz inhibits IICR by a different mechanism than Bcl-2, which may be to produce a conformational change of the IP3R that leads to decreased affinity for IP3 or impaired signal transmission to the channel region (or both), or to create steric hindrance limiting access of IP3 to the IP3-binding site.

The diverse relationships between Bcl-2 family proteins and IP3Rs underscore the complexity of their role in the regulation of IICR. Structure-function analyses of the IP3R indicate that the coupling of IP3 binding to the opening of the Ca2+-release channel depends on conformational changes in the backbone of the IP3R and direct interactions between its N and C termini (12). Thus, binding of regulatory proteins to the N terminus (3), the middle of the regulatory region (7, 10), or the C terminus (4) of the IP3R can potentially modulate IICR. Although some redundancy among Bcl-2 family proteins is possible, the presence of different binding sites on the IP3R may enable fine-tuning of IICR and different regulation under various cellular conditions.

Not all Bcl-2 family proteins have been investigated with respect to Ca2+ signaling, so additional interactions among Bcl-2 family proteins and IP3Rs are likely to be discovered. Bcl-2 family proteins differ substantially in sequence and structure, and IP3Rs have multiple sites for potential interactions. Whether and how these interactions affect IICR should be considered in the context of the multiple mechanisms that control IP3R activity, as these may interact in additive, synergistic, or antagonistic ways. In addition, future research should attempt to distinguish between direct and indirect effects on the IP3R, keeping in mind that direct effects also could depend on the presence of other interacting proteins (6) or posttranslational modifications such as phosphorylation (3). Indirect effects have been proposed for Bcl-2 (9) and Bok (10), and Bcl-2 has additional binding partners that could affect Ca2+ signaling, including Beclin 1 and Bax Inhibitor–1. Bcl-2 family proteins therefore could act as scaffolds anchoring IP3Rs to other regulatory proteins. Thus, any one Bcl-2 family protein could have multiple effects with respect to IICR.

We should expect the discovery of additional mechanisms by which Bcl-2 family proteins modulate intracellular Ca2+ signals. These mechanisms may not necessarily affect cell life or death, but they may also control other cellular processes such as cell movement, as highlighted by Bonneau et al. (3). The intimate and complex interactions among IP3Rs and Bcl-2 family proteins are therefore of great relevance for a broad array of biological functions and dysfunctions, including development, cancer, and neurodegeneration.

References and Notes

Acknowledgments: I thank H. De Smedt, G. Bultynck, and C. W. Distelhorst for helpful discussions and for critically reading this manuscript. Work perfomed in the author’s laboratory on the topic was supported by the Research Foundation–Flanders.
View Abstract

Navigate This Article