28.7: Calcium Signaling

Search Fundamentals of Biochemistry

Learning Goals 

(Learning goals written by Claude, Sonnet 4.6, Anthropic)

Ca²⁺ Homeostasis: Buffering, Storage, and Membrane Transport

• Explain why maintaining low basal cytosolic Ca²⁺ (~100 nM) is essential for its function as a second messenger, and describe the three main buffering strategies cells employ: sequestration by soluble EF-hand proteins (e.g., calmodulin, calbindin), active transport into organelles by ATP-driven pumps (SERCA, PMCA), and ion exchange via NCX.
• Describe the structural and functional differences between SERCA and PMCA pumps—including their transmembrane topology, Ca²⁺:ATP stoichiometry, and regulation by calmodulin binding and PKA/PKC phosphorylation—and explain how these differences suit each pump to its specific cellular role.
• Explain the mechanism of Store-Operated Ca²⁺ Entry (SOCE): describe how ER Ca²⁺ depletion is sensed by STIM1 (via its EF-hand domain), how STIM1 then activates the plasma membrane ORAI1/CRAC channel, and how this pathway sustains prolonged Ca²⁺ signaling, including NFAT-mediated gene transcription.
• Describe how mitochondria contribute to Ca²⁺ homeostasis through Mitochondrial Associated Membranes (MAMs), where IP3R3 on the ER, VDAC1 on the outer mitochondrial membrane, and the MCU complex on the inner mitochondrial membrane coordinate Ca²⁺ transfer that influences cellular metabolism, autophagy, and apoptosis.

Calmodulin and CaMKII: Ca²⁺ Signal Transduction

• Describe calmodulin's four EF-hand Ca²⁺-binding sites, explain how sequential Ca²⁺ binding drives the apo→semi-open→holo conformational transition, and explain how the resulting exposure of hydrophobic surfaces on the central amphiphilic helix enables calmodulin to bind and regulate diverse target proteins, including CAMKs, PMCA, and ion channels.
• Explain the activation mechanism of CaMKII: describe how Ca²⁺/calmodulin binding relieves autoinhibition, exposes the active site, and enables trans-autophosphorylation of Thr286 within the dodecameric holoenzyme, generating a Ca²⁺-independent autonomous activity that persists after calmodulin dissociation until PP1/PP2A-mediated dephosphorylation restores the inactive state.
• Distinguish between amplitude-encoded and frequency-encoded Ca²⁺ signaling, and explain why CaMKII's dodecameric architecture and autonomous activation make it particularly well suited to act as a molecular frequency decoder for Ca²⁺ oscillations in neurons.

Ca²⁺ Oscillations: Integration with Other Pathways and Computational Modeling

• Explain how GPCR→phospholipase C→IP3 signaling generates cytosolic Ca²⁺ oscillations through a Ca²⁺-induced Ca²⁺ release (CICR) mechanism, including the biphasic regulation of IP3R by cytosolic Ca²⁺ (activation at low [Ca²⁺] and inhibition at high [Ca²⁺]), and describe how SERCA and PMCA pumps restore basal levels between spikes.
• Using the Goldbeter mathematical model as a framework, explain how only two interacting variables (cytosolic Ca²⁺ concentration Z and sequestered pool concentration Y) are sufficient to generate sustained oscillations without requiring oscillation in IP3 itself, and identify which model parameters (v0, Vm2, Vm3, beta) most strongly influence oscillation frequency and amplitude.

The following is adapted directly and modified from Sharma et al. Biomedicines 2021, 9(9), 1077; https://doi.org/10.3390/biomedicines9091077. Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Introduction

Ca²⁺ is central to numerous cellular processes and functions. Its chemical features, including a low hydration energy, high polarizability, relative flexibility of coordination sites and bond length, and a large concentration gradient across cellular membranes (100 nM intracellular to 2 mM extracellular) due to low intracellular levels, make it the ion of choice at the core of cellular signaling in prokaryotes and eukaryotes alike.

In studying calcium ion signaling, we will focus on four key areas:

• Buffering of intracellular Ca²⁺ ion concentrations: Basal low levels must be maintained to allow transient increases to act as signals. Ca²⁺ ions are, hence, no different from other second messengers like cAMP, for example. What matters is the rise from a basal level to a threshold concentration level that allows binding to signaling proteins and subsequent signal transmission.
• Storage of intracellular Ca²⁺: Calcium ions are stored in organelles such as the ER, mitochondria, and lysosomes. The ions must be released in response to specific signals and then returned to the storage organelle to maintain basal Ca²⁺ levels.
• Signaling pathways activated by Ca²⁺ ions: We have seen many pathways stimulated by increases in second messengers and by phosphorylation of lipid and protein molecules within interconnected pathways. We will return to several previously studied pathways to examine how they integrate with Ca²⁺ signaling.
• Ca²⁺ binding proteins and their binding partners in signaling pathways: We will focus on one key Ca²⁺ binding protein, calmodulin (CAM), and the kinase it activates, the Ca2+/CAM protein kinases (CAMKs).

The next two sections are adapted and modified from Sharma et al. Biomedicines 2021, 9(9), 1077; https://doi.org/10.3390/biomedicines9091077. Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Buffering of intracellular Ca²⁺ ion concentrations

Calcium ions, like the hydronium ion, must be buffered in cells. Otherwise, their potential as a signaling agent would be compromised. The mechanisms adopted by cells for intracellular Ca²⁺ buffering involve sequestration by special proteins, as shown in Figure \(\PageIndex{1}\).

Figure \(\PageIndex{1}\): Types of intracellular Ca²⁺ buffers.

Intracellular Ca²⁺ levels are regulated by binding to specific proteins or by sequestration within distinct cellular compartments. The three main ways intracellular Ca²⁺ is buffered are depicted in Figure 1. These include soluble or unbound proteins that are found in the cytosol or inside organelles, membrane proteins (generally Ca²⁺ channels, ATP-driven pumps (SERCA or PMCA), and ion exchangers (NCX) inside organelles like the endoplasmic reticulum (ER), mitochondria, acidic vesicles (mainly lysosomes and Golgi bodies), or organelle junctions (endoplasmic reticulum-plasma membrane (ER-PM), endoplasmic reticulum-mitochondria, or endoplasmic reticulum-lysosomes). The major players regulating inter-organellar Ca²⁺ transfer are IP₃R (inositol-3,4,5-triphosphate receptor), NCX (sodium-Ca²⁺ exchanger), ORAI1/CRACM1 (Ca²⁺ release-activated modulator 1), PMCA (plasma-membrane Ca²⁺ ATPase), SERCA (sarco-endoplasmic reticulum Ca²⁺ ATPase), STIM1 (stromal interaction molecule 1), SOCE (store-operated Ca²⁺ entry), TPC1/2 (two-pore channel), TRP (transient receptor potential), and VDAC (voltage-dependent anion channel). We will discuss some of them below.

These proteins sequester cytosolic Ca²⁺ upon sensing elevated Ca²⁺ levels and relay the associated cellular messages. Other proteins that act as intracellular Ca²⁺ buffers are present in lipid bilayers, plasma membranes, or organelle membranes, including pumps and transporters. Apart from these proteins, intracellular Ca²⁺ is regulated by inter-organellar transport and the influx of Ca²⁺ ions from the extracellular space.

Storage of intracellular Ca²⁺ - Proteins

Soluble and Unbound Intracellular Proteins: Calmodulin, Calbindin, and Calretinin

Nonmembrane-associated proteins inside a cell can act as both Ca²⁺ sensors and buffers. Most of these proteins have an EF-hand motif that allows Ca²⁺ ions to bind, triggering changes in protein folding and influencing downstream or linked cellular pathways. Calmodulin (CaM) is one of the best-studied and ubiquitously expressed Ca²⁺-sensing proteins known to play a key role in intracellular Ca²⁺ homeostasis. It is a prototype for intracellular Ca²⁺ sensors. It has a 148-amino-acid structure with two Ca²⁺-binding sites in two separate lobes, each with two EF-hand motifs. The lobes are connected in the holo (Ca²⁺-bound form that binds other proteins. N- and C-terminal alpha-helices with a Ca²⁺ coordination loop in between, providing affinity for Ca²⁺ ion docking and sequestration. The ability of CaM to transmit a change in free intracellular Ca²⁺ levels into a signal depends on the conformational flexibility of the Ca²⁺-dependent (apo) form. CaM can exist in a Ca²⁺-free closed conformational state (Apo-CaM), a semi-open (Ca₂-CaM), or an open state (Holo-CaM or Ca₄-CaM) after Ca²⁺-binding, as shown in Figure \(\PageIndex{2}\).

Figure \(\PageIndex{2}\): Overview of calmodulin structure and Ca²⁺ binding conformations.

Figure \(\PageIndex{3}\) shows an interactive iCn3D model of human Holo-calmodulin with 4 bound Ca²⁺ ions (1CLL)

Figure \(\PageIndex{3}\): Holo-calmodulin with 4 bound Ca²⁺ ions (1CLL). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...8nT5HXgNJ4ymc6

Side chains interacting with one Ca²⁺ ion in an EF-hand are shown in the model. The central helix is colored by hydrophobicity, with green indicating greater hydrophobicity. This amphiphilic helix can bind target proteins in this region through nonpolar interactions.

Figure \(\PageIndex{4a}\) shows the change in conformation in the first Ca⁺² site (D20-E31, 1CFD, Xenopus laevis) on binding Ca⁺² (D20-E31, 1CLL, human).  The two Ca²⁺ sites (not the entire protein sequence) were aligned in this image using PyMOL.  Carbon atoms are shown in cyan, and the green sphere is Ca⁺². Other atoms are shown with normal CPK colors. 

Figure \(\PageIndex{4b}\): Conformational changes in the first Ca⁺² site (D20-E31, 1CFD, Xenopus laevis) on binding Ca⁺² (D20-E31, 1CLL, human). 

Note the different orientations of the carboxylate side chains as they move to interact with the Ca²⁺ ions.

When applied to all the Ca²⁺ binding sites, these localized conformational changes ultimately lead to large conformational changes in the entire protein.

Figure \(\PageIndex{4b}\) shows the conformational change in the cartoon representation of apo-CAM (cyan, 1CFD, Xenopus laevis) on binding four Ca⁺² ions (magenta, 1CLL, human).  The amino acid sequences of calmodulins from both species are identical.  Both chains were aligned and displayed with Pymol.

Figure \(\PageIndex{4b}\): Conformational changes in the cartoon representation of apo-CAM (cyan, 1CFD, Xenopus laevis) on binding four Ca⁺² ions (magenta, 1CLL, human).

In the holo-form, a fully formed central helix connects to the two lobes of a "dumbbell-shaped" structure.

Differential Ca²⁺ binding to the two lobes of CaM enables this protein to rapidly buffer a wide range of free intracellular Ca²⁺. The presence of methionine residues in its lobes and the plasticity of the central linker in its structure also provide CaM with properties to function as an adaptor protein in intracellular Ca²⁺ signaling. CaM can bind to several targets or effector molecules at variable distances and in multiple orientations to mediate changes in intracellular Ca²⁺ signaling. Major effector proteins regulated by CaM binding and relevant to Ca²⁺ homeostasis include EGFR, PI3K, and connexins.

CaM is required for spatial and temporal regulation of [Ca²⁺], as evident by its role in modulating (activating or inactivating) Ca²⁺ pumps (such as PMCA and SERCA) and Ca²⁺ channels (such as CaV1.3, TRPV5 and 6, ORAI). CaM also acts via serine/threonine kinases known as Calmodulin-activated Kinases (CaMKs) to influence cellular processes such as proliferation (for example, centrosome duplication at G1/S or the anaphase-to-metaphase transition via CaMKII). We will discuss those in detail below.

Integral Membrane protein molecular buffers: SERCA, PMCA, NCX, and TRP

Integral membrane protein Ca²⁺ buffers primarily translocate free Ca²⁺ between domains and organelles. These mainly comprise ion exchangers, channels, and ATP-driven pumps. SERCA, Sarcoendoplasmic Reticulum Ca²⁺ ATPase, is an ATP-dependent ion pump known to significantly maintain free cytosolic Ca²⁺ concentration via actively pumping the ion into the endoplasmic reticulum (or sarcoplasmic reticulum in muscle cells). They share a general structure that includes 10-pass transmembrane helices and three cytoplasmic domain lobes, as shown in Figure \(\PageIndex{5}\)

Figure \(\PageIndex{5}\): Sarcoplasmic endoplasmic Ca²⁺-ATPase structure (Panel A) and mechanism of Ca²⁺ ion transport (Panel B).

P-type Ca²⁺-ATPases also exist within the plasma membrane and maintain cytosolic Ca²⁺ levels by transferring them into the extracellular space. The Plasma Membrane Ca²⁺ ATPases (PMCAs) can transport one Ca²⁺ ion per ATP molecule, whereas SERCA transports two Ca²⁺ ions per ATP molecule. The general structure of such Ca²⁺ transporters comprises 10 transmembrane segments with large cytosolic loops TM 1–2 and TM 3–4, and cytosolic N- and C-termini tails are shown in Figure \(\PageIndex{6}\), panel A below.

Figure \(\PageIndex{6}\): The general structure of such Ca²⁺ transporters. (A) is the Plasma membrane Ca²⁺-ATPase (PMCA) and (B) shows the Na⁺–Ca²⁺ exchanger (NCX)

The cytoplasmic region of PMCA (left) contains three loop structures that serve as binding sites for signaling molecules such as CAM. They also have phosphorylation sites for additional regulation. The C-terminal tail contains additional CAM sites and a PKA site. It has a PDZ domain that can anchor the protein to cytoskeletal components. Differential RNA processing leads to variations in the amino acid sequence in this region and, hence, in binding specificity. The binding of CaM reverses auto-inhibition of the pump due to conformational shifts, which displace the C-tail from the cytosolic loops. Other means of autoinhibition reversal include phosphorylation of the C-tail (Ser/Thr residues) by protein kinase A or C, proteolytic cleavage of the C-tail, or dimerization via the C-terminus.

Transient Receptor Potential (TRP) channels function similarly in neurons, epithelial cells, and immune cells. The Mammalian TRP channel superfamily is composed of 28 family members belonging to six subfamilies—TRPC (Canonical), TRPA (Ankyrin), TRPM (Melastatin), TRPV (Vanilloid), TRPP (Polycystin), and TRPML (Mucopilin)—that differ in their sensitivity to various sensory stimulations and affinity for cation (including Ca²⁺ ions) sequestration. Commonly, TRP family members share a structure with six transmembrane domains, intracellular N- and C-termini, and a pore-forming TM 5–6 loop. The cytoplasmic C-terminus of each subunit is a site for protein interaction and post-translational modification. The C-tail of these channels can have PDZ protein binding domains (TRPV and C), sites for interaction with G-proteins (Gq/11)/calmodulin/PLCβ, ADP-ribose binding (NUDIX; TRPM2), or a PLC-interacting kinase (PLIK; TRPM6 and 7) domain.

TRP channels act as activators, integrators, and downstream effectors of Ca²⁺ signaling at the plasma membrane and in intracellular compartments. Many members of the TRPC subfamily are activated by DAG (diacylglycerol), which is produced by PLC β- or γ-mediated cleavage of phosphatidylinositol 4,5-bisphosphate (PIP₂) after ligand binding at GPCRs or RTKs. TRPP1/2, TRPA1, TRPM8, and TRPV1-4 are all expressed on the ER membrane. At this site, PLC-independent activation of TRP channels (such as TRPV1) is thought to induce ER Ca²⁺ release via inositol triphosphate receptors (IP3Rs), thereby triggering bulk entry of extracellular Ca²⁺ into the cell. On the flip side, cytosolic Ca²⁺ regulates the activity of TRP channels in response to physiological stimuli. This regulatory effect is usually mediated by CaM binding (inhibition of TRPV5 and TRPV6, and sensitization of TRPV3) and indirectly by CaM-binding kinase II (CaMKII). These activities are shown in Figure \(\PageIndex{7}\).

Figure \(\PageIndex{7}\): TRP Channels in Signaling. DAG, Diacylglycerol; GPCRs, G-protein coupled receptors; IP₃, inositol 1,4,5-triphosphate; ORAI (Ca²⁺ release activated Ca²⁺ modulator; RTKs, Receptor-tyrosine kinases; ROCE, Receptor-Operated Ca²⁺ Entry; SOC, store-operated Ca²⁺; SOCE, Store-Operated Ca²⁺ Entry; STIM, stromal interaction molecule; PIP₂, phosphatidylinositol 4,5-bisphosphate; PKC, protein kinase C; TRP, transient receptor potential; VDCCs, voltage-dependent Ca²⁺ channels.

Storage of intracellular Ca²⁺ ions - organelles

Endoplasmic Reticulum: STIM, ORAI, IP₃Rs, and TRPC1 in SOCE and SOCIC Ca²⁺ Entry Models

The endoplasmic reticulum (ER) is the largest and most dynamic organelle reservoir of intracellular Ca²⁺ and is therefore central to many signaling processes for protein synthesis, folding, and post-translational modifications. In contrast to the cytosol, ER Ca²⁺ ion levels can range from 100 μM to 1 mM based on the cell type. ER and other intracellular organelles buffer excessive cytosolic Ca²⁺ by housing Ca²⁺-binding proteins (e.g., calreticulin in the ER) and by active transport (e.g., SERCA pumps in the ER).

• Depletion of Ca²⁺ from the ER lumen activates an indirect mode of Ca²⁺ entry into the organelle, which is called Store-Operated Ca²⁺ Entry (SOCE) or Ca²⁺ Release Activated Ca²⁺ (CRAC) entry. It is activated when plasma membrane receptors, such as PLC-coupled GPCRs (not voltage-gated channels), trigger Ca²⁺ release from the organelle.
• Exhaustion of the intraluminal ER Ca²⁺ ion store following such prolonged release is then sensed by STIM (Stromal Interaction Molecule) tethered to the ER membrane and relayed to the plasma membrane's CRAC channels.

Figure \(\PageIndex{8}\) shows the domain structures of STIM 1/2, ORAI 1-3, and IP3Rs (panel A) and the mechanism of Ca²⁺ influx into the cell (panel B).

Figure \(\PageIndex{8}\): Role of STIM< ORAI and IP3Rs in Ca²⁺ influx.

IP3 Receptors (IP3Rs), upon IP3 stimulation, open and allow Ca²⁺ influx from the organelle lumen into the cytoplasm. After activation of IP3Rs on the ER membrane by ligand IP3 and cytosolic Ca²⁺ from activation of phospholipase C, STIM dimers are activated once the luminal Ca²⁺ concentration drops below basal levels. These receptors provide intracellular Ca²⁺ ions for downstream Ca²⁺ signaling, including NFAT-mediated transcription. The red semi-circle in the ER represents high luminal Ca²⁺ levels, the pink semi-circle is for moderately low Ca²⁺ ion concentration, and the pale semi-circle indicates extremely low Ca²⁺ concentration. CC, coiled-coil; NFAT, nuclear factor of activated T-cells; SAM, sterile alpha motif; SOAR, STIM1 Orai1-activating region; TM, transmembrane.

Mitochondria and Acidic Vesicles (Mainly Lysosomes)

Mitochondria also play a critical role in maintaining Ca²⁺ ion levels in the cytosol and endoplasmic reticulum. They are found mostly aggregated around the nucleus and store intracellular Ca²⁺ at levels similar to those in the cytosol (0.1 μM). The electrochemical proton gradient (membrane potential, Ψmt = −150 to −180 mV) and close association with the ER are the key factors responsible for mitochondrial intracellular Ca²⁺ uptake. The free movement of small molecules (less than 5 kDa) from the outer mitochondrial membrane (OMM) into the inner mitochondrial space and their impermeability across the latter generate a high electrochemical proton gradient for ATP synthesis. This gradient simultaneously draws Ca²⁺ ions from the cytosol.

Transfer of Ca²⁺ ions from the ER to mitochondria occurs at specialized microdomains or contact sites known as Mitochondrial Associated Membranes (MAMs). These are characterized by the ER and OMM, which are 10–25 nm apart and are strewn with clusters of channels, transporters, exchangers, and tethering proteins that facilitate Ca²⁺ ion transfer. IP3Rs localized at the ER side of the MAMs release Ca²⁺ ions that gate voltage-dependent anion channels (VDACs) on the OMM. VDACs (1, 2, and 3) are 30-kDa polypeptides with a 19-stranded beta-barrel structure that regulate the flux of metabolites (polyvalent anions such as ADP and ATP) across the outer mitochondrial membranes. These channels transport cations, including Ca²⁺, more readily than anions like chloride. Due to voltage-dependent electrostatic gating, the ion selectivity and flux across VDACs change between open and closed states. Figure \(\PageIndex{9}\) shows coupled mitochondrial and lysosomal effects on intracellular Ca²⁺ signaling.

Figure \(\PageIndex{9}\): Mitochondrial and lysosomal impact on intracellular Ca²⁺ signal.

Primary components of Ca²⁺ signaling at the mitochondrial-associated membranes (MAMs) include IP₃R3 on the endoplasmic reticulum, VDAC1 on the outer mitochondrial membrane, and the MCU complex on the inner mitochondrial membrane [151,154,156,161]. Transport of Ca²⁺ ions from the ER to mitochondria plays a crucial role in cellular metabolism (autophagy), cell survival (during unfolded protein response and cell death), lipid production, and distribution. 

While IP3 acts as the dominant Ca²⁺-mobilizing messenger, cADPR (cyclic ADP-ribose) and NAADP  (nicotinic acid adenine dinucleotide phosphate) are also known to modulate intracellular Ca²⁺ stores. cADPR evokes Ca²⁺ release from the ER by acting on ryanodine receptors (RyRs; the counterpart of IP3Rs in myocytes and co-expressed in other cell types). NAADP releases Ca²⁺ from acidic and/or secretory vesicles such as lysosomes and endosomes. In most mammalian cells, lysosomes comprise ~5 percent of cell volume and store intracellular Ca²⁺ at levels similar to those in the ER (0.5 mM). Due to their relatively small size compared with ER, lysosomes release nearly undetectable amounts of intracellular Ca²⁺ in response to an NAADP trigger.

Signaling pathways activated by Ca²⁺ ions

It is daunting for readers and writers to introduce myriad new signaling pathways. Instead, we will show how Ca²⁺ signaling fits into other pathways we have already discussed. A summary showing how Ca²⁺ signaling integrates with other pathways is shown in Figure \(\PageIndex{10}\).

Figure \(\PageIndex{10}\): https://www.creative-diagnostics.com...ng-pathway.htm

An Overview of the Calcium Signaling Pathways

Ca²⁺ signaling, as described above, requires ion buffer, organelle storage, and Ca²⁺ protein pumps and channels. The concentration (amplitude) and frequency of Ca²⁺ release affect signaling. Figure \(\PageIndex{10}\) shows the importance of upstream signaling through GPCRs, phospholipase C, RTKs, and IP3/DAGs. Ca²⁺ also enters the nucleus via IP3 receptors (IP3Rs) and ryanodine receptors (RYR). An important family of cytoplasmic transcription factors, the Nuclear factor of activated T-cells (NFAT), is activated by calcium ion signaling and is important in immune responses and the development of the muscle and nervous systems.

As described above, Ca²⁺ release from the ER is sensed by integral ER membrane proteins called STIMs. These bind Ca²⁺ ions as a buffering system, but the STIM-bound Ca²⁺ ions are also released if most of the ER calcium is depleted. This led to their self-association and subsequent activation of ORAI1, a component of the CRAC complex in cell membranes, which allows extracellular Ca²⁺ ions to enter the cell via store-operated calcium entry (SOCE). Sufficient calcium now accumulates in the cell to activate the transcription factor NFAT through dephosphorylation by calcineurin (PP2B), also abbreviated CaN in Figure \(\PageIndex{10}\). NFAT then translocates into the nucleus and activates gene transcription. It has also been shown that nuclear calcium ions can directly activate the cAMP response element-binding protein (CREB), a transcription factor that drives gene transcription. In addition, the CAM:CAMKII complex can translocate into the nucleus. Calcium signals also activate the ERK1/2-MAPK cascade.

Ca²⁺ binding proteins and their binding partners in signaling pathways

We have already described the key calcium-binding protein, calmodulin. On binding Ca²⁺, it undergoes a profound conformational change that allows it to interact with a family of key signaling kinases called CAM and Ca²⁺/CAM-Dependent Protein Kinases (CAMKs).

Ca²⁺/calmodulin-dependent protein kinase (CAMK) is a key signaling protein activated through the actions of Ca²⁺-calmodulin. Activated CAMK is a Ser/Thr kinase. There are many types of CAMKs. We will focus on multifunctional CAMKs that can phosphorylate multiple target proteins. These are important in learning and memory, metabolism, and gene transcription. As with other kinases, they have catalytic and regulatory domains. Some, like CAMK II, have association domains that enable the formation of CAMK II multimers. In addition, they must have a CAM binding domain. As with all kinases, the CAMKs must be able to switch from an inactive to an active form.

CAMKI has a catalytic and substrate-binding domain, and an autoinhibitory domain that blocks the active site. It is activated by the upstream kinase CAMKK (a naming scheme similar to the MAPK cascade) upon Ca²⁺ binding to CAM. CAMKI is found in the cytoplasm and nucleus and helps regulate transcription, the cell cycle, hormone production, cell differentiation, actin filament organization, and neurite outgrowth.

We will focus our attention on CAMKII, which has four isoforms (α, β, γ, and δ). It is activated by binding of Ca²⁺/CAM, which promotes autophosphorylation. After that, it is active in the absence of CAM. It is important for learning and memory, synaptic formation in neurons, and the regulation of sarcoplasmic reticulum Ca²⁺ transport in skeletal muscle.

They have an N-terminal catalytic domain and a C-terminal association domain that facilitates multimer formation into large holoenzymes with 12 or 14 CAMK monomeric subunits (a homomer or heteromer). These two domains are separated by a linker/regulatory domain containing a CAM-binding site, an autoinhibitory region, and key Ser and Thr side chains that are targets for phosphorylation. Figure \(\PageIndex{11}\) shows the domain structure of the CAMKII monomer (a), the overall structure of a homododecamer (b), and the mechanism for activation of kinase activity (c).

Figure \(\PageIndex{11}\): CamKII - structure and activation. Zhang, X., Connelly, J., Levitan, E.S. et al. Calcium/Calmodulin–Dependent Protein Kinase II in Cerebrovascular Diseases. Transl. Stroke Res. 12, 513–529 (2021). https://doi.org/10.1007/s12975-021-00901-9. Creative Commons Attribution 4.0 International License, http://creativecommons.org/licenses/by/4.0/.

T253, T286, and T305/306 are targets of autophosphorylation. M281/282 are also sites for oxidative modification. The C-terminal association domain allows multimer formation. It has a variable region that differentiates CaMKII subtypes. Panel (b) shows the multimer that forms on interactions of multiple association domains on different CAMKIIs. Panel (c) shows that binding of CAM promotes the phosphorylation of key residues, including T286 (through autophosphorylation).

In the absence of Ca²⁺/CAM, the  T286 amino acid forms interactions with the catalytic domain to maintain the inactive conformation. The regulatory domain effectively autoinhibits the kinase domain. Upon formation of the Ca²⁺/CAM/CAMKII complex, a conformational change releases the catalytic domain from autoinhibition, exposing the active site. Each subunit in the dodecamer is activated separately. T286 is now free to be "autophosphorylated" by an adjacent active subunit. Once phosphorylated, pT286 prevents the autoinhibitory region from rebinding to the catalytic domain, even after CAM dissociates. At this point, CAMKII is active in the absence of Ca²⁺/CAM.

Phosphorylation of T286 also regulates its binding to target proteins, thereby promoting their phosphorylation. In addition, CAMKII can autophosphorylate at T254 and T306, thereby further affecting its activity. T306 is only autophosphorylated after CAM dissociates, and the enzyme is autonomously active. Dephosphorylation by PP1 and PP2A returns the enzyme to an inactive state.

Figure \(\PageIndex{12}\) shows an interactive iCn3D model of a single subunit of human Ca²⁺ Calmodulin- Dependent Kinase II Holoenzyme (3SOA)

Figure \(\PageIndex{12}\): Single subunit of human Ca²⁺ Calmodulin- Dependent Kinase II Holoenzyme (3SOA). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...S4XAczeqqPE5Y9

The catalytic domain is shown in orange, the association domain in green, and the linker domain in light cyan. The CAM binding site is the spacefill light cyan site in the linker domain. The side chains of Thr 253 and Thr 286, phosphorylation sites, are shown in spacefill CPK colors and labeled. The side chains of Cys 280 and Met 281, the redox-regulation site, are shown in spacefill CPK colors and labeled. Bosutinib (shown in ball and stick) is an inhibitor of the BCR-ABL and src tyrosine kinase used to treat chronic myelogenous leukemia. It is also an inhibitor of CAMKII, which binds in the catalytic subunit's ATP-binding site. Finally, the activation loop is shown in red.

Figure \(\PageIndex{13}\) shows an interactive iCn3D model of Human Ca²⁺ Calmodulin- Dependent Kinase II Holoenzyme (3SOA)

Figure \(\PageIndex{13}\): Human Ca2+ Calmodulin- Dependent Kinase II Holoenzyme (3SOA). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...DPiWLo5VrgiLp9

The responses of most protein kinases we have studied depend on the concentration of a binding ligand (such as Ca²⁺). Imagine the ligand modulating the size (amplitude) of the dose-response curve.  The CAMKII dodecamer also responds to the frequency of Ca²⁺ waves or spikes as they are released from intracellular organelles. This is important in neurons, where CAMKII phosphorylates ion channels that control neuronal responses. If the frequency of Ca²⁺ spikes reaches a certain threshold level, the enzyme no longer depends on Ca²⁺.

Modeling of Ca²⁺ signaling

Model for signal-induced Ca²⁺ oscillations and their frequency encoding through protein phosphorylation.

Reference paper: ALBERT GOLDBETER, GENEVIVE DUPONT, AND MICHAEL J. BERRIDGE , Proc. Natl. Acad. Sci. USA, Vol. 87, pp. 1461-1465, February 1990, Biophysics

In response to hormones and neurotransmitters, cyclic intracellular oscillations/spikes (as a function of time) and spatial waves of cytoplasmic Ca²⁺ arise.  The Ca²⁺ ions derive from opening cell membrane channels and, more importantly, from releasing and recapturing the ions from intracellular compartments such as the endoplasmic reticulum (ER).  Signaling through membrane-bound GPCRs can activate phospholipase C, generating inositol 1,4,5-trisphosphate (IP3 or InsP₃) as a key second messenger.  IP3 can interact with the IP3 receptor (IP3R) on ER membranes, releasing intracellular Ca²⁺ stores into the cytoplasm and leading to oscillations in its concentration.   These processes are shown in Figure \(\PageIndex{14}\) below.

Figure \(\PageIndex{14}\): Ca²⁺ oscillations in response to inositol trisphosphate (IP₃) increase, with and without Ca²⁺ influx from extracellular space.  Catacuzzeno L, Franciolini F. Role of KCa3.1 Channels in Modulating Ca²⁺ Oscillations during Glioblastoma Cell Migration and Invasion. Int J Mol Sci. 2018 Sep 29;19(10):2970. doi: 10.3390/ijms19102970. PMID: 30274242; PMCID: PMC6213908.  Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

 

Panel (A) bottom shows a drawing illustrating the hormone-mediated production of IP3, which activates the IP₃ receptor (IP3R) to release Ca²⁺ from the endoplasmic reticulum (ER). The biphasic effects of cytosolic Ca²⁺ on IP₃ receptor gating (the basic mechanism for Ca²⁺ oscillations), whereby Ca²⁺ modulates positively the receptor at low [Ca²⁺] but negatively at high [Ca²⁺], are also illustrated. Top, Ca²⁺ oscillations. Note the decaying trend of Ca²⁺ spikes due to the absence of Ca²⁺ influx from the extracellular space.

Panel (B) shows  Ca²⁺ influx apparatus from extracellular spaces through ER-depletion by activated Orai channels on the plasma membrane (bottom), which generates sustained Ca²⁺ oscillations (top). 

Ca²⁺ levels fall through Ca²⁺ pumps on the sarcoplasmic/endoplasmic reticulum (SERCA), which moves the ion into the SR/ER or the plasma membrane (PMCA) that then moves it to the outside of the cell.  The SERCA and PMCA Ca²⁺ pumps and the ER STIM (Stromal Interaction Molecule that acts as a calcium sensor) proteins are not shown in Figure 14 above.  The STIM1 protein is involved in store-operated Ca²⁺ entry (SOCE).  This allows Ca²⁺ influx after intracellular stores are depleted.  The STIM1 protein has an EF-hand, so it binds Ca²⁺ and acts as a sensor.  When Ca²⁺ is depleted, STIM1 moves from the ER to the cell membrane and activates ORA1, a subunit of the Ca²⁺ release-activated Ca²⁺ (CRAC) channel, promoting Ca²⁺ influx into the cell.

In various cells, hormonal or neurotransmitter signals elicit a train of intracellular Ca²⁺ spikes. The analysis of a minimal model based on Ca²⁺-induced Ca²⁺ release from intracellular stores shows how sustained oscillations in cytosolic Ca²⁺ may develop due to a rise in inositol 1,4,5-trisphosphate (InsP₃) triggered by external stimulation. This rise elicits a certain amount of Ca2+ release from an InsP₃-sensitive intracellular store. The subsequent rise in cytosolic Ca²⁺ triggers the release of Ca²⁺ from a second store insensitive to InsP₃. The model shows how signal-induced Ca²⁺ oscillations might be effectively encoded by their frequency through phosphorylation of a cellular substrate by a protein kinase activated by cytosolic Ca²⁺.

The release of intracellular Ca²⁺ by IP3 can be broken down into three steps:

• agonist binding to GPCRs to activate the Phospholipase C pathway to produce the second messengers IP3 and DAG;
• IP3 induces some Ca²⁺ release from intracellular stores in the SR/ER.  This can be called the Primer Step, characterized by a rate V1 that produces cytosolic Ca²⁺ at concentration [Z], thereby priming the next step. This step has a rate of v1β.
• Released Ca²⁺ induces more Ca²⁺ release, a self-amplifying positive feedback step, which causes the Ca²⁺ spike. 

A simplified view of this model is shown in Figure \(\PageIndex{15}\) below.

Figure \(\PageIndex{15}\): Schematic representation of the one-pool model based on CICR with Ca2+-stimulated degradation of IP3. Lloyd, C.M., Lawson, J.L., Hunter, P.J., and Nielsen, P.F. The CellML Model Repository. Bioinformatics. 2008 September;24(18):2122-2123 (accessed 4/19/23; 5:40 am EDT).  Creative Commons Attribution 3.0 Unported License.

A simple model can be constructed to account for cytosolic Ca²⁺ fluxes.  The steps (more mathematically defined than described above) include

• IP3 causes a triggering a constant flux of Ca²⁺ into the cytosol, v1β ,to produce cytosolic [Ca²⁺] = Z (the primer)
• cytosolic Ca²⁺ flows into an IP3-insensitive sequestered pool (concentration Y) with rate v2 to keep low levels of cytosolic Ca²⁺  
• Spikes arise when the sequestered pool Y releases Ca²⁺ back into the cytosol at a rate v3, which cytosolic Ca²⁺ activates. 

Goldbeter et al. (Proc. Natl. Acad. Sci. 87, 1461-1465 (1990) constructed a mathematical model for the oscillations/spikes. In addition to the parameters mentioned above, they included three more:

• vo, describing the influx of extracellular Ca²⁺ into the cytosol, and k, describing the efflux of cytosolic Ca²⁺ from the cell.  These are controlled by the Ca²⁺ pumps (SERCA, PMCA, et al.) described above. 
• kf, which describes the passive leak of Y into Z.

The Ca²⁺ oscillations are then based on a self-amplified release of Ca²⁺ from intracellular stores. 

Here are their equations:

• \( dZ/dt=v_{0}+v_{1}beta - v_{2}+v_{3}+k_{f}Y-kZ \)
• \( dY/dt=v_{2}-v_{3}+v_{3}-k_{f}Y \)
• \(v_{2}=V_{m2}\frac{Z^{n}}{K^{n}_{2}+Z^{n}} \)
• \(v_{3}=V_{m3}\frac{Y^{m}}{K^{m}_{R}+Y^{m}}.\frac{Z^{p}}{K^{p}_{A}+Z^{p}} \)

We'll use Vcell to plot the following species: 

• Species Z: Concentration (μM) of cytosolic Ca²⁺

• Species Y: Concentration (μM) of Ca²⁺ stored in the InsP₃-insensitive pool. 

The model is based on code from EBI-Biomodels: https://www.ebi.ac.uk/biomodels/BIOMD0000000098.

---

MODEL

Calcium ion oscillation in the cell without oscillations in IP3

Here are the adjustable parameters:

• v0: influx of Ca²⁺ into the cell, uM/s.

• v1: influx of Ca²⁺ into the cell from the InsP₃ receptor, uM/s.

• beta: saturation function of the receptor for InsP₃ , unitless.

• Vm2: maximum rate of Ca²⁺ pumping into the intracellular InsP₃-insensitive store,  uM/s.

• Vm3: maximum rate of Ca²⁺ released from the intracellular store, uM/s;

• K2: Threshold constant for Ca²⁺ pumping, uM.

• n: Hill function cooperativity coefficient for Ca²⁺ pumping into the store, unitless.  

• m: Hill function cooperativity coefficient for Ca²⁺ pumping from the store, unitless.

• Ka: threshold constant for activation, uM.

• kf: efflux of cytosolic Ca²⁺ from the cell, 1/s.

• k:  influx of extracellular Ca²⁺ into the cytosol, 1/s.

• p denotes the degree of cooperativity of the activation process, unitless.

Select Load [model name] below

 Select Start to begin the simulation.

Interactive Element

Select Plot to change Y axis min/max, then Reset and Play  |  Select Slider to change which constants are displayed |  Select About  for software information.

Move the sliders to change the constants and see changes in the displayed graph in real time.

After loading the GoldBeter model, select the 'Start' button below to simulate the model. Adjust the parameter sliders below the plot to see how they affect Ca²⁺ concentrations (Z, Y). The simulator only displays twelve parameters at a time. To choose others, select the 'Slider' button on the side and chose up to twelve parameters to adjust.   

Time course model made using Virtual Cell (Vcell), The Center for Cell Analysis & Modeling, at UConn Health.  Funded by NIH/NIGMS (R24 GM137787); Web simulation software (miniSidewinder) from Bartholomew Jardine and Herbert M. Sauro, University of Washington.  Funded by NIH/NIGMS (RO1-GM123032-04)

---

 

Adjust the parameter sliders below the plot to see how they affect Ca²⁺ concentrations (Z, Y). The simulator displays only 12 parameters at a time. To choose others, press the 'Slider' button on the side and select up to 12 parameters to adjust.   

Only two variables, Y and Z, and some intra-connections are required to generate the oscillations, which do NOT require oscillation in the second messenger, IP3. One can imagine that cytosolic Ca²⁺ oscillations might also elicit oscillatory activity of protein kinases activated by it. 

Questions:

• Does v0 change the oscillation frequency?

• What other parameters affect the frequency?

• How does Vm2 affect the oscillations?

Summary

(Summary written by Claude, Sonnet 4.6, Anthropic)

Calcium ions are versatile and ubiquitous second messengers whose signaling power derives from the steep concentration gradient between the extracellular space (~2 mM) and the cytosol (~100 nM). Maintaining this low basal cytosolic level—while allowing rapid, transient, spatially localized Ca²⁺ rises to serve as signals—requires an elaborate buffering and storage infrastructure. Soluble EF-hand proteins such as calmodulin, calbindin, and calreticulin sequester free Ca²⁺ in the cytoplasm and organelle lumens. Integral membrane pumps (SERCA, PMCA) actively transport Ca²⁺ into the ER/sarcoplasmic reticulum or out of the cell, respectively, while the Na⁺/Ca²⁺ exchanger (NCX) and TRP channels provide additional flux pathways. When ER stores become depleted, the EF-hand-containing STIM1 protein senses the drop in luminal Ca²⁺, migrates to ER-plasma membrane junctions, and activates ORAI1/CRAC channels to allow sustained extracellular Ca²⁺ entry—a process called Store-Operated Ca²⁺ Entry (SOCE). Mitochondria participate via specialized contact sites (MAMs), where IP3R, VDAC, and the mitochondrial calcium uniporter (MCU) coordinate Ca²⁺ transfer, which influences metabolic activity, autophagy, and apoptotic signaling.

The key Ca²⁺ sensor protein calmodulin (CaM) undergoes a dramatic apo→holo conformational change upon binding four Ca²⁺ ions through its paired EF-hand motifs, exposing hydrophobic surfaces that enable binding to a broad array of target proteins, including pumps, channels, and the CaM-activated kinases (CAMKs). CaMKII, organized as a 12- or 14-subunit holoenzyme, is activated when Ca²⁺/CaM relieves intramolecular autoinhibition and allows trans-autophosphorylation of Thr286 between adjacent subunits. This phosphorylation locks the enzyme in an active state even after Ca²⁺ falls and CaM dissociates, making CaMKII a molecular memory device that integrates the frequency of Ca²⁺ oscillations—a property central to synaptic plasticity and learning and memory in neurons.

Ca²⁺ signaling does not operate in isolation but is deeply integrated with GPCR, phospholipase C, PKC, MAPK, and calcineurin/NFAT pathways. Cytosolic Ca²⁺ oscillations arise through Ca²⁺-induced Ca²⁺ release (CICR), in which IP3-triggered Ca²⁺ release from the ER stimulates further Ca²⁺ release in a self-amplifying positive feedback loop, while SERCA pumps restore basal levels between spikes. The Goldbeter mathematical model demonstrates that sustained oscillations can emerge from the interaction of just two variables—cytosolic Ca²⁺ (Z) and sequestered pool Ca²⁺ (Y)—without requiring oscillations in IP3 itself. Computational modeling with VCell reveals how parameters governing pump rates, threshold constants, and cooperativity coefficients determine oscillation frequency and amplitude, illustrating how quantitative approaches are indispensable for understanding the complex dynamics of Ca²⁺ signaling networks.