2. Pancreatic β-Cell Signal Transduction and Ca²⁺ Homeostasis

In the pancreatic β-cell, metabolism–secretion coupling describes the molecular mechanism linking nutrient sensing and signaling to insulin secretion. This process relates to the consensus model and additional coupling factors (including both triggering and amplifying pathways) of glucose-stimulated insulin secretion ^[18][19][22,23]. Glucose-stimulated insulin secretion is relatively well characterized and requires the sequential activation of several biological processes (Figure 1).

Figure 1. Consensus model of the signal transduction pathway of pancreatic β-cells. Metabolism–secretion coupling of β-cells requires the sequential activation of glycolysis, mitochondrial oxidative metabolism and Ca²⁺ entry through the plasma membrane. Glucose stimulates glycolysis and pyruvate production. Pyruvate triggers mitochondrial metabolism and the formation of the reduced form of nicotinamide adenine dinucleotide NADH (by the TCA cycle), which is the fuel for the respiratory complexes (1,2,3,4), enabling ATP production by ATP-synthase (5). ATP inhibits the K_ATP channel, inducing membrane depolarization (ΔΨ↓) and Ca²⁺ entry through voltage-gated Ca²⁺ channels, promoting insulin secretion. Ca²⁺ is taken up in parallel by mitochondria via the mitochondrial Ca²⁺ uniporter (MCU) and facilitates sustained insulin secretion. The amplifying pathway of metabolism–secretion coupling is co-generated by additive coupling factors. Additional players contributing to Ca²⁺ homeostasis are mentioned in the main text.

Glucose enters the β-cell by glucose-mediated transporters (GLUT). In the cytosol, it is metabolized by glycolysis to generate pyruvate, which is taken up by mitochondria. Mitochondrial pyruvate is metabolized by the tricarboxylic acid (TCA) cycle, which generates reducing equivalents NADH (reduced nicotinamide adenine dinucleotide) and FADH2 (reduced flavin adenine dinucleotide), which are substrates of the mitochondrial respiratory chain. Activation of mitochondrial respiration leads to mitochondrial ATP synthesis and thus to an increased cytosolic ATP-to-ADP ratio, which induces the closure of plasma membrane K_ATP channels and promotes plasma membrane depolarization. This opens voltage-gated plasma membrane Ca²⁺ channels, leading to an increase in cytosolic Ca²⁺ concentration, which finally triggers insulin exocytosis by activating Ca²⁺-sensitive granule-resident proteins (e.g., synaptotagmin-7 ^[20][24]). The amplifying pathways of metabolism–secretion coupling are contributed by additive coupling factors, and mitochondria have been characterized as a source of coupling factors ^[9][18][9,22].

The coupling between nutrient stimulation and hormone secretion is closely linked to Ca²⁺ homeostasis in the pancreatic β-cell ^{[7][21][22][23]}[7,25,26,27]. Therefore, insulin secretion is driven by electrical activity and oscillations of intracellular Ca²⁺ concentrations. However, in addition to K_ATP channels and voltage-dependent Ca²⁺ channels ^[24][28], other plasma membrane channels and intracellular stores have been shown to be involved in insulin secretion in pancreatic β-cells ^[22][25][26,29]. In particular, store-operated Ca²⁺ channels, which are voltage-independent Ca²⁺ channels activated upon depletion of the endoplasmic reticulum Ca²⁺ stores, and transient receptor potential channel 1 (TRPC1), have been indicated to be involved in insulin secretion ^[26][27][30,31]. For store-operated Ca²⁺ channels, Orai1 has been identified as the main protein that conducts the previously described Ca²⁺ release-activated current (I_CRAC). The activity of Orai1 channels is tightly controlled by the endoplasmic reticulum membrane protein stromal interacting molecule 1 (STIM1), which acts as an endoplasmic reticulum Ca²⁺ sensor and translocates, upon endoplasmic reticulum depletion, to endoplasmic reticulum/plasma membrane regions, where Orai1 is clustered ^[27][31]. In addition, mobilization of intracellular Ca²⁺ from the endoplasmic reticulum has been suggested to potentiate glucose-stimulated hormone secretion ^[28][32], and the Type 2 ryanodine receptor (RyR2) has been proposed to play a crucial role in regulating insulin secretion and glucose homeostasis ^[29][33]. Moreover, an atypical Ca²⁺ leak has been observed in the endoplasmic reticulum, specifically in pancreatic islets and β-cells. This continuous Ca²⁺ efflux from the endoplasmic reticulum was modulated by GSK3β-dependent phosphorylation of presenilin-1 and promoted mitochondrial activation ^[30][31][34,35]. Additional intracellular acidic compartments may contribute to the local modulation of β-cell Ca²⁺ homeostasis and thus β-cell function, including insulin granules and other acidic stores (e.g., lysosomes) ^[22][26].

The last player in pancreatic β-cell Ca²⁺ homeostasis and signal transduction is represented by the mitochondrial network. Therefore, mitochondria are intracellular organelles that take up and release Ca²⁺, promoting the sensing and shaping of cytosolic Ca²⁺ signals ^[32][36]. The role of mitochondrial Ca²⁺ signaling in energized mitochondria of the pancreatic β-cell is emerging as a biological process of critical importance to pancreatic β-cell function and is highlighted in the next section.

3. The Mitochondrial Calcium Uniporter and Its Existence in Pancreatic β-Cells

Despite the undisputed role of cytosolic Ca²⁺ elevation in triggering insulin secretion in the pancreatic β-cell, it is also accepted that such a rise in itself does not sustain insulin secretion ^[33][34][35][18,37,38]. Therefore, mitochondria have been demonstrated to contribute to robust insulin secretion by triggering additional regulatory factors, and it has been proposed that mitochondrial Ca²⁺ plays a crucial role as a receiver and generator of the signals essential for metabolism–secretion coupling ^[9][36][9,39]. The discovery and molecular definition of the MCU ^[14][37][14,40], the transporter which mediates the transport of Ca²⁺ in the mitochondrial matrix under physiological conditions, is shedding light on the role of mitochondrial Ca²⁺ elevation in different tissues, including the pancreatic β-cell (see Section 4).

The MCU complex is a low-affinity, high-capacity Ca²⁺ uniporter embedded in the inner mitochondrial membrane with an approximate in vitro Ca²⁺ binding affinity estimated to be between 10 and 70 µM ^{[38][39][40][41]}[41,42,43,44] in different tissues. The transport of Ca²⁺ ions into the mitochondrial matrix is electrochemically driven by a strong electrical gradient (~180 mV) through the ion-impermeable inner mitochondrial membrane ^[42][45]. As reported for many other tissues, in the pancreatic β-cell, resting-state mitochondrial Ca²⁺ levels are also very close to cytosolic Ca²⁺ levels, indicating tight regulation of the MCU ^[33][18]. Permeabilized INS-1 cells (a cell-line model of pancreatic insulin-secreting cells) display significant mitochondrial Ca²⁺ uptake when perfused with buffers containing a Ca²⁺ concentration of only 150 nM ^[43][46], which is below the general threshold of MCU activation ^[44][47]. This study suggests a slightly higher affinity configuration of the β-cell MCU, compared with the MCU of other tissues. However, in permeabilized β-cells perfused with a range of different Ca²⁺ buffers, mitochondrial Ca²⁺ uptake was more efficient above 2 µM ^[45][48]. These data indicate that although some studies demonstrated a lower Ca²⁺ activation threshold, the MCU of β-cells also behaves as a low affinity, high capacity Ca²⁺ transport system, with similar properties to those reported in other tissues. In any case, the mitochondrial Ca²⁺ concentration in glucose-stimulated β-cells was reported to reach only 600–800 nM ^[13][33][46][13,18,49], mirroring cytosolic Ca²⁺ events. Purinergic activation and potassium-induced depolarization generate cytosolic Ca²⁺ transients around one micromolar, with the corresponding mitochondrial Ca²⁺ rises reaching nearly 5 µM ^[36][39]. These results indicate that, on average, β-cell mitochondria barely reach micromolar Ca²⁺ concentrations in the matrix during physiological activation.

The MCU complex consists of at least six subunits, each of which plays an individual role in orchestrating mitochondrial Ca²⁺ uptake ^[38][41]. The MCU subunit is a 40 kDa protein that, together with its paralog, MCUb ^[47][50], has been proposed to form a tetramerizing pore which penetrates the inner mitochondrial membrane ^[48][51]. In order to channel Ca²⁺ ions into the mitochondrial matrix, it requires a third subunit known as the essential MCU regulator (EMRE), a 10 kDa protein ^[47][49][50,52]. EMRE is a single-transmembrane protein whose transmembrane helix connects it to the MCU subunit ^[50][53]. The MCU and MCUb share 50% sequence similarity and have opposite effects on mitochondrial Ca²⁺ uptake ^[47][50]. While the MCU has a promoting effect on mitochondrial Ca²⁺ uptake, MCUb exerts a negative effect on mitochondrial Ca²⁺ uptake ^[51][54]. The ratio of MCU and MCUb varies among tissues and is potentially influenced by metabolic impairments, leading to the hypothesis that an altered MCU/MCUb ratio may affect β-cell function ^[52][53][54][55,56,57]. When Ca²⁺ in the intermembrane space exceeds ~0.6 µM, it initiates activation of the MCU complex via the MCU gatekeeper protein paralogs mitochondrial calcium uptake 1 and 2 (MICU1 and MICU2) ^[41][55][44,58]. MICU2 is a 50 kDa protein and shares approximately 25% sequence identity with the 54 kDa protein MICU1 ^[39][42]. Neither MICU1 nor MICU2 contain transmembrane domains that would link them to the MCU subunit, so it has been proposed that MICU1 is linked to the MCU via EMRE through electrostatic interaction ^[55][58]. However, a more recent study revealed an additional direct binding of MICU1 to the highly conserved DIME motif of the MCU ^[56][57][59,60]. In addition, another study showed an interaction between MICU1/MICU2 and cardiolipin ^[55][58]. Through cysteine residues, MICU2 forms an intermembrane space-facing heterodimer with MICU1 via disulfide bonds ^[48][51]. Both gatekeeper proteins contain EF-hand motifs (consisting of a helix E, a loop and another helix F), with a Ca²⁺ binding affinity of ~0.3 µM for MICU1 and a Ca²⁺ binding affinity of ~0.6 µM for the MICU1–MICU2 dimer ^[55][58].

Genetic ablation of MICU2 lowers the required Ca²⁺ concentration in the intermembrane space for MICU1-regulated MCU activation, resulting in increased mitochondrial Ca²⁺ uptake ^[49][52]. On the other hand, ablation of MICU1 prevents indirect binding of MICU2 to the MCU, resulting in unregulated Ca²⁺ entry into the mitochondria ^[58][59][61,62]. Studies with MICU1 knockout as well as with MICU1 knockdown have described an inverse relationship between extra-mitochondrial Ca²⁺ concentrations and mitochondrial Ca²⁺ uptake ^[59][60][62,63]. This suggests that the absence of MICU1 at low/resting extra-mitochondrial Ca²⁺ concentrations leads to increased mitochondrial Ca²⁺ uptake, while high extra-mitochondrial Ca²⁺ concentrations are associated with decreased mitochondrial Ca²⁺ uptake. The final subunit, which has been proposed to contribute to the structure of the MCU complex is a scaffolding factor called mitochondrial calcium uniporter regulator 1 (MCUR1), which can interact with the MCU but not with MICU1 ^[61][64]. However, there are two possible interpretations of the direct or indirect role of MCUR1 in modulating mitochondrial Ca²⁺ uptake. Several studies have demonstrated a regulatory effect of MCUR1 on MCU complex activity, noting reduced mitochondrial Ca²⁺ uptake in the absence of MCUR1 and a decrease in reducing equivalents as well as ATP generation ^[62][63][64][65,66,67]. Alternatively, MCUR1 has been suggested to have a potential role as a Complex IV assembly factor in the mitochondrial respiratory chain ^[65][68]. Interestingly, the relative abundance of distinct MCU components has been reported to be different in distinct cell types or tissues ^[31][35], and it has been proposed that the specific stoichiometry of the MCU subunits may define the functional characteristics of the channel, including Ca²⁺ permeation across the pore, the activation threshold and the cooperativity. Consistent with this hypothesis, it has been reported that in parallel with the distinct relative abundance of MCU components ^[31][35], MCU activity also varies greatly among cell types ^[66][69]. Although the stoichiometry of the different MCU subunits and their relative abundance in different tissues, including in the pancreatic β-cell, are largely unknown, we tentatively speculate that these cells may have a specific configuration of the MCU that reduces the Ca²⁺ activation threshold.

Pancreatic β-cells have been demonstrated to express a functional MCU, and some studies have investigated the role of MCU subunits on insulin secretion ^[15][16][17][15,16,17] (Figure 2).

Figure 2. The mitochondrial calcium uniporter (MCU) comprises distinct pore-forming (MCU/MCUb) and associated proteins (essential MCU regulator (EMRE), mitochondrial calcium uptake (MICU) 1, MICU2, MICU3 and mitochondrial calcium uniporter regulator 1 (MCUR1)). The subunits that have already been studied in models of the pancreatic β-cell are rendered in green. (A) Model of the closed conformation of MCU. (B) Following glucose stimulation, the entry of Ca²⁺ into the matrix space is mediated by MCU. IMM, inner mitochondrial membrane; IMS, intermembrane space.

Not surprisingly, the existence of MCU and MICU1 subunits in pancreatic β-cell models has been reported and functionally validated (see ^{[15][16][17][52]}Section 4) [15,16,17,55]. In addition, a study recently published as a preprint in bioRxiv but not yet peer-reviewed proposes the existence of MICU2 subunits in rat and human insulin-secreting cell lines as well as in mouse islets (Mitochondrial Clearance of Ca²⁺ Controls Insulin Secretion; https://doi.org/10.1101/830323 (accessed on 1 February 2021); Vishnu, Hamilton, Bagge, Wernersson, Cowan, Barnard, Sancak, Kamer, Spégel, Fex, Tengholm, Mootha, Nicholls and Mulder). Further studies are expected to consolidate the specific role of the other subunits of the MCU in the pancreatic β-cell (Figure 2) and fully define the structure of this transporter. However, the crucial role of the MCU complex in the mitochondrial Ca²⁺ transport of the pancreatic β-cell and in the pancreatic β-cell metabolism–secretion coupling is already well established and will be discussed in the next section.

(References would be added automatically after the entry is online)