Regulation of Ca2+ release through inositol 1,4,5-trisphosphate receptors by adenine nucleotides in parotid acinar cells

Hyung Seo Park, Matthew J. Betzenhauser, Yu Zhang, David I. Yule


Secretagogue-stimulated intracellular Ca2+ signals are fundamentally important for initiating the secretion of the fluid and ion component of saliva from parotid acinar cells. The Ca2+ signals have characteristic spatial and temporal characteristics, which are defined by the specific properties of Ca2+ release mediated by inositol 1,4,5-trisphosphate receptors (InsP3R). In this study we have investigated the role of adenine nucleotides in modulating Ca2+ release in mouse parotid acinar cells. In permeabilized cells, the Ca2+ release rate induced by submaximal [InsP3] was increased by 5 mM ATP. Enhanced Ca2+ release was not observed at saturating [InsP3]. The EC50 for the augmented Ca2+ release was ∼8 μM ATP. The effect was mimicked by nonhydrolysable ATP analogs. ADP and AMP also potentiated Ca2+ release but were less potent than ATP. In acini isolated from InsP3R-2-null transgenic animals, the rate of Ca2+ release was decreased under all conditions but now enhanced by ATP at all [InsP3]. In addition the EC50 for ATP potentiation increased to ∼500 μM. These characteristics are consistent with the properties of the InsP3R-2 dominating the overall features of InsP3R-induced Ca2+ release despite the expression of all isoforms. Finally, Ca2+ signals were measured in intact parotid lobules by multiphoton microscopy. Consistent with the release data, carbachol-stimulated Ca2+ signals were reduced in lobules exposed to experimental hypoxia compared with control lobules only at submaximal concentrations. Adenine nucleotide modulation of InsP3R in parotid acinar cells likely contributes to the properties of Ca2+ signals in physiological and pathological conditions.

  • ATP
  • multiphoton microscopy
  • intracellular Ca2+

agonist-induced elevations in the cytosolic calcium concentration ([Ca2+]i) are intimately involved in initiating saliva secretion, the primary physiological function of parotid gland acinar cells (36). Activation of Ca2+-dependent effectors is important for both the fluid and protein component of saliva. The primary aqueous component of primary saliva is generated as an increase in Ca2+ acts initially on a Cl-dependent conductance expressed exclusively on the apical plasma membrane (35, 41). Recently, a promising candidate for the molecular identity of this conductance is a member of the Anoctamin-1 (ANO-1/TMEM16A) family of Cl channels (41). Additionally, the hyperpolarized membrane potential necessary for continued Cl secretion is maintained by the action of Ca2+ on K+ channels of the IK and BK families (4244). The increase in luminal Cl and negative potential pulls Na+ through the paracellular pathway. The subsequent increase in osmotic potential draws water into the lumen to create an isoosmotic NaCl-rich solution. Exocytosis of secretory granules containing the protein components of saliva can also be triggered by increases in cytosolic Ca2+, presumably in analogy to neurosecretion, by acting through synaptotagmin homologs expressed in the acinar cells (13, 46).

The most important physiological activator of Ca2+ signaling in salivary glands is acetylcholine released from parasympathetic neurons following olfactory and gustatory stimulation (2, 3, 36). Activation of muscarinic receptors on the acinar cells results in the Gαq-coupled activation of phospholipase C and the production of inositol 1,4,5-trisphosphate (InsP3). InsP3 then acts on inositol 1,4,5 trisphosphate receptors (InsP3R) expressed on the endoplasmic reticulum (ER). InsP3R are particularly abundant on ER immediately juxtaposed to the apical plasma membrane (28), and rapid Ca2+ release is invariably initiated at the apical Ca2+ release sites followed by the subsequent globalization of the Ca2+ signal (23, 28, 47). The temporal and spatial information in this signal are ideally suited both for initially activating the apical Cl conductance and exocytotic machinery followed by subsequent opening of basolaterally localized K+ channels. All three isoforms of the InsP3R are expressed in mouse parotid acinar cells, but only the type-2 and type-3 isoforms are essential for function (22, 23, 28). This fact is demonstrated by a study that revealed that a transgenic animal null for both the type-2 and type-3 InsP3R is refractory to muscarinic receptor stimulation because of failure to mount a Ca2+ signal on stimulation (22). Indeed, the central role of these particular InsP3R isoforms in exocrine secretion is exemplified by the observation that the animal dies soon after weaning primarily through a failure to produce saliva, resulting in the inability to consume food (22).

Given the importance of the specific temporal and spatial characteristics of the signal, it is probable that modulation of Ca2+ release through InsP3 impacts Ca2+ signaling and thus gland function in physiological and pathological situations. Indeed, InsP3Rs serve as signal integrators, receiving multiple inputs that ultimately impact the activity of the channel (38, 39, 50). For example, InsP3Rs are regulated by numerous factors including most importantly Ca2+ itself but also by cellular ATP levels, phosphorylation by various kinases, and through a multitude of protein-protein interactions (20, 38, 39). As a function of distinct primary sequence of individual InsP3R isoforms, regulation can occur in a subtype-specific manner, and thus the overall effect of a particular route of regulation of Ca2+ release will reflect the net effect on the complement of receptor subtypes expressed in an individual cell. For example, in parotid acinar cells, Ca2+ release through InsP3R is markedly potentiated following activation of protein kinase A (10, 11). This is likely a consequence of phosphorylation of both InsP3R-1 and InsP3-2, as phosphorylation of these isoforms has been shown to augment channel activity (5, 45). Although phosphoregulation by PKA is a prime example of the InsP3R serving as a “node” to sample crosstalk between two ubiquitous signaling pathways, modulation of InsP3R by ATP and other cellular nucleotides potentially provides coupling between the metabolic status of the cell and Ca2+ release in physiological and potentially in pathological situations (17, 19, 30, 50). This form of regulation was recognized soon after the initial characterization of the InsP3R, and subsequently it has been demonstrated that all isoforms are subject to this form of modulation but with distinct characteristics including nucleotide potency and the molecular sites of action (6, 7, 50). In salivary glands, variations in cellular nucleotide levels could “fine tune” the degree of Ca2+ release during secretion when the glands are metabolically active. In addition, this form of regulation may be particularly important under hypoxic conditions where ATP levels can decrease precipitously (1, 4). In the present study, we have therefore characterized the effects of nucleotides on Ca2+ release through InsP3R in parotid acinar cells. We report that Ca2+ release is potently modulated by adenine nucleotides including ATP and ADP with characteristics typical of InsP3R-2 playing a dominant role over the other isoforms present. In addition, under hypoxic conditions, Ca2+ release was markedly reduced. These data indicate that, in salivary gland acinar cells, nucleotides are important modulators of Ca2+ release under physiological conditions and furthermore that a decrease in ATP levels may impact Ca2+ signaling under pathological situations.



Fluo-4 AM was purchased from Teflabs. All other materials were obtained from Sigma Chemical. The InsP3R-2-null animals (29) were a kind gift from Dr. Ju Chen (University of California, San Diego, CA).

Preparation of parotid acinar cells.

Single cells or small clusters of parotid acinar cells were isolated from freshly dissected parotid glands from NIH Swiss mice (25 g) by sequential digestion with (single cells) or without (cell clusters) trypsin followed by collagenase as previously described (23, 47, 48). Following isolation, cells were resuspended in 1% BSA containing BME supplemented with 2 mM glutamine, penicillin/streptomycin and incubated at 37°C and gassed with 5% CO2-95% O2 until ready for use. All procedures were approved by the University of Rochester, University Committee on Animal Resources.

Permeabilized cell Ca2+ measurements.

Cells were loaded with 10 μM fura-AM (Teflabs) for 1 h at room temperature. Cells were then allowed to adhere to Cell-Tak-coated coverslips (BD Biosciences) at the bottom of a small-volume perfusion chamber (Warner). Cells were permeabilized by superfusion at a rate of ∼4 ml/min for 1–2 min with 40 μM of the saponinifying agent, β-escin in intracellular medium (ICM) containing 125 mM KCl, 19 mM NaCl, 10 mM HEPES, 1 mM EGTA (pH 7.3). Permeabilized cells were washed in ICM without β-escin for 15 min to facilitate removal of cytosolic dye. Cells were superfused in ICM containing 0.650 μM CaCl2 (free [Ca2+] of 200 nM), 1.4 mM MgCl2, and 3 mM Na2ATP to activate sarco-ER Ca2+-ATPase (SERCA) and load the intracellular Ca2+ stores. The free [Ca2+] was maintained at a constant 200 nM throughout all experimental maneuvers. Free [Ca2+] was verified by fluorimetric fura-2 measurements using buffer solutions with or without 5 mM ATP. The emission of the dye above 505 nm following excitation at 340 and 380 nm (exposure for 15 ms, once every 3 s) was recorded using a TILL Photonics imaging system as previously described (8). Superfusion of 40 mM β-escin resulted in reduction of ∼80% of the initial Furaptra fluorescence as cytoplasmic dye was lost to the superfusate. The remaining fluorescence represents dye trapped within intracellular organelles including the ER. Intracellular organelles were subsequently loaded with Ca2+ by activation of SERCA pump activity. Typically, a new steady-state level of fluorescence was established after 2–3 min of superfusion. The catalytic transport cycle of SERCA involves a phosphor-enzyme intermediate, and the phosphorylation and dephosphorylation of the enzyme during Ca2+ transport are markedly enhanced by Mg2+ (16, 25, 51). Therefore, 1 min before addition of InsP3 and throughout application of InsP3 in the test period, SERCA pump activity was effectively disabled by removal of Mg2+ from the superfusate. Over this time period, intracellular Ca2+ remained stable, and this maneuver allowed measurement of the essentially unidirectional flux of Ca2+ from the stores after InsP3 addition. Rates of Ca2+ release were generally estimated after InsP3 addition, by fitting the initial 30% decrease from the original fluorescence ratio to a single exponential function (26) (GraphPad Prism). In experiments where, because of very low release rates, these conditions were not fulfilled, the entire declining phase was fit. R2 values were >0.95.

Lobule preparation and multiphoton imaging.

Parotid salivary gland lobules were isolated following injection of saline beneath the capsule with a 27-gauge needle. Using a binocular dissecting scope, individual lobules <1 mm3 were removed with scissors essentially as described for pancreatic lobules (49). The lobules were loaded with fluo-4 AM for intracellular Ca2+ measurements (5 μM) or Calcein-AM for visualization of structure (5 μM) by incubation for 30–60 min. Lobules were placed in a low-volume perfusion chamber (Warner Instruments), maintained in position by using a wire grid in a weighted metal ring, and perfused in HEPES-buffered physiological saline (HBS) containing 137 mM NaCl, 0.56 mM MgCl2, 4.7 mM KCl, 1 mM Na2HPO4, 10 mM HEPES, 5.5 mM glucose, and 1.26 mM CaCl2 at pH 7.4. Multiphoton imaging was performed using an Olympus BX61WI upright microscope coupled to an Olympus Fluoview 1000 multiphoton/confocal (FV1000MP) microscope equipped with Spectra Physics Mai Tai (DeepSee) mode-locked Ti/sapphire tunable laser. Imaging was performed using a ×25 water-immersion objective (XL Plan N; na = 1.05). The laser was tuned to provide excitation at 810 nm, and emitted light was separated using a 565-nm dichroic mirror followed by emission filters HG525/50 and HQ605/50 in front of the detectors to visualize green and red emitted fluorescence, respectively. An individual plane, typically at a depth of 80–150 μm into the slice, was scanned at 1 Hz for measurement of intracellular Ca2+. A Z series through the preparation followed by analysis of a reconstructed maximum projection image was used to analyze lobule structure.


Initial experiments were performed to characterize the effects of ATP on InsP3-induced Ca2+ release in parotid acinar cells. First, Ca2+ release was monitored over a range of [InsP3] in either the absence or the presence of 5 mM [ATP]. In the absence of ATP, Ca2+ release could be readily detected following exposure to 0.3 μM InsP3. The rate of Ca2+ release was, however, markedly enhanced by inclusion of 5 mM Na+-ATP in the superfusate (Fig. 1A, individual trace, and 1B, representing pooled data). At higher [InsP3] even in the absence of ATP the rate of Ca2+ release was increased, reaching a maximum at 10 μM InsP3 (Fig. 2G). The inclusion of 5 mM ATP further increased the rate of Ca2+ release at all submaximal [InsP3] (Fig. 1, C and D, pooled data) but was without effect at maximal [InsP3] (Fig. 1, E and F, pooled data). A summary of the effects of 5 mM ATP on the initial rate of InsP3-induced Ca2+ release is shown in Fig. 1G and illustrates that 5 mM ATP resulted in an approximately twofold shift in the EC50 for InsP3-induced Ca2+ release (1.0 ± 0.16 μM, vs. 1.8 ± 0.12 μM InsP3, absence and presence of ATP, respectively).

Fig. 1.

Effect of ATP on Ca2+ release rate in parotid acini. A, C, and E: representative experiments where Ca2+ release is monitored in the presence and absence of 5 mM ATP. Effect of ATP on a low inositol 1,4,5-trisphosphate concentration ([InsP3]) (A), on a ∼EC50 [InsP3] (C), and on a saturating [InsP3] (E). B, D, and F: corresponding pooled data for low, EC50, and saturating [InsP3], respectively. Experiments with each [InsP3] were repeated at least 6 times and represent the data from >25 cells for each point. For comparison between experimental runs, the data were normalized to the initial ratio for each experiment. G: Ca2+ release rate constants calculated from fits of the data shown in B, D, and F.

Fig. 2.

ATP dependence of InsP3-induced Ca2+ release. A: pooled normalized data for the effect of various [ATP] on Ca2+ release evoked by 0.3 μM InsP3. N = 9 experimental runs >40 cells for each point. B: effect of ATP on Ca2+ release rate constants calculated from these data. The Kd for the effect of ATP is ∼8 μM.

Next, the ATP dependence of the enhancement of InsP3-induced Ca2+ release was determined at a sub-maximal [InsP3]. Figure 2A shows pooled data for the indicated [ATP] following addition of 0.3 μM InsP3. Ca2+ release was markedly augmented at 3 μM ATP, but no further increase in release rate was observed with [ATP] above 30 μM. Figure 2B shows the mean Ca2+ release rate derived from fitting the initial phase of release and yields an EC50 of 8.3 ± 0.5 μM. ATP had almost identical effects on Ca2+ release initiated by higher yet still submaximal [InsP3] (EC50 9.8 ± 0.6 μM; 0.6 μM InsP3, see Fig. 3, B and C). Regulation by ATP of InsP3R-induced Ca2+ release in parotid acinar cells does not require ATP hydrolysis because similar modulation of release could be readily supported by 5′-adenylyl (β,γ-methylene) diphosphonate (AMP-PCP), a nonhydrolysable form of ATP (EC50 8.0 ± 1.2 μM AMP-PCP at 0.6 μM InsP3) (Fig. 3A, averaged traces, and 4C, pooled release rates). These data are in agreement with previous studies performed in tissues predominately expressing InsP3R-1 and suggest that nucleotide binding does not require catalytic activity of the InsP3R and that binding likely regulates the receptor in an allosteric manner (9, 19).

Fig. 3.

Nonhydrolysable ATP analogs and other adenine nucleotides enhance InsP3-induced Ca2+ release. A: pooled normalized data for the effect of various 5′-adenylyl (β,γ-methylene) diphosphonate concentrations ([AMP-PCP]) on Ca2+ release evoked by 0.6 μM InsP3. N = 5 experimental runs >30 cells for each point. B: pooled normalized data for the effect of various nucleotides on Ca2+ release evoked by 0.6 μM InsP3. GTP (10 mM) did not support enhanced Ca2+ release, whereas Ca2+ release was augmented by ADP (5 mM) and AMP (10 mM) in addition to ATP (5 mM). C: comparison of the concentration dependence of the effects of the various nucleotides to enhance the InsP3-induced Ca2+ release rate constants. N = 5–9 experimental runs >30 cells for each point.

The original report documenting photoaffinity labeling of ATP to the cerebellar InsP3R showed that other nucleotides could compete for binding and thus could potentially modulate Ca2+ release (18, 30). We therefore next investigated whether binding of these nucleotides could also modulate InsP3-induced Ca2+ release in parotid acinar cells. As shown in Fig. 3, B and C (pooled data), although less potent than ATP, ADP was as efficacious and increased the Ca2+ release rate stimulated by a submaximal [InsP3] to levels similar to that achieved in the presence of ATP (EC50 119 ± 12.7 μM). In contrast, AMP only marginally increased Ca2+ release at concentrations as high as 10 mM (EC50 2.54 ± 0.4 mM), and GTP was without effect (Fig. 3B). These data suggest that in a cellular context modulation of Ca2+ release may reflect competition by various nucleotides at binding site(s) on the InsP3R.

The characteristics of adenine nucleotide modulation of Ca2+ release in parotid acinar cells are reminiscent of the properties described for the InsP3R-2 when expressed in isolation even given the expression of all InsP3R family members in this gland (6). Specifically, ATP appears not to be an obligate ligand required for maximal Ca2+ release at saturating [InsP3], and the modulation of release occurs with relatively high affinity. We next tested the hypothesis that the properties of InsP3R-2 dominate over other isoforms expressed by studying Ca2+ release from parotid acinar cells prepared from InsP3R-2-null (knockout, KO) animals (29). Ca2+ release in InsP3R-2 KO acini was markedly less sensitive to InsP3, and absolute Ca2+ release rates at all [InsP3] were dramatically lower compared with wild-type (WT) animals (pooled data in Fig. 4G). These data are consistent with InsP3R-2 possessing the highest functional affinity for InsP3 and highlight the dominant role of InsP3R-2 in establishing the distinct characteristics of Ca2+ release in parotid acinar cells compared with other exocrine cells. Similar to WT acini, Ca2+ release in KO acini was potentiated in the presence of 5 mM ATP (Fig. 4, A–F). However, in contrast to WT acini, Ca2+ release was augmented significantly by ATP at all [InsP3]; indeed even at a saturating [InsP3] (30 μM), maximal release rates could only be achieved in the presence of ATP (Fig. 4, E–G). The relative ATP sensitivity of Ca2+ release was next tested in KO animals at a submaximal [InsP3]. Notably, the ability of ATP to augment Ca2+ release was markedly reduced in KO acini. Specifically the EC50 for the effect of ATP was reduced >50 fold from ∼8 μM to 560 ± 56 μM (Fig. 5) (WT vs. KO, respectively). In total, data obtained in the KO animal are consistent with a dominant role of InsP3R-2 in establishing both the relative sensitivity to InsP3 and the characteristics of ATP regulation of Ca2+ release in parotid acinar cells.

Fig. 4.

Effect of ATP on Ca2+ release rate in InsP3R-2-null parotid acini. A, C, and E: representative experiments where Ca2+ release is monitored in the presence and absence of 5 mM ATP in acini isolated from InsP3R-2-null animals. Effect of ATP on a low [InsP3] (A), on a ∼EC50 [InsP3] (C), and on a saturating [InsP3] (E). B, D, and F: corresponding pooled data for low, EC50, and saturating [InsP3], respectively. N = 6 experimental runs >35 cells for each point. G: Ca2+ release rate constants in the presence and absence of 5 mM ATP. Dashed lines shown for comparison data obtained from wild-type (WT) animals presented in Fig. 2. Note that InsP3-induced Ca2+ release appears markedly less sensitive to InsP3 although in contrast to WT animals Ca2+ release is augmented even at saturating [InsP3].

Fig. 5.

ATP dependence of InsP3-induced Ca2+ release in InsP3R-2-null parotid acini. A: pooled normalized data for the effect of various [ATP] on Ca2+ release evoked by 10 μM InsP3. B: effect of ATP on Ca2+ release rate constants calculated from these data. N = 7 experimental runs >40 cells for each point. The EC50 for the effect of ATP is ∼560 μM compared with 8 μM in WT animals. Dashed lines show data obtained from WT animals presented in Fig. 3 for comparison.

In a final series of experiments, we investigated the consequences of ATP depletion on Ca2+ signaling in acinar cells in situ in parotid gland lobules using multiphoton microscopy. This preparation has the distinct advantage that no enzymatic digestion is used to prepare the lobules and the architecture of the gland is largely undisturbed. Figure 6A shows a transmitted laser light image of a lobule loaded with fluo-4 and image using MP excitation (Fig. 6B). The images clearly illustrate that the preparation retains the characteristic polarized nature of the gland. Under control conditions, fluo-4 AM loaded lobules were constantly perfused with HBS buffer gassed with 100% O2 and then stimulated with various concentrations of the muscarinic agonist carbachol (CCh). The initial peak height of the response was monitored as a measure of the initial InsP3-induced Ca2+ release (Fig. 6C). These signals were compared with lobules, which were incubated for 30 min in HBS without added glucose and gassed with 100 N2; conditions previously shown to mimic hypoxia and result in a significant reduction in cellular [ATP] in other exocrine cell types (4). Stimulation of lobules incubated in N2 sparged buffer resulted in significantly reduced Ca2+ release at submaximal [CCh] (Fig. 6, C and D). These data are consistent with effects of ATP on Ca2+ release occurring at submaximal [InsP3] through altering the functional sensitivity of the dominant InsP3R-2.

Fig. 6.

Ca2+ signals in lobules of parotid gland are reduced under conditions where ATP is depleted. A: transmitted laser light image of a parotid lobule. B: image of the same fluo-4-loaded lobule imaged using MP excitation at a depth of ∼100 μm from the surface of the tissue. C: Ca2+ signals following exposure of lobules to low (a), medium (b), and high concentrations of carbachol (CCh) (c), either in lobules maintained in O2 gassed HEPES buffered solution (black traces) or in N2 sparged solutions without added glucose. D: pooled data. CCh-evoked Ca2+ signals following exposure to submaximal CCh concentrations were significantly reduced. N = 4 lobules for each condition. *P > 0.05.


Agonist-stimulated Ca2+ release is fundamentally important for the primary physiological functions of parotid gland acinar cells. It is likely that the particular spatial and temporal characteristics of the Ca2+ signal are the key features that lead to the appropriate activation of effectors responsible for secretion (23, 27, 28, 47, 48). In turn, the major step defining the initiation of these processes is the integrated regulation by multiple factors of Ca2+ release through InsP3R (10, 23, 28, 47). In this study, we have investigated how cellular levels of adenine nucleotides might contribute to this regulation. A principle finding of this study is that only Ca2+ release stimulated by submaximal [InsP3] is subject to regulation by ATP. This finding is also consistent with only Ca2+ release elicited by lower concentrations of CCh being reduced by exposure of the lobules to conditions that lead to lowered cellular ATP levels. ATP regulation of submaximal release, together with the high ATP affinity, is typical and practically diagnostic of ATP regulation of InsP3R-2 when studied in isolation or in cells that predominately express this InsP3R isoform (6, 37). Thus it appears that, despite expression of all three isoforms of InsP3R in parotid acini and approximately equal levels of InsP3R-2 and InsP3R-3, the properties of the InsP3R-2 are dominant. This conclusion is strengthened by the observation that, in the absence of InsP3R-2 in transgenic animals, the characteristics of regulation are dominated by properties of the remaining InsP3R-3 including marked decrease in affinity and ATP now functioning as an obligate coagonist, essential for maximal release. These properties are qualitatively reminiscent of data obtained studying Ca2+ release in pancreatic acinar cells (37). However, in WT parotid acini the EC50 for ATP modulation of InsP3-induced release was fourfold lower than in pancreatic acini (8 vs. 30 μM) but very similar in both tissues in InsP3R-2-null animals. These data may reflect that parotid acini are reported to express a higher density of InsP3 binding sites than pancreas (23) and would suggest that it is the complement of InsP3R-2 in particular that is relatively greater in parotid acinar cells.

Cellular levels of ATP are generally thought to be in the range 0.5–2.0 mM, a vast majority in the form of Mg-ATP. Given the high affinity for the effect of ATP on InsP3-induced Ca2+ release in parotid acinar cells, it is possible that this form of regulation represents tonic, allosteric modulation of the receptor necessary for the proper signaling conformation of the receptor. In this scenario, it is unlikely that ATP levels would fluctuate enough to decrease to levels whereby the InsP3 would respond to normally occurring variations in ATP. Although little consensus exists in the literature, reports have suggested that InsP3-induced Ca2+ release is, however, not modulated by MgATP but instead by free ATP (32, 33). In this case, free levels of ATP in the tens of micromolar might be expected to exert a dynamic physiologically relevant regulation of Ca2+ release by acting as a sensor of the metabolic status of the cells, allowing fine tuning of the degree of Ca2+ release. This form of regulation may be particularly important where InsP3Rs are localized in intimate association with mitochondria or conversely at sites of high ATP consumption as a consequence of abundant SERCA or plasma membrane Ca-ATPase (PMCA) distribution (14, 15). Notably, the latter scenario occurs in salivary acinar cells as the distribution of InsP3R overlaps with both SERCA and PMCA in the extreme apical domain of the acinus (27).

Our studies demonstrate that InsP3-induced Ca2+ release in parotid acini is also modulated by other adenine nucleotides including ADP and AMP in addition to ATP. These data are similar to studies in cerebellum and some cultured cell lines (19, 31). Although ADP is less potent than ATP, it is fully efficacious and indeed was shown to support a higher maximal Ca2+ release rate than ATP itself. Because the levels of free ADP are very difficult to measure and are seldom reported, it is, however, difficult to predict whether regulation by ADP is physiologically important. Nevertheless as the ratio of ATP/ADP changes during oxidative phosphorylation (21) and also during periods of high metabolic demand, it is likely that the degree of regulation will be determined by the ratio of nucleotides locally present at release sites at a particular time.

Nucleotide regulation of InsP3-induced Ca2+ release might also be impacted under pathological situations in salivary glands in which ATP/ADP levels are compromised. For example, ATP levels have been shown to fall precipitously when tissue is subjected to hypoxic conditions (1, 4). Indeed, parotid lobules exposed to experimental hypoxia showed decreased peak Ca2+ signals when challenged with submaximal muscarinic receptor stimulation. Depletion of ATP would be predicted to compromise Ca2+ handling by both SERCA and PMCA and likely increase oxidative stress in acini, nevertheless, the decrease in peak height in these experiments is consistent with compromised Ca2+ release at the level of the InsP3R. Hypoxic conditions may exist in solid tumors of salivary glands as a result of abnormal vascularization of the malignant tissue and the low ATP level exacerbated by reduced production through the reliance of cancer cells on glycolysis rather than mitochondrial respiration (40). Hypoxia and a decrease in ATP levels may also occur in salivary glands following therapy for head and neck cancers as evidenced by the marked increase in gland autophagy following irradiation (12, 24). Our data would predict that InsP3R-mediated Ca2+ release, in particular in response to submaximal stimulation, would be reduced as a function of low ATP levels in disease states. Because Ca2+ uptake by mitochondria plays an important role in maintaining homeostatic ATP production (34), reduced Ca2+ release may function to further reduce ATP levels favoring increased autophagy. Ultimately, decreased Ca2+ release would be predicted to directly impact the physiological activation of ion channels and may contribute to the marked reduction in fluid secretion observed under pathological conditions.


The work was supported by NIH grant DE014756 to D. I. Yule.


No conflicts of interest, financial or otherwise, are declared by the authors.


The authors thank Lyndee Knowlton for excellent technical support throughout the study.

Present affiliations: H. Park, Dept. of Physiology, College of Medicine, Konyang University, Daejeon, South Korea; M. Betzenhauser, Dept. of Physiology & Cellular Biophysics, Columbia University Medical School, New York, NY.


  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.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
View Abstract