Interactions of endoplasmic reticulum and mitochondria Ca²⁺ stores with capacitative calcium entry

Abstract

Thiamine dependent enzymes are diminished in Alzheimer’s disease (AD). Thiamine deficiency in vitro and in rodents is a useful model of this reduction. Thiamine interacts with cellular calcium stores. To directly test the relevance of the thiamine dependent changes to dynamic processes in AD, the interactions must be studied in cells from patients with AD. These studies employed fibroblasts. Mitochondrial dysfunction including reductions in thiamine dependent enzymes and abnormalities in calcium homeostasis and oxidative processes occur in fibroblasts from Alzheimer’s Disease (AD) patients. Bombesin-releasable calcium stores (BRCS) from the endoplasmic reticulum (ER) are exaggerated in fibroblasts from patients with AD bearing a presenilin-1 (PS-1) mutation and in control fibroblasts treated with oxidants. ER calcium regulates calcium entry into the cell through capacitative calcium entry (CCE), which is reduced in fibroblasts and neurons from mice bearing PS-1 mutations. Under physiological conditions, mitochondria and ER play important and interactive roles in the regulation of Ca²⁺ homeostasis. Thus, the interactions of mitochondria and oxidants with CCE were tested. Inhibition of ER Ca²⁺-ATPase by cyclopiazonic acid (CPA) stimulates CCE. CPA-induced CCE was diminished by inhibition of mitochondrial Ca²⁺ export (−60%) or import (−40%). Different aspects of mitochondrial Ca²⁺ coupled to CPA-induced-CCE were sensitive to select oxidants. The effects were very different when CCE was examined in the presence of InsP₃, a physiological regulator of ER calcium release, and subsequent CCE. CCE under these conditions was only mildly reduced (20–25%) by inhibition of mitochondrial Ca²⁺ export, and inhibition of mitochondrial Ca²⁺ uptake exaggerated CCE (+53%). However, t-BHP reversed both abnormalities. The results suggest that in the presence of InsP₃, mitochondria buffer the local Ca²⁺ released from ER following rapid activation of InsP₃R and serve as a negative feedback to the CCE. The results suggest that mitochondrial Ca²⁺ modifies the depletion and refilling mechanism of ER Ca²⁺ stores.

INTRODUCTION

Thiamine dependent enzymes are diminished in Alzheimer’s disease (AD). Rodent thiamine deficiency (TD) has been used to model the mild impairment of metabolism that occurs in AD [Karuppagounder et al.,2009]. TD exaggerates plaque and tangle formation in mouse models [Karuppagounder et al.,2009] and elevating thiamine levels diminish plaques, tangles and memory deficits [Pan et al., 2010]. An understanding of the consequences of the reduction of thiamine dependent enzymes is important for understanding the pathophysiology of AD and for developing new therapies. Reduction of the thiamine dependent enzyme alpha-ketoglutarate dehydrogenase (KGDHC) either with an inhibitor or by genetic manipulation reveal that another consequence of diminished activity of a thiamine dependent enzyme is an alteration in the calcium stores in the endoplasmic reticulum. Thus, neurons taken from mice deficient in KGDHC have exaggerated stores of ER calcium whether the neurons are cultured from embryos or adults, just as in fibroblasts from patients with AD [Gibson et al., 2012].

Whether this change occurs and is important in AD is more difficult to answer. Since the calcium change is dynamic one cannot measure this property in autopsy brain. A commonly used model to study disease processes is cultured fibroblasts. Indeed, fibroblasts were used by Dr. Butterworth in pioneering studies in the 1980s in which he looked at thiamine dependent enzymes in Leigh’s disease in fibroblasts. Surprisingly, the same abnormalities in calcium homeostasis that we observed by reducing a thiamine dependent enzyme in mouse brains occurs in fibroblasts from AD patients. BRCS from the endoplasmic reticulum (ER) are exaggerated in fibroblasts from patients with AD bearing a presenilin-1 (PS-1) mutation [Ito et al., 1994] and in control fibroblasts treated with specific oxidants [Huang et al., 2005]. The two oxidants employed in these studies were: (1) tert-Butyl-hydroperoxide (t-BHP) which produces the radicals tert-butyloxyl (t-bu-OS) and t-butylperoxyl (t-bu-OOS) and (2) 3-morpholinosydnonimine (SIN-1), which is commonly used to produce various forms of nitrogen monoxides that react with O₂.⁻ to form peroxynitrite. A more detailed discussion is provided in [Huang et al., 2005]. The goal of the current study is to understand the consequences of these changes on cellular calcium regulation. Considerable research has been accomplished in understanding the increase in calcium in fibroblasts bearing presenilin-1 mutations leading to AD [Nelson et al., 2010]. However, these mechanisms only apply for patients bearing presenilin mutations. Thus, these interactions need be better understood in non-genetic forms of AD. The best cells to accomplish this are cells with a human genetic background (i.e., fibroblasts).

Calcium dynamics and the response of cells to oxidants are modified by thiamine [Huang, Chen and Gibson, 2010]. Specific oxidants can induce the same changes in calcium dynamics that occur in fibroblasts from patients with AD (an exaggeration of BRCS) [Huang, Chen and Gibson, 2005]. Further, the responses of cells to oxidants are modified by thiamine. Thiamine diminishes ER calcium stores, but does not alter CCE [Huang, Chen and Gibson, 2010].

The current studies test how these changes in ER calcium interact to regulate cellular calcium and the effects of oxidants on these responses. The mitochondria and ER are linked by proteins (mitochondrial associated proteins) that have been linked to proteins that underlie some genetic forms of AD [Area-Gomez et al., 2009]. In addition, to the interaction between the ER and mitochondria there is a link between these processes and the ability to fill cellular calcium pools. The depletion of ER Ca²⁺ content triggers Ca²⁺ influx through plasma membrane Ca²⁺ channels, a process known as capacitative calcium entry (CCE) [Putney and Ribeiro, 2000]. CCE is reduced in fibroblasts and neurons from mice bearing PS-1 mutations [Yao et al., 2000; Leissring et al., 2000]. Thus, genetic mutations that cause AD lead to increase intracellular calcium stores and an attenuation in the refilling mechanism CCE. Presenilin-1 mutations increase levels of ryanodine receptors and calcium release in PC12 cells and cortical neurons [Chan et al., 2000]. Studies suggest that ER Ca²⁺ disruption in AD is related to an enhancement of ryanodine receptor regenerative process [Stutzmann et al., 2006]. The ryanodine receptor is one of the Ca²⁺ -release channels located on the ER which have been implicated in Ca²⁺ -induced Ca²⁺ release activity (CICR). In addition, oxidation of RyR results in activation of CICR [Hidalgo et al., 2000].

Mitochondrial dysfunction may underlie altered calcium homeostasis in AD. Mitochondria may have a fundamental role in regulating CCE through bidirectional Ca²⁺-dependent crosstalk between mitochondria and CCE [Glitsch et al., 2002]. Mitochondria are often located near Ca²⁺ release channels on the ER [Simpson et al., 1997] or Ca²⁺ influx channels in the plasma membrane and modulate store-operated or CCE [Hoth et al., 1997]. Thus, mitochondria contribute to cellular Ca²⁺ homeostasis in many cell types by taking up and releasing Ca² which propagates and synchronizes Ca²⁺ signals [Pozzan et al., 1994]. Whether mitochondria act as a Ca²⁺ sink or as a Ca²⁺ relay mechanism, or whether the increase in mitochondrial Ca²⁺ an essential step to activate CCE is unclear [Glitsch et al., 2002].

The Ca²⁺ buffering capacities of the ER and mitochondria can regulate CCE [Malli et al., 2003]. InsP₃-mediated pathways regulate CCE and Ca²⁺ uptake and release from mitochondria [Kojima and Ogata, 1989]. Inhibition of the mitochondrial Na⁺/Ca²⁺ exchanger with CGP37157 results in the increase of [Ca²⁺]_mito and reduces the ability of mitochondria to buffer subplasmalemmal Ca²⁺ and leads to an increase of BK-sensitive Ca²⁺ channels activity and a decrease in CCE [Malli et al, 2003]. In addition, CGP37157 prevents complete Ca²⁺ refilling of the ER during cell stimulation with an InsP₃-generate agonist [Malli et al., 2005].

Mitochondrial Ca²⁺ uptake is operated by a high-capacity, low-affinity Ca²⁺ uniporter, which can be inhibited specifically by Ru360. At sites where mitochondria are in close contact with the ER [Rizzuto et al., 1998], micro-domains of high cytosolic Ca²⁺ are generated by Ca²⁺ release from ER through InsP₃ receptors [Rizzuto et al., 1999]. These high local cytosolic Ca²⁺ concentrations lead to increases in mitochondrial Ca²⁺ accumulation that correlate with increases in cytosolic Ca²⁺ [Rizzuto et al., 1993] [Rizzuto et al., 1998]. Ru360 has no effect on Ca²⁺ release from ER [Seguchi et al., 2005]. Partial inhibition of Ca²⁺ uptake by Ru360 increases the [Ca²⁺]_i transient in ventricular myocytes, and propagates the Ca²⁺ release. This effect of Ru360 did not appear to be due to altered ER Ca²⁺ content but may contribute to local control of Ca²⁺ release [Huang et al., 2004].

The present studies tested the interactions of mitochondria, ER Ca²⁺ and CCE under control conditions as well as with oxidants or InsP₃.

MATERIALS AND METHODS

The supplies were from the indicated companies: Cell culture reagents (GIBCO; Grand Island, NY); 3-morpholinosydnonimine (SIN-1) (Molecular Probes; Eugene, Oregon); bombesin, tert-butyl-hydroxyperoxide (t-BHP) (Sigma Chemical, St Louis, MO); CGP17157 and Ru360 (Calbiochem, San Diego, CA).

A human skin fibroblast cell line from a young male control (8399) was purchased from Coriell Cell Repository (Camden, NJ). Cells were maintained exactly as described in our published protocol [Stutzman et al., 2006].

Bombesin releasable calcium stores (BRCS)

Fluorescent intracellular calcium images were monitored as described previously [Huang et al., 2005; Huang et al., 2004]. Fibroblasts were loaded with 2 μM Fura-2 AM in BSS for one hr at room temperature, were rinsed twice with Ca²⁺-free BSS and [Ca²⁺]_i was monitored on the stage of an inverted Olympus IX70 microscope at room temperature with a Delta Scan System from PTI (Photon Technology International, Lawrenceville, NJ). Excitation wavelengths were alternated between 350 and 378 nm (band pass 4 nm) and emission was monitored at 510 nm with a Hamamatsu C2400 SIT camera at 5 sec intervals. Each value was the average of 32 images taken within 5 sec. Standard images of Fura-2 solutions with minimum and maximum [Ca²⁺]_i were taken at the end of each day’s experiment to calculate the intracellular calcium concentrations.

Capacitative Calcium Entry (CCE) measurement

After cells were preincubated in Ca²⁺-free media with or without bombesin and the ER Ca²⁺ -ATPase inhibitor cyclopiazonic acid (CPA), CaCl₂ (2.5 mM) was added. In Ca²⁺-free media, bombesin will release ER Ca²⁺ from InsP₃ sensitive stores, and CPA without bombesin will release InsP₃ insensitive Ca²⁺ stores. The resulting increase in calcium is a measurement of CCE.

Statistical analysis

All data are expressed as mean ± SEM. A Student’s t-test was used to compare two variables. For multiple variable comparisons, data were analyzed by a one-way analysis of variance (ANOVA) followed by a Student Newman-Keul’s test.

RESULTS

Characterization of CPA activated CCE

Fibroblasts were incubated in the absence of extracellular Ca²⁺, the ER Ca²⁺ was released with CPA, and then Ca²⁺ was added to initiate Ca²⁺ re-entry as a measure of CCE. CPA depletes ER Ca²⁺ stores by reversibly inhibiting ER Ca²⁺-ATPase that replenishes ER Ca²⁺. CPA gradually increased cytosolic Ca²⁺ in a dose-dependent manner by inhibiting ER Ca²⁺ uptake and allowing the Ca²⁺ leak from ER (Fig 1). Although 2 μM CPA did not elevate the cytosolic Ca²⁺ significantly, it activated subsequent CCE markedly (Fig 1). Thus, 2 μM CPA was used for the subsequent experiments most to minimize the complications form increases in cytosolic Ca²⁺.

Figure 1. Characterization of CPA activated CCE.

Typical temporal profiles of the CCE are shown in subsequent figures. Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. CPA (0, 2, 10, 50 μM) was added 4 min after basal [Ca²⁺]_i measurement, and after an additional 4 min CaCl₂ (2.5 mM) was added. The top panel shows experimental paradigms. The bottom panel shows the integrations of the [Ca²⁺]_i peak over the 3 min interval after CPA (255–480 sec) and after calcium addition (495–720 sec). Data are means ± SEM (n=51–62 cells). Asterisks indicate values vary significantly (p<0.05) from the control group by ANOVA followed by Student Newman Keul’s test.

Blockade of mitochondrial Ca²⁺ exporter by CGP37157 diminished CCE

Fibroblasts were incubated with in the absence of extracellular Ca²⁺ prior to treatment with CGP37157 (0, 20, 40 μM) an inhibitor of the Na⁺/Ca²⁺ exchanger the enzyme responsible for mitochondrial calcium export. CPA was then added and subsequently Ca²⁺ was added to a final concentration of 2.5 mM (CCE). Blockade of the mitochondrial Ca²⁺ exporter by CGP37157 did not elevate cytosolic Ca²⁺ significantly, but reduced subsequent CCE following CPA treatment in a dose-dependent manner to 80% (20 μM) or 37% (40 μM) of control (Fig 2).

Figure 2. Inhibition of mitochondrial exporter by CGP37157 diminished CCE induced by CPA.

Blockade of mitochondrial Ca²⁺ exporter by CGP37157 diminished CCE. Fibroblasts were loaded with Fura 2AM for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. CGP37157 (20, 40 μM), an inhibitor of the mitochondrial Na⁺/Ca²⁺ exchanger blocker was added after 1 min of basal [Ca²⁺]i measurements, and CPA (2 μM) was added after 5 min. After an additional 2 min, CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 45–133 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 3 min interval after calcium addition. Data are means ± SEM (n=45–133 cells). Asterisks indicate values vary significantly (p<0.05) from the control group by ANOVA followed by Student Newman Keul’s test.

Inhibition of mitochondrial Ca²⁺ uptake by Ru360 impaired CCE

Whether mitochondrial Ca²⁺ uptake regulates ER Ca²⁺ and subsequent CCE were tested by using an inhibitor of the mitochondrial Ca²⁺ uptake Ru360. Fibroblasts were incubated in the absence of extracellular Ca²⁺ prior to treatment with the Ca²⁺ uptake inhibitor Ru360 (0, 20 μM). CPA was added and subsequently Ca²⁺ was added (final concentration of 2.5 mM). Ru360 did not alter basal calcium nor CPA-induced Ca²⁺ release, whereas it reduced the subsequent CCE (−39%) (Fig 3).

Figure 3. Inhibition of mitochondrial Ca²⁺ uptake by Ru360 diminished CCE induced by CPA.

Inhibition of mitochondrial Ca²⁺ uptake by RU360 impaired CCE. Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. The mitochondrial Ca²⁺ uptake blocker Ru360 (0, 20 μM) was added 2 min after basal [Ca²⁺]_i measurements, and 4 min later CPA (2 μM) was added. Two min later CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 50–51 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 3 min interval after calcium addition (495–720 sec). Data are means ± SEM (n=50–51 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

t-BHP did not alter the CCE induced by CPA

Fibroblasts were incubated in the absence of extracellular Ca²⁺. The ER Ca²⁺ was released with CPA (2 μM) prior to addition of t-BHP (100 μM), and then various concentrations of Ca²⁺ (0.625, 1.25 and 2.5 mM) were added to initiate CCE. The magnitude of CPA-induced elevation in CCE depended on the extracellular Ca²⁺ concentrations. However, CCE was not modified by t-BHP at any concentration of calcium (Fig 4).

Figure 4. t-BHP did not alter CCE induced by CPA.

Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. CPA (2 μM) was added after 1 min of basal [Ca²⁺]_i measurement, and after an additional 3 min, t-BHP (100 μM) was added. Three min later different concentrations of CaCl₂ (final concentrations of 0.625, 1.25 or 2.5 mM) were added. The top panel shows the tracings taken from 44–87 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 3 min interval after calcium addition (380 - 660 sec). Data are means ± SEM (n= 44–87 cells).

t-BHP but not SIN-1 attenuated the impairment of CCE by CGP37157

Previous studies demonstrated that tBHP increases BRCS just as occurs in AD [Huang, Chen and Gibson, 2005]. The current studies extended that to CCE. The selective effects of oxidants were tested with a higher concentration of CPA. As shown in Figure 1, higher concentrations of CPA caused a larger increase in CCE which provides more opportunity for experimental manipulation. Cells were incubated in the absence of extracellular Ca²⁺. CGP37157 was added 1 min prior to addition of CPA (10 μM). Oxidants (t-BHP or SIN-1) were then added, and finally Ca²⁺ was added to activate CCE. The mitochondrial Ca²⁺ exporter inhibitor CGP37157 enhanced CPA-induced release Ca²⁺ from ER (+129 % of CPA alone) and t-BHP further enhanced this Ca²⁺ release from ER by 209 % of CPA treatment but SIN-1 did not alter the reduction by CGP37157 treatment (Fig 5). However, activation of CCE after CPA treatment was reduced by blockade of mitochondria Ca²⁺ exporter by CGP37157 (−60%) (Figs 2, 5). This inhibitory effect of CGP was reversed by t-BHP (−22%) but not by SIN-1 (Fig 5).

Figure 5. t-BHP but not SIN-1 attenuated the impairment of CCE by CGP37157.

Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. The CPA (10 μM) was added 1 min after CGP37157 (40 μM), t-BHP (100 μM) or SIN-1 (500 μM) was added 3 min later. After an additional 2 min CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 21–29 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 5 min interval after CPA (10 μM) or 3 min after calcium addition (495–720 sec). Data are means ± SEM (n=21–29 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

Blockade of mitochondrial Ca²⁺ uptake reduced CPA-induced CCE

Fibroblasts were incubated in the absence of extracellular Ca²⁺. SIN-1 was added prior to addition of the mitochondrial Ca²⁺ uptake inhibitor Ru360, then CPA was added, and finally Ca²⁺ was added to activate CCE. The mitochondrial Ca²⁺ uptake inhibitor RU360 reduced activation of CCE by CPA treatment by 38%, whereas SIN-1 abolished this impairment and the CCE was same as the control group (Figs 3, 6).

Figure 6. Inhibition of CPA-induced mitochondrial Ca²⁺ uptake by RU360 was reversed by SIN-1.

SIN-1 abolished the inhibition of CPA-induced CCE by the mitochondrial Ca²⁺ uptake inhibitor RU360. Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. RU360 (0, 20 μM) was added 2 min after SIN-1 (500 μM). After an additional 4 min, CPA (2 μM) was added, and 2 min later CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 42–51 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 3 min interval after calcium addition (495–660 sec). Data are means ± SEM (n= 42–51 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

In the presence of InsP₃, blockade of the mitochondrial Ca²⁺ exporter had no effect on BRCS but impaired CCE

Under physiological conditions, CCE is generally induced following depletion of ER calcium stores by agonists that induce InsP₃ which binds to the InsP₃ receptor to release Ca²⁺ from the ER. Fibroblasts were incubated in the absence of extracellular Ca²⁺. CGP37157, inhibitor of the mitochondrial Ca²⁺ release (Na⁺/Ca²⁺ exchanger) was added prior to addition of bombesin, which depletes ER Ca²⁺. CPA and then Ca²⁺ were added to activate CCE. CGP37157 (20 or 40 μM) did not alter basal Ca²⁺ or Ca²⁺ release by bombesin, but reduced subsequent CCE (20% and 25%) (Fig 7). Thus, the impairment of CCE by CGP37157 was much less in the presence of InsP₃ (−20%; Fig 7) than that in the absence of InsP₃ agonist (−63%; Fig 2). Thus, the impairment of CCE by a blocker of calcium export from the mitochondria was attenuated by InsP₃.

Figure 7. In the presence of InsP₃, blockade of mitochondrial Ca²⁺ release had no effect on BRCS but impaired CCE.

Fibroblasts were loaded with Fura 2AM for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. The CGP37157 (0, 20, 40 μM) was added after 1 min of basal [Ca²⁺]i measurements, and bombesin (2 μM) was added after 3 min. After an additional 3 min, CPA (4 μM) was added, and 3 min later CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 49–66 cells. The bottom panel shows the integrations of the [Ca²⁺]_i peak over the 2–3 min interval after bombesin (255–480 sec) or calcium addition (495–720 sec). Data are means ± SEM (n = 49–66 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

In the presence of InsP₃, blockade of mitochondrial Ca²⁺ uptake exaggerated CCE

Fibroblasts were treated with the mitochondrial Ca²⁺ uptake inhibitor Ru360 prior to bombesin treatment, and subsequent re-addition Ca²⁺. Ru360 (10 or 20 μM) did not alter basal calcium (data not shown) but increased Ca²⁺ release by bombesin (+81%) and exaggerated the subsequent CCE significantly (+53%) (Fig 8). This is in striking contrast with the results in the absence of InsP₃ where Ru360 diminished CCE (Fig 3).

Figure 8. In the presence of InsP3, blockade of mitochondrial Ca²⁺ uptake, exaggerated BRCS and subsequent CCE.

In the presence of InsP₃, blockade of mitochondrial Ca²⁺ uptake by Ru360 exaggerated BRCS and subsequent CCE. Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. Mitochondria Ca²⁺ uptake blocker Ru360 (0, 10, 20 μM) was added after 1 min basal [Ca²⁺]_i measurements, and bombesin (1 μM) was added after 3 min. After additional 3 min, CPA (2 μM) was added, and 3 min later CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 74–91 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 2–3 min interval after bombesin or calcium addition. Data are means ± SEM (n=74–91 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

In the presence of InsP₃, impairment of CCE by blockade of mitochondrial Ca²⁺ release was attenuated by t-BHP

Fibroblasts were incubated in the absence of extracellular Ca²⁺ and were then treated with t-BHP one min prior to addition of inhibitor of mitochondria Ca²⁺ release. Bombesin was added to deplete ER Ca²⁺. CPA was added and finally Ca²⁺ was added to activate CCE. The experiments show the effect of t-BHP on the results in Fig 7. CGP37157 did not alter BRCS but diminished subsequent CCE 29% (Fig 9). Under the conditions, t-BHP enhanced BRCS (+139%) which replicates our previous studies [Huang et al., 2005]. t-BHP also exaggerated the CCE (+34%) (Fig 9) and this increase was also blocked by CGP37157 (Fig 9).

Figure 9. In the presence of InsP₃, impairment of CCE by mitochondrial Ca²⁺ exporter inhibitor was attenuated by t-BHP.

Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. CGP37157 (20 μM) was added 1 min after t-BHP (100 μM) treatment, and bombesin (1 μM) was added after 3 min. After an additional 2 min, CPA (2 μM) was added, and 2 min later CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 69–116 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 3 min interval after bombesin or calcium addition. Data are means ± SEM (n=69–116 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

In the presence of InsP₃, t-BHP attenuated the exaggeration of BRCS and CCE due to inhibition of mitochondrial Ca²⁺ uptake

Fibroblasts were incubated in the absence of extracellular Ca²⁺, and were treated with t-BHP one min prior to addition of inhibition of mitochondrial Ca²⁺ uptake. Bombesin was then added to deplete ER Ca²⁺. CPA was added and finally Ca²⁺ was added to activate CCE. The experiment shows the effects of oxidant on the results in Fig 8. Blockade of mitochondrial Ca²⁺ uptake exaggerated BRCS and CCE by 82% and 53% of control, respectively (Fig 8, 10). In addition, t-BHP exaggerated BRCS and CCE by 139 % and 34% of control, respectively (Fig 9, 10). Interestingly, in the presence of InsP₃, the oxidant t-BHP abolished the exaggeration of BRCS and CCE due to the inhibition of mitochondrial Ca²⁺ uptake (Fig 10).

Figure 10. In the presence of InsP₃, t-BHP diminished exaggeration of BRCS and CCE due to the inhibition of mitochondrial Ca²⁺ uptake.

Fibroblasts were loaded with Fura 2AM (2 μM) for 60 min. After loading, the media were changed to calcium free BSS, and 3 min later the calcium measurements were initiated. The mitochondrial Ca²⁺ uptake blocker Ru360 (20 μM) was added 2 min after t-BHP (100 μM), and bombesin (1 μM) was added after 2 min. After additional 2 min, CPA (2 μM) was added, and 2 min later CaCl₂ (final concentration of 2.5 mM) was added. The top panel shows the tracings taken from 69–106 cells. The bottom panel shows the integration of the [Ca²⁺]_i peak over the 3 min interval after bombesin or calcium addition. Data are means ± SEM (n = 69–106 cells). Different letters indicate values vary significantly (p<0.05) from the other groups by ANOVA followed by Student Newman Keul’s test.

DISCUSSION

The present studies demonstrated that CPA induced Ca²⁺ release from ER, activated CCE, and that this process was regulated by mitochondrial Ca²⁺ release and uptake. Under normal condition, ER Ca²⁺ ATPase activity balances the passive leak of ER Ca²⁺ into the cytosol. Inhibition of this Ca²⁺ pump by CPA causes the release of ER Ca²⁺ and increases cytosolic Ca²⁺. The concentration dependent increase in cytosolic calcium after CPA indicates that greater inhibition of the Ca²⁺ ATPase leads to greater release of ER calcium. The release of ER calcium activates CCE even in the absence of increases in cytosolic free calcium (i.e. at low CPA concentrations) (Fig 1). Under this condition, impairment of the mitochondrial Ca²⁺ export system (Fig 2) or uptake system (Fig 3) attenuated CCE. The inhibitory effect of CGP37157 on CCE suggests that mitochondrial Ca²⁺ release has a direct impact on the activation of CCE (Fig 2, Fig 11). The reduction of CCE by Ru360 (−40%) without an alteration in cytosolic calcium suggests that mitochondrial calcium uptake is directly linked to CCE (Fig 3). Thus, mitochondria play an important role in the regulation of the CCE following depletion of the ER Ca²⁺. The results suggest that Ca²⁺ release from mitochondria contributes to the activation of CCE either directly or indirectly.

Fig 11. In the absence of InsP₃ agonist.

In the absence of InsP₃ producing agonist, CCE can be regulated by passive leak of ER Ca²⁺ (1), mitochondria Ca²⁺ release -refilling to ER (2), Ca²⁺ uptake by mitochondria (3) Ca²⁺ influx directly across plasma membrane (4).

CCE can be regulated at several possible sites (Figure 11. The –number after the figure number refers to the number on the figure): including passive leak of ER Ca²⁺ (Fig 11-1), mitochondrial Ca²⁺ release-refilling to ER (Fig 11-2), Ca²⁺ uptake by mitochondria (Fig 11-3), Ca²⁺ influx across plasma membrane (Fig 11-4). When the ER Ca-ATPase is inhibited, a reduction of mitochondrial Ca²⁺ release markedly enhanced cytosolic Ca²⁺ elevation and reduced subsequent CCE (− 56%). Thus, the results suggest that mitochondrial Ca²⁺ release serve as ER refilling mechanism (Fig 11-2). In other words, the mitochondria Ca²⁺ directly fill the ER stores, which is the signal for the CCE. Ru360 impaired the CCE (−40%) without altering cytosolic Ca²⁺. Thus suggests that mitochondria uptake Ca²⁺ influx directly affects CCE. Perhaps the mitochondria regulate localized calcium concentrations (Fig 11-3). However, in the absence of CPA, the addition of Ca²⁺ increased cytosolic free calcium also activated CCE (19.8 % of that with CPA), suggesting that Ca²⁺ influx is independent of either the mitochondria or the ER (Fig 11-4). Therefore, the results suggest that CCE can be regulated by passive leak of ER Ca²⁺ (Fig 11-1), mitochondrial Ca²⁺ release to refill ER calcium (Fig 11-2), Ca²⁺ uptake by mitochondria (Fig 11-3), Ca²⁺ influx across the plasma membrane (Fig 11-4).

Selective oxidants affect different sites of regulation of CCE. The present studies showed that t-BHP had no effect on CCE (Fig 4). These results suggest that the t-BHP must promote calcium entry either by facilitating calcium export from the mitochondria or by facilitating calcium entry into the ER from other sources. This effect is oxidant selective because SIN-1 is ineffective in this regard. These results are in agreement with our previous studies [Huang et al., 2005] which showed that t-BHP but not SIN-1, increased ER calcium stores. On the other hand, SIN-1 did overcome the inhibition of CCE that was caused by inhibition of mitochondrial calcium uptake (Fig 6). Taken together the results suggest that CCE regulated by mitochondrial Ca²⁺ release-refilling to ER (Fig 11-2) is sensitive to t-BHP, while CCE regulated by mitochondrial Ca²⁺ uptake can be modified by the action of SIN-1 (Fig 11-3). CCE regulated by ER Ca²⁺ passive leak pathway is not sensitive to t-BHP (Fig 4, Fig 11-1). The results suggest that Ca²⁺ release from mitochondria contributes to the activation of CCE either directly or indirectly and can be modified by select oxidants.

These interactions are modified in the presence of the InsP₃ producing agonist bombesin (Figure 12). Impairment of mitochondrial Ca²⁺ release by CGP37157 impaired CCE (Fig 7), suggesting that mitochondrial Ca²⁺ play a role to regulate CCE. CGP37157 prevents Ca²⁺ refilling of the ER during cell stimulation with an InsP₃-producing agonist [Malli et al., 2005]. The impairment of CCE by CGP37157 was much less in the presence of InsP₃-producing agonist (−25 %) than in its absence (−56 %). In the absence of InsP₃ producing agonist, ER Ca²⁺ was not up taken by mitochondria. However, mitochondria can uptake and release Ca²⁺ during cell stimulation with InsP₃-produdcing agonists, thereby generating subplasmalemmal microdomains of low Ca²⁺ that sustain activity of CCE [Malli et al., 2005]. Thus, mitochondrial Ca²⁺ uptake from the local transient Ca²⁺ release from ER via activation of InsP₃ receptor (Fig 12-1) is a major source for refilling ER by mitochondria Na⁺/Ca²⁺ exchanger (Fig 12-2). Other studies that directly measured mitochondrial Ca² show that in the absence of InsP₃ agonist, CGP37157 has no effect on mitochondrial Ca²⁺, while in the presence of InsP₃ agonist, CGP37157 leads to a long-elevation of mitochondrial Ca²⁺ [Malli et al., 2005].

Fig 12. In the presence of InsP₃ agonist.

In the presence of InsP₃ producing agonist, activation of InsP₃R evoked cytosolic Ca²⁺ release from ER that are transmitted to the mitochondria (1). The mitochondrial Ca²⁺ refills ER (2). If inhibition of the mitochondrial Ca²⁺ uptake by Ru360, the transient Ca²⁺ will facilitate rate of CICR and RyR(3) to empty more of the ER Ca²⁺ stores and activated more CCE (4).

The results show a cross talk between mitochondrial Ca²⁺, ER calcium and CCE. Impairment of mitochondrial Ca²⁺ uptake exaggerated Ca²⁺ release from ER and enhanced CCE activity following depletion of the ER Ca²⁺ by stimulation with InsP₃ producing agonist (Figs 8, 10,11, 12). The results with Ru360 in the presence (Fig 11) and absence (Fig 12) of InsP₃ producing agonist are opposite. Cytosolic Ca²⁺ signals elicited by activation of InsP₃R or RyR are transmitted to the mitochondria [Rizzuto et al., 1998, Rizzuto et al, 1993; Hajnoczky et al., 1995] (Fig 12-1). Thus, calcium moves from the ER to the mitochondria without mixing with the cytosol. Furthermore, mitochondrial Ca²⁺ transport modulates the spatio-temporal pattern of cytosolic Ca²⁺ responses evoked by InsP₃. As shown in Fig 12, mitochondrial Ca²⁺ uptake limits the rate of propagation of calcium induced calcium release (CICR) which leads to activation of RyR. Therefore, disabling mitochondrial Ca²⁺ uptake will increase the rate of CICR which requires Ca²⁺ released locally by InsP₃ acting in the ER Ca²⁺ stores (Fig 12-3). The greatest emptying of ER Ca²⁺ occurred with the combination of InsP₃R and CICR (Figs 8,10 as indicating by higher BRCS), and this produced higher CCE (Figs 8, 10; Fig 12-4). Thus, mitochondrial Ca²⁺ uptake is tightly coupled to Ca²⁺ release following activation of InsP3R-or ryanodine receptor. InsP₃ binding protein suppresses the Ca²⁺ release mediated by the InsP₃Rs that provides Ca²⁺ for mitochondrial Ca²⁺ uptake [Lin et al., 2005].

In the presence of InsP₃, t-BHP reversed the abnormalities caused by the disruptions of uptake or release of mitochondrial Ca²⁺ (Fig 9, 10, Figure 12), suggesting t-BHP-induced ROS signaling regulated this mitochondrial and ER Ca²⁺ signals. Oxidant t-BHP exaggerated Ca²⁺ release from ER and elevated CCE replicated our previous studies [Huang et al., 2005]. t-BHP independently exaggerated the elevation of Ca²⁺ release from ER and impaired CCE by blocking mitochondrial Ca²⁺ release. The t-BHP-induced exaggeration in CCE may be due to an oxidant-induced exaggeration in the ability of bombesin to deplete InsP₃ sensitive calcium stores (i.e. increased BRCS) and subsequent CICR. The ryanodine receptor channel is a thiols protein, which is S-nitrosylated and results to a progressive channel activation [Xu et al., 1998]. However, oxidation of additional thiols (or of another class of thiols) produces irreversible activation can lead to loss of control [Xu et al., 1998]. The disruption that ER Ca²⁺ regulation in AD has been suggested to be related to an enhancement of ryanodine receptor [Stutzmann et al., 2006]. In the presence of InsP₃, the mitochondria buffer the elevation of Ca²⁺ by the InsP₃ receptor, and refill the ER Ca²⁺. If the Ca²⁺ released by InsP₃ is not up taken by mitochondria, Ca²⁺ will facilitate the rate of CICR. In addition, t-BHP facilitated InsP3 and RyR activation and compensated mitochondria-ER refilling process (Fig 12-5).

t-BHP interrupted cross talk between ER and mitochondria Ca²⁺ stores, which in turn altered CCE. Ca²⁺ signals between ER and mitochondria occur during conditions that increase InsP₃. InsP₃ receptors have been identified as an important component of ER-mitochondria interactions [Boehning et al., 2003]. Cytochrome c, after its initial Ca²⁺ dependent release from the mitochondria translocates to the ER and interacts with InsP₃ receptors, which further augment Ca²⁺ release. The diminished ER Ca²⁺ augments CCE. During stimulation, mitochondria release cytochrome C and InsP₃ promotes release of ER calcium and together they activate apoptosis [Boehning et al., 2003]. Phospholipase C and polyphosphoinositides are required for the activation of the CCE [Broad et al., 2001].

Summary.

An understanding of the interaction of cellular calcium stores with oxidants and InsP₃ is necessary to determine how these are altered in AD, and to develop therapies directed at the calcium dysregulation in AD. Mitochondrial Ca²⁺ plays an important role in the regulation of the CCE following depletion of the ER, and the mechanisms differ in the presence and absence of InsP₃ producing agonist. Inhibition of mitochondrial Ca²⁺ uptake impaired the CCE in the absence of InsP₃ producing agonist, whereas it enhanced CCE activity in the presence of InsP₃ producing agonist. The results suggest that in the presence of InsP₃, mitochondria buffer the local Ca²⁺ released from ER following rapid InsP₃Ca²⁺ induced release, and serve as a negative feedback to the CCE. In the presence of InsP₃, blocking mitochondrial calcium release and uptake impaired and exaggerated CCE, respectively and these abnormalities could be reversed by t-BHP. On the other hand, impaired CCE by inhibition of mitochondrial Ca²⁺ uptake in the absence of InsP₃ could be diminished by SIN-1. Thus, selective oxidants regulated the interactions of mitochondria with CCE whether or not the InsP₃ producing agonist is present. Now that these interactions have been defined in a control line, the changes with AD can be delineated and strategies to ameliorate the abnormalities can be developed..

Abbreviations

BRCS

Bombesin-releasable calcium store

BSS

balanced salt solution

CCE

capacitative calcium entry

CGP37157

7-chloro-3,5-dihydro-5-phenyl-1H-4,1-benzothiazepine-2

CPA

cyclopiazonic acid

[Ca²⁺]_i

cytosolic free calcium concentration

DMEM

Dulbecco’s modified Eagle’s medium

ER

endoplasmic reticulum

Fura-2AM

fura-2-acetoxymethyl ester

PBS

phosphate-buffered saline

Ru360

ruthenium amino binuclear complex

SIN-1

3-morpholinosyndnonimine

t-BHP

tert-butyl-hydroxyperoxide