Mitochondrial Ca²⁺ Uptake Increases Ca²⁺ Release from Inositol 1,4,5-Trisphosphate Receptor Clusters in Smooth Muscle Cells

Abstract

Smooth muscle activities are regulated by inositol 1,4,5-trisphosphate (InsP₃)-mediated increases in cytosolic Ca²⁺ concentration ([Ca²⁺]_c). Local Ca²⁺ release from an InsP₃ receptor (InsP₃R) cluster present on the sarcoplasmic reticulum is termed a Ca²⁺ puff. Ca²⁺ released via InsP₃R may diffuse to adjacent clusters to trigger further release and generate a cell-wide (global) Ca²⁺ rise. In smooth muscle, mitochondrial Ca²⁺ uptake maintains global InsP₃-mediated Ca²⁺ release by preventing a negative feedback effect of high [Ca²⁺] on InsP₃R. Mitochondria may regulate InsP₃-mediated Ca²⁺ signals by operating between or within InsP₃R clusters. In the former mitochondria could regulate only global Ca²⁺ signals, whereas in the latter both local and global signals would be affected. Here whether mitochondria maintain InsP₃-mediated Ca²⁺ release by operating within (local) or between (global) InsP₃R clusters has been addressed. Ca²⁺ puffs evoked by localized photolysis of InsP₃ in single voltage-clamped colonic smooth muscle cells had amplitudes of 0.5–4.0 F/F₀, durations of ∼112 ms at half-maximum amplitude, and were abolished by the InsP₃R inhibitor 2-aminoethoxydiphenyl borate. The protonophore carbonyl cyanide 3-chloropheylhydrazone and complex I inhibitor rotenone each depolarized ΔΨ_M to prevent mitochondrial Ca²⁺ uptake and attenuated Ca²⁺ puffs by ∼66 or ∼60%, respectively. The mitochondrial uniporter inhibitor, RU360, attenuated Ca²⁺ puffs by ∼62%. The “fast” Ca²⁺ chelator 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid acted like mitochondria to prolong InsP₃-mediated Ca²⁺ release suggesting that mitochondrial influence is via their Ca²⁺ uptake facility. These results indicate Ca²⁺ uptake occurs quickly enough to influence InsP₃R communication at the intra-cluster level and that mitochondria regulate both local and global InsP₃-mediated Ca²⁺ signals.

Introduction

Smooth muscle functions to regulate many activities including blood flow in vascular blood vessels, peristaltic motion in the gastrointestinal tract, and rhythmic contractions of the uterus during labor. Transient increases in the cytosolic Ca²⁺ concentration ([Ca²⁺]_c)² provides the major trigger for contraction to regulate these activities. Other smooth muscle cell functions are also regulated by Ca²⁺ including transcription, growth, and apoptosis. Stimuli acting on the cell may increase [Ca²⁺]_c by influx of the ion from the extracellular medium, release from intracellular stores or both. The sarcoplasmic reticulum (SR) is an intracellular Ca²⁺ storage organelle and contains two Ca²⁺ release channels, the ryanodine receptor (RyR) and the inositol 1,4,5-trisphosphate receptor (InsP₃R). Release from RyR is triggered by a small influx of Ca²⁺ across the plasma membrane in a process known as Ca²⁺-induced Ca²⁺-release, which occurs in cardiac and skeletal muscle or in smooth muscle when the SR contains excessive levels of Ca²⁺ (1–3). In non-contracting cell types and smooth muscle, Ca²⁺ release from the SR occurs predominantly via InsP₃R and is triggered by hormonal stimulation of G protein-coupled receptors.

InsP₃R are not distributed uniformly throughout the SR but exist as clusters composed typically of 25–60 channels (4, 5). Ca²⁺ release from a cluster of InsP₃R, a Ca²⁺ puff, is attributed to the activation of a small number of InsP₃R within the cluster (4, 5). Ca²⁺ puffs are spatially localized Ca²⁺ transients that are short in duration and are considered the elementary building blocks of Ca²⁺ release from InsP₃R (6–10). Ca²⁺ puffs have been observed in several cell types including Xenopus oocytes, rat basophilic leukemia, glial, PC12, and smooth muscle cells where the localized release forms microdomains of [Ca²⁺]_c, which exceed that of the bulk cytoplasm (6–8, 10–12). Propagation of the Ca²⁺ response through the cell occurs when Ca²⁺ released from one InsP₃R cluster diffuses to other neighboring clusters and, in the presence of InsP₃, activates them in a Ca²⁺-induced Ca²⁺-release-like process (7, 8, 13, 14). Thus, activation of InsP₃R may generate a variety of Ca²⁺ responses that include localized Ca²⁺ puffs or cell-wide (global) responses in the form of Ca²⁺ oscillations and propagating Ca²⁺ waves.

Mitochondria, in addition to their ATP generating facility, take up Ca²⁺ from the cytosol. Global InsP₃-mediated Ca²⁺ release is modulated by mitochondrial Ca²⁺ uptake in smooth muscle and other cell types, which include HeLa cells, Xenopus oocytes, PC12 cells, and cultured oligodendrocytes (15–20). Mitochondrial Ca²⁺ uptake and the increase in mitochondrial Ca²⁺ concentration ([Ca²⁺]_mit), which normally occurs during global InsP₃-mediated increases in [Ca²⁺]_c is prevented by depolarizing the mitochondrial membrane potential (ΔΨ_M) in HeLa and rat basophilic leukemia-2H3 mast cells (21, 22). Inhibition of mitochondrial Ca²⁺ uptake has various effects on global InsP₃-mediated Ca²⁺ release. Mitochondrial Ca²⁺ uptake may negatively regulate InsP₃-mediated Ca²⁺ release in cultured hepatocytes and astrocytes. Here the Ca²⁺ wave amplitude or velocity or both increased when mitochondrial Ca²⁺ uptake was prevented by depolarizing ΔΨ_M (23, 24). In other cell types mitochondrial Ca²⁺ uptake positively affects InsP₃-mediated Ca²⁺ release so that preventing uptake inhibited Ca²⁺ responses in smooth muscle, astrocytes, and HeLa cells and increasing mitochondrial Ca²⁺ uptake augmented InsP₃-mediated Ca²⁺ wave amplitude and velocity in Xenopus oocytes (15–17, 19, 20, 25).

The differences in mitochondrial regulation of global InsP₃-mediated Ca²⁺ release in various tissues may arise from the localization of sites of mitochondrial Ca²⁺ uptake relative to sites of SR Ca²⁺ release. Proximity of sites of mitochondrial Ca²⁺ uptake to InsP₃R clusters may determine whether Ca²⁺ uptake prevents either inactivation or complete activation of InsP₃R. The localization of mitochondria will also determine whether the organelle may only regulate the global signal (e.g. waves) alone or may additionally regulate local (e.g. puffs) Ca²⁺ signals. For example, localization of mitochondria between InsP₃R clusters will enable the organelle to regulate global but not local Ca²⁺ signals. On the other hand, mitochondria positioned at InsP₃R clusters will enable the organelle to regulate local and global Ca²⁺ signals. To determine whether mitochondria regulate local or global Ca²⁺ signals, the effect of depolarizing ΔΨ_M, or inhibiting the mitochondrial uniporter, to prevent mitochondrial Ca²⁺ uptake, on localized release of Ca²⁺ from InsP₃R clusters (Ca²⁺ puffs) has been examined. InsP₃-mediated Ca²⁺ puffs were evoked by localized flash photolysis of caged InsP₃ in colonic smooth muscle cells. Ca²⁺ puff sites were uncoupled by using EGTA to prevent Ca²⁺ release from one InsP₃R cluster activating adjacent InsP₃R clusters and generate a global Ca²⁺ wave. Inhibition of mitochondrial Ca²⁺ uptake attenuated the magnitude of Ca²⁺ puffs as well as global InsP₃-mediated Ca²⁺ signals. These results indicate that mitochondrial Ca²⁺ uptake influences the InsP₃R communication at the intra-cluster level by regulating the amount of Ca²⁺ released from a cluster of InsP₃R before the released Ca²⁺ diffuses between InsP₃R clusters. Mitochondrial Ca²⁺ uptake appears to prevent a negative feedback effect of high [Ca²⁺]_c on InsP₃R activity within a cluster to prolong Ca²⁺ release from the SR. Support for this conclusion was found in experiments which show the “fast” Ca²⁺ chelator 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA) also prolonged Ca²⁺ release via InsP₃R clusters. Together, these results indicate that sites of mitochondrial Ca²⁺ uptake are located in proximity to InsP₃R to influence the local and global response to InsP₃-mediated Ca²⁺ release.

EXPERIMENTAL PROCEDURES

Cell Isolation

Male guinea pigs (350–500 g) were humanely killed by cervical dislocation followed by immediate exsanguination in accordance with the guidelines of the Animal (Scientific Procedures) Act UK 1986. A segment of intact distal colon (∼5 cm) was transferred to oxygenated (95% O₂, 5% CO₂) physiological saline solution comprising (mm): 118.4 NaCl, 25 NaHCO₃, 4.7 KCl, 1.13 NaH₂PO₄, 1.3 MgCl₂, 2.7 CaCl₂, and 11 glucose (pH 7.4). Following removal of the mucosa and longitudinal muscle layer from the tissue, single smooth muscle cells, largely from circular muscle, were enzymatically dissociated (16). All experiments were carried out at room temperature (20 ± 2 °C).

Electrophysiology

Cells were voltage-clamped using conventional tight-seal whole cell recording methods (26). The extracellular solution contained (mm): 80 sodium glutamate, 40 NaCl, 20 tetraethylammonium chloride, 1.1 MgCl₂, 3 CaCl₂, 10 HEPES, and 30 glucose (pH 7.4 with NaOH). The pipette solution contained (mm): 85 Cs₂SO₄, 20 CsCl, 1 MgCl₂, 30 HEPES, 3 MgATP, 2.5 pyruvic acid, 2.5 malic acid, 1 NaH₂PO₄, 5 creatine phosphate, 0.5 guanosine phosphate, and 0.025 caged inositol 1,4,5-trisphosphate (InsP₃) trisodium salt. Pyruvic acid and malic acid were present to maintain mitochondrial activity. The high concentration of HEPES was to ensure pH control during mitochondrial depolarization. Creatine phosphate and ATP were to maintain [ATP] during the experiments. Whole cell currents were measured using an Axopatch 200B (Axon Instruments, Union City, CA), low-pass filtered at 500 Hz (8-pole Bessel filter; Frequency Devices, Haverhill, MA), digitally sampled at 1.5 kHz using a Digidata interface and pClamp (version 8; Axon Instruments) and stored for analysis. In the majority of experiments, as specified under “Results,” EGTA (250 μm to 1 mm) was added to the pipette solution to buffer the [Ca²⁺]_c. EGTA was added to uncouple InsP₃R clusters (27). The slow binding kinetics (K_on = ∼5 μm⁻¹ s⁻¹) of the buffer prevents EGTA from significantly influencing the release of Ca²⁺ within a cluster (28). In other experiments, as specified under “Results,” BAPTA (250 μm) replaced EGTA in the pipette solution to buffer the [Ca²⁺]_c. The fast binding kinetics of BAPTA (K_on = 100–1000 μm⁻¹ s⁻¹) influences Ca²⁺ release within an InsP₃R cluster (28).

Imaging

Single, freshly isolated colonic smooth muscle cells were loaded with the Ca²⁺-sensitive dye fluo 3 acetoxymethylester (AM) (10 μm) and wortmannin (10 μm; to prevent contraction) for at least 20 min before the start of the experiment and then allowed to settle for 10 min. 30 min was sufficient to allow intracellular esterases to hydrolyze the AM moiety. Two-dimensional [Ca²⁺]_c images were obtained using a wide-field digital imaging system. Single cells were illuminated at 488 nm (bandpass 14 nm) from a monochrometer (Polychrome IV, T.I.L.L. Photonics, Martinsried, Germany) and imaged through an oil-immersion objective (×40 UV 1.3 NA; Nikon UK, Surrey, United Kingdom). Excitation light was passed via a fiber optic guide through a 485-bandpass (15 nm) filter and a fieldstop diaphragm and reflected off a 505-nm long-pass dichroic mirror. Emitted light was guided through a 535-nm barrier filter (bandpass 45 nm) to an intensified, cooled, frame transfer CCD camera (Pentamax Gen IV, Roper Scientific, Trenton, NJ) operating in “virtual chip” mode with program WinView 32 (Roper Scientific, Trenton, NJ). Full-frame images (160 × 160 pixels), with a pixel size of 720 nm at the cell, were acquired at 50 frames/s. In some experiments the software program Metafluor (Molecular Devices, Wokingham, UK) was used to obtain longer periods of data acquisition. In these experiments sampling rates of ∼10 frames/s were used during and 1 frame/s between Ca²⁺ transients. Ca²⁺ imaging data were recorded on a personal computer. Electrophysiological measurements and imaging data were synchronized by recording, on pClamp, a transistor transistor logic output from the CCD camera, which reported its readout status together with the electrophysiological information.

In some experiments, cells were loaded with the Ca²⁺-sensitive dye fluo 4-AM (10 μm) and the ΔΨ_M-sensitive dye tetramethylrhodamine ethyl ester (TMRE) (10 nm) so that [Ca²⁺]_i and ΔΨ_M, respectively, could be imaged near simultaneously (15). Here cells were illuminated at 475 and 560 nm, respectively, and light was passed via a fiber optic guide through a dual 483/553-nm bandpass filter (bandpass 15 and 20 nm, respectively), through an ND4 neutral density filter, and reflected off a custom-made dual long-pass dichroic mirror (transmissive in ranges 505–540 and 577–640 nm, and reflective from 490 to below 300 nm; Chroma, Rockingham, VT). The emitted light was guided through a dual bandpass 518/594-nm barrier filter (bandpass 25 and 18 nm, respectively) to the CCD camera.

Localized Flash Photolysis

The output of a xenon flashlamp (Rapp Optoelecktronic, Hamburg, Germany), used to uncage InsP₃, was passed through a UG-5 filter to select ultraviolet light, focused, and merged into the excitation light path through a fiber optic bundle and long-pass dichroic mirror at the lens part of the epi-illumination attachment of the microscope. The diameter of the fiber optic together with the lens magnification determined the area (spot size ∼20 or ∼125 μm) of InsP₃ photolysis (29). A photolysis region of ∼20 μm diameter was used to evoke InsP₃-mediated Ca²⁺ release to allow for greater flexibility in photolysis spot placement relative to the patch clamp electrode to prevent uncaging of InsP₃ within the patch pipette. The output intensity of the flash lamp was altered in the Ca²⁺ puff experiments to between 20 and 100% (0.025–0.19 milliwatts) of the maximum output to control the amount of InsP₃ that was uncaged and determined empirically in each cell.

Data Analysis

Images were analyzed using the program Metamorph 7.1.3 (Molecular Devices, Wokingham, UK). Fluorescence images were initially background subtracted and smoothed using a median average of 3 × 3 pixels. Changes in fluorescence were expressed as ratios (F/F₀ or ΔF/F₀) of fluorescence counts (F) relative to baseline (control) values (taken as 1) before stimulation (F₀). The average baseline value over the 100 frames occurring before flash photolysis of caged InsP₃ was subtracted from peak height (ΔF/F₀). Full width at half-maximum amplitude (FWHD) was used to determine the duration of the puff at half its peak value. The time to peak of the Ca²⁺ puff was measured as the time required to increase from 10 to 90% of maximal peak amplitude. The decay of the Ca²⁺ puff was measured as the time required to decline from 90 to 10% of maximal amplitude. The delay in the onset of the Ca²⁺ puff after photolysis of InsP₃ was measured as the time required for [Ca²⁺]_c to increase by 0.2 F/F₀ from baseline values.

Summarized results are expressed as mean ± S.E. of n cells. A paired or unpaired Student's t test was applied to the raw data, as appropriate; p < 0.05 was considered significant.

Drugs and Chemicals

Drugs were applied by addition to the extracellular solution. Concentrations in the text refer to the salts, where appropriate. Fluo 3-AM, fluo 4-AM, and TMRE were purchased from Invitrogen and caged InsP₃-trisodium salt from SiChem GmbH (Bremen, Germany). All other reagents were purchased from Sigma.

RESULTS

In voltage-clamped, single smooth muscle cells flash photolysis of caged InsP₃ (25 μm) at 60-s intervals produced transient elevations in the free intracellular Ca²⁺ concentration ([Ca²⁺]_c) throughout the photolysis region (i.e. global increases) (Fig. 1). Upon photolysis of InsP₃, [Ca²⁺]_c, measured by ΔF/F₀, increased to 3.91 ± 0.63 (n = 6). The rise in [Ca²⁺]_c from InsP₃ occurs exclusively as a result of InsP₃R activity; InsP₃-mediated Ca²⁺ release does not activate further Ca²⁺ release via RyR (30, 31).

FIGURE 1.

Mitochondrial depolarization decreased the magnitude of InsP₃-mediated Ca²⁺ release. At −70 mV, locally photolyzed caged InsP₃ (25 μm) (↑, C) in a ∼20 μm diameter region (A, bright spot in left-hand panel, see also whole cell electrode, left side) evoked approximately reproducible rises in Ca²⁺ in a freshly isolated colonic myocyte (B and C). Depolarization of ΔΨ_M by the protonophore CCCP (1 μm) (used with oligomycin; 6 μm) inhibited InsP₃-mediated Ca²⁺ increases (B and C). [Ca²⁺]_c and ΔΨ_M are shown near simultaneously using fluo 4 and TMRE, respectively. Mitochondria appeared as punctuate areas of fluorescence because of their ΔΨ_M and were imaged before (A, middle panel) and after (A, right-hand panel) superfusion of CCCP. When ΔΨ_M was depolarized TMRE dissipated and redistributed throughout the cell. The [Ca²⁺]_c images (B) are derived from the time points indicated by the corresponding numbers in C. [Ca²⁺]_c changes in B are expressed by color; dark blue, low and yellow, high [Ca²⁺]_c. Flash photolysis of InsP₃ (↑, C) at ∼60-s intervals generated approximately comparable [Ca²⁺]_c increases (C). InsP₃ continued to be photolyzed at ∼60-s intervals during superfusion of CCCP and oligomycin to depolarize ΔΨ_M, which inhibits mitochondrial Ca²⁺ uptake (horizontal bar; C). Measurements were made from a 3 × 3 pixel box located within the flash area. Mitochondrial depolarization with CCCP plus oligomycin did not alter SR Ca²⁺ content: RyR activation by caffeine (CAF, 100 μm, applied by localized pressure ejection) evoked comparable Ca²⁺ release before, immediately after, and 15 min after superfusion of CCCP plus oligomycin (1 and 6 μm, respectively), despite inhibition of Ca²⁺ increases evoked by InsP₃ (D).

Mitochondria accumulate Ca²⁺ in these cells following release of the ion via IP₃R (15). The contribution of mitochondrial Ca²⁺ uptake to the magnitude of global InsP₃-mediated SR Ca²⁺ release was examined. Mitochondrial Ca²⁺ uptake was prevented by collapsing the mitochondrial electrochemical (ΔΨ_M) gradient using the protonophore CCCP. Because, in CCCP, the mitochondrial ATPase operates in reverse direction oligomycin was also included to inhibit the F₁F₀-ATPase and prevent ATP depletion. ATP (3 mm) and phosphocreatine (5 mm) were present in the patch pipette filling solution. When mitochondrial Ca²⁺ uptake was prevented in CCCP and oligomycin (1 and 6 μm, respectively), the InsP₃-mediated Ca²⁺ increase was inhibited (see Fig. 1) (15, 16). The InsP₃-mediated Ca²⁺ increase was 3.91 ± 0.63 ΔF/F₀ in control and 1.33 ± 0.47 ΔF/F₀ (n = 6) after CCCP and oligomycin (i.e. 34% of control (p < 0.05)). The inhibition of the IP₃-mediated Ca²⁺ release is unlikely to be explained by a reduction in the SR Ca²⁺ content by CCCP and oligomycin. The SR Ca²⁺ content, as assessed by the extent of Ca²⁺ release by the RyR agonist caffeine was unaffected by CCCP plus oligomycin (Fig. 1D). RyR and IP₃R share access to a single common Ca²⁺ store in these cells (32). Together, mitochondrial Ca²⁺ uptake regulates global InsP₃-mediated Ca²⁺ release.

To generate a global increase in Ca²⁺, Ca²⁺ released via one cluster of InsP₃R activates other adjacent InsP₃R clusters in a Ca²⁺-induced Ca²⁺-release-like manner to summate into a cell-wide Ca²⁺ rise (8, 33, 34). Mitochondria may control the global rise in [Ca²⁺]_c evoked by InsP₃ by regulating the Ca²⁺ signal that propagates among InsP₃R clusters. Alternatively mitochondria may regulate the Ca²⁺ signal, which occurs within a single InsP₃R cluster. In the former, mitochondria may only regulate global Ca²⁺ signals. In the latter, both local and global signals will be controlled by mitochondrial activity. The question arises do mitochondria influence InsP₃-mediated release by operating within or between InsP₃R clusters?

To address this question, mitochondrial regulation of the amplitude and duration of Ca²⁺ release from a single cluster of InsP₃R was examined. Ca²⁺ puffs, the fluorescence manifestation of the discrete release of Ca²⁺ from a single cluster of InsP₃R, were evoked by low energy (∼20–50% of maximum) flash photolysis of caged InsP₃ (25 μm). Ca²⁺ puffs differed from a global rise in [Ca²⁺]_c in that they were localized events that occurred within a small region of the flash area and their duration was much shorter than for a global release of Ca²⁺. In contrast, when there was a global rise, [Ca²⁺]_c increased uniformly throughout the flash area. The following criteria identified Ca²⁺ rises as “puffs”: a change in peak amplitude of between 1 and 3 ΔF/F₀, time to peak of 50 to 100 ms, and a duration at half-maximum amplitude (FWHD) of 100 to 200 ms. These criteria are derived from those previously reported in smooth muscle, Xenopus oocytes, and HeLa cells (7, 10, 13, 14, 35).

In initial experiments Ca²⁺ puffs were observed infrequently and over only a narrow range of [InsP₃] above which they were rapidly summated to produce a global rise in Ca²⁺ response. To overcome this restricted range and evoke Ca²⁺ puffs consistently, a low concentration of the slow Ca²⁺ buffer, EGTA, was added to the cells via the patch pipette. EGTA prevents InsP₃-mediated Ca²⁺ puffs from coalescing into a global Ca²⁺ rise because the buffer prevents the ion from reaching neighboring InsP₃R clusters (27). As a first step in these experiments, the effect of [EGTA] (between 250 μm and 1 mm) on Ca²⁺ puffs was examined. At [EGTA] (250–500 μm) Ca²⁺ puffs of a similar magnitude to those reported elsewhere and in non-buffered cells were measured (Fig. 2). Above 500 μm EGTA significantly attenuated the amplitude of Ca²⁺ puffs and, in some cells, they could not be evoked. EGTA (300 μm) at a concentration similar to that used in experiments examining Ca²⁺ puffs in Xenopus oocytes was used in the present study to enable Ca²⁺ puffs to be evoked and measured (27, 36). EGTA (300 μm) neither affects the magnitude nor slows the kinetics of Ca²⁺ puffs (27, 37).

FIGURE 2.

Ca²⁺ puffs may occur with different latencies of onset after photolysis of InsP₃. At −70 mV, localized photolysis of caged InsP₃ (25 μm) (↑, B) in a ∼20-μm diameter region triggered several Ca²⁺ puffs (A and B). Five localized Ca²⁺ puffs were evoked within the photolysis site; the time of onset for each Ca²⁺ puff was variable. Ca²⁺ puff 1 occurred 160 ms after photolysis of InsP₃ and was located 22.5 μm away from Ca²⁺ puff 2, which occurred 200 ms after photolysis. The measurement of delay in Ca²⁺ puff onset after photolysis of InsP₃ was determined by how long it took for the [Ca²⁺]_c to increase by 0.2 F/F₀ from baseline. The [Ca²⁺]_c images (A) are derived from the time points indicated by the corresponding lowercase letters in B. [Ca²⁺]_c changes in A are expressed by color: dark blue, low and light blue, high [Ca²⁺]_c. Measurements were made from 3 × 3 pixel boxes located at the center of each puff (not shown). To record Ca²⁺ puffs single colonic myocytes were buffered using EGTA (300 μm). This did not affect the amplitude, nor duration of the puff but did allow for puffs to be recorded using a larger range of concentrations of InsP₃. The large increase in fluorescence at time 0 is the flash artifact.

Characteristics of Ca²⁺ Puffs

In each cell there were typically one to five individual puffs evoked within the photolysis region (∼20 μm diameter). The latency between photolysis of caged InsP₃ and onset of the Ca²⁺ puff, peak amplitude, time to peak, FWHD, and decay time were measured (Table 1). The amplitude of Ca²⁺ puffs varied between 0.5 and 4.0 ΔF/F₀ and averaged 1.94 ± 0.24 ΔF/F₀ (n = 36 from 12 cells). The average time to peak (10–90% interval) of Ca²⁺ puffs was 57 ± 8 ms, FWHD was 112 ± 16 ms, and the decay time (90–10% interval) was 267 ± 40 ms. Ca²⁺ puffs also had variable latency between the time of photolysis of InsP₃ and when the rise in [Ca²⁺]_c occurred (Fig. 2). In the representative cell shown in Fig. 2, five Ca²⁺ puffs occurred after photolysis of InsP₃. Fig. 2A, a and b, show two Ca²⁺ puffs (puff 1 and puff 2) that are 22.5 μm apart; the peak of Ca²⁺ puff 1 preceded that of puff 2 by 40 ms. The majority of cells analyzed displayed 1–2 puff sites and the average onset latency was ∼64 ± 12 ms (n = 36 from 12 cells).

TABLE 1.

Comparison of the amplitude, time to peak, full width at half-maximum amplitude, decay time, and latency of onset of Ca²⁺ puffs under different conditions

The time to peak and decay time were determined from the changes that occurred between 10 and 90% of the peak ΔF/F₀ values.

Condition	Amplitude, ΔF/F0	Time to peak (10–90%)	FWHD	Decay time (90-10%)	Latency of onset	n
ms
EGTA buffered	1.94 ± 0.24	57 ± 8	112 ± 16	267 ± 40	64 ± 12	36 from 12 cells
BAPTA buffered	1.53 ± 0.54	1486 ± 156	1208 ± 334	4324 ± 435		12 from 4 cells
Not buffered	1.01 ± 0.17	419 ± 141	481 ± 76	1066 ± 279		3 from 3 cells

Ca²⁺ puffs at various sites were not of fixed and constant size but had a continuum of amplitudes presumably due to activation of a different number of InsP₃R within different clusters (13, 38). In the representative cell shown in Fig. 3 a large photolysis region (125 μm diameter) was used to evoke InsP₃-mediated Ca²⁺ release and generated puffs of various amplitudes at different sites. There were three individual Ca²⁺ puffs (labeled puff 1, 2, and 3); the largest increase in [Ca²⁺]_c occurred at puff 2. The increases in [Ca²⁺]_c, which occurred at the other sites are unlikely to be explained by diffusion of Ca²⁺ from puff 2 (Fig. 3, B and C). If the increases in [Ca²⁺]_c were due to diffusion, then a gradual decrease in the maximum change in [Ca²⁺]_c would be expected to have occurred from the site of Ca²⁺ release, the rate of Ca²⁺ increase would have decreased, and there would be a delay in the onset of the Ca²⁺ increase. At a location situated outside the flash site area, where InsP₃ was not photoreleased, no increase in [Ca²⁺]_c occurred (region 4). Although puffs at different sites had various amplitudes, within a site Ca²⁺ puffs could be reproducibly evoked and had consistent amplitudes (see Figs. 5C, 6C, and 7C). In the remaining experiments a photolysis region of ∼20 μm diameter was used to evoke InsP₃-mediated Ca²⁺ release to allow for greater flexibility in photolysis spot placement relative to the patch clamp electrode to prevent uncaging of InsP₃ within the patch pipette.

FIGURE 3.

InsP₃-mediated Ca²⁺ puffs may occur in multiple regions simultaneously. At −70 mV, localized photolysis of caged InsP₃ (25 μm) (↑, C) in a 125-μm diameter region (A, bright spot highlighted with a dotted line, see also whole cell electrode, left side) triggered Ca²⁺ puffs in an EGTA (300 μm)-buffered colonic myocyte (B and C). The Ca²⁺ puffs were of various amplitudes and occurred near simultaneously in different regions of the flash photolysis site. [Ca²⁺]_c did not increase in areas (region 4) located outside the photolysis site. The [Ca²⁺]_c images (B) are derived from the time points indicated by the corresponding lowercase letters in C. [Ca²⁺]_c changes in B are expressed by color: dark blue, low and light blue, high [Ca²⁺]_c. Measurements were made from 3 × 3 pixel boxes located at the center of each puff (not shown). The large increase in fluorescence at time 0 is the flash artifact.

FIGURE 5.

Ca²⁺ puffs are inhibited by 2-APB. At −70 mV, locally photolyzed caged InsP₃ (25 μm) (↑, C) in a ∼20-μm diameter region (A, bright spot in left-hand panel, see also whole cell electrode, left side) evoked Ca²⁺ puffs in an EGTA (300 μm)-buffered colonic myocyte (B and C). A second and third photolysis of InsP₃, each at ∼60-s intervals, at the same site generated an approximately comparable [Ca²⁺]_c increase (C). Superfusion of 2-APB (100 μm) abolished InsP₃-mediated Ca²⁺ puffs within ∼60 s (B and C). The [Ca²⁺]_c images (B) are derived from the time points indicated by the corresponding numbers in C. [Ca²⁺]_c changes in B are represented by color: dark blue, low and light blue, high [Ca²⁺]_c. Measurements were made from a 3 × 3 pixel box (A, right-hand panel, white square). The large increase in fluorescence at time 0 is the flash artifact.

FIGURE 6.

Depolarization of ΔΨ_M with CCCP/oligomycin inhibits Ca²⁺ puffs. At −70 mV, locally photolyzed caged InsP₃ (25 μm) (↑, C) in a ∼20-μm diameter region (A, bright spot in left-hand panel, see also whole cell electrode, left side) evoked Ca²⁺ puffs in an EGTA (300 μm)-buffered colonic myocyte (B and C). Note: there are two individual Ca²⁺ puff sites in response to photorelease of InsP₃; one site releases Ca²⁺ just before the other site. Flash photolysis of InsP₃ every ∼60 s generated approximately comparable [Ca²⁺]_c increases (C). Superfusion of CCCP and oligomycin (1 and 6 μm, respectively) while continuing to photolyze InsP₃ at ∼60 intervals, decreased the amplitude of InsP₃-mediated Ca²⁺ puffs (B and C). The [Ca²⁺]_c images (B) are derived from the time points indicated by the corresponding numbers in C. [Ca²⁺]_c changes in B are expressed by color: dark blue, low and light blue, high [Ca²⁺]_c. Measurements were made from a 3 × 3 pixel box (A, right-hand panel, white square). The large increase in fluorescence at time 0 is the flash artifact.

FIGURE 7.

Depolarization of ΔΨ_M with rotenone inhibits Ca²⁺ puffs. At −70 mV, locally photolyzed caged InsP₃ (25 μm) (↑, C) in a ∼20-μm diameter region (A, bright spot in left-hand panel, see also whole cell electrode, left side) evoked Ca²⁺ puffs in an EGTA (300 μm)-buffered colonic myocyte (B and C). Flash photolysis of InsP₃ every ∼60 s generated approximately comparable [Ca²⁺]_c increases (C). Superfusion of rotenone and oligomycin (5 and 6 μm, respectively), whereas continuing to photolyze InsP₃ at ∼60-s intervals, decreased the amplitude of InsP₃-mediated Ca²⁺ puffs (B and C). The [Ca²⁺]_c images (B) are derived from the time points indicated by the corresponding numbers in C. [Ca²⁺]_c changes in B are expressed by color: dark blue, low and light blue, high [Ca²⁺]_c. Measurements were made from a 3 × 3 pixel box (A, right-hand panel, white square). The large increase in fluorescence at time 0 is the flash artifact.

Ca²⁺ puffs could be evoked in the absence of the Ca²⁺ buffer EGTA. However, there was substantial variation in response to the same photolysis strength and Ca²⁺ puffs could only be evoked over a narrow range of [InsP₃] and flash intensities presumably because, in the absence of EGTA, Ca²⁺ could evoke release at neighboring InsP₃R clusters. Ca²⁺ puffs in non-buffered cells were evoked by low energy flash photolysis of a low concentration (6.25 μm) of caged InsP₃. In the representative cell, a single Ca²⁺ puff was evoked in response to photolysis of InsP₃ (Fig. 4). [Ca²⁺]_c was increased at this location by 0.88 ΔF/F₀. The average puff amplitude in non-buffered myocytes was 1.01 ± 0.17 ΔF/F₀ (n = 3). The average time to peak (10–90% interval) was 419 ± 141 ms, FWHD was 481 ± 76 ms, and the decay time (90–10% interval) was 1066 ± 279 ms (Table 1). The rise and decay of Ca²⁺ puffs from non-buffered cells were both slower than that measured in EGTA-buffered cells suggesting that EGTA facilitates diffusion of Ca²⁺ away from an InsP₃R cluster presumably by capturing the ion (37). Increasing the flash lamp intensity, which liberates a greater amount of InsP₃, triggered a global rise in Ca²⁺ (Fig. 4). The global [Ca²⁺]_c increase, evoked by the greater release of InsP₃, was 3.48 ΔF/F₀.

FIGURE 4.

InsP₃-mediated Ca²⁺ puffs and global Ca²⁺ increases. At −70 mV, locally photolyzed caged InsP₃ (6.25 μm) (↑, B) in a ∼20-μm diameter region (A, bright spot in left-hand panel, see also whole cell electrode, left side) evoked a Ca²⁺ puff that was preceded by a Ca²⁺“blip”-like event (A and B). Flash photolysis of caged InsP₃ in the area is indicated by the circle, in A each ∼60 s generated Ca²⁺ puffs (B, black line). When photolysis energy was increased to photolyze more InsP₃ a larger amount of Ca²⁺ release occurred throughout the region (B, red line). The Ca²⁺ puff was overlaid (B, right panel) for comparison. The [Ca²⁺]_c images (A) are derived from the time points indicated by the corresponding numbers in B. [Ca²⁺]_c changes in A are expressed by color: dark blue, low and light blue, high [Ca²⁺]_c. Measurements were made from a 3 × 3 pixel box located at the center of each Ca²⁺ release event (not shown). The large increase in fluorescence at time 0 is the flash artifact.

To ensure Ca²⁺ puffs arose from InsP₃-mediated Ca²⁺ release the effect of the InsP₃R blocker 2-aminoethoxydiphenyl borate (2-APB) was examined. In this series of experiments EGTA (300 μm) was again used to facilitate the measurement of consistent Ca²⁺ puffs. Approximately reproducible Ca²⁺ puffs were evoked by photolysis of caged InsP₃ (25 μm) (Fig. 5). Ca²⁺ puffs were abolished within 60 s by 2-APB (100 μm) (Fig. 5). Ca²⁺ puff amplitude was 2.04 ± 0.32 ΔF/F₀ before and 0.25 ± 0.02 ΔF/F₀ (n = 3) (p < 0.05) after application of 2-APB. Together these results suggest the localized Ca²⁺ increases evoked by InsP₃R in EGTA-buffered colonic myocytes are Ca²⁺ puffs and suitable for the examination of the influence of mitochondrial Ca²⁺ uptake on intra-cluster dynamics during InsP₃-mediated SR Ca²⁺ release.

Mitochondrial Control of Ca²⁺ Puffs

To determine whether or not mitochondria modulate InsP₃-mediated Ca²⁺ signaling by operating within or between InsP₃R clusters, the effect of inhibition of mitochondrial Ca²⁺ uptake on Ca²⁺ puffs was examined. The slow Ca²⁺ chelator EGTA (300 μm), as before, was used to prevent Ca²⁺ released from one InsP₃R cluster from activating neighboring InsP₃R clusters and generating a global Ca²⁺ rise. Ca²⁺ puffs were evoked by localized flash photolysis of caged InsP₃ at 60-s intervals (Fig. 6). When the cell was superfused with CCCP (1 μm; and oligomycin, 6 μm), to depolarize ΔΨ_M and inhibit mitochondrial Ca²⁺ uptake, Ca²⁺ puff amplitude decreased to 34% of control (1.34 ± 0.17 ΔF/F₀ before and 0.46 ± 0.05 ΔF/F₀ after CCCP, n = 5, p < 0.05; Fig. 6). Rotenone (5 μm; and oligomycin, 6 μm) a complex 1 inhibitor, which also depolarizes the ΔΨ_M and prevents mitochondrial Ca²⁺ uptake, inhibited SR Ca²⁺ release. Ca²⁺ puff amplitude was decreased to 40% of its original amplitude in rotenone (Fig. 7). Thus, Ca²⁺ puff amplitude was 2.02 ± 0.36 ΔF/F₀ before and 0.82 ± 0.15 ΔF/F₀ after rotenone application (n = 4, p < 0.05). Direct inhibition of mitochondrial Ca²⁺ uptake by Ru360 (10 μm) also inhibited Ca²⁺ puff amplitude to 38% of control values (from 1.32 ± 0.40 in control to 0.51 ± 0.03 in Ru360 ΔF/F₀, n = 9 puffs from 3 cells, p < 0.05). These results suggest that Ca²⁺ released from a cluster of InsP₃R is taken up by mitochondria before it diffuses to other neighboring clusters, i.e. mitochondrial Ca²⁺ uptake is involved in regulating InsP₃R Ca²⁺ release at the intra-cluster level.

The results suggest that mitochondrial Ca²⁺ uptake prolongs Ca²⁺ release and occurs rapidly to influence the kinetics of Ca²⁺ release within an InsP₃R cluster. To test this proposal further the cell was buffered with fast Ca²⁺ chelator BAPTA. Although BAPTA has a similar Ca²⁺ binding affinity to EGTA its faster binding kinetics (K_on = 100–1000 μm⁻¹ s⁻¹) allows BAPTA to capture the ion while within an InsP₃R cluster in a way similar to how mitochondrial Ca²⁺ uptake is proposed to function. A Ca²⁺ puff evoked in an EGTA-buffered myocyte is compared with a Ca²⁺ release event of the same magnitude in a BAPTA-buffered myocyte (Fig. 8). Although the average peak amplitude (1.53 ± 0.54 ΔF/F₀ in BAPTA) was similar to that in EGTA, the average time to peak (1486 ± 156 ms; n = 12 from 4 cells), a measure of the time course of release, was significantly prolonged in BAPTA-buffered myocytes (Table 1). These results suggest that restricting the [Ca²⁺]_c change at an InsP₃R cluster prolongs the time course of InsP₃-meditated Ca²⁺ release. These results are consistent with those in Xenopus oocytes; BAPTA prolonged the duration of InsP₃-mediated Ca²⁺ release and Ca²⁺ release events were no longer spatially discrete (28). This experiment further supports that mitochondrial Ca²⁺ uptake influences both global and local InsP₃-mediated Ca²⁺ signals.

FIGURE 8.

Localized InsP₃-mediated Ca²⁺ release in BAPTA- and EGTA-buffered myocytes. When myocytes are buffered with the fast Ca²⁺ chelator BAPTA, Ca²⁺ release is significantly prolonged. A Ca²⁺ puff in an EGTA (300 μm)-buffered myocyte (black trace) is compared with a Ca²⁺ release event of a comparable magnitude in a BAPTA (250 μm)-buffered myocyte (red trace). Myocytes were voltage-clamped at −70 mV and caged InsP₃ (25 μm) (↑) was locally photolyzed in a ∼20-μm diameter region (not shown). The rate of rise, a measure of the time course of Ca²⁺ release, was significantly increased when compared with an EGTA-buffered myocyte. The inset shows an expanded time scale to illustrate the differences in rate of rise of the Ca²⁺ events in the EGTA (black line)- and BAPTA (red line)-buffered myocytes. Measurements were made from 3 × 3 pixel boxes located at the center of each Ca²⁺ release event (not shown). The large increase in fluorescence at time 0 is the flash artifact.

DISCUSSION

InsP₃-sensitive Ca²⁺ release initiates at discrete sites on the SR. These sites contain a few tens of receptors from which the local increase in [Ca²⁺] is termed a puff. Ca²⁺ puffs are spatially restricted events and of short duration. Puffs are considered elementary release events and may interact and coalesce to generate a global release in Ca²⁺. In the present study Ca²⁺ puffs, evoked by photorelease of caged InsP₃, occurred in multiple regions of the cell either near simultaneously or after various latencies of onset. Ca²⁺ puffs at different locations had various amplitudes and durations. Yet, at an individual site Ca²⁺ puff amplitude was relatively constant in response to a given [InsP₃]. Ca²⁺ puffs occurred by the release of Ca²⁺ from InsP₃R. In support, puffs were blocked by the InsP₃R inhibitor 2-APB.

In smooth muscle, global InsP₃-mediated Ca²⁺ release is modulated by mitochondrial Ca²⁺ uptake. InsP₃-evoked Ca²⁺ release decreased when the driving force for mitochondrial Ca²⁺ uptake was collapsed by depolarization of ΔΨ_M with either CCCP or rotenone (15, 16, 25). How mitochondria regulate InsP₃-mediated Ca²⁺ signals was examined here. If Ca²⁺ uptake occurred at an InsP₃R cluster mitochondria would influence local Ca²⁺ signaling at the level of a puff. Alternatively, if Ca²⁺ uptake occurred between clusters then mitochondria would influence global but not local Ca²⁺ signals. These two possibilities were addressed by examining the influence of mitochondria on local Ca²⁺ signals. When mitochondrial Ca²⁺ uptake was prevented by depolarizing ΔΨ_M with CCCP or rotenone, Ca²⁺ puff amplitude decreased by 66 or 60%, respectively. When mitochondrial Ca²⁺ uptake was prevented directly by the uniporter antagonist Ru360, Ca²⁺ puff amplitude decreased by 62%. These results suggest that mitochondria regulate Ca²⁺ release by acting within an IP₃R cluster.

Ca²⁺ exerts positive and negative feedback effects on IP₃R. Positive feedback, which increases Ca²⁺ release, occurs over lower concentrations of the ion (∼<500 nm), whereas negative feedback dominates at higher [Ca²⁺]. Mitochondria because of their low affinity for the ion may limit only the negative feedback effect of Ca²⁺ on the IP₃-mediated Ca²⁺ release. In support of the occurrence of negative feedback the fast Ca²⁺ chelator BAPTA prolonged Ca²⁺ release in a way similar to how mitochondrial Ca²⁺ uptake appears to function to increase the InsP₃-mediated Ca²⁺ release. Together the results suggest mitochondrial Ca²⁺ uptake influences the amount of Ca²⁺ released from a cluster of InsP₃R and as a result will regulate both local and global InsP₃-mediated Ca²⁺ signals.

The question arises as to how mitochondria, by lowering [Ca²⁺] near IP₃R, increases the [Ca²⁺]_c derived from IP₃R activity. Large increases in [Ca²⁺]_c (∼1 μm) may generate a persistent reduction in activity in IP₃R, which resembles channel inactivation (29, 40). When IP₃R falls into this inactivated-like state Ca²⁺ release is terminated for many (up to 30) seconds (29, 40). Mitochondria by removing Ca²⁺ near IP₃R may prevent the ion from reaching a concentration high enough to induce the persistent inactivation of the channel. Ca²⁺ release may persist when mitochondria maintain a slightly lower [Ca²⁺] near IP₃R.

Other support for close proximity between sites of mitochondrial Ca²⁺ uptake and sites of SR Ca²⁺ release has come predominantly from experiments that measure increases in both [Ca²⁺]_c and mitochondrial Ca²⁺ concentration ([Ca²⁺]_mit) during InsP₃-mediated Ca²⁺ release (11, 17, 18, 41). In rat basophilic leukemia-2H3 mast cells buffering [Ca²⁺]_c with EGTA (100 μm) decreased the global [Ca²⁺]_c signal yet did not prevent an increase in [Ca²⁺]_mit (22). This result suggests that InsP₃R clusters are located to within 100 nm of sites of mitochondrial Ca²⁺ uptake. It may be that there are structural tethers that keep sites of mitochondrial Ca²⁺ uptake and SR Ca²⁺ release in close proximity (42). Indeed, structural evidence provides support for close association between mitochondria and SR membranes. Close associations between the SR and mitochondrial membranes have been noted in several cell types (21, 22, 43–45). For example, in unstimulated tracheal smooth muscle the majority of mitochondria completely enveloped and form multiple junctions with at least one contact point within ∼22 nm of the SR (43, 45). In HeLa cells there are also close contacts between the endoplasmic reticulum and mitochondria (5–20% mitochondrial surface) (21). Immunogold labeling of thin sections of Purkinje neurons demonstrated close associations between mitochondria and InsP₃R in stacked endoplasmic reticulum cisternae (46, 47). It is in these regions of close contact that Ca²⁺ released from the SR is proposed to be taken up by mitochondria via the Ca²⁺ uniporter (11, 48).

Although there is close proximity between SR and mitochondria, Ca²⁺ uptake may either positively or negatively regulate local InsP₃-mediated Ca²⁺ release. In oligodendrocyte progenitor cells, as in the present study, mitochondrial Ca²⁺ uptake positively affected (increased) InsP₃-mediated Ca²⁺ release. Depolarizing ΔΨ_M decreased the number of cells that exhibited methacholine-evoked Ca²⁺ puffs and shortened the duration of Ca²⁺ puffs in those cells that still responded (49). However, the changes in puff characteristics were attributed to a decrease in the amount of agonist-mediated InsP₃ produced after ΔΨ_M depolarization rather than feedback regulation of InsP₃R activity. Changes in InsP₃ concentration are an unlikely explanation for the present findings in smooth muscle because the concentration of inositide was increased by photolysis of caged InsP₃ rather than being generated by an agonist.

Alternatively, there may be a negative relationship between mitochondrial function and Ca²⁺ puff amplitude when InsP₃R clusters are in close proximity to sites of mitochondrial Ca²⁺ uptake. In rat basophilic leukemia-2H3 mast cells, although sites of mitochondrial Ca²⁺ uptake are located near InsP₃R, inhibition of mitochondrial Ca²⁺ uptake increased InsP₃-mediated Ca²⁺ release (22). In Xenopus oocytes the majority of mitochondria are located an average of ∼2.3 μm from Ca²⁺ puffs sites so that they are separated by a distance where presumably they do not take up Ca²⁺ during a Ca²⁺ puff (44). Yet, there are also a small subset of mitochondria that are in close proximity (<1.25 μm) to Ca²⁺ release sites. Upon sustained InsP₃ application, both Ca²⁺ puffs and Ca²⁺ wave initiation occurred less frequently at sites where mitochondria were in close proximity to the SR suggesting that this population of mitochondria function to prevent Ca²⁺ release (i.e. mitochondria support a negative feedback influence of InsP₃-mediated Ca²⁺ release) (44). Interestingly, the amplitude of the global InsP₃-evoked Ca²⁺ response decreased when mitochondrial Ca²⁺ uptake was prevented in Xenopus oocytes (i.e. in this case mitochondria appear to provide a positive feedback influence on Ca²⁺ release) (20). Initially, the effects of mitochondria on local and global signals may appear contradictory. Perhaps in Xenopus oocytes mitochondria may influence inter-cluster (positive feedback) as well as intra-cluster (negative feedback) communication.

Both the positive and negative effects of ΔΨ_M depolarization on InsP₃-mediated Ca²⁺ release were prevented when cells were buffered using BAPTA (23, 39). These results suggest mitochondrial effects on InsP₃-mediated Ca²⁺ release are exerted via Ca²⁺ uptake. For example, uncoupling InsP₃R clusters with BAPTA prevented the potentiation of InsP₃-mediated Ca²⁺ release, which occurred upon depolarizing ΔΨ_M in hepatocytes (23). BAPTA also restored InsP₃-mediated Ca²⁺ release, which had decreased after depolarizing ΔΨ_M in baby hamster kidney-21 cells (39). In the present study BAPTA prolonged InsP₃-mediated Ca²⁺ release. In both smooth muscle and Xenopus oocytes BAPTA appears to operate by binding to Ca²⁺ within an InsP₃R cluster and preventing intra-cluster negative feedback to prolong the SR Ca²⁺ release (28). At the concentration used (250 μm) BAPTA should “capture” Ca²⁺ ∼70–200 nm from the release site (28). Thus BAPTA, which both prolongs InsP₃-mediated Ca²⁺ release and prevents ΔΨ_M depolarization from affecting InsP₃-mediated Ca²⁺ release, supports a direct role of mitochondrial Ca²⁺ uptake in influencing InsP₃R communication at the intra-cluster level.

In conclusion, in colonic smooth muscle sites of mitochondrial Ca²⁺ uptake influence InsP₃-mediated Ca²⁺ release within an InsP₃R cluster. Presumably close contacts between mitochondria and InsP₃R act to prevent negative feedback inhibition of Ca²⁺ release at InsP₃R clusters. The interaction between mitochondria and SR will modulate Ca²⁺ signaling mechanisms from local Ca²⁺ puffs to global Ca²⁺ oscillations and waves.