INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PROPAGATING OSCILLATIONS of intracellular Ca²⁺ concentration ([Ca²⁺]_i) have been reported in response to acetylcholine (ACh) in porcine (6, 12, 17, 18, 23) and guinea pig (24) tracheal smooth muscle (TSM) cells. Such ACh-induced [Ca²⁺]_i oscillations in porcine TSM cells arise from repetitive release of sarcoplasmic reticulum (SR) Ca²⁺ through ryanodine-receptor (RyR) channels (6, 23). The propagation of [Ca²⁺]_i oscillations suggests that agonist stimulation triggers localized Ca²⁺ release that further stimulates release from adjacent regions, perhaps via a Ca²⁺-induced Ca²⁺ release (CICR) mechanism (2, 6, 23). Another important observation is that the propagation of [Ca²⁺]_i oscillations is initiated within a limited region of the TSM cell and propagation occurs in one direction (6, 23). The initiation of [Ca²⁺]_i oscillations in a limited region of the cell may reflect localized differences in RyR-channel distribution (11) and/or sensitivity to CICR.

Spontaneous, localized Ca²⁺ transients (Ca²⁺ sparks) have been observed in skeletal muscle fibers (8, 10, 25), cardiac myocytes (5, 14, 16), and vascular smooth muscle cells (15; also reviewed in Refs. 4, 21). Ca²⁺ sparks are thought to represent unitary Ca²⁺ release through RyR channels (5, 8, 25). Accordingly, the amplitude of Ca²⁺ sparks likely reflects the number of RyR channels that are more or less synchronous in their Ca²⁺ release and the frequency reflects channel kinetics. In a recent study, Sieck et al. (23) demonstrated the existence of Ca²⁺ sparks in porcine TSM cells, which also most likely represents SR Ca²⁺ release through RyR channels. Because propagating [Ca²⁺]_i oscillations in TSM cells also represent SR Ca²⁺ release through RyR channels, it is likely that there is a spatial and temporal relationship between the pattern of spontaneous Ca²⁺ sparks and agonist-induced [Ca²⁺]_i oscillations.

The temporal aspects of Ca²⁺ sparks have been characterized in several previous studies (8, 10, 14-16, 25) with line-scan confocal microscopy, with a temporal resolution of ~2 ms. These studies have provided important information on the amplitude and incidence of sparks. However, the spatial resolution of [Ca²⁺]_i measurements with line-scan confocal microscopy is relatively poor (~1 µm width with a ×60, 1.4-numerical aperture oil-immersion objective). Furthermore, evaluation of the temporal aspects of sparks and propagating [Ca²⁺]_i oscillations necessitates measurements from different parts of the cell. Therefore, line-scan confocal microscopy is inadequate to evaluate the spatiotemporal relationships between spontaneous Ca²⁺ sparks and agonist-induced [Ca²⁺]_i oscillations. In the present study, we used rapid real-time two-dimensional confocal imaging of Ca²⁺ sparks and [Ca²⁺]_i oscillations in TSM cells to determine the spatiotemporal relationships between these two phenomena.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell preparation. Porcine tracheae were obtained from a local abattoir, and TSM cells were isolated with previously described techniques (7). The dissociated cells were plated on collagen-coated glass coverslips and incubated for 1-2 h in a 5% CO₂ incubation chamber at 37°C. Based on trypan blue exclusion, cell viability of >90% was confirmed. A sample of cells dissociated from each animal was also processed with an anti-smooth muscle myosin antibody (Sigma Immunochemicals, St. Louis, MO) to estimate the relative proportion of smooth muscle myocytes (immunoreactive) and fibroblasts, which was found to be ~50:1.

Cells were incubated in 5 µM fluo 3-AM (Molecular Probes, Eugene, OR) for 30-45 min at 37°C. The cells were then washed in Hank's balanced salt solution (HBSS), and the coverslip was mounted on an open slide chamber (RC-25F, Warner Instruments, Hamden, CT). The tissue chamber was perfused at 2-3 ml/min at room temperature.

Real-time confocal imaging. Detailed descriptions of the real-time two-dimensional confocal-imaging technique have been recently published (18). Briefly, an Odyssey XL real-time confocal system (Noran Instruments, Middleton, WI) equipped with an Ar-Kr laser and mounted on a Nikon Diaphot microscope was used to visualize fluo 3-loaded TSM cells. Image size was set to 640 × 480 pixels, and the pixel area for a Nikon ×40, 1.3-numerical aperture oil-immersion objective lens was calibrated with a stage micrometer (0.06 µm²/pixel). A fixed combination of laser intensity and photomultiplier gain was set to ensure that pixel intensities within regions of interest (ROIs) ranged between 25 and 255 gray levels. To calibrate [Ca²⁺]_i, cells were exposed to 10 µM A-23187, a Ca²⁺ ionophore, at varying levels of extracellular Ca²⁺ ranging from 0 (HBSS with EGTA) to 10 µM, and fluorescence intensities were measured. The relationship between fluorescence intensity and [Ca²⁺]_i was found to be linear from 10 nM to 10 µM.

Images of fluo 3-loaded TSM cells were acquired at sampling frequencies ranging between 15 and 480 frames/s to evaluate the extent of frequency aliasing of the dynamic [Ca²⁺]_i response. We found that, in TSM cells, acquisition rates of 120 frames/s for Ca²⁺ sparks and 30 frames/s for [Ca²⁺]_i oscillations were sufficient to measure various parameters with adequate resolution and without frequency aliasing.

ROIs with a fixed dimension of 5 × 5 pixels (1.5 µm²) were defined within the boundaries of individual cells. The optical section thickness for confocal measurements was set to 1 µm by controlling the slit size on the Odyssey system. This corresponded to the optimal sectioning capability of the ×40 lens as determined in a previous study (19). The focus was adjusted such that measurements were obtained in a plane through the maximum thickness of a cell as far as possible. Overall, [Ca²⁺]_i measurements were obtained from a volume of 1.5 µm³. Each ROI represented 0.05-0.10% of the volume of a TSM cell. To determine intracellular heterogeneity in the dynamic [Ca²⁺]_i regulation, up to 8 ROIs were defined. The distances between these ROIs within a cell were measured with the length calibration for the ×40 lens. On-line [Ca²⁺]_i measurements were made with the Odyssey system for acquisition rates of 30 frames/s, whereas [Ca²⁺]_i measurements at higher acquisition rates were made post hoc from acquired images with an image-processing software package [ANALYZE, Mayo Biomedical Imaging Resource (20)].

In experiments where the spatial distribution of Ca²⁺ during sparks was determined, a hardware zoom of ×3 or ×4 was used such that the scanning dimensions were decreased by the same factor, but the image size was maintained. Obviously, this was likely to result in greater dye bleaching. Therefore, acquisitions were limited to ~1 min, at which time the extent of dye bleaching was estimated to be <5%. Images were acquired at 120 frames/s and processed to delineate Ca²⁺ distribution. The centroid of the distribution was then calculated as an index of the "origin" of the spark within the confocal plane with ANALYZE.

Characterization of [Ca²⁺]_i transients. The amplitude of Ca²⁺ sparks and oscillations was defined as the difference between the peak of the transient and the basal level of [Ca²⁺]_i. Rise time was normalized for amplitude, whereas fall time was normalized for the difference between the peak of the response and the basal [Ca²⁺]_i level at the end of the transient. The incidence of Ca²⁺ sparks was measured over 1-min intervals. The frequency of the [Ca²⁺]_i oscillations was measured as the inverse of the peak-to-peak interval between oscillations.

Effect of Ca²⁺ influx on Ca²⁺ sparks. To determine whether Ca²⁺ sparks are dependent on Ca²⁺ influx, TSM cells were exposed to nominally Ca²⁺-free HBSS, and changes in the incidence and amplitude of Ca²⁺ sparks were determined over a 15-min period. Ca²⁺ at 2.5 mM was then reintroduced into the extracellular medium, and the incidence and amplitude of sparks were reevaluated. As mentioned in Real-time confocal imaging, a potential confounding factor in these experiments was dye bleaching due to continued laser exposure. Therefore, images were acquired at 1-min intervals for 15-30 s.

Effect of ryanodine and caffeine on Ca²⁺ sparks. To determine whether Ca²⁺ sparks arise from SR Ca²⁺ release through RyR channels, TSM cells were exposed to 0.1, 1, and 10 µM ryanodine, and the changes in various parameters of Ca²⁺ sparks were evaluated. In a second set of experiments, TSM cells were exposed to 1, 10, and 50 µM caffeine, and the changes in various parameters of Ca²⁺ sparks were evaluated.

ACh-induced [Ca²⁺]_i oscillations. After evaluation of Ca²⁺ sparks, TSM cells were exposed to 1 µM ACh to induce [Ca²⁺]_i oscillations. A previous study (22) in porcine TSM has shown that the ACh concentration at which the response is 50% of maximum for the [Ca²⁺]_i response is ~1 µM. In a second set of experiments, after evaluation of Ca²⁺ sparks and [Ca²⁺]_i oscillations, TSM cells were washed in HBSS for 15 min, and the incidence and amplitude of Ca²⁺ sparks were reevaluated.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ca²⁺ sparks in TSM cells. Ca²⁺ sparks were observed in 35 TSM cells that were analyzed shortly (<2 h) after being plated on coverslips and in 22 cells >15 h after being plated. These cells represented 75-80% of the total number of cells studied.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Confocal image of a porcine tracheal smooth muscle (TSM) cell (A) loaded with fluorescent Ca2+ indicator fluo 3, displaying a region of Ca2+ sparks (box). Region in box has been magnified in B to display spatial and temporal aspects of Ca2+ sparks (red and yellow). Nos. in boxes, time in ms. Note short duration of sparks and quiescent periods between subsequent sparks. Bar, 0.5 µm.

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Image manipulation of Ca2+ spark data. XY images obtained from confocal microscope were stacked to yield a time series. Set of images was then viewed in Y vs. time plane at any given X. Line profiles of Ca2+ sparks in this new orientation provided data on time-dependent variation in spatial aspects and intensity of spark. However, unlike line scans obtained in other studies, line profiles could be simultaneously obtained from several regions within spark. Note variation in amplitude and apparent rate of occurrence in different line profiles (lines 1 and 2), suggesting that area of a spark is not fixed.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Ca2+ sparks in different regions of interest (ROIs) of a single porcine TSM cell. Incidence of sparks in adjacent regions was correlated but was not correlated across larger distances. There was considerable interregional variation in spark amplitude and occurrence.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Fusion of Ca2+ sparks. Ca2+ sparks often appeared in bursts of 2-6 sparks followed by periods of quiescence. On some occasions, sparks summated into a large [Ca2+]i response followed by a return to baseline and a continuation of spark activity.

View larger version (68K): [in this window] [in a new window]   Fig. 5.   Summation of Ca2+ sparks within an ROI of a TSM cell. Individual sparks (red) occurring with <30-ms delay resulted in a summation that yielded a greater elevation in intracellular Ca2+ concentration ([Ca2+]i; yellow). Nos. in boxes, time in ms. On some occasions, this summation triggered a propagating [Ca2+]i wave (data not shown). Bar, 0.5 µm.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Range of Ca2+ spark amplitudes. Histogram analysis of range of spark amplitudes within an ROI (summarized across 10 cells) revealed a multimodal distribution, typically with 3 peaks corresponding to approximate multiples of 1st peak.

Effect of Ca²⁺ influx on Ca²⁺ sparks. In 25 TSM cells, the incidence of Ca²⁺ sparks was initially unaffected by inhibition of Ca²⁺ influx either by exposing cells to nominally free extracellular Ca²⁺ (Fig. 7) or by blocking Ca²⁺ influx through voltage-dependent Ca²⁺ channels with 100 nM nifedipine. However, after ~5 min, Ca²⁺ sparks disappeared in the absence of Ca²⁺ influx, possibly as a result of a depletion of SR Ca²⁺ stores.

View larger version (46K): [in this window] [in a new window]   Fig. 7.   Role of extracellular and intracellular Ca2+ on Ca2+ sparks. Ca2+ sparks persisted for several minutes after removal of extracellular Ca2+, suggesting that sparks arise from intracellular stores (A). However, with continued exposure to Ca2+-free extracellular medium, sparks eventually disappeared after 10-15 min. During ongoing spark activity in presence of extracellular Ca2+, exposure to low concentrations of ryanodine resulted in increased spark activity (B). High ryanodine concentrations inhibited spark activity (data not shown).

Effect of ryanodine and caffeine on Ca²⁺ sparks. In response to 0.1 µM ryanodine, the incidence of Ca²⁺ sparks in 20 TSM cells increased (443 ± 88% of control value; P < 0.05; Fig. 8) and so did the amplitude (529 ± 67% of control value; P < 0.05). The fusion of individual Ca²⁺ sparks precluded a more rigorous analysis. In contrast, exposure to 1 µM ryanodine did not change either the frequency or amplitude (94 ± 26 and 101 ± 66%, respectively, of control values). However, exposure to 10 µM ryanodine induced an [Ca²⁺]_i transient and inhibited spark activity. Exposure to 1, 10, and 50 µM caffeine all increased the incidence of Ca²⁺ sparks in 15 TSM cells (135 ± 5, 288 ± 23, and 370 ± 26%, respectively, of control values; P < 0.05). The amplitude of the sparks was significantly changed with 50 µM caffeine (385% of control value; P < 0.05).

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Effect of ryanodine on Ca2+ sparks. In response to 0.1 µM ryanodine, incidence of Ca2+ sparks as well as amplitude increased compared with control values. In contrast, exposure to 1 µM ryanodine did not change either frequency or amplitude (compared with control value). * Significant difference from control value, P < 0.05.

ACh-induced [Ca²⁺]_i oscillations. The spatial and temporal patterns of ACh-induced [Ca²⁺]_i oscillations were studied in 35 TSM cells where multiple foci of Ca²⁺ sparks were observed. Within each oscillation, the first detectable change in [Ca²⁺]_i always occurred at a site where spontaneous Ca²⁺ sparks had been previously recorded. In >80% of these cells, the intracellular site that previously displayed the highest incidence of Ca²⁺ sparks was also the origin of the first significant change in [Ca²⁺]_i above basal levels on ACh exposure (Fig. 9). After the rise in [Ca²⁺]_i at this initiation site, the oscillation spread toward other parts of the cell. In some cases, two [Ca²⁺]_i waves were initiated, typically from the long ends of the cell, and propagated independently toward the center of the cell. In these instances, the origins of the two [Ca²⁺]_i waves were also sites where a high incidence of Ca²⁺ sparks had been previously observed.

View larger version (20K): [in this window] [in a new window]   Fig. 9.   Relationship between Ca2+ sparks and ACh-induced [Ca2+]i oscillations. In a cell with ongoing Ca2+ sparks, exposure to ACh resulted in initiation of [Ca2+]i oscillations. Site of initiation of [Ca2+]i oscillations was intracellular focus with highest spark incidence.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Real-time confocal imaging was used to examine the spatial and temporal relationships of Ca²⁺ sparks to propagating [Ca²⁺]_i oscillations in porcine TSM cells. Ca²⁺ sparks displayed relatively constant rise times and amplitudes within a focus of a TSM cell, but across different foci, these attributes displayed considerable variability. Individual sparks were often grouped together and coupled across adjacent regions. Fusion of individual sparks produced large local elevations in [Ca²⁺]_i that triggered a propagating [Ca²⁺]_i wave in an unstimulated cell. The incidence of sparks was increased by both ryanodine and caffeine but was largely unaffected by removal of extracellular Ca²⁺. These data suggest that Ca²⁺ sparks in TSM cells represent SR Ca²⁺ release through RyR channels. Exposure to ACh triggered repetitive, propagating [Ca²⁺]_i oscillations that originated from foci with a high incidence of Ca²⁺ sparks. The [Ca²⁺]_i oscillations disappeared with removal of ACh, but Ca²⁺ sparks reappeared. These results indicate that agonist-induced [Ca²⁺]_i oscillations represent a spatiotemporal integration of local Ca²⁺-release events through RyR channels in TSM cells.

In the present study, we used rapid confocal imaging of Ca²⁺ sparks in TSM cells. Although the temporal resolution of this imaging technique was comparable to that used with line scans in previous studies (8, 10, 14-16, 25), a distinct advantage was that in addition to the temporal aspect, two-dimensional information on Ca²⁺ distribution within the area of the spark could also be obtained. With this feature, we found that the area occupied by a spark varied from event to event within a sparking region. This observation would be entirely missed by a line scan where the shifting of the spark area would only appear as a change in the apparent amplitude of the spark. These findings are also of significance with regard to the proposed mechanisms underlying Ca²⁺ sparks. Previous observations in skeletal muscle fibers, cardiac myocytes, and vascular smooth muscle cells (1, 5, 8, 15, 25) led to suggestions that Ca²⁺ sparks represent elemental or unitary Ca²⁺ release through RyR channels (5, 13, 15, 25). The fact that both ryanodine and caffeine modulated the incidence and amplitude of Ca²⁺ transients in TSM cells also indicates the involvement of RyR channels. The concept of Ca²⁺ sparks representing elemental Ca²⁺ release was also generally supported by the relatively constant amplitude and rise time of individual Ca²⁺ sparks in porcine TSM cells. More importantly, the modes of the spark amplitude distribution were found to be multiples of a basic amplitude, resembling quantal neurotransmitter release at neuromuscular junctions. Accordingly, these data suggest that individual sparks represent all-or-none SR Ca²⁺ release that can occasionally fuse into larger events.

If Ca²⁺ sparks are elemental units of [Ca²⁺]_i regulation, it may be expected that the spatiotemporal patterns of Ca²⁺ sparks in different tissues reflect the kinetics of Ca²⁺ regulatory mechanisms such as release, reuptake, and passive diffusion. In a previous study on vascular smooth muscle cells, Nelson et al. (15) reported Ca²⁺ sparks that displayed rise and fall times of the same order of magnitude as those observed in the present study on TSM cells. Cannell et al. (3) reported that Ca²⁺ sparks in cardiac myocytes lasted for 100-200 ms. In a separate study, Prakash et al. (19) recently observed Ca²⁺ sparks in rat cardiac myocytes, where the amplitude of sparks was comparable to that observed in TSM cells, but the rise and fall times were considerably shorter, with 10- to 35-ms rise times and 60- to 100-ms fall times. These data suggest that the differences in temporal aspects of Ca²⁺ sparks between tissues most likely reflect the kinetics of the release and reuptake mechanisms. In this regard, the shifting in the area occupied by the spark may reflect heterogeneities in the kinetics of RyR channels within a group of channels that contribute to a spark.

Multiple foci for Ca²⁺ sparks were frequently observed in individual TSM cells. Adjacent regions of Ca²⁺ sparking were often coupled, whereas more distant regions were not. This observation suggests that localized SR Ca²⁺ release may induce Ca²⁺ release from surrounding regions, perhaps via CICR. Indeed, we frequently observed groups of three to four individual Ca²⁺ sparks separated by periods of quiescence. These events may represent localized facilitation of sparking from different groups of RyR channels. The spatial limitation of foci may be due to SR Ca²⁺ reuptake acting as a barrier to the initiation of a propagating [Ca²⁺]_i oscillation from the region of sparking (5). In other cases, we observed larger [Ca²⁺]_i responses, with individual sparks superimposed on both the rising and falling phases of the larger response. Similar events have been observed previously in cardiac myocytes (5). These events may represent facilitation of Ca²⁺ release from a larger SR store, most likely via CICR.

In the present study, we observed that, in many TSM cells, regions of increased incidence of Ca²⁺ sparks corresponded with the site of initiation of propagating ACh-induced [Ca²⁺]_i oscillations. These regions also displayed spontaneous summation of individual sparks, leading to larger [Ca²⁺]_i transients. In most cases, the amplitudes of the spontaneous, summated responses were comparable, if not identical, to those of ACh-induced responses. Therefore, our results suggest that Ca²⁺ sparks in TSM cells may arise from "trigger" sites that reflect areas of high RyR-channel density, as suggested by Lesh et al. (11), in vascular smooth muscle and/or sensitivity and act as "primers" for agonist stimulation. However, other potential mechanisms cannot be ruled out. For example, in previous studies (6, 23), we used a -escin-skinned TSM cell preparation to demonstrate that ACh-induced [Ca²⁺]_i oscillations require inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃]-induced SR Ca²⁺ release at least for initiation, although the steady-state phase is not affected by inhibitors such as heparin. Accordingly, the site of initiation may also reflect higher density and/or sensitivity of muscarinic receptors and Ins(1,4,5)P₃-receptor channels in the SR. Furthermore, the distribution of different receptors and channels is likely to play a role in the interaction between Ins(1,4,5)P₃-induced SR Ca²⁺ release and oscillations through RyR channels. These issues need to be examined in future studies.

A study (15) in vascular smooth muscle has suggested that Ca²⁺ sparks may be important in the control of resting membrane potential via Ca²⁺-activated K⁺ channels. In this scenario, increased spark incidence would actually lead to a relaxation of smooth muscle as a result of membrane hyperpolarization. Accordingly, it may be expected that regions with a higher incidence of Ca²⁺ sparks would play a greater role in hyperpolarization of the membrane, making the cell less responsive to agonist stimulation. Whether sparks locally regulate membrane potential in TSM cells, as has been shown in vascular smooth muscle, is not known. In the present study, we did not attempt to determine the location of Ca²⁺ sparks relative to the plasma membrane because the optical section thickness of 1 µm and the XY resolution of 0.25 µm were suboptimal. Nonetheless, a previous study (9) in airway smooth muscle has indicated that Ca²⁺-activated K⁺ channels do not significantly contribute to membrane potential. In this regard, the functional significance of Ca²⁺ sparks may vary between airway and vascular smooth muscle.

In conclusion, the results of the present study support a hypothesis that Ca²⁺ sparks represent Ca²⁺-release events from finite SR Ca²⁺ pools via RyR channels. Ca²⁺ sparks may arise from regions with RyR channels of high sensitivity or high density, which also serve as initiation sites for agonist-induced [Ca²⁺]_i oscillations.