|Home | About | Journals | Submit | Contact Us | Français|
In this article, we review the biophysical basis and functional implications of a novel Ca2+ signal (called “Ca2+ sparklets”) produced by Ca2+ influx via L-type Ca2+ channels (LTCCs) in smooth muscle. Ca2+ sparklet activity is bimodal. In low activity mode, Ca2+ sparklets are produced by random, brief openings of solitary LTCCs. In contrast, small clusters of LTCCs can function in a high activity mode that create sites of continual Ca2+ influx called “persistent Ca2+ sparklets”. Low activity and persistent Ca2+ sparklets contribute to Ca2+ influx in arterial, colonic, and venous smooth muscle. Targeting of PKCα by the scaffolding protein AKAP150 to specific sarcolemmal domains is required for the activation of persistent Ca2+ sparklets. Calcineurin, which is also associated with AKAP150, opposes the actions of PKCα on Ca2+ sparklets. At hyperpolarized potentials, Ca2+ sparklet activity is low and hence does not contribute to global [Ca2+]i. Membrane depolarization increases low and persistent Ca2+ sparklet activity, thereby increasing local and global [Ca2+]i. Ca2+ sparklet activity is increased in arterial myocytes during hypertension, thus increasing Ca2+ influx and activating the transcription factor NFATc3. We discuss a model for subcellular variations in Ca2+ sparklet activity and their role in the regulation of excitation-contraction coupling and excitation-transcription coupling in smooth muscle.
L-type Ca2+ channels (LTCCs) play a critical role in the regulation of excitability, contraction, and gene expression in smooth muscle [1-6]. Ca2+ influx through these channels produce subcellular Ca2+ signals called “Ca2+ sparklets” [7, 8]. In this review, we examine strategies for the recording, detection, and analysis of Ca2+ sparklets. Furthermore, we discuss recent work indicating that Ca2+ sparklet activity varies within the sarcolemma of smooth muscle cells and examine the potential mechanisms for these regional variations in LTCC activity. We conclude the review with a model that incorporates these findings and evaluate their implications on the regulation of local and global [Ca2+]i as well as gene expression in smooth muscle under physiological and pathophysiological conditions.
First, we would like to discuss some technical considerations related to the recording of Ca2+ sparklets. This is important because optical recording of Ca2+ sparklets and related Ca2+ influx events is a relatively new approach with great potential, but with associated complexities that are worth discussing for those interested in using these approaches. Due to space limitations, our discussion will be brief. Readers interested in a more extensive discussion of these issues are encouraged to read other papers by our group [8-10] and excellent articles by Ian Parker's group [11, 12].
Recent advances in the imaging field have allowed investigators to record relatively small Ca2+ signals from single sarcolemmal Ca2+ channels with unprecedented speed and spatial resolution [9, 11]. One particular example is the development of electron multiplying charged-coupled device (EMCCD) cameras. These cameras have quantum efficiencies >90% and are thus capable of detecting photons emitted from a single fluorescence molecule. EMCCD cameras can be used to acquire ultra fast (e.g. 100-500 Hz) 2-dimensional images of dim fluorescence events. This is particularly attractive for investigators interested in imaging rare and low amplitude Ca2+ signals like Ca2+ sparklets, as one could image relatively large areas of the cell (compared to a confocal line-scan imaging) at relatively fast acquisition rates and with high sensitivity. Indeed, EMCCD camera-based imaging systems — confocal and evanescent field total internal reflection fluorescence (TIRF) microscope — have recently been used to image Ca2+ influx events via single Ca2+ channels in multiple cell types [7, 8, 11, 13-16]. Figure 1A shows images of Ca2+ sparklets in arterial and portal vein smooth muscle obtained with an EMCCD camera (Navedo and Santana, unpublished work). Ca2+ sparklets have also been recorded in colon smooth muscle using TIRF microscopy .
To record Ca2+ sparklets, we use the following experimental conditions [8-10]. We begin experiments by allowing acutely dissociated smooth muscle cells to settle and adhere to the coverslip of a perfusion chamber. This is especially critical for those experiments involving TIRF microscopy in which one needs a relatively large portion of the cell within the evanescent field created during TIRF in order to increase the probability of observing Ca2+ sparklet sites. We strongly recommend that cells are patch-clamped in the whole-cell configuration so that membrane potential, and hence the open probability of LTCCs and the driving force for Ca2+ influx are controlled. Using the whole-cell patch-clamp technique in its conventional configuration has an additional advantage: it allows the introduction into the cell of a solution containing the relatively fast fluorescent Ca2+ indicator Fluo-5F (200 μM; penta-potassium salt) and the slower and non-fluorescent Ca2+ buffer EGTA (10 mM).
We use this combination of Fluo-5F and EGTA to record Ca2+ sparklets for two important reasons. First, Ca2+ influx events produce a bright fluorescence signal that is confined to the volume near the site of entry . This happens because as Ca2+ enters the cell it preferentially binds the fast Fluo-5F molecules, thus increasing their fluorescence. However, this fluorescence is short-lived, as Ca2+ dissociates from Fluo-5F and binds to the more abundant and non-fluorescence EGTA. Using this combination of Fluo-5F and EGTA, we have found that the average area of a Ca2+ sparklet was small: ≈0.8 μm2. Second, by using high concentrations of EGTA, global [Ca2+]i is maintained relatively low, thus decreasing background fluorescence. This, in combination with the thin optical section (≈100 nm with a 1.45 NA lens and 488 nm excitation light ) of the evanescent field during TIRF microscopy minimizes background fluorescence, thus increasing the signal-to-noise ratio of this system and hence one's ability to detect small amplitude Ca2+ signals.
A limitation of TIRF microscopy is that imaging is restricted to the region of the cell within the fixed evanescent field that decays exponentially from the coverslip. The use of a confocal microscope could overcome this limitation. Note, however, that the thicker optical section of the confocal microscope — typically about 1 μm in the z-axis (≈10-fold thicker than in TIRF) — will increase background fluorescence, potentially limiting detection of small Ca2+ sparklets events.
Although we , and others , have been able to record Ca2+ sparklets in the absence of an exogenous Ca2+ buffer such as EGTA in smooth muscle cells, at least in our hands, recording small Ca2+ sparklet events under these experimental conditions has been very difficult. Thus, whenever possible, we recommend that the combination of Fluo-5F and EGTA is used to record Ca2+ sparklets.
Molecular identification of ion channels is based on pharmacological, biophysical, and molecular biological criteria. Thus, Ca2+ sparklets meet all the generally accepted pharmacological, biophysical, and molecular biological criteria used to identify LTCCs. Ca2+ sparklets are activated by the dihydropyridine agonist Bay-K 8644 and are eliminated by the dihydropyridine antagonists nifedipine and nisoldipine [10, 14, 15]. Simultaneous recordings of Ca2+ signals and LTCC currents indicate that Ca2+ sparklets are associated with an inward Ca2+ current (Figure 1B). Importantly, Ca2+ sparklets have similar voltage dependencies of activity and amplitude as LTCCs. Furthermore, Ca2+ sparklets in tsA-201 cells expressing L-type Cav1.2 channels reproduce all the basic features of native Ca2+ sparklets including block by nifedipine, activation by Bay-K 8644, bimodal gating modalities, amplitude of quantal event and voltage dependencies (Table 1) . Finally, arterial myocytes expressing a mutant Cav1.2 channel that is insensitive to inhibition by dihydropyridines [19, 20], produced dihydropyridine-insensitive low activity and persistent Ca2+ sparklets . This finding is important because it eliminates the possibility that Ca2+ sparklets are produced by a TRP or store-operated channel in arterial myocytes.
As Heping Cheng's group demonstrated for Ca2+ sparks , manual detection of small amplitude Ca2+ signals such as Ca2+ sparks and Ca2+ sparklets is difficult and prone to bias. Thus, we developed an algorithm for the automatic detection of Ca2+ sparklets [8, 9]. The software, written in IDL language, is available to any investigator interested in studying Ca2+ sparklets and can be requested via email.
Briefly, the program uses spatial, temporal, and amplitude criteria for the detection of Ca2+ sparklet sites and events. The program has an extremely low (< 5%) false-positive rate of Ca2+ sparklet detection. We highly recommend that this or similar automatic detection methods [21-24] be used over a visual detection approach to facilitate the identification and later classification of Ca2+ sparklet sites. Not using an automatic detection algorithm may lead to underestimation of Ca2+ sparklet events and to human bias towards rare, high amplitude, and high activity events .
Using the imaging and analytical approaches described above, we recorded Ca2+ sparklets in the presence of 2 mM external Ca2+ at -70 mV and physiological -40 mV . Figure 2A show an all-points histogram generated from representative blank TIRF [Ca2+]i records. This histogram shows a large peak at a change in [Ca2+] of 0, which represents the mean basal [Ca2+]i. This peak could be fitted with a Gaussian function with a mean of 0 and standard deviation (SD) of 6.1 nM. The width of this peak is useful because it provides an indication of the “noise” of our system, which we used to set a threshold of mean basal of Δ[Ca2+]i (i.e. 0) + 3 * SD (i.e. 18 nM) for the detection of Ca2+ sparklets. Readers will appreciate that the width of this peak, and hence the amplitude detection threshold, would critically depend on experimental conditions. For example, conditions that increase background fluorescence (e.g. absence of intracellular EGTA or epifluorescence imaging) would increase the baseline SD, therefore decreasing one's ability to resolve small amplitude Ca2+ sparklets.
At -70 mV and with 2 mM external Ca2+, Ca2+ sparklet amplitude range from 18 to 280 nM . Note, however, that under these experimental conditions the vast majority of these events had amplitudes close to or at the amplitude detection threshold of 18 nM, suggesting frequent sub-threshold events. This is the reason why we decided to increase the driving force for Ca2+ influx by increasing external Ca2+ from 2 to 20 mM. In Figure 2B, we show an all-points histogram generated from representative Ca2+ sparklet records obtained using 20 mM external Ca2+. The histogram could be fit with a multi-component Gaussian function with a quantal unit of Ca2+ influx of ≈35 nM. Furthermore, note that the amplitude of the various peaks of the histogram progressively decreases with amplitude, suggesting that Ca2+ entry through Ca2+ sparklet sites is quantal in nature and that multi-quantal events “likely arise from random overlapping openings of adjacent” LTCCs .
Two observations suggest that quantal Ca2+ sparklets likely result from the opening of a single LTCC. First, simultaneous recordings of Ca2+ signals and membrane currents indicate that quantal Ca2+ sparklets are associated with elementary LTCC currents (Figure 1B) . Second, consistent with these imaging and electrophysiological data, analysis of the signal mass (i.e. total Ca2+ charge in coulombs) of Ca2+ sparklets suggest that, at least in principle, these signals could be produced by the opening of a single LTCC. Briefly, the smallest and briefest Ca2+ sparklets detected at -40 mV with 2 mM external Ca2+ had a signal mass of 7 fC (Table 1) . This signal mass could be produced by a single LTCC channel producing a -0.2 pA current  with a duration of 35 ms: 0.2 × 10-12 C/s * 0.035 s = 7 × 10-15 C.
An interesting feature of Ca2+ sparklets is that their activity is heterogeneous [8, 9]. Tour et al.  and McCarron et al.  recently reported similar findings in Hela cells expressing Cav1.2 channels and colonic smooth muscle, respectively. This has important functional implications, but also poses analytical challenges. Let us deal with the later issue first. We developed a method for the quantitative analysis of sparklet activity. Two important features of Ca2+ sparklets helped us develop an analytical framework for the quantification of Ca2+ sparklet activity: (a) Ca2+ sparklet activity is quantal and (b) under TIRF and with Fluo-5F/EGTA these records resemble single channel recordings. Thus, analogous to single channel analysis we decided to quantify Ca2+ sparklet activity as nPs, where n is the number of quantal levels and Ps is the probability of Ca2+ sparklet occurrence. Details of this analysis could be found elsewhere . An example of this type of analysis is shown in Figure 2C. Briefly, the first step is to determine the quantal unit of Ca2+ influx using an approach similar to the one developed by del Castillo and Katz  for neurotransmitter release. To estimate nPs, we import Ca2+ sparklet records into Clampex 9 software. Individual Ca2+ sparklet events can be detected using this program's “threshold detection analysis” using no duration constraints and the quantal unit of Ca2+ entry determined above as a starting point for event detection. Traces can then be fitted (green lines, Figure 2C) with these initial parameters.
This analysis revealed that Ca2+ sparklet activity was bimodal [8, 9]. Thus, Ca2+ sparklets were grouped three categories. By default, silent sites had no activity (nPs = 0). Note that a silent Ca2+ sparklet site represents a site that is typically inactive, but can be activated by an agonist. Second, low activity Ca2+ sparklet sites had a relatively low probability of activation (nPs between 0.0-0.2). Third, high activity, persistent Ca2+ sparklet sites (nPs > 0.2) were characterized by frequent LTCC openings. Examination of the amplitude and duration histograms of Ca2+ sparklet sites suggest that low activity sparklets are produced by rare, brief openings of single LTCCs (i.e. single mode amplitude histogram) while persistent Ca2+ sparklets are produced by frequent, longer openings of a small clusters of LTCCs (i.e. multi modal amplitude histogram) [8, 9].
Ca2+ sparklet amplitude and duration is highly variable . Thus, the nPs analysis provides a better indicator of Ca2+ influx than amplitude and duration plots alone. We suggest that amplitude and duration histograms are accompanied by nPs data for each site analyzed. An attractive feature of calculating the nPs value of a sparklet site is that it is proportional to the nPo (Po = open probability) of the underlying Ca2+ channel(s). L-type Ca2+ channels in mode 1 and mode 2 have mean open times of about 2 and 11 ms, respectively  (Table 1). Thus, if as proposed , LTCCs in persistent Ca2+ sparklet sites are in mode 2 and have an open time ≥11 ms, the nPs values of these sites at acquisition rates >90 Hz should approximate the nPo of the channels in these sites. Because the mean open time of LTCCs in mode 1 is ≈2 ms, acquisition rates of 1 KHz or higher are required to convert their nPs to nPo values.
Quantitative approaches and automatic detection algorithms like the one described above are absolutely necessary for the detection, analysis, and classification of Ca2+ sparklet activity. Indeed, in our opinion, in the absence of an nPs or similar quantitative approach to determine Ca2+ sparklet activity, conclusions about the frequency or type of Ca2+ sparklet site in any particular cell are difficult to interpret  as it may lead to biased classification of Ca2+ sparklet events.
The finding that sparklet activity varies within cells is interesting because Cav1.2 channels are broadly distributed throughout the sarcolemma of smooth muscle [8, 25, 28]. Indeed, we performed a Chi square (χ2) test to test the null hypothesis that the spatial distribution of Ca2+ sparklet sites followed a Poisson distribution (i.e. was random) . Although the specific location of Ca2+ sparklets differed between cells, the χ2 tests indicated there were a significant difference between the spatial distribution of Ca2+ sparklets and the predicted Poisson distribution in each cell. This indicates that the spatial distribution of Ca2+ sparklets within a cell is not random.
Work by McCarron et al.  gives credence to this view. Although, unlike Navedo et al. , they did not perform a statistical test for randomness of the spatial distribution of Ca2+ sparklet sites, they found that in colon smooth muscle membrane depolarization activated what seems like a persistent Ca2+ sparklet (whether or not this is a persistent sparklet site is unclear since the authors did not quantify the activity of this or other sites) within a small region of the sarcolemma (≈1 μm), while the rest of the analyzed regions of the membrane remained optically silent. This suggests that, as in arterial myocytes, Ca2+ sparklet activity in colon smooth muscle cells is not uniform throughout their sarcolemma. Thus, analysis of the gating and spatial distribution of Ca2+ sparklets leads to two fundamentally different conclusions regarding these local Ca2+ signals that should not be confused: while quantal analysis of Ca2+ sparklets suggests that their gating is predominantly stochastic, independent (i.e. multi-modal amplitude histogram with a progressively decreasing probability of multi-quantal sparklet occurrence); the spatial distribution of Ca2+ sparklet activity is not random (i.e. non-Poisson distribution), being higher at specific sites within the sarcolemma of smooth muscle cells .
It is important to note that although the analysis described above suggests that Ca2+ sparklet gating is predominantly stochastic, the possibility that small clusters of LTCCs in persistent sparklet sites could gate in a coordinated, non-stochastic manner cannot be completely ruled out. Future studies should examine the whether at least a fraction of persistent Ca2+ sparks are produced by coupled gating of small clusters of adjacent LTCCs.
What are the mechanisms underlying regional variations in Ca2+ sparklet activity? The diagram in Figure 3 summarizes a model for heterogeneous Ca2+ sparklet function in smooth muscle. It proposes that regional variations in Ca2+ sparklet activity are not necessarily due to clustering of functional Cav1.2 channels, as immunofluorescence, electrophysiological, and imaging data suggest that these channels are broadly distributed throughout the sarcolemma of smooth muscle cells [8, 25, 28]. Instead, membrane targeting of PKCα by the scaffolding protein AKAP150 (the rodent ortholog of human AKAP79) to specific regions of the surface membrane likely underlies heterogeneous Ca2+ sparklet activity in these cells . Local targeting of PKCα by AKAP150 to specific regions of the surface membrane likely underlies heterogeneous Ca2+ sparklet activity in these cells . Local targeting of PKCα by AKAP150 to specific regions of the surface membrane allows this kinase to phosphorylate nearby LTCCs thus promoting persistent Ca2+ sparklet activity. Two important observations support this view . First, AKAP150 and PKCα co-localize in specific regions at or near the sarcolemma of arterial myocytes. Second, loss of AKAP150 prevents PKCα targeting to the sarcolemma of arterial myocytes and thus induction of persistent Ca2+ sparklet activity.
AKAP150 also targets calcineurin to the surface membrane . In our model calcineurin and PKCα have opposing actions on Ca2+ sparklets and the level of Ca2+ sparklet activity varies regionally depending on the relative activities of PKCα and calcineurin . Recent work by John Scott's lab suggests a potential mechanism for local Ca2+-dependent activation of PKCα . They found that as Ca2+ enters the cell it recruits Ca2+/calmodulin to AKAP150 . This releases PKCα from the AKAP150 complex allowing it to phosphorylates nearby targets including LTCCs. Because in this model PKCα is associated with AKAP150 in the sarcolemma of arterial myocytes and hence readily available for activation by a nearby local increase in Ca2+, a global increase in [Ca2+]i is not necessary to induce the translocation of this kinase to the membrane.
LTCCs are critical regulators of [Ca2+]i in smooth muscle [4, 5]. It is therefore important to consider the impact of Ca2+ sparklets on [Ca2+]i in these cells. We will focus on whether Ca2+ sparklets influence global [Ca2+]i as their impact on local [Ca2+]i is evident. Determining the relative contribution of low activity and persistent Ca2+ sparklets to global Ca2+ would be relatively easy if a specific blocker of each one of these events were available. There are none.
LTCCs antagonists block low activity and persistent Ca2+ sparklets. Thus, a multi-pronged approach must be used to determine the contribution of low activity and persistent Ca2+ sparklets to global [Ca2+]i. As a first step, we determined the signal mass associated with Ca2+ sparklets (Figure 4). The Walsh's  and Singer's  groups developed a method for estimating the signal mass of local Ca2+ signals in coulombs from 2d images. The strength of this approach is that one could attribute signal mass to specific Ca2+ sparklet sites from which we have previously determined their nPs (i.e. activity).
The signal mass of low and high nPs Ca2+ sparklets was measured at −70 mV and physiological -40 mV using 2 mM external Ca2+ (Figure 4) . The median signal mass of low and high nPs, persistent Ca2+ sparklets at −70 mV was 63 and 189 fC, respectively (Figure 4A and C). As expected for an LTCC, membrane depolarization increased the signal mass of Ca2+ sparklets. At -40 mV, the median signal mass of low and high nPs Ca2+ sparklets was 292 and 452 fC, respectively (Figure 4B and C). These signal mass values could be used to provide an indication of the potential impact of low and high activity Ca2+ sparklets on global [Ca2+]i using the following equation:
where ΔQCa is the change in calcium influx (in coulombs, C; as determined by the signal mass analysis), F is the Faraday constant, V is the accessible cytosolic volume, and B is the buffering capacity of the cell. Assuming that smooth muscle cells have an accessible cytosolic volume of 0.9 pl and a buffering capacity of 80, we determined that at −70 mV median low and high nPs Ca2+ sparklets could increase [Ca2+]i by as much as 5 and 14 nm, respectively . As expected, membrane depolarization increases Ca2+ sparklet activity [7, 10, 15]. At −40 mV, median low and high nPs sites could increase global [Ca2+]i by as much as 21 and 38 nm, respectively . Because the driving force for Ca2+ entry decreases with depolarization, this voltage-induced increase in Ca2+ entry is likely to be due to an increase in the activity of LTCCs. These data indicate that Ca2+ influx via low activity and persistent Ca2+ sparklets is voltage dependent and could influence local and global [Ca2+]i at physiological -40 mV and 2 mM external Ca2+.
Do Ca2+ sparklets contribute to global [Ca2+]i at hyperpolarized membrane potentials (i.e. -70 mV)? Because low nPs Ca2+ sparklets are rare, the probability of coincidental activation of independent events is low . Thus, it is reasonable to predict that infrequent, random, non-overlapping Ca2+ sparklets with a signal mass similar to that of the mode (100 pC) or median of the distribution would have a small (< 5 nM) impact on global [Ca2+]i at -70 mV. However, because the signal mass of persistent Ca2+ sparklets at this potential is larger than that of low nPs events, the predicted impact of these events on global [Ca2+]i is potentially larger than the typical low nPs sparklet. Persistent Ca2+ sparklets with a signal mass similar to that of the mode (100 pC) or median of the distribution could have a relatively small (< 14 nM) impact on global [Ca2+]i at -70 mV.
To provide further support to this conclusion, we modeled the effects of LTCC inhibition on global Ca2+ measured in a cell loaded with Fluo-5F using epifluorescence or TIRF microscopy (Figure 5). As above, the model assumes 1 persistent Ca2+ sparklet contributing to a steady-state increase in [Ca2+]i of 14 nM Ca2+ (median signal mass) to [Ca2+]i at -70 mV . In the model, noise was simulated assuming a Gaussian distribution using experimentally determined values from arterial myocytes loaded with Fluo-5F and imaged in TIRF or epifluorescence mode using an 60× (1.45 NA) lens and an EMCCD camera. As expected, the standard deviation of the basal fluorescence signal was larger in epifluorescence (24 nM) than in TIRF (6 nM) imaging, due to the thicker optical section in epifluorescence than in TIRF mode. We generated all-points histograms from [Ca2+]i records undergoing a decrease of 14 nM [Ca2+]i in epifluorescence (Figure 5A) and TIRF mode (Figure 5B). Note that both histograms could be fitted with single Gaussian function, suggesting that a change in [Ca2+]i of <14 nM cannot be distinguished from baseline using Fluo-5F and epifluorescence or TIRF microscopy. These findings are consistent with the observation by McCarron et al.  that application of 1 μM nisoldipine did not change global [Ca2+]i in colonic smooth muscle at -70 mV. Importantly, these findings indicate that due to the relatively low frequency of persistent Ca2+ sparklets at hyperpolarized potentials, epifluorescence imaging cannot be used to determine their contribution to global [Ca2+]i.
As noted above, we demonstrated that depolarization from -70 to -40 mV increases the signal mass of low activity and persistent Ca2+ sparklets due to an increase in Ca2+ sparklet activity . This brings up the following question: What is the relative contribution of low and high nPs sites to global [Ca2+]i in smooth muscle at the physiological membrane potential of -40 mV? Answering this question is difficult due to the absence of specific blockers of low activity and persistent Ca2+ sparklets. Two recent studies in AKAP150  and PKCα null  myocytes provide insights on this issue.
Loss of AKAP150 and PKCα expression eliminated persistent but not low activity Ca2+ sparklets. Indeed, we found that the dihydropyridine-sensitive component of steady-state L-type Ca2+ current and the global [Ca2+]i at -40 mV was ≈50% lower in cells lacking AKAP150 or PKCα expression. On the basis of these findings and the signal mass data above, we proposed that at the membrane potential of -40 mV and with 2 mM external Ca2+, persistent Ca2+ sparklets (i.e. AKAP150 and PKCα-dependent events) contribute ≈50% of the dihydropyridine-sensitive component of Ca2+ influx and could therefore be involved in the activation of myosin light chain kinase during excitation-contraction coupling in smooth muscle.
Our Ca2+ sparklet model also provides insight into the mechanisms by which LTCC agonists like Bay-K 8644 increases Ca2+ influx into arterial myocytes . Note that application of Bay-K 8644 induced persistent Ca2+ sparklet activity in tsA-201 expressing Cav1.2 and PKCα. In contrast, tsA-201 cells co-expressing Cav1.2 channels only, Bay-K 8644 increases the frequency and duration of quantal Ca2+ sparklets, but did not induce persistent Ca2+ sparklet activity. Based on our model, we propose that application of Bay-K 8644 to arterial myocytes increases Ca2+ sparklet activity by increasing the mean open time and/or open probability of LTCCs. If any of these events occurs in the vicinity of an AKAP150-targeted, Ca2+-sensitive PKCα, it could activate this kinase. Once activated, PKCα can induce multiple adjacent LTCCs to operate in a persistent gating mode (i.e. mode 2), thus causing further Ca2+ influx, which could presumably maintain PKCα activity and hence persistent Ca2+ sparklet activity. Bay-K 8644 has little or no effect on previously active (i.e. during control conditions) persistent Ca2+ sparklet sites presumably because the majority of the LTCCs in these clusters were already in mode 2 before application of the drug. An interesting implication of our model is that while Ca2+ sparklets could be evoked by voltage or pharmacological means wherever Cav1.2 channels are expressed, persistent Ca2+ channel activity would only occur in regions of the cell membrane where Cav1.2, AKAP150, and PKCα are located.
LTCC activity is increased in arterial smooth muscle during the development of hypertension . Accordingly, the number and probability of activation of low activity and persistent Ca2+ sparklet sites — not an increase in the amplitude of quantal Ca2+ sparklets or their duration — underlies increased Ca2+ influx via LTCCs into arterial myocytes during hypertension. This is the result of an increase in Cav1.2 channel expression  and PKCα activity in arterial smooth muscle during hypertension .
The increase in persistent Ca2+ sparklet activity had important functional consequences in arterial myocytes during hypertension: they activate the transcription factor NFATc3 via calcineurin . Thus, the Ca2+ sparklet model described above has important consequences on Ca2+ influx and gene expression (Figure 3). Activation of PKCα induces persistent Ca2+ sparklet, producing an increase in local Ca2+ influx that activates nearby AKAP150-targeted calcineurin. Upon activation, calcineurin dephosphorylates NFATc3 allowing this transcription factor to translocate into the nucleus of arterial myocytes where it modulates gene expression (Figure 3). Calcineurin and NFATc3 activities are low in arterial smooth muscle under physiological conditions  because of low levels of persistent Ca2+ sparklet activity (≈1 persistent Ca2+ sparklet site per cell)  and relatively high nuclear NFATc3 export rates . However, during hypertension, an increase in PKCα activity increases persistent Ca2+ sparklet activity thereby increasing calcineurin activity and consequently NFATc3 nuclear import rate, which if coupled with a decrease in export rate , would lead to high nuclear accumulation of this transcription factor. This, in turn, leads to down-regulation of Kv2.1 transcript expression [2, 37]. As suggested previously [2, 38, 39], down-regulation of Kv2.1 decreases voltage-gated K+ currents and contributes to arterial smooth muscle depolarization during hypertension, which indirectly increases Ca2+ sparklet activity, global [Ca2+]i, and tone (Figure 3). Consistent with this model, loss of AKAP150 and PKCα not only eliminated persistent Ca2+ sparklet activity, but also protected against the development angiotensin II-induced hypertension .
The work discussed in this review suggests multiple strategies for the recording, detection, and later analysis of Ca2+ sparklets. Implementation of these approaches has revealed an interesting feature of LTCCs: their opening probability can vary regionally within the sarcolemma of smooth muscle cells. This is not the result of massive clustering of functional LTCCs, as multiple independent studies suggest they are broadly distributed throughout the sarcolemma of smooth muscle cells. Instead, heterogeneous Ca2+ sparklet activity results from regional targeting of PKCα and calcineurin by AKAP150 as well as the relative activities of this kinase and phosphatase.
Although it is persistent Ca2+ sparklets are critical for excitation-contraction coupling and excitation-transcription coupling in smooth muscle multiple questions about these Ca2+ events remain unanswered. For example, what are the relationships among sarcoplasmic reticulum Ca2+ load, ryanodine receptor function, and Ca2+ sparklet activity? What is the spatial and functional relationship between PKCα and Ca2+ sparklet activity in living cells? Is the number of persistent Ca2+ sparklet per cell limited by the level of AKAP150 expression in it? What are the factors determining the size of Ca2+ sparklet sites? What is the relationship between gating modalities and Ca2+ influx in individual LTCCs within a persistent Ca2+ sparklet site? Finally, what is the relationship between persistent Ca2+ sparklet activity and NFATc3 translocation? Future studies should address these important questions.
The work from the author's laboratory was supported by the AHA and NIH.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.