Calcium sparklets regulate local and global calcium in murine arterial smooth muscle
- Gregory C. Amberg1,
- Manuel F. Navedo1,
- Madeline Nieves-Cintrón1,
- Jeffery D. Molkentin2 and
- Luis F. Santana1
- 1Department of Physiology and Biophysics, University of Washington Box 357290, Seattle, WA 98195, USA2Children's Hospital Medical Center for Molecular Cardiovascular Biology, 3333 Burnet Avenue MLC7020, Cincinnati, OH 45229-3039, USA
- Corresponding author L. F. Santana: Department of Physiology and Biophysics, University of Washington Box 357290, Seattle, WA 98195, USA. Email: santana{at}u.washington.edu
Abstract
In arterial smooth muscle, protein kinase Cα (PKCα) coerces discrete clusters of L-type Ca2+ channels to operate in a high open probability mode, resulting in subcellular domains of nearly continual Ca2+ influx called ‘persistent Ca2+ sparklets’. Our previous work suggested that steady-state Ca2+ entry into arterial myocytes, and thus global [Ca2+]i, is regulated by Ca2+ influx through clusters of L-type Ca2+ channels operating in this persistently active mode in addition to openings of solitary channels functioning in a low-activity mode. Here, we provide the first direct evidence supporting this ‘Ca2+ sparklet’ model of Ca2+ influx at a physiological membrane potential and external Ca2+ concentration. In support of this model, we found that persistent Ca2+ sparklets produced local and global elevations in [Ca2+]i. Membrane depolarization increased Ca2+ influx via low-activity and high-activity persistent Ca2+ sparklets. Our data indicate that Ca2+ entering arterial smooth muscle through persistent Ca2+ sparklets accounts for approximately 50% of the total dihydropyridine-sensitive (i.e. L-type Ca2+ channel) Ca2+ influx at a physiologically relevant membrane potential (−40 mV) and external Ca2+ concentration (2 mm). Consistent with this, inhibition of basal PKCα-dependent persistent Ca2+ sparklets decreased [Ca2+]i by about 50% in isolated arterial myocytes and intact pressurized arteries. Taken together, these data support the conclusion that in arterial smooth muscle steady-state Ca2+ entry and global [Ca2+]i are regulated by low-activity and PKCα-dependent high-activity persistent Ca2+ sparklets.
Resistance arteries regulate blood flow by constricting or dilating in response to changes in intraluminal pressure (Bayliss, 1902). Ultimately, the contraction or relaxation of the smooth muscle cells lining the walls of arteries is responsible for the determination of arterial diameter. Increasing intraluminal pressure causes gradual depolarization of arterial smooth muscle, which in turn increases the global free Ca2+ concentration ([Ca2+]i). Increasing [Ca2+]i initiates a signalling cascade where myosin light chain kinase activates the contractile apparatus resulting in smooth muscle contraction and arterial constriction (Harder, 1984; Somlyo & Somlyo, 1994; Knot & Nelson, 1998; Knot et al. 1998). Decreasing intraluminal pressure has the opposite effect.
Mechanisms linking membrane potential to global [Ca2+]i in smooth muscle have been the subject of intense investigation. Multiple lines of evidence suggest that voltage-dependent dihydropyridine-sensitive L-type Ca2+ channels are essential in transducing changes in membrane potential to changes in global [Ca2+]i in smooth muscle. (1) Blockade of L-type Ca2+ channels or removal of external Ca2+ decreases global [Ca2+]i to the same extent and prevents the development of myogenic tone (Harder et al. 1987; Knot & Nelson, 1998). (2) Disruption of L-type Ca2+ channel function by genetic means results in lower global [Ca2+]i (Wegener et al. 2004). (3) The voltage dependence of [Ca2+]i closely follows the voltage dependence of L-type Ca2+ channels (Fleischmann et al. 1994; Rubart et al. 1996; Knot & Nelson, 1998). (4) Increasing intravascular pressure in the presence of an L-type Ca2+ channel antagonist elicits the same magnitude of membrane depolarization as under control conditions, but without the accompanying increase in arterial wall [Ca2+]i and constriction (Harder, 1984; Knot & Nelson, 1998). On the basis of these findings, a general model has been proposed where increased intraluminal pressure depolarizes arterial smooth muscle, which increases the open probability of L-type Ca2+ channels in the sarcolemma of arterial myocytes. This increases Ca2+ influx and consequently global [Ca2+]i.
Recently, we used total internal reflection fluorescence (TIRF) microscopy to examine with high resolution the spatial organization of functional L-type Ca2+ channels in arterial myocytes (Navedo et al. 2005, 2006; Amberg et al. 2006). Using this approach, we observed discrete Ca2+ influx events – called ‘Ca2+ sparklets’ (Wang et al. 2001) – through L-type Ca2+ channels operating in two functional modes: low-activity and high-activity persistent Ca2+ sparklets. Low-activity Ca2+ sparklets were rare, occurred randomly throughout the sarcolemma, and were produced by the opening of solitary L-type Ca2+ channels. In contrast, persistent Ca2+ sparklets were produced by small clusters of seemingly coupled L-type Ca2+ channels operating in a high open probability mode that created local areas of nearly continual Ca2+ influx. An important distinction between low activity and persistent Ca2+ sparklets is that protein kinase Cα (PKCα) was required for persistent Ca2+ sparklets, but not low activity Ca2+ sparklets (Navedo et al. 2005, 2006). These findings prompted us to propose a Ca2+ sparklet-based model for the regulation of steady-state Ca2+ influx and [Ca2+]i in arterial smooth muscle. This model predicts that Ca2+ influx via low-activity and PKCα-dependent high-activity persistent Ca2+ sparklet sites modulates local and global [Ca2+]i in arterial myocytes.
At present this model remains virtually untested. For example, Ca2+ sparklets in arterial myocytes have largely been recorded at non-physiological voltages (−70 mV) and external Ca2+ concentrations (20 mm). Thus, the physiological significance of Ca2+ sparklets, their influence on [Ca2+]i, and the validity of the model described above remain unclear. Here, we tested the Ca2+ sparklet model of steady-state Ca2+ entry by examining the contribution of the two Ca2+ sparklets modalities (high and low) to [Ca2+]i at physiological Ca2+ and membrane potential. Our data indicate that in arterial smooth muscle, Ca2+ influx through protein kinase C (PKC)-dependent persistent Ca2+ sparklet sites is significant, accounting for ∼50% of the Ca2+ influx via L-type Ca2+ channels under physiologically relevant Ca2+ and membrane potential. These data suggest that Ca2+ entry through two functionally distinct populations (low and high activity) of L-type Ca2+ channels modulates local and global [Ca2+]i in arterial myocytes.
Methods
Isolation of arterial myocytes
Rats (Sprague-Dawley; ≈250 g) as well as wild-type and PKCα knockout (PKCα−/−) mice (≈25 g) (Braz et al. 2004) were killed with a lethal dose of sodium pentobarbital (250 mg kg−1; intraperitoneally) as approved by the University of Washington Institutional Animal Care and Use Committee. Myocytes were dissociated from cerebral and mesenteric arteries using standard enzymatic techniques described elsewhere (Amberg & Santana, 2003). After dissociation, cells were maintained in a nominally Ca2+-free Ringer solution until used. Thapsigargin (1 μm) was included in all solutions to eliminate Ca2+ release from intracellular stores during experimentation.
Patch-clamp electrophysiology and confocal [Ca2+]i imaging
We used the conventional whole-cell patch-clamp technique to control membrane voltage and record L-type Ca2+ currents using an Axopatch 200B amplifier. During experiments, cells were continuously superfused with a solution containing (mm): 140 NMDG, 5 CsCl, 1 MgCl2, 10 glucose, 10 Hepes and 2 CaCl2, adjusted to pH 7.4. Pipettes were filled with a solution composed of (mm): 87 caesium aspartate, 20 CsCl, 1 MgCl2, 5 MgATP, 10 Hepes, 10 EGTA and 0.2 fluo-5F, adjusted to pH 7.2 with CsOH. In some experiments, pipettes were filled with a solution similar to the one above, but with 50 μm fluo-5F and no added EGTA. A voltage error of 10 mV attributable to the liquid junction potential of these solutions was corrected for. Dihydropyridine-sensitive currents were isolated during analysis by digitally subtracting recordings made in the presence of nifedipine or isradipine (1 μm) from recordings obtained from control and Gö6976-exposed (200 nm) cells using pCLAMP 9.0 software. L-type Ca2+ currents were sampled at 20 kHz and low-pass filtered at 2 kHz. All voltage-clamp experiments were performed at room temperature (22–25°C).
In some experiments, [Ca2+]i was imaged using a Radiance 2100 confocal system (Cambridge, MA, USA) coupled to a Nikon TE300 inverted microscope equipped with a Nikon 20× (NA = 0.75; used for pressurized artery experiments, see below) or 60× (NA = 1.4; used in experiments involving isolated arterial myocytes) lens. We used the conventional configuration of the whole-cell patch clamp technique to dialyse fluo-5F into arterial myocytes and hold the membrane potential at −40 mV.
Small mesenteric arteries (types II and III; Su et al. 1998) from wild-type and PKCα−/− mice were removed and cleaned in normal Ringer solution. Isolated arteries were cannulated and mounted in a close-working-distance arteriograph, and the endothelium was disrupted by passing air bubbles through the artery. The arteriograph was placed on the stage of an inverted microscope and extralumenally perfused (3–6 ml min−1) at 37°C with a bicarbonate-based physiological saline solution composed of (mm): 119 NaCl, 4.7 KCl, 24 NaHCO3, 1.2 KH2PO4, 1.6 CaCl2, 1.2 MgSO4, and 11 glucose, with the pH set to 7.4 by bubbling with a gas mixture of O2 (95%) and CO2 (5%). After equilibration (20 min), intravascular pressure was increased to 80 mmHg for 20 min. To load the arteries with Ca2+ indicator, perfusion was halted and the membrane-permeable acetoxymethyl-ester form of fluo-4 (fluo-4 AM) was pipetted into the recording chamber at a final concentration of ≈50 μm. Following 30 min of incubation, perfusion was reinstated and the artery was allowed to re-equilibrate at 80 mmHg for an additional 30 min prior to experimentation. Arterial viability was tested by raising external K+ to 60 mm; arteries that failed to contract robustly in response to raised K+ were discarded.
Fluo-4 and fluo-5F fluorescence signals were converted to Ca2+ concentration units using the ‘Fmax’ equation (Maravall et al. 2000):
as described in detail previously (Dilly et al. 2006; Navedo et al. 2006). Briefly, F is fluorescence, Fmax is the fluorescence intensity of fluo-4/5F in the presence of saturating free Ca2+, Kd is the dissociation constant (fluo-5F = 1280 nm; fluo-4 = 800 nm), and Rf (fluo-5F = 286; fluo-4 = 150) is the indicator's Fmax/Fmin ratio. Fmin is the fluorescence intensity of fluo-4/5F in a solution where the Ca2+ concentration is 0. Kd and Rf values were determined in vitro using standard methods and are similar to those reported by others (Woodruff et al. 2002). Fmax was determined at the end of each experiment by exposing cells or arteries to the Ca2+ ionophore ionomycin (10 μm) and 20 mm external Ca2+.
TIRF microscopy
Ca2+ sparklets were recorded as previously described (Navedo et al. 2005, 2006). Briefly, we used a through-the-lens TIRF microscope built around an inverted Olympus IX-70 microscope equipped with an Olympus PlanApo (60×, NA = 1.45) oil-immersion lens and an Andor iXON CCD camera (South Windsor, CT, USA). Cells were loaded with the Ca2+ indicator fluo-5F, with excitation achieved by the 488 nm line of an argon laser. Excitation and emission light was separated with the appropriate set of filters. Images were acquired at 100–300 Hz. As before (Navedo et al. 2005, 2006), we determined the activity of Ca2+ sparklets by calculating the nPs of each Ca2+ sparklet site, where n is the number of quantal levels and Ps is the probability that a quantal Ca2+ sparklet event is active. Using this analysis we have grouped Ca2+ sparklet sites into three categories: silent (by default has an nPs of 0), low (nPs between 0 and 0.2), and high (nPs higher than 0.2). A detailed description of this analysis can be found in Navedo et al. (2006).
Signal mass analysis
We used the ‘signal mass’ approach developed by Zou et al. (2002, 2004) to determine the amount of Ca2+ flux associated with Ca2+ sparklets recorded using TIRF microscopy. Briefly, for this analysis the total fluorescence intensity (Ftotal) associated with a Ca2+ sparklet is determined from raw images by summing the fluo-5F fluorescence from all the pixels within an area of the image larger than the entire fluorescence signal produced by a Ca2+ sparklet. The change in Ftotal (ΔFtotal) is then determined by subtracting the total fluorescence before the channel(s) open from the total fluorescence at each time point during the opening. Signal mass was obtained by determining the peak of the integral of ΔFtotal trace over time (∫ΔFtotaldt) for each Ca2+ sparklet. To determine the Ca2+ influx associated with a Ca2+ sparklet we determined the relationship between the signal mass of a Ca2+ sparklet and the Ca2+ entry associated with the Ca2+ current (ΔQCa) underlying it. Signal mass and ΔQCa were determined from simultaneous recordings of Ca2+ sparklets and Ca2+ currents (see Fig. 1). For these recordings external Ca2+ was raised to 20 mm to increase the amplitude of L-type Ca2+ channel currents and hence Ca2+ sparklets; the concentration of NMDG was lowered 20 mm to offset the increase in external Ca2+. ΔQCa (in coulombs) was determined by integrating the Ca2+ current underlying the Ca2+ sparklet (∫ICa/(2e)dt). Note that time courses of simultaneously recorded Ca2+ sparklets and Ca2+ currents were tightly coupled (Fig. 1A). For comparative purposes ΔF/F0 Ca2+ sparklet traces are included. Consistent with this observation and in agreement with previous reports (Wang et al. 2001; Zou et al. 2002, 2004) we found that the relationship between Ca2+ sparklet signal mass and ΔQCa was linear (Fig. 1B). We therefore used the slope of this relationship to convert Ca2+ sparklet signal masses obtained with 2 mm external Ca2+ to ΔQCa values.
Ca2+ sparklet event duration analysis
We determined the duration of Ca2+ sparklet events. For this analysis, Ca2+ sparklets were recorded at −70 and −40 mV in the presence of 2 mm external Ca2+. Ca2+ sparklet event duration times were obtained from fits to Ca2+ sparklet records using the pCLAMP ‘threshold detection analysis’ used for activity analysis (nPs) briefly described above and in greater detail elsewhere (Navedo et al. 2006). An example of this analysis is included in the online Supplemental material (Fig. S2).
Our Ca2+ sparklet duration analysis assumes that the time course of these events accurately represents the time course of the underlying Ca2+ current. Experimental observations support this assumption. As shown in Fig. 1A and in our recent study (see Navedo et al. 2005), simultaneous recordings of Ca2+ sparklets and the underlying Ca2+ currents demonstrate that the time courses of the fluorescence Ca2+ sparklet signal and the underlying Ca2+ current are tightly matched. Thus, the rapid onset and offset of Ca2+ sparklets probably results from the opening and closing of the underlying Ca2+ channels and their duration reflects the duration of the Ca2+ entry event.
An interesting feature of Ca2+ sparklets, notably those with durations >10 ms, is that a plateau is reached shortly following activation. One consideration with respect to Ca2+ sparklet duration analysis is whether or not this plateau results from saturation of the Ca2+ indicator. This is relevant as saturation of the indicator would prevent us from accurately recording submembrane [Ca2+]i dynamics during a Ca2+ sparklet event. However, multiple lines of evidence suggest that the plateau phase of Ca2+ sparklets recorded using TIRF are not due to fluo-5F saturation. (1) Ca2+ sparklet amplitudes are quantal (Navedo et al. 2005, 2006; Parker, 2006) with the quantal unit being ≈34 nm at −70 mV with 20 mm Ca2+. Accordingly, quantal increases in [Ca2+]i can be observed during the plateau phase of a previously active Ca2+ sparklet event, presumably as additional channels within the cluster are coincidentally activated (see Fig. S1A). This observation empirically demonstrates that under our experimental conditions fluo-5F is not saturated during Ca2+ sparklet recordings and additional increases in Ca2+ entry can be detected. Note that 10-fold less external Ca2+ was used to record Ca2+ sparklets for the event duration analysis described here. (2) We found that Ca2+ sparklet amplitude (at the same site) increased by raising external Ca2+ from 2 to 20 mm (see Fig. S1B). These data were consistent with our previous study showing that Ca2+ sparklet amplitude depends on external Ca2+ (Navedo et al. 2005). (3) The relationship between Ca2+ influx and Ca2+ sparklet signal mass was linear over the range of Ca2+ influx observed in this study. Consistent with this, we have also observed a linear relationship between the Ca2+ sparklet amplitude and the simultaneously recorded underlying Ca2+ current (Navedo et al. 2005). Collectively, these data strongly suggest that it is unlikely that fluo-5F saturates during long Ca2+ sparklet events in the presence of 2 or 20 mm external Ca2+.
In addition, recent experimental and theoretical work by other groups suggests a probable mechanism for the ‘square-like’ shape of our Ca2+ sparklets during TIRF experimentation (Demuro & Parker, 2004, 2005; Shuai & Parker, 2005). Briefly, with TIRF microscopy, fluorescence is limited to a thin (∼100 nm) evanescent field that decays exponentially from the bottom of the coverslip. When a Ca2+ channel opens, Ca2+ enters and binds to cytosolic fluo-5F, which increases its fluorescence. Note, however, that as Ca2+-bound fluo-5F molecules diffuse away from the site of Ca2+ entry, many of these molecules will depart the evanescent field thus decreasing fluorescence. In this scenario, fluorescence at a Ca2+ sparklet site increases rapidly as the sarcolemmal channel opens and a plateau phase is reached when the rate of Ca2+ entry and Ca2+-bound fluo-5F diffusion out of the evanescent field are equal.
In our experiments the inclusion of the relatively slow Ca2+ buffer EGTA (on rate approximately 100-fold slower than fluo-5F) enhances this property of TIRF microscopy described above by further restricting fluo-5F fluorescence to the site of Ca2+ entry (within∼1 μm) (Zenisek et al. 2003). With EGTA present, Ca2+ entering the cell initially interacts with the faster fluo-5F, producing fluorescence, but then quickly (approximately 2 ms) (Zenisek et al. 2003) binds to the more abundant and non-fluorescent EGTA. Thus, in our TIRF experiments where fluo-5F and EGTA are present, [Ca2+]i signals are limited to the submembrane space near the mouth of L-type Ca2+ channel. A similar strategy has been used to record Ca2+ release sites in ventricular myocytes using confocal microscopy (Song et al. 1998). In our TIRF experiments, the rapid onset of Ca2+ sparklets is likely to be produced by Ca2+ entering the cell at a high rate, which then plateaus as Ca2+ influx and efflux (either free or bound to fluo-5F) from the evanescent field are balanced and Ca2+ binds to EGTA. The Ca2+ sparklet quickly terminates when the channel(s) closes and Ca2+ diffuses away from the channel and the evanescent field.
It is also important to note that two recent studies have reported [Ca2+]i signals resulting from the opening of N-type Ca2+ channels (Demuro & Parker, 2004) and acetylcholine receptors (Demuro & Parker, 2005) using TIRF microscopy (without intracellular EGTA). The [Ca2+]i records resulting from the opening of these channels were kinetically similar to the Ca2+ sparklets in our study. Together with the issues described above, these findings suggest that the kinetics (e.g. rapid onset, plateau) of Ca2+ sparklets are not unique and result from the optical properties of TIRF microscopy irrespective of the cell and Ca2+ channel studied. Thus, our Ca2+ sparklet duration analysis is justified as reasonable for the following reasons. First, the time course of Ca2+ sparklet events closely follows the time course (i.e. rapid on and offset) of the underlying Ca2+ current. Accordingly, the duration of Ca2+ sparklets is directly related to the duration of Ca2+ entry. Second, as noted above, the plateau phase of Ca2+ sparklets is likely to represent a balance between Ca2+ entry and diffusion of Ca2+ away from the evanescent field and the binding of Ca2+ to EGTA, and it is not due to fluo-5F saturation.
PKCα immunofluorescence labelling
Immunofluorescence labelling of dispersed myocytes was performed as previously described (Amberg & Santana, 2003) using a monoclonal antibody that recognizes the α isoform of PKC (Abcam, Cambridge, MA, USA). The secondary antibody was an Alexa-Fluor-488-conjugated rabbit anti-mouse (5 mg ml−1) from Molecular Probes. Cells were imaged using our confocal system. We tested the specificity of our labelling by performing a negative control where the primary antibody was substituted with PBS. PKCα-associated fluorescence was not detected under these conditions (data not shown). Surface and cytosolic PKCα-associated immunofluorescence was quantified by measuring the intensity of pixels above a set threshold defined as the mean fluorescence intensity outside the cells (i.e. background) plus three times its standard deviation. For each cell, pixel intensities of the entire surface membrane (≈1 μm in width) and a spatially equivalent section of cytosol were obtained and averaged. From these measurements we determined the ratio of surface to cytosolic PKCα and used this as an indicator of PKCα translocation and activity (Khalil & Morgan, 1991; Khalil et al. 1994).
Chemicals and statistics
Gö6976, PKC inhibitory peptide (PKCi) and ionomycin were from Calbiochem (San Diego, CA, USA); all other chemicals were from Sigma-Aldrich (St Louis, MO, USA) unless stated otherwise. Normally distributed data are presented as means ± standard error of the mean (s.e.m.). Two-sample comparisons were made using a Student's t test; multigroup comparisons were made by ANOVA, which, if necessary, was followed by Tukey's multicomparison test. Non-parametric statistical analyses (Mann-Whitney test) were used for non-normally distributed data. P values less than 0.05 were considered significant. Asterisks used in the figures indicate a significant difference between groups as indicated.
Results
Ca2+ influx via persistent Ca2+ sparklets is substantial and increases with depolarization
We used the ‘signal mass’ approach (Zou et al. 2002, 2004) to directly measure the Ca2+ influx associated with Ca2+ sparklets (ΔQCa) in dissociated cerebral arterial smooth muscle cells at −70 mV and the physiologically relevant membrane potential of −40 mV in the presence of 2 mm external Ca2+. All experiments in this study were performed in cells treated with the SR Ca2+ ATPase inhibitor thapsigargin (1 μm) to eliminate SR Ca2+ release. As described in Methods, for this analysis we first made simultaneous recordings of Ca2+ sparklets and Ca2+ currents. From these records we determined the relationship between the signal mass of a Ca2+ sparklet and the Ca2+ entry from the underlying Ca2+ current (ΔQCa). Ca2+ sparklet signal mass values were obtained by determining the peak of the integral of ΔFtotal over time (∫ΔFtotaldt) for each Ca2+ sparklet; ΔQCa (in coulombs) was determined by integrating the Ca2+ current underlying the Ca2+ sparklet. We then used the slope of the signal mass−ΔQCa relationship (Fig. 1B) to calculate the ΔQCa of Ca2+ sparklets recorded with 2 mm external Ca2+ at −70 and −40 mV.
Figure 2A shows representative ΔFCa records from low- and high-activity (nPs) Ca2+ sparklet sites at −70 and −40 mV in the presence of 2 mm extracellular Ca2+. We determined the nPs(where n is the number of quantal levels and Ps is the probability of Ca2+ sparklet occurrence) of Ca2+ sparklet sites as described in Methods and Navedo et al. (2006). These ΔFCa records show that Ca2+ sparklet duration and amplitude (i.e. activity) increased with membrane depolarization. Note that ΔFtotal was increased by nearly an order of magnitude upon depolarization to −40 from −70 mV as reflected by the smaller scale bar for the records at −40 mV.
We constructed histograms of ΔQCa amplitudes for low and high nPs Ca2+ sparklets at −70 mV (Fig. 2B). Because these data did not have a normal distribution, non-parametric statistical analyses were used. We found that the mode of the ΔQCa histograms of low (100 fC) and high (100 fC) nPs Ca2+ sparklets at −70 mV was similar. The histograms in Fig. 2B show that the ΔQCa of individual Ca2+ sparklets in low and high nPs sites at this voltage ranged from 9 to 485 fC and from 16 to 4870 fC, respectively. The median ΔQCa of low and high nPs Ca2+ sparklets at −70 mV was 63 (n = 58) and 189 fC (n = 34), respectively. Together, these data indicate that while the majority of Ca2+ sparklet events in low and high nPs Ca2+ sparklets involve similar amounts of Ca2+ entry (i.e. they have similar modes), high nPs sites produce Ca2+ sparklets with ΔQCa values higher than those in low nPs sites (i.e. they have a higher range).
Figure 2C shows histograms of the ΔQCa amplitudes of Ca2+ sparklets from low and high nPs sites at a physiological potential of −40 mV. As with Ca2+ sparklets at −70 mV, the modes of the low and high nPs histograms at −40 mV were similar (≈200 fC). The ΔQCa amplitudes of Ca2+ sparklets at −40 mV ranged from 7 to 2987 for low nPs Ca2+ sparklets, and from 50 to 5132 fC for high nPs Ca2+ sparklets. The median ΔQCa at −40 mV was larger (P < 0.05) for high (452 fC; n = 60) than low nPs Ca2+ sparklets (292 fC; n = 41). Note that membrane depolarization produced a shift in the mode and increased the range of the ΔQCa of Ca2+ sparklets in low and high nPs sites. Accordingly, the median ΔQCa of Ca2+ sparklets in low and high nPs sites was larger at −40 than at −70 mV (P < 0.05).
The signal mass analysis revealed multiple novel aspects of Ca2+ sparklets in arterial myocytes. First, the data suggest that Ca2+ influx via persistent high nPs Ca2+ sparklets is larger than for Ca2+ sparklets in low nPs sites. Second, high nPs sites have Ca2+ sparklets with a broader ΔQCa amplitude distribution than low nPs sites. This suggests that Ca2+ influx is higher with high nPs than low nPs Ca2+ sparklet sites because, while low nPs sites produce only low ΔQCa amplitude Ca2+ sparklets, high nPs sites produce low and high ΔQCa Ca2+ sparklets. Third, membrane depolarization increases Ca2+ influx via Ca2+ sparklets in low and high nPs sites. Interestingly, the relative increase in ΔQCa is higher in low than high nPs Ca2+ sparklets. Because the driving force for Ca2+ entry decreases with membrane depolarization, the resulting increase in Ca2+ influx is consistent with an increase in the open probability of L-type Ca2+ channels upon membrane depolarization.
The duration of high activity Ca2+ sparklet events is longer than low activity Ca2+ sparklet events and increases with membrane depolarization
One potential mechanism that could contribute to greater Ca2+ influx via high activity than via low activity Ca2+ sparklet sites is that the duration of high nPs Ca2+ sparklet events are longer than those of low nPs Ca2+ sparklet events. To investigate this possibility, we determined the duration of Ca2+ sparklet events in low and high nPs Ca2+ sparklet sites (Fig. 3). Details of this analysis as well as validation and justification of these approaches are included in Methods.
First, we analysed the duration of low and high nPs Ca2+ sparklet events at −70 mV. We found that at this potential, event duration histograms of low nPs Ca2+ sparklets were best fit with a single exponential function with a time constant (τ) of 26 ms (Fig. 3A, top). In contrast, event duration histograms of high nPs sparklets were best fit with a biexponential function with a fast (τ1) and slow time constant (τ2) of 30 and 75 ms, respectively (Fig. 3A, bottom).
Depolarization to −40 mV increased the duration of low and high nPs Ca2+ sparklet events with respect to −70 mV (Fig. 3B). As observed at −70 mV, the histogram of low nPs Ca2+ sparklet event durations at −40 mV was best fit with a single exponential function. Note, however, that the τ of low nPs Ca2+ sparklet event durations at this depolarized potential was increased to 41 ms. Thus, membrane depolarization from −70 to −40 mV increased the duration of low nPs Ca2+ sparklet events approximately 1.5-fold.
As at −70 mV, the duration of Ca2+ sparklets in high nPs sites at −0 mV was best fit with a biexponential function where τ1 = 35 ms and τ2 = 200 ms. At −40 mV the fast component of high nPs Ca2+ sparklet event durations was similar to that of the monoexponential low nPs sites (41 ms). However, the fast (τ1) and slow (τ2) components of high nPs sparklet events were of longer duration at −40 mV than those of high nPs sparklets at −70 mV. From these data we conclude that low nPs sites produce Ca2+ sparklet events of relatively short duration and high nPs sites produce Ca2+ sparklet events with short (fast τ1) and much longer durations (slow τ2). Furthermore, our data indicate that depolarization increases Ca2+ influx, at least in part, by prolonging low and high nPs Ca2+ sparklet event durations.
PKC-dependent persistent Ca2+ sparklets modulate local and global [Ca2+]i
We used the ΔQCa values determined above for low and high activity Ca2+ sparklets to estimate their impact on global [Ca2+]i in cerebral arterial myocytes with the following equation:
where ΔQCa is change in Ca2+ influx (in coulombs; as determined by the signal mass analysis), F is the Faraday constant (96 485 C mol−1), V is the accessible cytosolic volume (in litres), and BC is the buffering capacity of the cell.
We assumed an accessible cytosolic volume of 0.9 pl (Aaroson et al. 1988; Daub & Ganitkevich, 2000), a buffering capacity of 80 (Guerrero et al. 1994), and for simplicity, no Ca2+ extrusion. Although this last assumption is physiologically unrealistic, our calculation is still useful because it offers an upper boundary for the potential change in global [Ca2+]i by Ca2+ sparklet events. Using these values, we calculated that the median ΔQCa of low and high nPs Ca2+ sparklet events at −40 mV could produce global increases in [Ca2+]i of ∼21 and 38 nm, respectively. Thus, at least theoretically, Ca2+ influx via Ca2+ sparklets could influence local and global [Ca2+]i.
To further establish the link between Ca2+ sparklets and global [Ca2+]i, we simultaneously recorded Ca2+ sparklets and changes in global [Ca2+]i (Fig. 4). For these experiments EGTA was omitted from the pipette solution. In addition, we set the angle of the laser beam on our TIRF system slightly below the critical angle for total internal reflection. Under these conditions, light is refracted at a shallow angle from the coverslip, which results in a narrower excitation section than conventional epifluorescence, but thicker than the one obtained in TIRF mode (∼100 nm in our system). This allowed us to image deeper into the cell and hence record near-membrane and global cytosolic [Ca2+]i.
To compensate for the increased optical section and accompanying increase in background fluorescence, which decreases the signal-to-noise ratio of our recordings thus reducing our ability to detect low amplitude Ca2+ sparklets, we initiated our recordings in the absence of external Ca2+ to decrease global [Ca2+]i (i.e. background fluorescence). Ca2+ sparklets and global cytosolic [Ca2+]i were then recorded following introduction of Ca2+ (2 mm). Figure 4A shows concordant time courses of [Ca2+] at a single Ca2+ sparklet site (black trace) and the spatially averaged, global [Ca2+]i (red trace) after introduction of 2 mm external Ca2+. Note that Ca2+ sparklets were observed shortly after the introduction of 2 mm Ca2+. The initial Ca2+ sparklet in this trace (event ‘i’) did not change global [Ca2+]i as endogenous Ca2+ buffers shunt Ca2+ away from the fluorescent indicator. However, as global cytosolic [Ca2+]i increased and the buffers became loaded with Ca2+, individual Ca2+ sparklets (events ‘ii and iii’) become associated with increases in global [Ca2+]i. Interestingly, after reaching a plateau – probably reflecting a new steady-state level of global [Ca2+]i – Ca2+ sparklets (event ‘iii’) increased global [Ca2+]i transiently as Ca2+ extrusion mechanisms balanced influx. Similar results were obtained in five independent experiments. Note that in the presence of the L-type Ca2+ channel antagonist nifedipine (10 μm), Ca2+ sparklets were absent and the increases in global [Ca2+]i observed during the introduction of 2 mm external Ca2+ were abolished (Fig. 4B; n = 5). Marginal elevations in [Ca2+]i were occasionally observed in nifedipine-treated cells following introduction of 2 mm Ca2+. However, these small drifts in [Ca2+]i (see Fig. 4B) were not statistically different from background noise in the absence of external Ca2+ (P > 0.05, n = 5). This indicates that Ca2+ influx through L-type Ca2+ channels underlies Ca2+ sparklets and the rise in global [Ca2+]i observed during exposure to 2 mm external Ca2+.
Next, we tested the hypothesis that persistent Ca2+ sparklets contribute to [Ca2+]i in isolated cerebral arterial smooth muscle cells. We took advantage of our previous observation that high nPs, persistent Ca2+ sparklet activity is strictly dependent on PKC activity (Navedo et al. 2005, 2006). Specifically, pharmacological inhibition of the α isoform of PKC ablates persistent Ca2+ sparklet activity and persistent Ca2+ sparklet activity is absent in arterial myocytes from PKCα null mice. In contrast, PKC is not necessary for low nPs Ca2+ sparklets produced by random openings of solitary L-type Ca2+ channels. Thus, we used the PKC dependence of persistent Ca2+ sparklets to separate this gating modality from stochastic openings of solitary channels.
In these experiments, we dialysed arterial myocytes with a PKCα inhibitory peptide (PKCi; 100 μm) and recorded Ca2+ sparklets and global [Ca2+]i as before. Unlike control cells, and as previously reported by us (Navedo et al. 2005, 2006), Ca2+ sparklets were not observed following inhibition of PKCα. Furthermore, as with nifedipine, we did not observe the abrupt elevations in global [Ca2+]i, which we attributed to persistent Ca2+ sparklet-mediated Ca2+ influx under control conditions. Rather, when 2 mm external Ca2+ was applied to PKCi-dialysed cells, [Ca2+]i slowly and steadily increased, reaching a steady-state level about 60% lower than control cells (P < 0.05, n = 7; Fig. 4C and D). Taken together, these data support the view that PKC-dependent Ca2+ sparklets contribute to Ca2+ influx thereby regulating local and global [Ca2+]i in cerebral arterial smooth muscle.
PKC-dependent L-type Ca2+ channel activity contributes to arterial myocyte [Ca2+]i
If PKC-dependent L-type Ca2+ channel activity (i.e. persistent Ca2+ sparklets) contributes to whole-cell Ca2+ influx in arterial myocytes as our data suggest, then whole-cell L-type Ca2+ currents should be influenced by PKC activity. Accordingly, we voltage-clamped isolated cerebral arterial myocytes to determine if basal PKC activity enhances whole-cell L-type Ca2+ currents (ICa) in these cells. Figure 5A shows representative ICa records evoked by step depolarizations to +30 mV from a holding potential of −70 mV before and after application of the PKC inhibitor Gö6976 (200 nm). At this concentration, Gö6976 prevented translocation (i.e. activation) of PKCα to the sarcolemma of arterial myocytes (see Fig. S3 in online Supplemental material). Consistent with the hypothesis that basal PKC function modulates L-type Ca2+ channel function in cerebral arterial myocytes, Gö6976 reduced peak ICa densities from −1.6 ± 0.3 pA pF−1 under control conditions to −0.8 ± 0.2 pA pF−1 in the presence of Gö6976 (n = 6, P < 0.05), corresponding to a 50% decrease in ICa (Fig. 5B).
We then measured the PKC-dependent component of the L-type Ca2+ current (i.e. persistent Ca2+ sparklet) during ramp depolarizations (40 mV s−1) from −70 to +40 mV. Dihydropyridine-sensitive currents as shown in Fig. 5C were obtained by digital subtraction of voltage ramps made in the presence of either nifedipine or isradipine (1 μm for each) from control or Gö6976 (200 nm) exposed cells. Under control conditions, dihydropyridine-sensitive currents were evident early (i.e. just past −70 mV) during the ramp protocol and reached a peak near −10 mV. Note that the amplitude of the L-type current increased steadily from −70 to −30 mV (see the expanded trace in Fig. 5C), a voltage range which corresponds to those experienced in intact cerebral arteries (Knot & Nelson, 1998).
To quantify the amount of Ca2+ entering the cells over this voltage range, we integrated the dihydropyridine-sensitive current (i.e. ICa) between −70 and −30 mV. Under control conditions, −172 ± 18 pC (n = 5) of dihydropyridine-sensitive charge was detected between −70 and −30 mV. Following PKC inhibition with Gö6976 (200 nm), the residual ICa(i.e. non PKC-dependent persistent Ca2+ sparklet current) was reduced with respect to control conditions. Indeed, the amount of dihydropyridine-sensitive charge between −70 and −30 mV was reduced to −85 ± 7 pC (n = 5), which was approximately 50% of that of control (P < 0.05). These results suggest that about half of the Ca2+ entering arterial smooth muscle cells at physiological potentials is mediated by PKC-dependent persistent Ca2+ sparklets.
We used confocal microscopy to assess the contribution of dihydropyridine-sensitive, PKC-dependent Ca2+ entry (i.e. persistent Ca2+ sparklets) to global [Ca2+]i in dispersed arterial myocytes. The top trace in Fig. 5D shows global [Ca2+]i as a function of time in a control myocyte held at −40 mV during exposure to nifedipine (1 μm). As expected, L-type Ca2+ channel blockade produced a gradual decrease in [Ca2+]i until a new steady-state level was reached. Consistent with our electrophysiological data above, inhibition of PKC with Gö6976 (200 nm) reduced [Ca2+]i (Fig. 5D, bottom). While subsequent application of nifedipine decreased [Ca2+]i further, the total nifedipine-sensitive decrease in [Ca2+]i was only 33.5 ± 0.5% that of control (P < 0.05; n = 5) in the presence of Gö6976. These data are consistent with the notion that Ca2+ entry through PKC-dependent Ca2+ sparklets contributes to global [Ca2+]i in cerebral arterial smooth muscle.
PKC-dependent L-type Ca2+ channel activity contributes to arterial wall [Ca2+]i
Our data from isolated arterial myocytes suggest that steady-state [Ca2+]i in these cells is regulated by two functionally distinct populations of L-type Ca2+ channels, which correspond to low and high nPs(activity) Ca2+ sparklets. To expand on these findings, we pressurized fluo-4-loaded mesenteric arterial segments to 80 mmHg and examined the relationship between multimodal L-type Ca2+ channel function and global Ca2+ in intact arteries (Fig. 6A, left). Under control conditions, wild-type mesenteric arteries had a resting [Ca2+]i of 214 ± 7 nm (n = 10). Inhibition of PKC with Gö6976 (200 nm) gradually decreased arterial wall Ca2+ to 152 ± 7 nm (P < 0.05, n = 7), corresponding to an average reduction of 50 ± 5 nm. Subsequent application of isradipine (1 μm) reduced [Ca2+]i an additional 46 ± 8 nm (P < 0.05, n = 7) to a final level of 105 ± 5 nm. In independent experiments, isradipine (1 μm) alone lowered [Ca2+]i to 113 ± 6 nm (P < 0.05, n = 4), a level not significantly different from that observed with the combination of Gö6976 + isradipine (105 ± 5) (Fig. 6B and C). Note that in the presence of isradipine, Gö6976 (200 nm) had no effect on arterial wall Ca2+ (n = 4), suggesting that the effect of PKC inhibition on arterial wall [Ca2+]i resulted from a decrease in Ca2+ influx through L-type Ca2+ channels.
Next, we examined arterial wall [Ca2+]i in mesenteric arteries from PKCα-null mice (PKCα−/−; Fig. 6A, right) to unequivocally examine the contribution of PKCα to arterial wall Ca2+ and to further test the specificity of Gö6976. Under control conditions PKCα−/− arteries had [Ca2+]i of 175 ± 7 nm (n = 5), which was lower than that observed in wild type arteries under control conditions (P < 0.05). Interestingly, [Ca2+]i was not different between control PKCα−/− and wild-type arteries in the presence of Gö6976 (P > 0.05). Furthermore, while Gö6976 had no effect on [Ca2+]i in PKCα−/− arteries, isradipine lowered [Ca2+]i to 117 ± 10 nm, corresponding to a decrease in [Ca2+]i of 57 ± 10 nm (n = 5) (Fig. 6B and C). The decrease in [Ca2+]i induced by isradipine in PKCα−/− arteries and wild-type arteries in the presence of Gö6976 was not different (P > 0.05).
To further establish the link between PKCα activity and L-type Ca2+ channel function in arterial smooth muscle, we voltage-clamped isolated myocytes from wild-type and PKCα−/− mesenteric arteries and compared the density of ICa under control conditions (Fig. 6A, right). In agreement with our arterial wall [Ca2+]i data, the density of ICa in PKCα−/− myocytes (−0.9 ± 0.1 pA pF−1) was approximately half of that of wild-type myocytes (−1.5 ± 0.3 pA pF−1; P < 0.05, n = 9). It is possible that differences in L-type Ca2+ channel expression account for the differences in ICa densities observed. To examine this, we performed RT-PCR on arteries from wild-type and PKCα−/− mice. In agreement with our pharmacological data from wild-type mice presented above (see Fig. 5A and C), we found that amplification of Cav1.2 transcripts, the predominant L-type Ca2+ channel pore-forming subunit in arterial smooth muscle, was not different between WT and PKCα−/− arteries (n = 3; data not shown). These data suggest that the smaller ICa densities observed in PKCα−/− myocytes resulted from an absence of PKCα-dependent enhancement of ICa density as opposed to a difference in expression.
Discussion
In this study we have performed the first direct test of the Ca2+ sparklet model of Ca2+ entry in arterial myocytes (Navedo et al. 2005, 2006; Amberg et al. 2006). In doing so we provide the first quantitative analysis of Ca2+ influx via low activity and high activity, PKCα-dependent, persistent Ca2+ sparklet sites and demonstrate that Ca2+ sparklets modulate local and global [Ca2+]i in arterial smooth muscle. Furthermore, our data indicate that membrane depolarization increases Ca2+ influx in arterial myocytes by increasing Ca2+ entry via low activity and PKC-dependent, persistent Ca2+ sparklets. We also performed the first quantitative examination of the relative contribution of different L-type Ca2+ channel gating modalities (i.e. low activity versus PKCα-dependent, persistent Ca2+ sparklets) to steady state Ca2+ in arterial smooth muscle. Our data indicate that, on average, Ca2+ influx via high activity, persistent Ca2+ sparklets accounts for approximately 50% of the steady-state Ca2+ entry at physiological Ca2+ concentrations and membrane potentials in cerebral and mesenteric arteries.
A central tenet of the Ca2+ sparklet model is that steady-state Ca2+ influx is produced by the continual opening of small clusters of apparently coupled persistent L-type Ca2+ channels in addition to infrequent, stochastic, voltage-dependent openings of individual L-type Ca2+ channels. PKCα is required for persistent Ca2+ sparklet activity. L-type Ca2+ channel agonists (e.g. Bay K8644) and activators of PKC increase Ca2+ influx by promoting additional clusters of L-type Ca2+ channels to operate in persistent Ca2+ influx mode and by increasing the open probability of solitary channels.
The data presented here provide compelling support for this model and reveal new, unexpected, information regarding the mechanisms by which Ca2+ sparklets regulate [Ca2+]i in cerebral and mesenteric arterial smooth muscle. The first major finding in this study is that Ca2+ influx (i.e. ΔQCa) via Ca2+ sparklets sites impact local and global [Ca2+]i. As expected, the amount of Ca2+ entry associated with high nPs(activity) Ca2+ sparklet is larger than for low nPs sites at −70 mV and the physiological potential of −40 mV. Establishing precise contributions of individual low and high nPs Ca2+ sparklets to global [Ca2+]i is difficult as the conditions required (2 mm external Ca2+ and a holding potential of −40 mV) to measure physiologically relevant steady-state global [Ca2+]i with epifluorescence illumination make it difficult to record Ca2+ sparklets because of a reduced signal-to-noise ratio.
Analysis of the ΔQCa values determined above provide an indication of the potential impact of low and high activity Ca2+ sparklets on global [Ca2+]i. This analysis requires three assumptions: (1) an accessible cytosolic volume of 0.9 pl (Aaroson et al. 1988; Daub & Ganitkevich, 2000); (2) a buffering capacity of 80 (Guerrero et al. 1994); and (3) no Ca2+ extrusion during the Ca2+ sparklet event. Although the last assumption is unrealistic, our analysis is useful because it provides upper and lower boundaries for the potential impact of Ca2+ sparklets on [Ca2+]i. Our analysis reveals 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, Ca2+ influx via low and high nPs Ca2+ sparklets increased during depolarization from −70 to − 40 mV. 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-evoked increase in Ca2+ entry is likely to be due to an increase in the activity (i.e. open probability) of L-type Ca2+ channels. Taken together, these data indicate that Ca2+ influx via Ca2+ sparklets is voltage dependent and influences local and global [Ca2+]i.
Two additional independent sets of experiments support this conclusion. (1) Pharmacological and genetic (i.e. PKCα−/−) inhibition of PKCα eliminated persistent Ca2+ sparklets and decreased the DHP-sensitive component of the global [Ca2+]i by ∼50% in dissociated arterial myocytes and in intact pressurized arteries. (2) Ca2+ influx via ICa was reduced ∼50% during pharmacological or genetic inhibition of PKCα, which eliminates persistent Ca2+ sparklet activity (Navedo et al. 2005, 2006). Thus, on average, persistent Ca2+ sparklet activity contributes to ∼50% of the DHP-sensitive Ca2+ influx required for maintenance of steady-state [Ca2+]i under physiological conditions.
Our data (Navedo et al. 2005, 2006), as well as the work of others (Yang et al. 2005), strongly support the hypothesis that PKCα increases persistent Ca2+ sparklet activity by modulating L-type Ca2+ channel function independent of changes in membrane potential. However, it is important to note that PKC can potentially modulate Ca2+ sparklet activity indirectly as well by inducing membrane depolarization (Slish et al. 2002). Consequently, it is likely that in intact pressurized arteries PKC inhibition decreases L-type Ca2+ channel-mediated Ca2+ influx by decreasing persistent Ca2+ sparklet activity and by inducing membrane hyperpolarization. Note, however, that in dispersed arterial myocytes voltage clamped at −40 mV, PKC inhibition decreased L-type Ca2+ channel function resulting in a decrease in global Ca2+ (see Fig. 5D above).
Our analysis of the duration of Ca2+ sparklets revealed interesting and unexpected insights regarding the mechanisms underlying differences in Ca2+ influx via low and high nPs Ca2+ sparklets. As noted above, the event duration histogram of low nPs Ca2+ sparklets were best fit with an exponential function with τ ≈ 26 ms at −70 mV and τ ≈ 41 at −40 mV. Interestingly, we found that the event duration histograms of high nPs Ca2+ sparklets were best fit with the sum of two exponential functions with a τfast and τslow of ≈30 and 75 ms (at −70 mV) and ≈35 and 200 ms (at −40 mV), respectively. This analysis indicated that the duration of low and high nPs Ca2+ sparklets increased during membrane depolarization from −70 to −40 mV, which suggests the intriguing possibility that the L-type Ca2+ channels underlying Ca2+ sparklets operate in two gating modes. The L-type Ca2+ channels that open briefly and rarely underlie low nPs Ca2+ sparklet sites while high nPs Ca2+ sparklet sites are produced by L-type Ca2+ channels that either undergo frequent, but brief openings (i.e. bursts) and/or open for relatively long periods of time. These data suggest that the greater Ca2+ influx associated with high nPs sites at −70 and −40 mV result from longer event duration times than those of low nPs sites. Furthermore, our data indicate that membrane depolarization increases Ca2+ influx by increasing the duration of low and high nPs Ca2+ sparklets.
An important observation in this study is that basal PKCα activity is critical in regulating Ca2+ influx via persistent Ca2+ sparklets and global [Ca2+]i. Previous studies support this observation. For example, recent reports suggest that inhibition of basal PKC activity – which presumably decreases high nPs, persistent Ca2+ sparklet activity – decreases global [Ca2+]i and/or myogenic tone in mesenteric (Wesselman et al. 2001), coronary (Korzick et al. 2004) and cerebral arteries (Jarajapu & Knot, 2005). In addition, PKC inhibition has also been shown to modulate L-type Ca2+ channel function in myocytes from rabbit portal vein (Zhong et al. 2001) and human umbilical vein (Schuhmann & Groschner, 1994).
It should be noted, however, that examination of previous work revealed apparently contradictory results on the role of basal PKC activity on global [Ca2+]i and myogenic tone in arterial smooth muscle. For example, one study (Jarajapu & Knot, 2005) found that 3 μm, but not 500 nm, of the PKC inhibitor bisindolyl-maleimide (BIM) decreased myogenic tone in cerebral arteries. However, [Ca2+]i and PKC activity were not measured. In contrast, we have reported that BIM (500 nm) completely abolished persistent Ca2+ sparklet activity (Navedo et al. 2005). Furthermore, here we found that the PKC inhibitor Gö6976 (200 nm) inhibited PKCα activity and the specific PKCα inhibitory peptide – as well as Gö6976 – lowered [Ca2+]i. Although the reasons for these contradictory findings are unclear, we speculate that variations in basal PKCα activity and the effectiveness of inhibitors used may contribute to these differences. The absence of [Ca2+]i and PKC activity data in the aforementioned studies preclude us from answering these questions and from making direct comparisons between them. By measuring Ca2+ sparklets, global [Ca2+]i, and PKCα activity in wild-type and PKCα−/− arterial myocytes and intact pressurized arteries we provide a comprehensive examination of the role of this kinase on [Ca2+]i in these cells.
To conclude, we have provided the first direct test of the multimodal Ca2+ sparklet model of Ca2+ entry in arterial smooth muscle. Our data lend support to some of the basic features of this model and have revealed important new insights into mechanisms governing Ca2+ influx via Ca2+ sparklets in arterial myocytes. We believe that our data are the first to demonstrate that Ca2+ sparklets modulate local and global [Ca2+]i in arterial smooth muscle. Membrane depolarization increases Ca2+ influx, at least in part, by increasing the duration of low and high activity Ca2+ sparklets. Furthermore, our data suggest that by regulating persistent high nPs Ca2+ sparklet activity, PKCα plays an important role in the regulation of [Ca2+]i in these myocytes. Specifically, we found that PKCα-dependent persistent Ca2+ sparklet activity accounts for, on average, about 50% of the DHP-sensitive component of the global [Ca2+]i in cerebral and mesenteric arterial smooth muscle.
Acknowledgements
This work was supported by NHLBI grants HL085870, HL077115 and HL07828 and the American Heart Association (AHA 0635118 N). We thank Mr V. Scott Votaw for help with image analysis and Ms Jennifer Cabarrus for technical assistance.
Footnotes
-
(Resubmitted 6 November 2006; accepted 30 November 2006; first published online 7 December 2006)
- 2007 The Authors. Journal compilation © 2007 The Physiological Society
References
Figure 1. Simultaneous recordings of Ca2+ sparklets and the underlying ICa in arterial myocytes A, Ca2+ current (ICa), change in total fluorescence intensity (ΔFtotal), and ΔF/F0 records from two representative Ca2+ sparklet sites recorded at −70 mV (a and b, centre traces). The lower and upper traces show the integral of ΔFtotal and ICa, respectively. B, plot of the relationship between Ca2+ sparklets signal mass (i.e. peak ΔFtotal dt) and ΔQCa in arterial myocytes. The continuous line is the best least squares fit to the data using a linear equation y = mx + b, where m is the slope (555 peak ∫ΔFtotaldt units fC−1), and b is the y intercept.
Figure 2. Ca2+ influx is greater via persistent high nPs Ca2+ sparklets than via low nPs Ca2+ sparklets, and increases with membrane depolarization A, ΔFmax traces from representative low and high nPs (where n is the number of quantal levels and Ps is the probability that a quantal Ca2+ sparklet event is active) Ca2+ sparklet sites at −70 (top) and −40 mV (bottom). Amplitude histograms of ΔQCa from low and high nPs Ca2+ sparklet sites at −70 (B) and −40 mV (C). D, scatter plot of ΔQCa amplitudes from low and high nPs Ca2+ sparklet sites at −70 and −40 mV; the red line represents the median value for each column of values.
Figure 3. The duration of high nPs Ca2+ sparklet events is longer than low nPs Ca2+ sparklet events, and increases with membrane depolarization Histograms of the duration of low (top) and high (bottom) nPs Ca2+ sparklet events at −70 (A) and −40 mV (B). The continuous black line in the low nPs histograms is a fit to the data with the single exponential function y = Ae(−x/τ) + yi, where A is the amplitude (at −70 mV = 0.60; at −40 mV = 0.35), τ is the time constant (at −70 mV = 26 ms; at −40 mV = 41 ms), and yi is the y intercept (at −70 mV = 0.009; at −40 mV = 0.009). The continuous black line in the high nPs histograms is a fit to the data with the double exponential function y = A1e(−x/τ1) + A2e(− x/τ2) + yi, where A1 is the amplitude of the first component (at −70 mV = 0.30; at −40 mV = 0.17), A2 is the amplitude of the second component (at −70 mV = 0.05; at −40 mV = 0.003), τ1 is the fast time constant (at −70 mV = 30 ms; at −40 mV = 35 ms), τ2 is the slow time constant (at −70 mV = 75 ms; at −40 mV = 200 ms), and yi is the y intercept (at −70 mV = 0.001; at −40 mV = 0.001).
Figure 4. Ca2+ sparklets modulate local and global [Ca2+]i [Ca2+]i records from a control cell (A), a nifedipine (10 μm)-treated cell (B), and a cell dialysed with 100 μm PKCi (C), while in a nominally Ca2+-free solution and after the introduction of 2 mm Ca2+ external Ca2+. The red traces represent the time course of the spatially averaged, global [Ca2+]i; the black trace in the control cell represents the time course of [Ca2+]i at a Ca2+ sparklet site (i.e. local [Ca2+]i). D, plot of the mean ± s.e.m. steady-state global [Ca2+]i in control, nifedipine-treated, and PKCi-dialysed myocytes.
Figure 5. Basal PKCα activity increases ICa and [Ca2+]i in arterial myocytes A, representative ICa traces (evoked by a voltage pulse from −70 to +30 mV) before (top) and after application of the PKC inhibitor Gö6976 (200 nm; bottom). B, plot of the mean ± s.e.m. peak ICa before and after application of Gö6976. C, representative DHP-sensitive traces (ICa) evoked during a voltage ramp from −70 to +40 mV with a rate of depolarization of 40 mV s−1 before (top) and after application of Gö6976 (bottom). The continuous line marks the zero current level. The bar graph depicts the mean ± s.e.m. of the total charge (pC) associated with the DHP-sensitive ICa recorded under control conditions and after Gö6976. D, global [Ca2+]i records from a typical myocyte (holding potential −40 mV) before and after the application of nifedipine in the absence (i.e. control; top) and presence of Gö6976 (bottom). The bar graph plots the mean ± s.e.m. of the DHP-sensitive component of global Ca2+ in control and Gö6976-treated cells.
Figure 6. PKCα modulates arterial wall [Ca2+]i in intact pressurized arteries A, arterial wall [Ca2+]i in pressurized (80 mmHg) WT and PKCα−/− mesenteric arteries during application of Gö6976 (200 nm) and isradipine (1 μm). The ICa voltage-clamp records shown on the right were obtained from wild-type and PKCα−/− myocytes during step a depolarization from −70 to +30 mV. B, plot of the mean ± s.e.m. of [Ca2+]i under control conditions (baseline) and after Gö6976, isradipine, and Gö6976 + isradipine in wild-type and PKCα−/− arteries. C, plot of the mean ± s.e.m. of arterial wall [Ca2+]i under control conditions (baseline) and after application of Gö6976, isradipine, and Gö6976 + isradipine in WT and PKCα−/− arteries.
