Facilitation of cytosolic calcium wave propagation by local calcium uptake into the sarcoplasmic reticulum in cardiac myocytes

Authors


L. A. Blatter: Department of Molecular Biophysics and Physiology, Rush University Medical Center, 1750 W. Harrison Street, Chicago, IL 60612, USA. Email: lothar_blatter@rush.edu

Key points

  • • Cytosolic calcium (Ca2+) waves result from spontaneous release of Ca2+ from the sarcoplasmic reticulum (SR) Ca2+ store that occurs under Ca2+ overload conditions and can give rise to arrhythmias in the heart. The prevailing paradigm of Ca2+ wave propagation involves cytosolic Ca2+-induced Ca2+ release.
  • • A recent challenge to this paradigm proposed the requirement for an intra-SR ‘sensitization’ Ca2+ wave that primes release activation due to the luminal Ca2+ sensitivity of the release mechanism.
  • • We tested this hypothesis in cardiac myocytes with direct simultaneous high-resolution measurements of cytosolic and intra-SR Ca2+ using fluorescence confocal microscopy.
  • • We found that the increase in cytosolic Ca2+ at the wave front preceded release and depletion of SR Ca2+ in time, and during this latency period a transient increase of SR Ca2+ was observed at individual release sites that gave rise to a propagating intra-SR Ca2+ sensitization wave.
  • • The intra-SR sensitization wave depended on the activity of the sarco-endoplasmic reticulum Ca2+-ATPase (SERCA) and occurred by a mechanism where Ca2+ uptake by SERCA at the wave front facilitates propagation of cytosolic Ca2+ waves via luminal sensitization of the release mechanism, thus supporting a novel paradigm of a ‘fire-diffuse-uptake-fire’ mechanism for Ca2+ wave propagation.

Abstract  The widely accepted paradigm for cytosolic Ca2+ wave propagation postulates a ‘fire-diffuse-fire’ mechanism where local Ca2+-induced Ca2+ release (CICR) from the sarcoplasmic reticulum (SR) via ryanodine receptor (RyR) Ca2+ release channels diffuses towards and activates neighbouring release sites, resulting in a propagating Ca2+ wave. A recent challenge to this paradigm proposed the requirement for an intra-SR ‘sensitization’ Ca2+ wave that precedes the cytosolic Ca2+ wave and primes RyRs from the luminal side to CICR. Here, we tested this hypothesis experimentally with direct simultaneous measurements of cytosolic ([Ca2+]i; rhod-2) and intra-SR ([Ca2+]SR; fluo-5N) calcium signals during wave propagation in rabbit ventricular myocytes, using high resolution fluorescence confocal imaging. The increase in [Ca2+]i at the wave front preceded depletion of the SR at each point along the calcium wave front, while during this latency period a transient increase of [Ca2+]SR was observed. This transient elevation of [Ca2+]SR could be identified at individual release junctions and depended on the activity of the sarco-endoplasmic reticulum Ca2+-ATPase (SERCA). Increased SERCA activity (β-adrenergic stimulation with 1 μm isoproterenol (isoprenaline)) decreased the latency period and increased the amplitude of the transient elevation of [Ca2+]SR, whereas inhibition of SERCA (3 μm cyclopiazonic acid) had the opposite effect. In conclusion, the data provide experimental evidence that local Ca2+ uptake by SERCA into the SR facilitates the propagation of cytosolic Ca2+ waves via luminal sensitization of the RyR, and supports a novel paradigm of a ‘fire-diffuse-uptake-fire’ mechanism for Ca2+ wave propagation in cardiac myocytes.

Abbreviations 
AP

action potential

[Ca2+]i

cytosolic free calcium concentration

[Ca2+]SR

intra-sarcoplasmic reticulum free calcium concentration

CICR

calcium-induced calcium release

CPA

cyclopiazonic acid

ISO

isoproterenol

jSR

junctional sarcoplasmic reticulum

njSR

non-junctional sarcoplasmic reticulum

RyR

ryanodine receptor

SERCA

sarco-endoplasmic reticulum Ca2+-ATPase

SR

sarcoplasmic reticulum

Introduction

Under certain conditions such as Ca2+ overload, spontaneous Ca2+ release from the sarcoplasmic reticulum (SR) has been shown to propagate as regenerative Ca2+ waves in cardiac cells (Wier et al. 1987; Wier & Blatter, 1991; Cheng et al. 1996). Spontaneous Ca2+ waves are cellular events which, in the intact heart, are known to trigger lethal arrhythmias (Stern et al. 1988; Wakayama et al. 2005; Chelu & Wehrens, 2007). The current model of Ca2+ wave propagation is based on the mechanism of calcium-induced calcium release (CICR) from the ryanodine receptor (RyR) Ca2+ release channel and is also known as the ‘fire-diffuse-fire’ model (Keizer & Smith, 1998; Keizer et al. 1998). In this model a cluster of activated RyRs of the junctional SR (jSR) release Ca2+ (‘fire’) that diffuses through the cytosol to an adjacent neighbouring SR junction (‘diffuse’) where it is able to activate Ca2+ release (‘fire’) from this cluster of RyRs based on the receptor's intrinsic sensitivity to activation by cytosolic Ca2+. However, the open probability of the RyR is also determined by luminal Ca2+ (Györke & Györke, 1998; Györke et al. 2002) and the importance of luminal control of RyR Ca2+ release kinetics in cardiac myocytes under normal and pathological conditions is well documented (Shannon et al. 2000; Zima et al. 2008a,b; Domeier et al. 2009, 2010). The recognition of the importance of RyR regulation by luminal Ca2+ has led to the emergence of an amendment to the ‘fire-diffuse-fire’ paradigm for wave propagation. In this more comprehensive model, it is proposed that cytosolic Ca2+ wave propagation is in part driven by a ‘RyR sensitization’ wave front that moves through the SR, thereby luminally priming the RyR for activation by cytosolic Ca2+ (Keller et al. 2007). A key feature of this model is the necessity of SERCA activity for Ca2+ uptake into the SR to create local increases in [Ca2+]SR that act in tandem with the increases in [Ca2+]i allowing for facilitation of wave propagation (Keller et al. 2007). Furthermore, a computational model of [Ca2+]i and [Ca2+]SR kinetics during spontaneous Ca2+ wave propagation has lent support to the feasibility of such a mechanism (Ramay et al. 2010); however, direct experimental proof has yet to be established.

In this study, we tested the intra-SR sensitization wave hypothesis by direct simultaneous measurements of cytosolic ([Ca2+]i) and intra-SR ([Ca2+]SR) calcium signals during wave propagation in intact rabbit ventricular myocytes. [Ca2+]i and [Ca2+]SR were measured with the fluorescent probes rhod-2 and fluo-5N, respectively, using high-resolution confocal imaging. In summary, the increase in [Ca2+]i at the wave front preceded release of Ca2+ and depletion of the SR in time. During this latency period a transient increase of [Ca2+]SR was observed that could be identified at individual release junctions and depended on the activity of SERCA. Our data provide experimental evidence that local Ca2+ uptake by SERCA into the SR facilitates the propagation of cytosolic Ca2+ waves via luminal sensitization of the RyR. The data are consistent with the intra-SR RyR sensitization wave hypothesis (Keller et al. 2007) and support a novel paradigm of a ‘fire-diffuse-uptake-fire’ mechanism for Ca2+ wave propagation in cardiac myocytes.

Part of this work has been presented previously in abstract form (Maxwell & Blatter, 2012).

Methods

Solutions and chemicals

All chemicals and reagents were obtained from Sigma-Aldrich (St Louis, MO, USA), unless noted otherwise. Fluorescent calcium indicator dyes were from Molecular Probes/Life Technologies, Grand Island, NY, USA). Tyrode solution contained (in mm): 130 NaCl, 4 KCl, 7 CaCl2, 1 MgCl2, 10 d-glucose, 10 Hepes; pH 7.4 with NaOH. Isoproterenol (isoprenaline) (ISO) was dissolved in water and diluted to the working concentration (1 μm) in Tyrode solution. Cyclopiazonic acid (CPA) was dissolved in dimethyl sulfoxide and diluted to the working concentration (3 μm) in Tyrode solution. Ca2+ waves were recorded immediately after application of CPA, as this resulted in the abolishment of waves within approximately 45 s. All experiments were performed in elevated extracellular [Ca2+] (7 mm) to cause SR Ca2+ overload and induce spontaneous Ca2+ waves. To prevent movement of cells during recording 2,3-butanedione monoxime (10 mm) and blebbistatin (10 μm) were included in the Tyrode solution during all experiments.

Myocyte isolation

Ventricular myocytes were isolated from New Zealand White rabbits (14 animals; 2.5 kg, Myrtle's Rabbitry, Thompsons Station, TN, USA) (Domeier et al. 2009). Rabbits were anaesthetized with sodium pentobarbital (50 mg kg−1) and hearts were excised and mounted on a Langendorff apparatus. Hearts were retrogradely perfused with nominally Ca2+-free Tyrode solution for 5 min, followed by minimal essential medium Eagle (MEM) solution containing 20 μm Ca2+ and 45 μg ml−1 Liberase Blendzyme TH (Roche Applied Science, Indianapolis, IN, USA) for 20 min at 37°C. The left ventricular free wall was removed and digested for an additional 5 min in the enzyme solution at 37°C. Digested tissue was then minced, filtered and washed in a MEM solution containing 50 μm Ca2+ and 10 mg ml−1 bovine serum albumin. Isolated cells were kept in MEM solution with 50 μm Ca2+ at room temperature (22–24°C) until indicator dye loading and subsequent experimentation. All protocols were approved by the Institutional Animal Care and Use Committee of Rush University Chicago, and comply with US and UK regulations on animal experimentation (Drummond, 2009).

[Ca2+] measurements

To directly monitor [Ca2+]SR the SR was loaded with the low-affinity Ca2+ indicator fluo-5N by incubation of ventricular myocytes with 10 μm of the membrane permeant fluo-5N/AM together with 0.25% Pluronic F-127 in nominally Ca2+-free Tyrode solution for 2.5 h, followed by a 30 min wash, all at 37°C. To monitor cytosolic Ca2+ waves myocytes preloaded with fluo-5N were subsequently loaded with the spectrally distinct [Ca2+]i indicator rhod-2/AM (5 μm) for 10 min, followed by a 10 min wash.

Confocal microscopy (Nikon A1R, Nikon Corporation, Melville, NY, USA) was used to image [Ca2+]SR (fluo-5N) and [Ca2+]i (rhod-2). For simultaneous recording of [Ca2+]SR and [Ca2+]i, fluo-5N was excited at 488 nm and emission collected at 515 ± 15 nm, while rhod-2 was excited at 543 nm and emission collected at wavelengths >600 nm. Control experiments revealed no cross-talk between the two fluorescence channels. Ca2+ wave measurements were acquired from intact myocytes during rest from 0.8–1.5 Hz stimulation in line scan mode at 512 lines s−1 using a ×60 oil-immersion objective lens (NA = 1.49). The scan line was placed along the longitudinal axis of the cell with a pixel size of 0.15–0.24 μm. The empirically measured point-spread function of the system had full-width at half-maximum (FWHM) values of 0.28, 0.29 and 0.7 μm in the x, y and z (axial) dimensions.

Action potentials (APs) and global Ca2+ transients were elicited by electrical field stimulation of intact ventricular myocytes using a pair of platinum electrodes (voltage set at ∼50% above the threshold for contraction). Experiments were conducted at room temperature (22–24°C).

Data presentation

Line scan data and fluorescence traces are presented as individual observations representative of multiple recordings or as the average of the indicated number of individual recordings. Fluorescence traces were background subtracted and plotted as F/F0, where F0 is the initial resting fluorescence. Fluorescence traces were recorded from <0.5 μm wide regions and represent spatial averages from 2 or 3 neighbouring pixels in a line scan image (pixel sizes 0.15–0.24 μm). No additional filtering in space or time was applied to recordings from individual SR junctions. Summary data are presented as the mean ± SEM of n measurements. Statistical comparisons between groups were performed with Student's t test. Differences were considered statistically significant at P < 0.05.

Results

Latency between increase of [Ca2+]i and SR Ca2+ depletion during cytosolic Ca2+ waves: evidence for an intra-SR Ca2+ sensitization wave from single SR junctions

To test the intra-SR Ca2+ sensitization wave hypothesis (Keller et al. 2007; Ramay et al. 2010) for cytosolic Ca2+ wave propagation, SR Ca2+ was monitored directly using the low-affinity Ca2+ indicator fluo-5N entrapped within the SR, while [Ca2+]i was simultaneously imaged with the spectrally distinct Ca2+ dye rhod-2. We have used this method successfully and extensively for simultaneous monitoring of [Ca2+]i and [Ca2+]SR in cardiac myocytes (Zima et al. 2008b, 2010; Domeier et al. 2009, 2010; Bovo et al. 2011; Picht et al. 2011). Longitudinal line scans of ventricular cardiomyocytes provided visualization of single SR junctions as narrow bands of high fluorescence, making it possible to distinguish between the jSR and the adjacent non-junctional SR (njSR) that appeared as regions of lower intensity between the bright SR junctions. The line scan images in Fig. 1 show cytosolic [Ca2+]i and [Ca2+]SR during an electrically induced action potential (Fig. 1A) and during a spontaneous Ca2+ wave (Fig. 1B). Local fluorescence intensity profiles representing individual SR junctions were recorded from subcellular, 0.5 μm wide cytosolic ([Ca2+]i) and the corresponding jSR ([Ca2+]SR) regions (labelled a, b and c) and are presented as normalized (F/F0) [Ca2+]i and [Ca2+]SR signals. As expected, during an electrically induced AP the rise in [Ca2+]i at a particular junction coincided exactly with the beginning of the decline of [Ca2+]SR (dotted vertical line in Fig. 1Aa–c) at that same junction. Furthermore, the beginning of SR Ca2+ depletion and the rise of [Ca2+]i was homogeneous and synchronized throughout the entire myocyte. In contrast, [Ca2+]i and [Ca2+]SR measurements during a spontaneous Ca2+ wave revealed two interesting and related features of Ca2+ wave propagation not seen during AP-induced Ca2+ transients (Fig. 1B). First, at any particular junction [Ca2+]i rose prior to the decline in [Ca2+]SR resulting from activation of SR Ca2+ release, giving rise to a ‘latency’ period between cytosolic Ca2+ wave onset (line 1) and the decline in [Ca2+]SR (line 2). On average this latency period was 33.5 ± 4.0 ms under control conditions (n= 11; cf. Fig. 3B). Second, during this latency period a transient increase in local junctional [Ca2+]SR was observed (bottom traces in Fig. 1Ba–c).

Figure 1.

Comparison of [Ca2+]i and [Ca2+]SR signals during AP-induced Ca2+ transient and spontaneous Ca2+ wave at individual SR junctions 
Top: line scan images (x–t) of rhod-2 and fluo-5N fluorescence during an electrically induced action potential-dependent global Ca2+ transient (A) and during a spontaneous Ca2+ wave (B). Fluorescence signals were measured from ≤0.5 μm wide regions delineated by the black boxes labelled a, b and c and correspond to the normalized (F/F0) local single junction [Ca2+]i and [Ca2+]SR traces below the line scan images. Cells were stimulated at 1 Hz for recording the AP-dependent Ca2+ transient. The interval between lines 1–2 in (B) is referred to as the latency period, and also delineates the intra-SR sensitization Ca2+ transients. Here and in the following figures the horizontal dashed lines represent F/F0= 1.

Figure 3.

Changes in SERCA activity determine the amplitude of sensitization Ca2+ wave and duration of latency period 
A, simultaneous single junction [Ca2+]i (top) and [Ca2+]SR (bottom) measurements during a spontaneous Ca2+ wave after treatment with 1 μm isoproterenol (ISO; activation of SERCA) or 3 μm cyclopiazonic acid (CPA; inhibition of SERCA). ISO was applied 5 min prior to recording. Vertical lines mark the beginning of the increase in junctional [Ca2+]SR and the start of release. Summary graphs of latency periods (B) and amplitude of the sensitization Ca2+ transient (C) for untreated cells (CTL), and after ISO or CPA treatment. *P < 0.001 vs. CTL; #P < 0.001 vs. ISO. CTL: n= 11; ISO: n= 6; CPA: n= 5.

The Ca2+ dynamics during wave propagation shown in Fig. 1 were further analysed in Fig. 2. Here, simultaneous measurements of [Ca2+]i and [Ca2+]SR from two neighbouring release junctions (J1 and J2) during a spontaneous Ca2+ wave propagating from junction J1 (top) to junction J2 (bottom) are shown. Temporal alignment of the [Ca2+] traces from these two junctions show that a sequence of release, diffusion, and uptake can be identified during Ca2+ wave propagation. At line 1 [Ca2+]i begins to increase at J1 presumably due to diffusion from the previously activated neighbouring junction (J0). It is important to note that at this point no measureable Ca2+ release (see [Ca2+]SR signal at J1) has occurred at J1 that could have contributed to the local increase of [Ca2+]i at J1. This also marks the beginning of the latency period (as defined above) during which [Ca2+]SR transiently increased at J1. The latency period and associated increase in [Ca2+]SR continued until a measurable activation of Ca2+ release at this junction occurred (line 2) which marked the end of the latency period at J1 and coincided with the rise of [Ca2+]i at the next junction (J2) along the wave front (line 2, bottom panel). The rise of [Ca2+]i at J2 was brought upon by cytosolic diffusion of Ca2+ released from J1, leading to a local elevation of [Ca2+]i at J2 prior to activation of release at J2 (marked by line 3). During the latency period at J2[Ca2+]SR transiently increased as observed before at J1. Analogously, the wave continues to propagate from one junction to the next (J2→J3→…→Jn). The observed latency periods at J1 and J2 are remarkably similar as is the amplitude of the transient elevation of [Ca2+]SR, suggesting that a common mechanism is regulating release (and sensitization of release) at each junction along the wave front.

Figure 2.

Latency period and sensitization Ca2+ transient at two neighbouring SR junctions during spontaneous Ca2+ wave propagation 
Single junction fluorescence traces of [Ca2+]i and [Ca2+]SR during a propagating spontaneous Ca2+ wave measured simultaneously at two neighbouring SR junctions (J1 and J2). The wave propagates from J1 to J2. Vertical lines: 1, start of increase of [Ca2+]i and [Ca2+]SR at J1; 2, start of Ca2+ release and decrease of [Ca2+]SR at J1 and start of increase of [Ca2+]i and [Ca2+]SR at J2; 3: start of Ca2+ release and decrease of [Ca2+]SR at J2.

In conclusion, during cytosolic Ca2+ wave propagation, the cytosolic wave front is paralleled by a propagating intra-SR Ca2+ wave that is followed, with a latency of ∼35 ms, by the SR Ca2+ depletion wave. These results are consistent with an intra-SR Ca2+ sensitization wave that primes RyRs of individual release junctions to cytosolic CICR due to the inherent luminal Ca2+ sensitivity of the RyR. However, the question remains of what was the source of Ca2+ that was responsible for the transient rise of junctional [Ca2+]SR prior to release. The observation that [Ca2+]SR could increase at a given SR junction (J2, between 2 and 3) while at the same time Ca2+ was released and [Ca2+]SR decreased at the neighbouring junction (J1) argued against intra-SR Ca2+ diffusion as Ca2+ would have to diffuse against its concentration gradient. An alternative possibility, originally suggested by Keller et al. (2007), would be local sequestration of cytosolic Ca2+ at the wave front by SERCA. Therefore, in the next set of experiments we tested whether latency and magnitude of the intra-SR Ca2+ sensitization wave depended on SERCA activity.

SERCA-dependence of latency and magnitude of intra-SR sensitization wave

To test the role of SERCA in the transient increase of [Ca2+]SR and the latency period during wave propagation we pharmacologically altered the activity of SERCA. The β-adrenergic agonist isoproterenol (ISO) was used to increase SERCA activity which occurs through phosphorylation of the inhibitory protein phospholamban (Inui et al. 1986), while SERCA activity was reduced with the inhibitor CPA. Figure 3A shows representative [Ca2+]i and [Ca2+]SR measurements from single SR junctions in cells treated with ISO alone (left) or with ISO followed by exposure to CPA (right). The use of CPA to reduce SERCA activity required the previous exposure to ISO because in cells treated with CPA only spontaneous waves were abolished immediately. The application of 1 μm ISO to cells resulted in a decrease in the latency of SR depletion (Fig. 3B) and an increased amplitude of the sensitization transient (Fig. 3C) compared to untreated cells. Following ISO treatment CPA (3 μm) was applied. This resulted in an increase in the latency of SR depletion (Fig. 3B) and a decrease in the amplitude of the sensitization transient (Fig. 3C) compared to cells treated with ISO only. Due to the necessity to apply CPA immediately after ISO treatment to enable a 30–45 s time window where SERCA could be blocked partially but the remaining effects of the ISO treatment sustained Ca2+ wave activity, the observed amplitude of the sensitization transient remained higher than control. Because of these experimental constraints the amplitude was statistically compared to the ISO treatment in Fig. 3C.

To better illustrate the relatively small increases in junctional [Ca2+]SR seen prior to activation of SR Ca2+ release, normalized fluorescence traces from multiple SR junctions were averaged together to create the fluo-5N fluorescence traces shown in Fig. 4A and B. Figure 4A illustrates average normalized fluorescence traces from 6 jSR regions and their neighbouring 12 njSR regions. These traces show that the sensitization transient is only observed or resolved in the jSR regions, but not in the njSR regions less than 1 μm apart. This observation represents an important control experiment because it illustrates that the observed sensitization transient is not an artifact of cell motion as the wave propagates. Of note is the average wave collision trace shown in Fig. 4B (right and inset with line scan image of fluo-5N fluorescence). [Ca2+]SR was recorded at junctions where two spontaneous waves collided and annihilated. Interestingly, it shows nearly a doubling of the amplitude of the sensitization transient indicating that these particular junctions sequester Ca2+ from two approaching wave fronts.

Figure 4.

Average traces of junctional [Ca2+]SR
A, left: average trace from 6 individual junctional [Ca2+]SR (jSR) recordings under control conditions during Ca2+ wave propagation. Right: average non-junctional [Ca2+]SR (njSR) recordings obtained from both sides adjacent to the SR junctions at a distances of <1 μm (average of 12 recordings). Inset: fluo-5N line scan image revealing 3 individual SR junctions as bright fluorescence with non-junctional SR regions in between. B, average recordings of junctional [Ca2+]SR under control (CTL) conditions, after treatment with 1 μm ISO or 3 μm CPA, and from junctions where waves collided (line scan image of fluo-5N fluorescence during wave collision is shown on the right). Shaded areas mark the beginning of the increase in junctional [Ca2+]SR and the start of Ca2+ release. C, average wave propagation velocities under control conditions and in the presence of ISO, CPA and caffeine (250 μm). D, average rate of rise of [Ca2+]SR (d(ΔF/F0)/dt) during sensitization wave. CTL: n= 11; ISO: n= 6; CPA: n= 5; wave collision: n= 3. *P < 0.001 vs. CTL. #P < 0.05 vs. CTL.

The bottom panels of Fig. 4B show average results of ISO and CPA treatment on latency and sensitization transient amplitude. Furthermore, an inverse correlation between the latency period and the wave velocity (control wave velocity: 97.5 ± 3.9 μm s−1; n= 11) was found with decreased latency during faster waves after ISO treatment (134.6 ± 3.1 μm s−1; n= 6) and increased latency during slowed waves after CPA treatment (89.6 ± 1.1 μm s−1; n= 5) (Fig. 4C). Also included in this graph is the wave velocity during exposure to the RyR-sensitizing agent caffeine (250 μm) (O’Neill & Eisner, 1990; Porta et al. 2011). Under this condition, no sensitization transients were observed, presumably due to the increased sensitivity of the RyRs to CICR and the elevation of [Ca2+]i at the wave front being sufficient to drive wave propagation. This treatment resulted in the slowest wave propagation (70.2 ± 8.6 μm s−1; n= 5). Figure 4D shows the effect of ISO and CPA on the rate of rise of the intra-SR sensitization Ca2+ transient. During ISO stimulation [Ca2+]SR rose significantly faster consistent with an ISO-dependent stimulation of SERCA and increased uptake of Ca2+ into the SR which was completely abolished by CPA.

In support of the central role of SERCA activity for the generation of the intra-SR sensitization wave and the efficacy of Ca2+ wave propagation, the rate of rise of [Ca2+]SR during the sensitization transient increased nearly 4-fold when SERCA was activated by ISO (Fig. 4D), whereas CPA treatment decreased the rate to slightly below control levels.

In summary, our data show that SERCA activity is responsible for the intra-SR sensitization Ca2+ wave, and provide support to the notion that local Ca2+ uptake by SERCA into the SR facilitates the propagation of spontaneous Ca2+ waves in cardiac myocytes. These data provide support for a ‘fire-diffuse-uptake-fire’ model of facilitation of Ca2+ wave propagation.

Discussion

The data presented here show that SR Ca2+ uptake by SERCA is an integral part of the mechanism by which spontaneous Ca2+ waves propagate through cardiac myocytes. We provide evidence of a latency period between the rise in [Ca2+]i at the wave front and the activation of SR Ca2+ release at the same subcellular location. During this latency period [Ca2+]SR transiently increases and sensitizes the RyR from the luminal compartment to activation by cytosolic Ca2+. This transient elevation of [Ca2+]SR propagates as an intra-SR Ca2+ sensitization wave synchronized and in parallel with the cytosolic wave front. Manipulation of SERCA activity – stimulation by ISO treatment or inhibition with CPA – decreased or increased the latency period, respectively, and changed the amplitude of the sensitization wave. The data not only attest to a central role of SERCA for wave propagation as proposed by Keller et al. (2007), the results also reconcile the inherently low cytosolic Ca2+ sensitivity of the RyR (Meissner & Henderson, 1987; Cannell & Soeller, 1997) with the experimental data that [Ca2+]i rarely exceeds 1 μm at the wave front (Lipp & Niggli, 1993) by providing a mechanism by which luminal sensitization of the RyR to activation by cytosolic Ca2+ facilitates the propagation of spontaneous Ca2+ waves in cardiac myocytes.

If Ca2+ waves propagate only by CICR, changing the activity of SERCA should have no immediate effect on their propagation velocity (unless changes in SERCA activity lead to abrupt and significant changes of SR Ca2+ load and cytosolic [Ca2+] which in turn would affect Ca2+ wave probability and propagation velocity in opposite directions). Indeed, reports of the effect of regulators of SERCA activity on Ca2+ wave propagation and velocity are contradictory (Takamatsu & Wier, 1990; Lukyanenko et al. 1999; O’Neill et al. 2004; Keller et al. 2007). Nonetheless, rapid inhibition of SERCA slowed wave propagation immediately indicating that this effect is independent of any secondary effects of SERCA inhibition such as changes of SR Ca2+ load (Keller et al. 2007). Indeed, the abolishment of Ca2+ waves after exposure to the SERCA blocker clearly preceded a measurable decrease in [Ca2+]SR by tens of seconds (Zima et al. 2010; Bode et al. 2011), lending further support to the SERCA-dependent Ca2+ sensitization wave hypothesis. Additionally, in a SERCA2 knockout mouse wave development was shown to be less frequent and propagation velocity decreased compared to normal mice (Stokke et al. 2010). Our data demonstrate a clear reciprocal relationship between SERCA activity and wave velocity. Through luminal sensitization of RyRs to CICR, conditions that increase SERCA activity will increase the rate of rise of [Ca2+]SR (MacQuaide et al. 2007; see also Fig. 4D) and decrease the latency, and thus would be expected to increase wave propagation velocity, and vice versa. Our results using ISO and CPA to increase or decrease the activity of SERCA, respectively, support this hypothesis experimentally and suggest that the mechanism by which SERCA is able to affect wave velocity is the efficiency at which it is able to sequester Ca2+ into the SR from the cytosolic Ca2+ wave front and to build up the intra-SR Ca2+ sensitization wave.

Interestingly, our experimental results match predictions from a computational model of local SR Ca2+ dynamics during Ca2+ wave propagation in ventricular myocytes (Ramay et al. 2010). Specifically, the simulations from this computational model resulted in three important predictions that are in agreement with our experimental results: (1) local increases in [Ca2+]SR can occur ahead of a propagating wave; (2) these local increases do not result from diffusion within the SR, rather from Ca2+ released from an activated junction that diffuses through the cytosol to the next junction where it is sequestered into the SR by SERCA; (3) partial inhibition of SERCA attenuates these local increases in [Ca2+]SR potentially accounting for the slower Ca2+ wave propagation seen with SERCA inhibitors.

Our data show that a latency period and the transient increase in [Ca2+]SR that takes place during it are novel characteristics of spontaneous Ca2+ wave propagation; however, the question remains whether this increase in [Ca2+]SR and resulting luminal sensitization of the RyR represents an absolute requirement for the propagation of a Ca2+ wave. When low-dose (250 μm) caffeine was used to sensitize the RyR to cytosolic Ca2+ activation, Ca2+ waves were observed in the absence of any (detectable) sensitization Ca2+ transient, suggesting that under conditions of heightened sensitivity of the RyR to Ca2+, elevated [Ca2+]i may be sufficient to drive propagation, albeit at a slower velocity.

Our results presented here add to the growing list of studies that provide evidence for the importance of dynamic local changes in [Ca2+]SR in regulating SR Ca2+ release (reviewed in Sobie & Lederer, 2012). Many studies have shown the importance of luminal Ca2+-dependent changes in RyR gating in the regulation of Ca2+ release activation, termination, and refractoriness, and have shown that alterations and defects in this mechanism lead to spontaneous Ca2+ waves which play a role in arrhythmogenic activity (Berlin et al. 1989; Györke & Terentyev, 2008). In conclusion, several factors – including sensitization of CICR to [Ca2+]i, kinetics of refractoriness of SR Ca2+ release and the inherent larger rate of Ca2+ release at higher [Ca2+]SR (we estimate the change in SR Ca2+ during spontaneous Ca2+ waves to have a depletion amplitude of ∼400 μm based on our previous reports (Zima et al. 2008b; Domeier et al. 2009, 2010; Bovo et al. 2011) – allow for propagating Ca2+ waves in cardiac myocytes. Here, we demonstrate experimentally the importance of an additional factor consisting of the activity of the SERCA pump that sequesters additional Ca2+ into the SR at the wave front, generates an intra-SR Ca2+ sensitization wave that propagates in parallel with the cytosolic Ca2+ wave and sensitizes RyRs to CICR. We therefore propose that the widely accepted ‘fire-diffuse-fire’ paradigm of Ca2+ wave propagation be amended to a ‘fire-diffuse-uptake-fire’ model to account for the central role of SERCA in Ca2+ wave propagation.

Appendix

Author contributions

J.T.M. and L.A.B. contributed to the conception and design of the experiments, analysis and interpretation of the data, and writing of the article. Both authors have approved the final version of the manuscript. J.T.M. performed the experimental work.

Acknowledgements

We wish to thank Dr Seth Robia (Loyola University Medical Center) for helpful discussion, and Dr Vyacheslav Shkryl for help with the point-spread function measurements. This work was supported by National Institutes of Health Grants HL62231, HL80101 and HL101235, and the Leducq Foundation.

Ancillary