This article describes the detailed protocol and equipment necessary for dual optical mapping of transmembrane potential (Vm) and free intra-sarcoplasmic reticulum (SR) Ca2+ in the Langendorff-perfused rabbit heart. This method allows for direct observation and quantification of Vm and SR Ca2+ dynamics in the intact heart.
Sarcoplasmic reticulum (SR) Ca2+ handling plays a key role in normal excitation-contraction coupling and aberrant SR Ca2+ handling is known to play a significant role in certain types of arrhythmia. Because arrhythmias are spatially distinct, emergent phenomena, they must be investigated at the tissue level. However, methods for directly probing SR Ca2+ in the intact heart remain limited. This article describes the protocol for dual optical mapping of transmembrane potential (Vm) and free intra-SR [Ca2+] ([Ca2+]SR) in the Langendorff-perfused rabbit heart. This approach takes advantage of the low-affinity Ca2+ indicator Fluo-5N, which has minimal fluorescence in the cytosol where intracellular [Ca2+] ([Ca2+]i) is relatively low but exhibits significant fluorescence in the SR lumen where [Ca2+]SR is in the millimolar range. In addition to revealing SR Ca2+ characteristics spatially across the epicardial surface of the heart, this approach has the distinct advantage of simultaneous monitoring of Vm, allowing for investigations into the bidirectional relationship between Vm and SR Ca2+ and the role of SR Ca2+ in arrhythmogenic phenomena.
Dual optical mapping of intracellular Ca2+ and transmembrane potential (Vm) in the intact Langendorff-perfused heart has become a mainstay of investigations in cardiac electrophysiology, including mechanisms of arrhythmia and excitation-contraction coupling1-4. This approach has provided unprecedented knowledge into normal and abnormal electrophysiology and, importantly, into the bidirectional relationship between Vm and intracellular Ca2+. However, optical mapping of intracellular Ca2+ with high-affinity fluorescent indicators (such as Rhod-2 and Fluo-4) only reports on bulk changes in intracellular Ca2+ and is unable to distinguish whether these changes are due to transmembrane Ca2+ flux, release and reuptake into intracellular stores, or in most instances, some combination of both. Furthermore, high-affinity Ca2+ indicators have slow on-off kinetics and may not accurately report rapid changes in Ca2+ concentration5.
Each action potential triggers a rise in intracellular Ca2+, known as the intracellular Ca2+ transient (CaT). In the mammalian heart, approximately 70 – 90% of the total CaT is due to release of Ca2+ from the sarcoplasmic reticulum (SR) via opening of ryanodine receptors (RyRs)6. Within the SR, approximately half of the total Ca2+ is bound to calsequestrin (CSQ) and other intra-SR buffers7, which play an important role in SR Ca2+ homeostasis8,9. The amount of free SR Ca2+ dictates the driving force for SR Ca2+ release as well as gating of RyR, and therefore has a significant impact on the intracellular CaT. Furthermore, alterations in SR Ca2+ release or reuptake can, in turn, impact Vm via the electrogenic Na+-Ca2+ exchange, which may have arrhythmogenic consequences. Therefore, in addition to the CaT, monitoring of free SR Ca2+ can provide important insights into contractile and electrophysiological dysfunction.
Over the past several years, investigators have made significant advances in the monitoring of SR Ca2+ in isolated cardiac myocytes and from a single location on the intact heart. One such method requires rapid pulses of caffeine to open RyRs and the SR Ca2+ content is then inferred or calculated from the immediate rise in intracellular Ca2+10. Another intriguing approach uses low-affinity Ca2+ indicators, such as Fluo-5N11 or Mag-Fluo412, which bind to free SR Ca2+. These indicators have dissociation constants (Kd) in the range of 10 – 400 μM and therefore exhibit minimal fluorescence in the cytosol compared to the SR lumen, where the Ca2+ concentration ([Ca2+]SR) is in the millimolar range. Using low-affinity Ca2+ indicators, several aspects of SR Ca2+ cycling have been investigated at the level of the isolated myocyte, including fractional SR Ca2+ release and the mechanisms of Ca2+ alternans13,14. However, in order to fully understand the heterogeneous nature of SR Ca2+ cycling in the intact heart and the role of SR Ca2+ in spatially distinct arrhythmic phenomena, methods for imaging SR Ca2+ across the epicardial surface of the intact heart are required15.
This article describes methodology for dual optical mapping of free SR Ca2+ and Vm in the intact Langendorff-perfused rabbit heart with the low-affinity Ca2+ indicator Fluo-5N. In addition to revealing SR Ca2+ characteristics spatially across the epicardial surface of the heart, this approach has the advantage of simultaneous monitoring of Vm, allowing for investigations into the bidirectional relationship between Vm and SR Ca2+.
All procedures involving animals were approved by the Animal Care and Use Committee of the University of California, Davis, and adhered to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health.
1. Preparation
2. Harvesting, Perfusion, and Dye-loading of Rabbit Heart
3. Optical Mapping
Figure 1A shows a schematic diagram of the optical configuration for dual Vm and SR Ca2+ mapping. With this setup, there is complete spectral separation of the Vm and SR Ca2+ signals (Figure 1B). A diagram of the dual-loop perfusion system used for Fluo-5N dye loading is illustrated in Figure 1C. Figure 1D shows the horizontal orientation of the heart in the perfusion dish. Representative Vm and SR Ca2+ optical traces during continuous pacing from different locations on the epicardial surface of the heart are shown in Figure 2A. The optical field of view with this setup is 31 mm x 31 mm and the orientation of the heart within the field of view is shown in Figure 2B. Activation maps are also shown in Figure 2B and are constructed by selecting the activation time of each pixel and plotting this value on an isochronal map (note the typical delay of ~ 8 – 10 msec between the Vm and SR Ca2+ activation maps). Importantly, key action potential (AP) properties such as AP rise time (from 10% to 90% of the amplitude [Trise]), and AP duration at 80% repolarization (APD80), are not statistically different between hearts loaded with RH237 and Fluo-5N compared to those loaded with RH237 and Rhod-2 (Trise: 14.9 msec ± 1.9 msec vs. 14.2 msec ± 1.2 msec and APD80: 155.3 msec ± 5.8 msec vs. 156.9 msec ± 5.7 msec for Fluo-5N and Rhod-2, respectively; p >0.05 for both; n = 8/group), indicating minimal effect of Fluo-5N or dye loading on the AP. Figure 2C demonstrates the ability to observe and quantify relative changes in diastolic SR Ca2+ load and systolic SR Ca2+ release amplitudes at various pacing cycle lengths. Only relative changes in these parameters can be quantified, as absolute calibration ([Ca2+]SR) of the signal in the intact heart is not possible.
This approach is particularly useful for characterizing and quantifying pathological SR Ca2+ handling, as is shown in Figure 3. Following application of the β-adrenergic receptor agonist isoproterenol (100 nM), clear diastolic Ca2+ leak from the SR is observed (red arrows, Figure 3A). If the leak is large enough, Ca2+-mediated triggered activity may occur (Figure 3B).
Figure 1. System Setup for Dual Mapping of Vm and SR Ca2+. (A) Diagram of the optical setup, including necessary filters and dichroic mirror for proper spectral separation. (B) Imaging of hearts loaded with either Fluo-5N only (i) or RH237 only (ii) indicates complete spectral separation of the Vm and SR Ca2+ fluorescent signals. Ex: excitation, Em: emission. (C) Diagram of the dual-loop perfusion system for normal perfusion and Fluo-5N dye loading. (D) Image of a cannulated heart and ECG electrodes in the perfusion dish. Panels A and B reproduced with permission from Wang et al.15. Please click here to view a larger version of this figure.
Figure 2. Vm and SR Ca2+ Optical Traces during Continuous and Non-continuous Pacing. (A) Representative Vm and SR Ca2+ optical traces during continuous pacing at a cycle length of 300 msec. Signals are shown from the individual locations indicated in (B). Note the delay of ~ 8 – 10 msec between the Vm and SR Ca2+ activation times as is typical for normal excitation-contraction coupling. (B) Image of a heart within the mapping field of view (white arrowhead indicates pacing electrodes) and activation maps of Vm and SR Ca2+. (C) SR Ca2+ optical traces demonstrating changes in diastolic SR Ca2+ load and the amplitude of SR Ca2+ release following abrupt changes in pacing rate. All traces start with pacing at CL = 300 msec. Pacing rate is then abruptly changed to CL = 500 msec (i), CL = 400 msec (ii), or CL = 250 msec (iii). When longer CLs are initiated (i – ii), a larger SR Ca2+ release first occurs (due to the longer diastolic interval and recovery of RyRs) and the diastolic SR Ca2+ load slightly decreases to a new steady-state (horizontal dashed line). When a shorter CL is initiated, however, a smaller SR Ca2+ release amplitude occurs (due to the shorter diastolic interval and incomplete recovery of RyRs) and diastolic SR Ca2+ load increases to a new steady-state (horizontal dashed line). Alternation of SR Ca2+ release amplitude also occurs at CL = 250 msec. S: small release amplitude, L: large release amplitude. (CL: cycle length) Please click here to view a larger version of this figure.
Figure 3. Diastolic SR Ca2+ Leak and Subsequent Triggered Activity. (A) ECG, Vm, and SR Ca2+ traces during baseline (left) and following treatment with 100 nM isoproterenol (Iso, right). With Iso, significant diastolic SR Ca2+ leak is observed (red arrows). SR Ca2+ signal amplitudes have been normalized before and after Iso. The sino-atrial node was ablated to allow for pacing CL = 300 msec in both cases. (B) If SR Ca2+ leak is large enough, it may trigger focal activity (premature ventricular complexes, PVCs). In this example, PVCs are interrupting the normal sinus rhythm. Please click here to view a larger version of this figure.
The keys to successful Fluo-5N dye loading are the small-volume recirculating perfusion setup, which allows for a high Fluo-5N concentration without the need for large amounts of dye, the length of loading time (1 hr), and performing the loading at RT. If loading is performed at physiological temperatures, cellular enzymatic activity quickly cleaves the –AM tag when the dye crosses the cell membrane, trapping the dye molecules in the cytosol and not allowing them to cross the SR membrane. At RT, however, enzymatic activity is slowed and a sufficient amount of dye can cross both the cell membrane and SR membrane before the –AM tag is cleaved, trapping the dye in the SR.
Even with optimal loading conditions, however, signal amplitudes and fractional fluorescence changes (ΔF/F) with this approach will be lower than with traditional optical mapping of Vm and intracellular Ca2+ with high-affinity dyes. For example, with the optical setup used here that consists of LED excitation light, macroscopic objectives, large filters and dichroic mirror (5 cm x 5 cm), and CMOS cameras, fractional fluorescence changes (ΔF/F) are on the order of 1 – 2% for the Fluo-5N signal and 2 – 3% for the RH237 signal15. The fractional fluorescence will, of course, depend on the specific optical configuration, light source, and cameras of a particular setup, but even under optimal conditions the Fluo-5N signal is significantly lower than what is typically observed for recordings of intracellular Ca2+ with high-affinity dyes. For example, on the same setup, ΔF/F is on the order of 20 – 30% for Rhod-2 when green (545 nm) excitation light is used18. The green excitation light is also closer to the excitation peak of RH237, thus the ΔF/F of RH237 is also larger in this case, around 4 – 5%. Thus, for those labs already performing dual optical mapping of Vm and intracellular Ca2+, care should be taken to optimize all components of the optical setup to ensure maximal ΔF/F for the Fluo-5N signal.
With this approach, investigators can precisely probe SR Ca2+ dynamics and its impact on Vm in ways that are not possible with existing approaches. For example, quantitative measurements of SR Ca2+ leak (as in Figure 3A) and the precise time constant of SR Ca2+ recovery (tau –a measure of SR Ca2+-ATPase [SERCA] activity) can be performed. Using existing approaches for optical mapping of intracellular Ca2+ with high-affinity dyes, these measurements may be ‘contaminated’ with transmembrane Ca2+ flux (i.e., contributions from L-type Ca2+ channels and the Na+-Ca2+ exchanger). Furthermore, high-affinity Ca2+ indicators have inherently slower kinetics than low-affinity indicators5, thus Fluo-5N signals are expected to report on precise SR Ca2+ movements with high fidelity and essentially no time delay between actual Ca2+ changes and corresponding changes in fluorescence.
The ability to investigate the detailed mechanisms by which SR Ca2+ release and reuptake contribute to arrhythmogenic phenomena is another advantage of this approach. For example, using this methodology, we recently reported how refractoriness of SR Ca2+ release contributes to the onset of Ca2+ alternans and that during ventricular fibrillation, SR Ca2+ release remains nearly continuously refractory15. This was the first report of SR Ca2+ dynamics during fibrillation and highlights the key advantage of SR Ca2+ imaging in the intact heart: spatially distinct and complex phenomena such as spatially discordant alternans and fibrillation can be studied. Unfortunately, this type of information cannot be gleaned from isolated cells, where fibrillation is not possible, or from methods that only record from a single location on the intact heart12.
Furthermore, simultaneous optical mapping of Vm and free SR Ca2+ in the intact heart may provide a novel tool for assessing abnormal Ca2+ handling in pathological conditions, including heart failure. The abnormalities in SR Ca2+ regulation, including activation and termination of SR Ca2+ release, diastolic SR Ca2+ leak, and SR Ca2+ uptake – all major steps in Ca2+ cycling and all potentially altered in heart failure – can be directly studied. In addition, this methodology could also be a useful tool for assessing the effects of novel therapeutic interventions, such as RyR stabilization19 and SERCA modulation20.
This approach is not without limitations, however, including the low ΔF/F as discussed above and the inability to precisely calibrate fluorescence values to exact [Ca2+]SR concentrations. Unfortunately, variability in dye loading, non-uniform excitation and emission light due to the curved surface of the heart, and photo-bleaching and dye leak from the SR throughout the course of the experiment make it extremely difficult to calibrate Fluo-5N signals to exact [Ca2+]SR in the intact heart. Previous in vitro studies in physiological buffer solutions and permeabilized isolated myocytes have reported such calibrations and have indicated that changes in Fluo-5N fluorescence are proportional to changes in [Ca2+]SR and the fluorescence is not expected to saturate at physiological [Ca2+]SR levels11. Thus, investigators should choose the method of Ca2+ mapping (low- or high-affinity dye) based on their specific experimental objectives and physiological investigations.
The authors have nothing to disclose.
This work was supported in part by the US National Institutes of Health (R01 HL 111600) and the American Heart Association (12SDG9010015).
NaCl | Fisher Scientific | S271-1 | Component of Tyrode's solution |
CaCl2 (2H2O) | Fisher Scientific | C79-500 | Component of Tyrode's solution |
KCl | Fisher Scientific | S217-500 | Component of Tyrode's solution |
MgCl2 (6H2O) | Fisher Scientific | M33-500 | Component of Tyrode's solution |
NaH2PO4 (H2O) | Fisher Scientific | S369-500 | Component of Tyrode's solution |
NaHCO3 | Fisher Scientific | S233-3 | Component of Tyrode's solution |
D-Glucose | Fisher Scientific | D16-1 | Component of Tyrode's solution |
95% O2 5% CO2 | AirGas | carbogen | For oxygenation and pH of Tyrode's solution |
Blebbistatin | Tocris Bioscience | 1760 | Excitation-contraction uncoupler |
RH237 | Biotium | 61018 | Voltage-sensitive dye |
Fluo-5N AM | Invitrogen | F-26915 | Low-affinity Ca2+ indicator; Alternative: Invitrogen F-14204; Loading must be performed at room temperature |
Pluronic F127 | Biotium | 59004 | For Ca2+ indicator loading; Warm until the solutiion is clear before use |
Dimethyl sulphoxide (DMSO) | Sigma-Aldrich | D2650 | For dissolving blebbistatin and dyes |
Filter | EMD Millipore | NY1104700 | 11um in-line filter |
Pressure Transducer | WPI | BLPR2 | For measuring perfusion pressure |
Transbridge Transducer Amplifier | WPI | SYS-TBM4M | For transducing/amplifing pressure signal; PowerLab may also be used with appropriate BioAmp |
PowerLab 26T | ADInstruments | For continuous recording of pressure and ECG signals | |
THT Macroscope | SciMedia | Macroscopic optical setup. Details: 0.63x objective (NA=0.31), 2x condensing objective, resultant field of view = 3.1×3.1 cm, depth of focus = ~1.5mm | |
MiCam Ultima-L CMOS | SciMedia | Optical mapping cameras | |
Precision LED Spot Light | Mightex | PLS-0470-030-15-S | LED light source |