Fig 1.
A. A schematic plot of the WG model. The outer solid line represents the membrane. The space between the dashed line and the solid line is the submembrane space which is divided into two compartments: the junctional (JXN) and subsarcolemmal (SL) space. The JXN space (grey) is the cleft adjacent to the junctional sarcoplasmic reticulum (JSR, green), and the SL space (light grey) is the remaining submembrane space. NSR is the network sarcoplasmic reticulum. B. The AP of the control case.
Fig 2.
EADs driven by Vm oscillations.
A. Phase diagram showing different EAD behaviors in the α(PCa) − α(GKs) plane for α(kmax) = 2, α(JCaslmyo) = 2.64, and . The grey and white regions represent repolarization failure and normal repolarization, respectively. Insets show two representative APs. Three types of EADs are observed, as colored dark yellow, green, and magenta. B. A representative AP and associated currents in the dark yellow region (i.e., EADs driven by Vm oscillations). The parameter set is α(GKs) = 1.16 and α(PCa) = 5.4, as marked by the black circle in A. The dashed vertical lines indicate the takeoff moments of the two EADs. The grey patch indicates the range of [Ca]sub clamped used in the [Ca]sub-clamp simulations. C. An example showing Vm and [Ca]sub versus time with [Ca]sub being clamped at the lowest (red) and highest (blue) levels.
Fig 3.
EAD behaviors under S1S1 and S1S2 pacing protocols.
A. Phase diagram obtained with the S1S1 protocol. The parameter set is the same as Fig 2A. S1S1 interval = 3 s. The grey region is determined using the criterion of APD being greater than the S1S1 interval. The colored regions are different EAD regions classified by the same way as in Fig 2A. B. Phase diagram obtained with the S1S2 protocol. The EAD classification is done using the S2 beat. The S1S1 interval is 1 s and the S1S2 interval is 2 s. C. An example showing Vm and [Ca]sub versus time from the regime of Vm-driven EADs with the S1S1 protocol. α(GKs)=0.8 and α(PCa)=4.4. The traces in the right panels are those when [Ca]sub is clamped in the last pacing beat. D. An example showing Vm and [Ca]sub versus time from the regime of Vm-driven EADs with the S1S2 protocol. α(GKs)=0.8 and α(PCa)=4.6. The traces in the right panels are those when [Ca]sub is clamped in the S2 beat.
Fig 4.
EADs driven by Ca oscillations alone.
α(kmax) = 7, α(JCaslmyo) = 3.6, and α(INCX)=2.2. A. Phase diagram in the α(PCa) − α(GKs) plane. EADs occur in the magenta region. B. An example showing Vm, [Ca]sub, ICa,L, and INCX versus time from the EAD region [α(GKs) = 0.56, α(PCa)=2.8, marked as star in A]. The first two vertical dashed lines indicate the takeoff of the first two EADs and the third vertical dashed line marks the peak of the EAD prior to the last one. C. The same case as in B but [Ca]sub is clamped at two different levels in the plateau (grey patch in the [Ca]sub panel in B). D. The same case as in B but Vm is clamped at two different levels in the plateau (grey patch in the Vm panel in B). E. Upper panel: Vm and [Ca]sub versus time when ICa,L is clamped at a constant in the plateau (dashed line in the ICa,L panel in B). Lower panel: Vm and [Ca]sub versus time when INCX is clamped at a constant in the plateau (dashed line in the INCX panel in B).
Fig 5.
EADs driven by oscillations caused by the Ca-ICa,L-Vm-Ca feedback loop.
A. Vm, [Ca]sub, ICa,L, and INCX versus time for a representative case [α(GKs) = 1.6 and α(PCa) = 8.2, square in Fig 2A]. The left vertical dashed line indicates the peak and the right one indicates the valley of [Ca]sub. B. Vm and [Ca]sub versus time when [Ca]sub is clamped at the lowest (red), middle (green), and highest (blue) levels during the plateau phase. C. Vm and [Ca]sub versus time when Vm is clamped at lowest (red) and highest (blue) levels in the plateau phase. D. [Ca]sub versus Vm from simulations using the Vm clamp protocol in which Vm is switched from -80 mV to different clamped levels. Black dots: INCX = 0. Red dots: ICa,L = 0. The inset is a schematic plot of the feedback loop.
Fig 6.
EADs driven by a large Ca transient.
α(kmax) = 7, α(JCaslmyo) = 1, and α(INCX) = 1.2. A. Phase diagram showing different mechanisms of EAD on the α(PCa) − α(GKs) plane. The Ca transient driven EADs occur in the cyan region. Other colors are the same as indicated in the phase diagram in Fig 2A. B. Vm, [Ca]sub, ICa,L, and INCX versus time for a representative case [α(GKs) = 1 and α(PCa) = 1, diamond in A]. The vertical dashed line indicates the EAD takeoff moment. C. Same as B but [Ca]sub is clamped to 0.5 times of the trace in B (highlighted by red). The blue dashed Vm trace in the Vm panel is the original one for comparison. D. Same as B but ICa,L is reduced linearly from the EAD takeoff moment. E. Same as B but INCX is reduced linearly from the EAD takeoff moment.
Table 1.
Mechanisms of EADs in different AP models.
For each model, a total of 10000 randomly selected parameter sets is simulated, and the number of parameter sets for which the APs exhibit EADs is indicated in the parentheses.