Optogenetic quantification of cardiac excitability and electrical coupling in intact hearts to explain cardiac arrhythmia initiation

Abstract

Increased cardiac excitability and reduced electrical coupling promote cardiac arrhythmia and can be quantified by input resistance (R_m), pacing threshold (I_thr), and cardiac space constant (λ). However, their measurement in the heart was not feasible because the required homogenous current injection cannot be performed with electrical stimulation. We overcame this problem by optogenetic current injection into all illuminated cardiomyocytes of mouse hearts in different action potential phases. Precisely triggered and patterned illumination enabled measuring R_m and λ, which both were smallest at diastole. Pharmacological and depolarization-induced reduction of inwardly rectifying K⁺ currents (I_K1), gap junction block, and cardiac infarction reduced I_thr, showing the importance of high I_K1 density and intact cardiomyocyte coupling for preventing arrhythmia initiation. Combining optogenetic current injection and computer simulations was used to classify pro- and anti-arrhythmic mechanisms based on their effects on R_m and I_thr and allowed to quantify I_K1 inward rectification in the intact heart, identifying reduced I_K1 rectification as anti-arrhythmic concept.

Fig. 1.. Optogenetic determination of input resistance (R_m) in intact hearts.

(A and B) Representative AP trace with pacing (red, ~350 μW/mm², 5 to 10 ms) and subthreshold (blue, 15 to 40 μW/mm², 20 ms) light stimulation (465 nm). (C) Averaged APs with second subthreshold light pulses of delays between 10 and 240 ms. (D and E) Membrane potential change (D, ΔE) calculated as difference between averaged APs with and without subthreshold light pulse and corresponding ChR2 current (E, I_ChR2) computed by a ChR2 gating model for the different delays. (F) ChR2 current of a ventricular cardiomyocyte evoked by high (red, 315.5 μW/mm²) and subthreshold (blue, 13.6 μW/mm²) illumination analog to the stimulation in (A) to (E). (G to J) Representative averaged AP (I) and corresponding values of R_m (J) calculated as ratio of maximum membrane potential change (G) and ChR2 current (H) [(A) to (J), from one representative heart or cardiomyocyte]. (K) R_m at AP plateau, 70 and 90% of repolarization (APD70 and APD90) and diastole [repeated measures one-way analysis of variance (ANOVA), Tukey’s multiple comparisons posttest, N = 10, P < 0.0001]. Means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 2.. Effect of I_K1 reduction on cardiac excitability.

(A and B) Representative, averaged APs (A) and R_m for different delays (B) of the subthreshold light pulse with and without BaCl₂ (10 μM). (C to F) Quantification of the effect of I_K1 block on R_m during the first phase of final repolarization [(C), at APD70, two-tailed, paired t test, N = 8, P = 0.39] and diastole [(D), two-tailed, paired t test, N = 8, P = 0.0016], RMP [(E), two-tailed, unpaired t test, N = 4, n = 38-41, p = 0.25], and maximum upstroke velocity [(F), maximum dV/dt, two-tailed, unpaired t test, N = 4, n = 38 to 41, P = 0.53]. (G) DADs resulting from fast pacing (cycle length of 70 to 140 ms) without (black), with caffeine (1 mM, red), and with BaCl₂ additionally to caffeine (blue). (H) Statistical analysis of DAD amplitude (repeated measures one-way ANOVA, Tukey’s multiple comparisons posttest, N = 7, P = 0.0003). (I) Representative AP traces of a S1-S2 protocol (left, last S1 showed) followed by a premature stimulus (S2) of variable delay (120 to 260 ms) and traces of the transition from sub (black) to supra-threshold (red) optical stimulation (right). (J and K) Thresholds for optogenetic pacing (PT) for different delays of S2 (10-ms binning) after APD90 (J) and statistics of diastolic I_thr determined 135 ms after APD90 [(K), two-tailed, paired t test, N = 6, P = 0.02] with and without BaCl₂. Means ± SEM. *P < 0.05 and **P < 0.01; n.s., not significant; Mohms, megohms.

Fig. 3.. Effect of RMP depolarization on cardiac excitability.

(A) Representative, averaged, paced (red bar) AP traces without (black) and with low-intensity illumination during the diastole (9.4 to 33.6 μW/mm², light blue bar) and subthreshold pulses to determine R_m (dark blue bar). (B to E) Relationship between depolarization of RMP (ΔRMP) and change of R_m (B, ΔR_m) or I_thr (D, ΔI_thr) with different colors indicating individual hearts. Statistical analysis of diastolic R_m [(C), N = 6, P = 0.021] and I_thr [(E), N = 5, P = 0.0015] for 3.5- and 7-mV RMP depolarization (repeated measures one-way ANOVA, Tukey’s multiple comparisons posttest). Means ± SEM. *P < 0.05 and **P < 0.01.

Fig. 4.. Contribution of diastolic ion currents to excitability in a computational model of a mouse ventricular cardiomyocyte.

(A to C) Effect of scaling I_K1, Na⁺,K⁺-ATPase current (I_NaK), sodium background current (I_Nab), voltage-dependent sodium current (I_Na), and inward rectification (ir) of I_K1 on RMP (A), diastolic R_m (B), and I_thr (C). (D) Relative change in R_m in response to RMP depolarization of 3.5 mV by adding a cation background current (I_leak) mimicking I_ChR2 for different scaling factors of diastolic currents. Dotted, black line indicates the experimentally measured increase in R_m of 8.7%. (E to G) Relationships between relative changes in RMP, R_m, and I_thr predicted by the model with ir*1 (dotted lines) and ir*0.37 (solid lines) for reduction of I_K1 (blue) and increase in I_leak (red). Gray line indicates effect of 3.5 mV RMP depolarization on R_m (E). Experimental data from Figs. 2 (blue, means ± SEM) and 3 (red) are shown as dots.

Fig. 5.. Effect of Na⁺ and multichannel blockers on cardiac excitability.

(A to D) Representative, averaged APs (A), maximum upstroke velocity [(B), maximum dV/dt, N = 5, P < 0.0001], I_thr [(C), N = 5, P = 0.0022], and diastolic R_m [(D), N = 5, P = 0.73] without and with the sodium channel blocker lidocaine (10, 30 μM, repeated measures one-way ANOVA, Tukey’s multiple comparisons posttest). (E) Relationship between change of R_m (ΔR_m) and I_thr (ΔI_thr) resulting from sodium current reduction in simulations (solid line) and experiments (dots). (F and G) Simulated relationship between degree of I_Na block and increase in pacing threshold (F) and R_m (G) for increasing concentrations of the multichannel blockers amiodarone and dronedarone and isolated I_Na block. (H) Summary of the relationship between ΔR_m and ΔI_thr for isolated modifications of indicated ion channels, I_K1 inward rectification (ir), and the multichannel blockers from simulations in the calibrated cardiomyocyte model. Means ± SEM. **P < 0.01 and ***P < 0.001.

Fig. 6.. Optogenetic determination of cardiac space constant (λ) in the intact heart.

(A) Concept for determination of λ. Theoretical example of depolarization in one-dimensional space (left, red) resulting from illumination restricted in space (blue) with high (solid, 1 mm) and low (dotted, 0.5 mm) λ. The magnitude of depolarization is calculated by convolution (f ∗ g) of the distribution of light (f) with a weight function (g, right) which depends on λ (solid, 1 mm; dotted, 0.5 mm). (B) Averaged APs with different illumination sizes of the subthreshold light pulses (blue bar, 20 ms) applied after a global pacing pulse (red bar, 5 to 10 ms). (C and D) Relationship between subthreshold membrane potential change (ΔE) and size of illumination area fitted with the convolution function f ∗ g [(C); LA, left atrium; LV, left ventricle] and diastolic λ obtained from fitting λ as parameter of g upon decreasing the illumination area in longitudinal (blue) and transverse (green) direction of fiber orientation [(D), two-tailed, unpaired t test, N = 6, P < 0.0001]. (E) Longitudinal λ at plateau (Plat.), 70% of repolarization (APD70), and diastole (Diast.; repeated measures one-way ANOVA, Tukey’s multiple comparisons posttest, N = 5, P = 0.008). (F and G) Diastolic, longitudinal λ [(F), N = 5, P = 0.032] and I_thr [(G), N = 6, P = 0.0028] without and with the gap junction blocker carbenoxolone (CBX; 10 μM, two-tailed, paired t test). (H to J) Change in diastolic R_m (H) and λ (I) during successive increase of BaCl₂ concentration up to 10 μM in one representative heart and normalized changes in R_m and λ (J) fitted with a power function (solid black, a = 1.002, 99% confidence interval: [0.986, 1.018], b = 0.572, 99% confidence interval: [0.487, 0.656]). Square root relationship shown with dotted line. Colors indicate different hearts. Means ± SEM. *P < 0.05, **P < 0.01, and ****P < 0.0001.

Fig. 7.. Increased cardiac excitability after acute myocardial infarction.

(A and B) Image of the left ventricle (A) after cryoinfarction (white) at the heart apex (scale bar, 1 mm) and pacing threshold (PT) before and after infarction for different regions (B). The border of infarction is indicated with dashed line, and pattern of remote myocardium (RM), border zone (BZ; pattern −2), and infarction (MI) are highlighted in blue, red, and purple, respectively [(A) and (B)]. (C and D) Statistical analysis of I_thr before [(C), two-tailed, paired t test, N = 9, P = 0.38] and after infarction [(D), two-tailed, paired t test, N = 9, P = 0.0037]. (E) RMP before (∅MI) and after cryoinfarction recorded at different distances from the border of infarction (ordinary one-way ANOVA, N = 2 to 3, n = 9 to 11, P = 0.27). (F) Change in I_thr (ΔI_thr) of RM and BZ without and with assuming 3.5-mV RMP depolarization after infarction compared to ΔI_thr due to 3.5-mV RMP depolarization alone (∅MI, data from Fig. 3E) (ordinary one-way ANOVA, N = 5 to 9, P < 0.0001, calculated on the basis of values in picoamperes per picofarad). Means ± SEM. **P < 0.01, ***P < 0.001, and ****P < 0.0001.