A. ECMO circuit consisting of a venous reservoir (VR), roller pump (RP), miniature membrane oxygenator (MO) between arterial (femoral artery, FA) and venous (right atrium via external jugular (EJ) systems perfused in a retrograde fashion. B. Timeline of experiment (CA=cardiac arrest) followed by 15 mins downtime and then ECMO initiation. C. An example arrhythmic sudden cardiac arrest protocol. D. Ordinate for D-F is the same, and shown in 1D only. Zoomed in view of the beginning of Rapid Ventricular Pacing (RVP) at 12 Hz pacing over the LV apex leading to hypotension. BP (black) is shown on top with the scale bar on the left, and heart rate (far field EKG; gray) is shown underneath. E. Both ventricular fibrillation (VF) followed by pulseless electrical activity (PEA) later in time in a zoomed in view. F. Zoomed in view of the heart off ECMO after full resuscitation. G, H. Representative PV loops from pre (G) and post-arrest (H) with ECMO resuscitation in the same animal. The ESPVR (solid line) and EDPVR (dotted line) are superimposed. The ESPVR slope was 1.38 and 0.6 mmHg/μl.

A. Isolated twitching myocyte before (top) and during twitch in response to electrical activation demonstrating carbon fiber bending. B The sarcomere length (μm) in response to 1 Hz pacing vs time as CF stretch was applied to three different endpoint lengths demonstrates the cellular Frank-Starling effect. Between each step, the CFs were returned to baseline. Each stretch produced increased sarcomere length change. C. Carbon fiber bending (CFB) (proportional to force) vs time demonstrating the increase in both diastolic (passive) and systolic (active) single cell force, respectively. D. Stretch induced peak systolic (open squares) and baseline diastolic (filled squares) tension from a control cell. The linear fits to the data represented the EDLTR (black line) and ESLTR (gray line). In this case, the ESLTR was 22 μN/μm (R²=0.96), and the EDLTR was 6.6 μN/μm (R²=0.95). E. Stretch induced peak systolic (open squares) and baseline diastolic (filled squares) tension from a PMAD cell. In this case, the ESLTR was 11 μN/μm(R²=0.92), and the EDLTR was 6 μN/μm (R²=0.98). F. PRSW_cell vs afterload_cell force for control (open squares) and PMAD (filled squares), n= 8 cells from 3 hearts and 8 cells from 2 hearts for control and PMAD, respectively. Exponential fits to the data were done using the equation FSG= y0+A1e(−x/t1). For control and PMAD, respectively, y0, A1, t1 were: 0.94 and 1.47, 9.5 and 2.7, 0.78 s and 0.54 s. R² were 0.96 and 0.92. N= 8 cells from 3 hearts and 8 cells from 2 hearts for control and PMAD, respectively. G. Force frequency relationship (FFR) for pooled BSF (μm) vs pacing frequency for control (black squares) and PMAD (open squares) plotted with SE, n= 8 cells from 3 hearts and 8 cells from 2 hearts for control and PMAD, respectively. H. Example traces for simultaneous CICR (bottom, gray) and cellular force (top, black) measurements, shown for an isolated cell electrically paced at 1 Hz. The SL was (in μm) ^~1.87,^~1.94 and ^~2.0 (at slack length, first, and second stretch, respectively).

A. Example CICR transients measured by Fluo-5f in control compared against PMAD, and also PMAD in which arrest was generated in the presence of KN93 (PMAD-KN93) or ryanodine (PMAD-Rya), as well as a control animal exposed to IV infusion of epinephrine (Epi) for 30 minutes. Cells were paced steadily at 1Hz. B. Pooled analysis of ΔF/Fpeak for all conditions tested. The ΔF/Fpeak was significantly higher in animals exposed to epinephrine infusion and cardiac arrest. Ryanodine and KN93 reduced ΔF/Fpeak to control levels in PMAD ventricular myocytes. WT was no different than sham. n=114, 135, 89, 16, 17, 9, 11, 15, 11, 31, 20, 51, 32, 119, 19, 11 cells for control, PMAD, PMADAIP, PMAD-AIP-IN, PMAD-KN93, PMAD-KN92 PMAD-Rya, Epi infusion, control-KN93 (Cntr-KN93), control ryanodine (Cntr-Rya), epi infusion with ryanodine (Epi-Rya), PMAD-Langendorff-Saline (PMAD-Lan-Sal), PMAD-Langendorff-PMAD blood (PMAD-Lan-Blood), Brainstem herniation (BSH), BSH-ryanodine (BSH-Rya), and PMAD-H89. Cntr-Sham experiments demonstrated no difference compared with control. P values compared to control for PMAD, Epi infusion, BSH, PMAD-Lan-Blood were 6.4e-24, 2.2e-18, 2e-25, and 2e-14 (corrected for multiple comparisons). C. Example CICR transients (red) and sarcomere shortening (black) prior to and after carbon fiber stretch. Top set showed stretch followed by a delayed calcium release (black arrow) initiating fibrillatory behavior in both traces. Lower panel showed 30 seconds later termination of “arrhythmia” through rapid pacing. D. Example CICR transients from two different PMAD cells. Top panel showed CICR in response to electrical stimulation followed delayed calcium transients. Bottom panel showed spontaneous calcium transients competing for paced rhythm. E. Pooled analysis of velocity of contraction (Vc) vs sarcomere length for control and PMAD. Baseline (filled) and stretched (open) for pairs of similar stretches from ^~1.8 μm ^~ 2.0 μm. PMAD (red) than control (black); n= 8, 9 cells from 3 and 4 hearts for PMAD and control hearts respectively. F. Pooled data comparing unloaded %SL vs ΔF/Fpeak of CICR under control, PMAD, epinephrine infusion in animal, but not cells (EPI inf), and epinephrine in isolated cells (EPI cell), but not in whole animal), as well as PMAD in the presence of Alda-1 (PMAD-Alda, n=25 cells from 4 animals). N for control, PMAD, and Epi infusion are the same as above. G. Baseline Force (BSF) in example cells paced steadily at 1 Hz in control, PMAD, PMADKN93 and PMAD-Rya cells. H. Pooled data for BSF measurements at 1 Hz steady-state pacing in control, PMAD in the presence and absence of KN93 or ryanodine. n= 8, 8, 5, 4 cells for control, PMAD, PMAD-KN93, PMAD-Rya.

A. Plot of force (z-axis), pacing frequency (x-axis) and calcium expressed as ΔF/F_peak (y-axis) for control cells at slack length loaded with Fluo-5f and simultaneously stretched with CF. n=4 cells from 2 control hearts. The curve showed a steep curve shifted to lower ΔF/Fpeak. The color represents force and is coded from low to high as from blue to red, B. Similar plot in post arrest cells. n=4 cells from 2 control hearts. By comparison to panel A, the curve is shifted to a flatter curve despite also being shifted to higher ΔF/Fpeaks (rightward) at all pacing frequencies. Force was color coded with the same palette as in panel A, and all scale bars were identical.

A. Example fluorescence measurements of CICR (y-axis F/F vs x-axis time). In vivo fluorescence imaging was performed 3 or 4 weeks after GCaMP6f delivery in rat (3w P.D., 4w P.D.). B. Example image of raw fluorescence seen through fiber optic on heart by camera at baseline and peak fluorescence during a calcium transient. C. Example cardiac arrest experiment from a different GECI-rat while measuring CICR fluorescence continuously through experiment (y-axis F/F vs x-axis time). Shown were measurements at baseline, after cardiac arrest, on ECMO demonstrating delayed calcium release initiating fibrillatory behavior similar to what was seen in isolated cells, and finally off ECMO.

A and B. Example Western blots showing CaMKII-dependent phosphorylation and protein level of CaMKII, phospholamban and RyR2 in Control group, PMAD model (30 minutes of cardiac arrest and 30 minutes of ECMO reperfusion) and PMAD in the presence of AIP (A) and Alda-1 (B), both were given 30 minutes before the onset of cardiac arrest. C-I. Quantitative summary of Western blot analysis. CaMKII-dependent phosphorylation of CaMKII at T287 (C) and phosphorylation of PLN at T17 (E) increased significantly in PMAD and inhibited by AIP while the protein level of CaMKII (D) and PLN (F) did not change significantly. CaMKII autophosphorylation (PT287) in PMAD was also inhibited by Alda-1(C). CaMKII-dependent RyR2 phosphorylation at S2814 did not vary considerably in PMAD or with AIP or Alda-1 treatment, illustrated as PS2814/RyR2 in G. However, RyR2 protein level surged 5-fold in PMAD compared with Control group (RyR2/GAPDH in H) therefore overall pRyR2 PS2814 was also shown to increase dramatically (normalized to GAPDH in I). Neither AIP nor Alda-1 treatment reduced the protein level or total RyR2 phosphorylation at S2814 (H, I). J. Western blot and quantitative assessment showed that PKA phosphorylation of PLN at S16 and PLN protein level were indifferent between PMAD and Control group. K. Representative Western blot and quantitative summary showed that CaMKII was activated after PMAD but not epinephrine infusion for 30 minutes. Graph data represent the Mean ± SEM of each group. *, p<0.05; **, p<0.005; ns, not-significant. p-values were reported by ordinary one-way ANOVA with multiple comparisons (C-I and K) or t-test (J).

A. Myocardial O₂^•– production, measured by lucigenin, was mildly increased in hearts treated with epinephrine, but markedly elevated after cardiac arrest (^†P < 0.01). The NADPH oxidase inhibitor DPI and the NOS inhibitor L-NMMA did not change O₂^•– generation compared to uninhibited controls, however the ALDH2 inhibitor Alda-1 did inhibit O₂^•– (*P < 0.01). B. Dihydroethidium (DHE) fluorescence microscopy allowed topographic assessment of O₂^•– production in the ventricular wall, and showed mild increase in O₂^•– production with epinephrine infusion, compared with marked increase after cardiac arrest (^†P < 0.01). Alda-1 inhibited O₂^•– generation. Mito-TEMPO, used as the mitochondria-targeted antioxidant with superoxide and alkyl scavenging properties, confirmed that the majority of O₂^•– being generated was from mitochondria (compared to uninhibited *P < 0.01). Myocytes exhibited green autofluorescence. White arrows denoted endothelial O₂^•– producing cells. Scale bar: 5 μm.

Alda-1 improved in vivo contractility and recovery rate to spontaneous circulation after cardiac arrest. A. Summary data for ESPVR for control, PMAD, PMAD-Alda (Alda-1) p=0.006. B. Summary data for PRWV for control, PMAD, PMAD-Alda (Alda-1), p=0.04. C. Summary data for CI for control, PMAD, PMAD-Alda (Alda-1)D. Summary data for tau for control, PMAD, PMAD-Alda (Alda-1). P-values in graphs are with respect to control using post-hoc analysis. Comparisons between groups performed using Kruskall-Wallis nonparamateric test. E. Proposed mechanism for post-arrest myocardial dysfunction, mediated by calcium overload, CaMKII, and oxidative stress (ROS). Catecholamines released during cardiac arrest and the ischemic/reperfusion process triggered cytosolic Ca²⁺ influx and SR Ca²⁺ release/uptake. The activation of RyR2 (by oxidative stress or other unknown mechanism**) also increased CICR. Elevated cytosolic Ca²⁺ -activated CaMKII further enhanced CICR and lead to mitochondrial Ca²⁺ overload. This Ca²⁺ overload contributed to excess ROS (O₂^•–) and reactive aldehyde (4-HNE) generation, more CaMKII activation, and other cellular damages. AIP inhibited CaMKII autophosphorylation but did not improve cardiac function after arrest. Alda-1 reduced oxidative stress, inhibited CaMKII auto-phosphorylation and improved cardiac function.