CaMKII-Dependent Contractile Dysfunction and Pro-Arrhythmic Activity in a Mouse Model of Obstructive Sleep Apnea

Figures

• 

  Figure 1. Study flowchart. C57BL/6 wildtype, MMVV knock-in, and CaMKIIδ knock-out mice were randomly assigned to either control or OSA induction by PTFE injection. After 8 weeks, echocardiography, ROS measurements, and epifluorescence microscopy were performed, whereby some experiments were performed in the same subject.

• 

  Figure 2. (A) Original M-mode echocardiographic traces of left ventricle. (B) Mean left ventricular ejection fraction per animal at baseline (CTRL groups: n = 9 WT, n = 8 MMVV, n = 13 CaMKII-KO; PTFE groups: n = 25 WT, n = 6 MMVV, n = 10 CaMKII-KO; ANOVA p = n.s.). (C) Mean left ventricular ejection fraction per animal after 8 weeks, # indicates p < 0.05 vs. WT PTFE at baseline. (D) Before–after presentation of PTFE group data from (B,C). (E) Heart rate during echocardiography from (B,C). (F) Animal body weight at baseline and after 8 weeks. Statistical comparisons are based on ANOVA, mixed-effects analysis with Holm–Sidak’s post hoc correction, and paired t-test, as appropriate.

• 

  Figure 3. (A) Original confocal frame scans of isolated ventricular cardiomyocytes loaded with the cytosolic ROS fluorescence dye CellRoxTM Orange. (B) Fluorescence intensity over time, normalized to baseline (p < 0.001 for difference in linear regression slopes). (C) Mean data per animal for the increase in CellRoxTM Orange fluorescence intensity normalized to baseline fluorescence (ΔF/F0) at 10 min after start of recording (n = 7 vs. 7). (D) Original confocal frame scans of isolated ventricular cardiomyocytes loaded with the mitochondrial ROS fluorescence dye MitoSoxTM Red. (E) Fluorescence intensity over time, normalized to baseline (p < 0.001 for difference in linear regression slopes). (F) Mean data per animal for the increase in MitoSoxTM Red fluorescence intensity normalized to baseline fluorescence (ΔF/F0) at 10 min after start of recording (n = 7 vs. 7). Statistical comparisons are based on Student’s t-test.

• 

  Figure 4. (A) Original Ca transients recorded in Fura-2 AM loaded isolated ventricular cardiomyocytes (1 Hz). The red arrow indicates a non-stimulated pro-arrhythmic event in a WT PTFE cell. (B) Mean Ca transient amplitude, diastolic Ca (C), relaxation time to 70% (D) and 90% of baseline (E) per animal (n = 21 WT CTRL, n = 11 MMVV CTRL, n = 22 WT PTFE, n = 8 MMVV PTFE). (F) Frequency of non-stimulated pro-arrhythmic events (as in (A), n = 19 WT CTRL, n = 10 MMVV CTRL, n = 20 WT PTFE, n = 6 MMVV PTFE). Statistical comparisons are based on ANOVA with Holm–Sidak’s post hoc correction.

• 

  Figure 5. (A) Original Ca transients recorded in Fura-2 AM loaded isolated ventricular cardiomyocytes before and after 30 s cessation of electrical stimulation (post-rest test). (B) Mean Ca transient amplitude ratio per animal after the pause/before the pause as a measure of diastolic SR Ca leakage (n = 18 WT CTRL, n = 11 MMVV CTRL, n = 19 WT PTFE, n = 7 MMVV PTFE). (C) Original recordings of Ca transients induced by rapid caffeine application (red arrows), mean data per animal shown in (D) (n = 12 WT CTRL, n = 4 MMVV CTRL, n = 13 WT PTFE, n = 4 MMVV PTFE). Statistical comparisons are based on ANOVA with Holm–Sidak’s post hoc correction.