Atrial GIRK Channels Mediate the Effects of Vagus Nerve Stimulation on Heart Rate Dynamics and Arrhythmogenesis

Figures

• 

  FIGURE 1. Characterization of an atrial-specific Girk-/- mouse line. (A,B) SLNCre(–) and SLNCre(+) mice were crossed with the Cre-dependent fluorescent reporter strain, Ai14-tdTomato. Representative sections stained with DAPI (Left), and associated tdTomato fluorescence (Right), in hearts from SLNCre(–) (Top) and SLNCre(+) (Bottom) offspring. (C) Representative whole-cell currents (Vhold = –70 mV) evoked by carbachol (CCh, 10 μM) in a high-K+ bath solution in adult SAN cells from SLNCre(–):Girk1fl/fl and SLNCre(+):Girk1fl/fl mice. Scale bars: 1 nA/10 s. (D,E) Summary of CCh-induced currents in adult SAN cells (t47 = 11.3; ∗∗∗P < 0.001; n = 24–25/genotype) and ventricular myocytes (VM; t21 = 0.6; P = 0.54; n = 11–12/genotype) from SLNCre(–):Girk1fl/fl and SLNCre(+):Girk1fl/fl mice.

• 

  FIGURE 2. Impact of atrial GIRK channel ablation on HR and HRV responses to carbachol. (A) Representative segments of ECG recordings from anesthetized SLNCre(–):Girk1fl/fl and SLNCre(+):Girk1fl/fl mice, at baseline (Top) and following administration of CCh (1.0 mg/kg IP; Bottom). (B) Summary of HR data at baseline (base) and after CCh injection for SLNCre(–):Girk1fl/fl (n = 7) and SLNCre(+):Girk1fl/fl (n = 8) mice. Two-way ANOVA analysis revealed an interaction between genotype and treatment (F1,13 = 38.6; P < 0.001). Symbols: ∗P < 0.05 and ∗∗∗P < 0.001, vs. baseline (within genotype); ###P < 0.001 vs. SLNCre(–):Girk1fl/fl (within treatment). (C) Summary of HRV data at baseline (base) and after CCh injection for SLNCre(–):Girk1fl/fl (n = 7) and SLNCre(+):Girk1fl/fl (n = 8) mice. HRV was calculated as the ratio of standard deviation of RR to mean RR (see section “Materials and Methods”). Two-way ANOVA analysis revealed an interaction between genotype and treatment (F1,13 = 5.4; P < 0.05). Symbols: ∗P < 0.05 vs. baseline (within genotype); ##P < 0.01 vs. SLNCre(–):Girk1fl/fl mice (within treatment).

• 

  FIGURE 3. Impact of whole-heart and tissue-specific GIRK channel ablation on baseline HR and HRV. (A) Summary of baseline HR data. There were no significant differences in baseline HR between wild-type (n = 10) and Girk4-/- (n = 10) mice (t18 = 1.0; P = 0.32), SLNCre(–):Girk1fl/fl (n = 6) and SLNCre(+):Girk1fl/fl (n = 5) mice (t9 = 1.1; P = 0.29), or MLC2VCre(–):Girk1fl/fl (n = 6) and MLC2VCre(+):Girk1fl/fl (n = 8) mice (t12 = 0.7; P = 0.51). (B) Summary of baseline HRV data. There were no significant differences in HRV between wild-type (n = 10) and Girk4-/- (n = 10) mice (t18 = 1.2; P = 0.23), SLNCre(–):Girk1fl/fl (n = 5) and SLNCre(+):Girk1fl/fl (n = 5) mice (t8 = 1.1, P = 0.32), or MLC2VCre(–):Girk1fl/fl (n = 6) and MLC2VCre(+):Girk1fl/fl (n = 8) mice (t12 = 0.32; P = 0.76).

• 

  FIGURE 4. Impact of whole-heart and tissue-specific GIRK channel ablation on VNS-induced decrease in HR. (A,B) Representative ECG segments from anesthetized SLNCre(–):Girk1fl/fl (Top) and SLNCre(+):Girk1fl/fl (Bottom) mice prior to (PRE, Left), during (ON, Middle), and after (POST, Right) VNS. The small spikes observed during VNS (ON) correspond to electrical stimulation artifacts. (C) Relative change in HR (ΔHRON) during VNS. A significant difference in ΔHRON was observed between wild-type (n = 9) and Girk4-/- (n = 10) mice (t17 = 6.5; ∗∗∗P < 0.001), as well as SLNCre(–):Girk1fl/fl (n = 6) and SLNCre(+):Girk1fl/fl (n = 5) mice (t9 = 3.9; ∗∗P < 0.01). No difference was seen between MLC2VCre(–):Girk1fl/fl (n = 6) and MLC2VCre(+):Girk1fl/fl (n = 8) mice (t12 = 0.95; P = 0.36). (D) Relative change in HR (ΔHRON) during VNS, in the presence of atropine (2.0 mg/kg IP). There were no significant differences between wild-type (n = 4) and Girk4-/- (n = 6) mice (t8 = 0.6; P = 0.59), SLNCre(–):Girk1fl/fl (n = 6) and SLNCre(+):Girk1fl/fl (n = 4) mice (t8 = 0.2; P = 0.82), and MLC2VCre(–):Girk1fl/fl (n = 6) and MLC2VCre(+):Girk1fl/fl (n = 4) mice (t8 = 0.2, P = 0.88).

• 

  FIGURE 5. Impact of whole-heart and tissue-specific GIRK channel ablation on VNS-induced increase in HRV. (A) Typical RR tachograms extracted from ECG recordings while VNS was on, comparing beat-to-beat changes in RR intervals between SLNCre(–):Girk1fl/fl (Top) and SLNCre(+):Girk1fl/fl (Bottom) mice. (B) Summary of mean HRV data at baseline and during VNS. Two-way ANOVA analysis revealed an interaction between genotype and treatment between wild-type (n = 8) and Girk4-/- (n = 10) mice (F1,16 = 20.3; P < 0.001), and between SLNCre(–):Girk1fl/fl (n = 6) and SLNCre(+):Girk1fl/fl (n = 5) mice (F1,9 = 12.8; P < 0.01). There was a significant main effect of treatment (F1,12 = 44.0; P < 0.001) between MLC2VCre(–):Girk1fl/fl (n = 6) and MLC2VCre(+):Girk1fl/fl (n = 8) littermates, but no main effect of genotype (F1,12 = 4.2; P = 0.06) or interaction between genotype and treatment (F1,12 = 3.2; P = 0.10). Symbols: ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001, vs. PRE (within genotype); ###P < 0.001, wild-type vs. Girk4-/- (within treatment); ##P < 0.01, SLNCre(–):Girk1fl/fl vs. SLNCre(+):Girk1fl/fl (within treatment). (C) Summary of mean HRV data at baseline and during VNS, following atropine…

• 

  FIGURE 6. Impact of whole-heart and tissue-specific GIRK channel ablation on VNS-induced arrhythmogenesis. (A) Representative segments of ECG recordings of arrhythmias (bigeminy, skipped beats, bradycardia, and tachycardia) observed during VNS. (B) Total number of mice exhibiting arrhythmic episodes during VNS. Fisher’s exact test revealed a significant difference in arrhythmia incidence between wild-type (n = 10) and Girk4-/- (n = 10) mice (∗∗P < 0.01), and SLNCre(–):Girk1fl/fl (n = 6) and SLNCre(+):Girk1fl/fl (n = 5) mice (∗P < 0.05), but no difference between MLC2VCre(–):Girk1fl/fl (n = 6) and MLC2VCre(+):Girk1fl/fl (n = 8) mice (P = 0.14).