AMPK-mediated HCN4 channel phosphorylation contributes to age-related intrinsic bradycardia

Figures

• 

  Figure 1.. AMPK activation inhibits I f in female and male mouse pacemaker cells. (A–D) Whole-cell recordings of If in 3-mo-old female (A and B) and 3-mo-old male (C and D) mouse SAN cells in control (black) and after 4-h treatment with AMPK activator (AICAR 1 mM) (red). Data are obtained from N = 3 females and N = 3 males. (A and C) Representative current traces (left) and activation curves (right). Best fitting with the Boltzmann equation yielded the following values: (A) V1/2 = −91.0 ± 1.67; s = 8.2 ± 0.5 mV (n = 14) and V1/2 = −89.5 ± 1.69 mV; s = 7.5 ± 0.4 mV (n = 13) (n.s.); (C) V1/2 = −88.9 ± 1.76; s = 9.2 ± 0.5 mV (n = 12) and V1/2 = −88.2 ± 1.65 mV; s = 9.1 ± 0.7 mV (n = 15) (n.s.), for control and AICAR-treated cells, respectively. (B and D) Steady-state I/V relations; bar graphs on the left show the distributions of current densities at −125 mV. Predicted means ± SEM assuming GLMM (Gamma distribution) were (B) −35.4 ± 4.03 (n = 14) and −24.2 ± 2.87 pA/pF (n = 13) (P = 0.0982, n.s.); (D) −44.6 ± 5.49 (n = 12) and −25.2 ± 2.77 pA/pF (n = 15) (P = 0.0030, **) for control and AICAR-treated cells, respectively. The difference in current density in female vs. males did not reach…

• 

  Figure S1.. Western blot analysis of the ratio pAMPK/AMPK. (A) Left: Representative western blot of AMPK and pAMPK in HEK293T cells under control conditions and following 4-h incubation in AICAR-containing solution (1 mM). Right: Quantification of active AMPK (pAMPK) protein expression related to the total AMPK (unpaired t test, P = 0.038, n = 5). (B) Real-time PCR from HEK293T cells transfected with hHCN4 from three technical replicates, indicating no significant fractional reduction in hHCN4 mRNA levels after AICAR treatment (paired t test, P = 0.5238, n.s.). Source data are available for this figure: SourceData FS1.

• 

  Figure 2.. AMPK activation reduces membrane expression of HCN4 channels in HEK293 cells. Whole-cell recordings from HEK293T cells transfected with hHCN4 in control (black) and after 4-h treatment with the AMPK activator AICAR (1 mM) (red). (A) Representative current traces (left) and activation curves (right). Best fitting with the Boltzmann equation yielded V1/2 = −76.1 ± 1.3 (n = 18) and −75.0 ± 1.6 mV (n = 16) (unpaired t test, P = 0.6105, n.s.); s = 11.8 ± 0.8 and 11.6 ± 0.6 mV, in control and AICAR-treated cells, respectively. (B) Steady-state I/V relations; bar graphs on the left show the distributions of current densities at −130 mV. Mean ± SEM values were −95.3 ± 12.6 (n = 36) and −38.9 ± 5.29 pA/pF (n = 34) in control and AICAR-treated cells, respectively (GLM Gamma, P < 0.0001, ****). Details of statistical analysis are shown in Table S2.

• 

  Figure S2.. AICAR-induced IHCN4 inhibition in HEK293T vs HEK293F cells. Fully activated IHCN4 I/V curves normalized to capacitance measured from hHCN4-transfected HEK293T (left) or HEK293F cells (right) under control conditions (black) and following 4-h incubation in AICAR-containing solution (1 mM) (red). Linear fitting led to normalized conductance values of 0.704 (n = 16) and 0.418 nS/pF (n = 13) for HEK293T and 1.035 (n = 20) and 0.501 nS/pF (n = 23) for HEK293F cells, in control conditions and following AICAR incubation, respectively. The current density reduction due to AICAR treatment was thus 40.6% in HEK293T and 51.6% in HEK293F cells.

• 

  Figure S3.. FACS experiments on HEK293T cells transfected with a plasmid carrying the construct GFP-hHCN4-HA-tag. The green GFP signal reports in this case the synthesis of hHCN4 and the red HA-tag the positive insertion of the channel in the plasma membrane. Left: Representative histogram of normalized red fluorescence intensity signal in green positive population. Right: Plot of the geometric means of six technical replicates. APC geometric means were 3,135 ± 367 in control runs, and 2,542 ± 430 after AICAR treatment (paired t test, P = 0.044).

• 

  Figure 3.. Removal of AMPK activation abolishes its action. (A and B) Top: Representative traces of current density recorded from HEK293F cells. Data were obtained from day-matched cells in control conditions or after treatment. (A) Bottom: Mean fully activated I/V curves normalized to capacitance measured under control conditions (black) and after 4-h incubation with either AICAR 1 mM (red) or Compound C 30 µM (blue); normalized conductance values from linear fitting were 1.09, 0.49, and 1.21 nS/pF, respectively. (B) Bottom: Similar set of measurements made under control conditions (black), and after 4-h incubation with either AICAR 1 mM (red) or the combination AICAR 1 mM + Compound C 30 µM (blue); normalized conductance values were 0.98, 0.53, and 0.99 nS/pF, respectively. See Tables 1 and S2 for statistical data analysis.

• 

  Figure S4.. SDS-Page analysis of AICAR action on HCN4 protein phosphorylation. (A) SDS-PAGE gel loaded with HCN4 proteins purified from transfected HEK293F cells under control conditions and after AICAR treatment. The quaternary structure of HCN4 is not completely dissolved in a denaturing gel (Saponaro et al., 2021a; Saponaro et al., 2021b). Therefore, DTT was added to favor the disruption of HCN4 tetramers and the consequent isolation of the monomers. Circled bands were analyzed with nLC-ESI-MS/MS Q Exactive HF. (B) Table of all residues found to be phosphorylated in MS analysis. Source data are available for this figure: SourceData FS4.

• 

  Figure 4.. Involvement of hHCN4 serine 1157 in AMPK-mediated channel modulation. (A and B) In A and B, upper panels are typical current records, and lower panels are fully activated I/V relations, normalized to cell capacitance, from HEK293F cells in control conditions (black) and after AICAR treatment (red). Data from cells transfected with wild-type hHCN4 (left in A and B) are compared with data from cells transfected with S1157A or S1157D mutant channels (right in A and B, respectively). Linear fitting of I/V curves yielded the following normalized conductance values (nS/pF): (A) 1.06, 0.52 (left) and 1.04, 1.01 (right); (B) 1.02, 0.54 (left) and 1.08, 1.05 (right) for wild-type and AICAR-treated cells, respectively. In each experiment, control and mutant channels data are from the same day-matched transfection protocol. Statistical analysis and data comparisons are shown in Tables 1 and S2.

• 

  Figure 5.. hHCN4 serine 1158 alone does not directly contribute to AMPK-mediated channel modulation. (A and B) In A and B, upper panels are typical current records, and lower panels are fully activated I/V relations, normalized to cell capacitance, from HEK293F cells in control conditions (black) and after AICAR treatment (red). Data from cells transfected with wild-type hHCN4 (left in A and B) are compared with data from cells transfected with S1158A or S1158D mutant channels (right in A and B, respectively). Linear fitting of I/V curves yielded the following normalized conductance values (nS/pF): (A) 1.08, 0.70 (left) and 1.27, 0.80 (right); (B) 1.10, 0.50 (left) and 1.02, 0.59 (right) for wild-type and AICAR-treated cells, respectively. Statistical analysis and data comparisons are shown in Tables 1 and S2.

• 

  Figure 6.. Loss of AMPK action on double mutant S1157D/S1158D. Representative current records (top) and fully activated I/V relations (bottom) normalized to cell capacitance, from HEK293F cells in control conditions (black) and after AICAR treatment (red). While cells transfected with wild-type hHCN4 (left) show a standard response to AICAR treatment, cells transfected with S1157D/S1158D double-mutant channels in day-matched experiments (right) have a reduced basal current and are unresponsive to AICAR. Linear fitting of I/V curves yielded the following normalized conductance values (nS/pF): 1.20, 0.63 (left) and 0.68, 0.60 (right) for wild-type and AICAR-treated cells, respectively. Statistical analysis and data comparisons are shown in Tables 1 and S2.

• 

  Figure 7.. Age dependence of AMPK action of I f in pacemaker cells. (A–D) Whole-cell recordings of If in 3-mo-old (young) (A and B) and 24-mo-old (old) (C and D) mouse SAN cells in control (black) and after 4-h treatment with AMPK activator (AICAR 1 mM) (red). Data are obtained from N = 3 young and N = 4 old mice. (A and C) Representative current traces (left) and activation curves (right). Best fitting yielded the following values (mV): (A) V1/2 = −89.9 ± 0.81 (n = 21) and −87.8 ± 0.88 (n = 18) (n.s.); s = 8.0 ± 0.4 and 7.4 ± 0.3, for controls and AICAR-treated cells, respectively; (C) V1/2 = −89.5 ± 0.78 (n = 23) and −90.3 ± 0.97 (n = 15) (n.s.); s = 8.0 ± 0.4 and 8.5 ± 0.7 for control and AICAR-treated cells, respectively. (B and D) I/V relations; bar graphs on the left show the distributions of current densities at −125 mV. Mean ± SEM values were (pA/pF) as follows: (B) −42.8 ± 5.34 (n = 21) and −27.9 ± 3.48 (n = 21) (P = 0.0037, **); (D) −25.4 ± 2.91 (n = 23) and −26.5 ± 3.37 (n = 16) (P = 0.9894, n.s.), for control and AICAR-treated cells, respectively. The difference between current densities from old and young mice in control conditions was significant (P = 0.0112, *). Details of…

• 

  Figure 8.. AMPK is constitutively activated in old, but not in young, mice. (A–D) Whole-cell recordings of If in 3-mo-old (young) (A and B) and 24-mo-old (old) (C and D) mouse SAN cells in control (black) and after 4-h treatment with AMPK inhibitor (Compound C 30 µM for 4 h) (blue). Data are obtained from N = 3 young and N = 5 old mice. (A and C) Representative current traces (left) and activation curves (right). Best fitting yielded the following values (mV): (A) V1/2 = −86.2 ± 1.30 (n = 18) and −86.4 ± 1.42 (n = 14) (n.s.); s = 8.3 ± 0.4 and 10.2 ± 0.7; (C) V1/2 = −83.6 ± 1.04 (n = 24) and −82.5 ± 1.19 (n = 17) (n.s.); s = 8.7 ± 0.3 and 9.6 ± 0.4, for control and Compound C–treated cells, respectively. (B and D) I/V relations; bar graphs on the left show the distributions of current densities at −125 mV. Mean ± SE values (pA/pF) were as follows: (B, young) −45.3 ± 6.94 (n = 19) and −44.1 ± 7.24 (n = 16) (P = 0.9986, n.s.); and (D, old) −29.1 ± 3.55 (n = 27) and −47.1 ± 6.37 (n = 20) (P = 0.0110, *), for control and Compound C–treated cells, respectively. Current densities from young mice in control conditions and from old mice after Compound C treatment did not differ significantly (P =…