Hypertrophic cardiomyopathy mutations increase myofilament Ca2+ buffering, alter intracellular Ca2+ handling, and stimulate Ca2+-dependent signaling

Figures

• 

  Figure 1.. Recombinant adenoviral infection, protein expression, and localization in guinea pig left ventricular cardiomyocytes. A shows the representative GFP expression in cardiomyocytes infected with recombinant adenovirus at ∼1000 MOI 48 h after transfection. B shows the relative protein expression of FLAG-tagged human recombinant WT/R92Q cTnT, WT/R145G cTnI, and WT/D175N α-TM compared with endogenous guinea pig subunits probed using anti-cTnT, anti-cTnI, and anti-α-TM primary antibodies, respectively. The third lane of each gel in B shows that endogenous protein in uninfected control cardiomyocytes is higher than endogenous levels from infected cardiomyocytes, therefore suggesting that recombinant protein replaces the endogenous at the myofilament. The bar graph in C shows the relative infection ratio of each virus as a percentage of green cells, whereas D shows the relative expression levels of endogenous (dark gray bars) versus recombinant (colored bars) protein, calculated from densitometery measurements of doublet bands in B multiplied by the relative expression levels of each virus calculated in C. E shows Western blots for cell lysates expressing each different recombinant…

• 

  Figure 2.. HCM mutations acutely decrease the apparent Kd of cytosolic Ca2+ buffering. Cardiomyocytes expressing WT or mutant cTnT/cTnI/α-TM were simultaneously assessed to measure [Ca2+]i by fura2 fluorescence and NCX current (to calculate [Ca2+]total) by voltage clamping upon the rapid application of 10 mm caffeine. A, plots of Δ[Ca2+]i versus Δ[Ca2+]total for cardiomyocytes containing WT cTnT (n = 20) versus R92Q cTnT (n = 23), WT cTnI (n = 9) versus R145G cTnI (n = 11), and WT α-TM (n = 12) versus D175N α-TM (n = 14) reveal increased buffering capacity in cells containing HCM mutants. The red lines show best fits to the following: [Ca2+]total = {Bmax*[Ca2+]i/(Kd + [Ca2+]i)} + Bmin. B, bar graph comparing mean Kd values for the fits calculated in A. C, bar graph showing the relative estimated changes in buffering capacity (calculated from relative Kd and Bmax values for individual cells undergoing the buffing protocol), where mutant calls have been compared with WT-infected cells set to 1. D, cytosolic buffering curves over a wide range of Ca2+ concentration drawn using the Kd and Bmax values calculated in A. Comparison of WT with HCM mutant: **, p < 0.01; *, p < 0.05.

• 

  Figure 3.. Unloaded sarcomere shortening and Ca2+ transient measurements in adenovirally infected guinea pig left ventricular cardiomyocytes. A, intracellular Ca2+ transients of cardiomyocytes loaded with 1 mm fura2 at 0.5 Hz pacing, expressing HCM-causing mutations cTnT R92Q, cTnI R145G, and α-TM D175N compared with cardiomyocytes infected with the equivalent WT protein. B and C, the average diastolic [Ca2+]i and time to 50% Ca2+ reuptake, respectively. D, the corresponding unloaded sarcomere shortening measurements at 1-Hz pacing. E and F, the average basal sarcomere length and time to 50% relaxation, respectively. The pairwise sarcomere shortening of fura2-loaded cardiomyocytes acquired at 0.5 Hz is shown in Fig. S7. Each curve was averaged from multiple cells (n) taken from at least four separate cell preparations; total n numbers are given in the legends of each plot. Comparison of WT with HCM mutant: ***, p < 0.001; *, p < 0.05.

• 

  Figure 4.. HCM mutant-infected cardiomyocytes have increased NCX current. A, representative traces from voltage-clamp recordings during the application of 10 mm caffeine. B, the average peak amplitude for WT cTnT (n = 24), cTnT R92Q (n = 23), WT cTnI (n = 8), cTnI R145G (n = 8), WT α-TM (n = 15), and α-TM D175N (n = 19). Comparison of WT with HCM mutant: *, p < 0.05.

• 

  Figure 5.. HCM mutant-infected cardiomyocytes have increased SR load, fractional SR Ca2+ release, and SERCA2 activity via CaMKII-dependent PLN phosphorylation. A, representative caffeine-induced [Ca2+]total transients generated by the direct perfusion of 10 mm caffeine after a 5-s pause of pacing comparing WT and HCM mutant-transfected cardiomyocytes, where each transient amplitude has been adjusted from [Ca2+]i (inset) to [Ca2+]total using the buffering curves in Fig. 1A. Bar graphs show the SR load of [Ca2+]total (B), fractional release of Ca2+ (calculated from the [Ca2+]total caffeine transient amplitudes in B and Table S5B subtracted from the preceding [Ca2+]total transient amplitude shown in Table S5B) (C), and the SERCA2 activity (calculated by the subtraction of [Ca2+]total τ-decay constants (total) from the [Ca2+]total caffeine transient (NCX) τ-decay rates Table S5B) (D). The extracted parameters from [Ca2+]i baseline and caffeine transients are also tabulated in Table S5A. E, representative Western blots of total and phosphothreonine 17 PLN. The bar graph shows the densitometric quantification of phosphorylation increase (n = 5). All preparations were paced for 8 h at 0.5 Hz. **,…

• 

  Figure 6.. HCM mutant-infected cardiomyocytes have increased RyR leak/load relationships and CAMKII activation of RyR. SR Ca2+ load and Ca2+ leak via the RyR were assessed by the sequential perfusion of 1.8 mm CaCl2 (A), Na+- and Ca2+-free solution containing 1 mm tetracaine for 50 s (B), the direct application of 10 mm caffeine (C), 1.8 mm CaCl2 for 100 s (D), Na+- and Ca2+-free solution for 50 s (E), and a direct application of 10 mm caffeine (F). A, representative experimental traces for WT cTnT (gray) and cTnT R92Q (purple) cardiomyocytes. All mutants tested were analogous to cTnT R92Q, whereas all control and WT-infected cells tested resembled WT cTnT. B, the observed RyR-dependent leak rate. C, the buffering-adjusted caffeine transient amplitude taken after perfusion with either 0Ca0Na (−) or with 0Na0Ca solution containing 1 μm tetracaine (+). D, the leak/load ratio, calculated as the leak rate divided by the caffeine transient amplitude at the end of each experiment. Each bar graph is an average of 30 cells, 15 of which were acquired as shown in A or B, whereas a further 15 were acquired by reversing the sequence of 0Na0Ca solution with or without 1 μm tetracaine to prevent errors…

• 

  Figure 7.. HCM mutant-infected cardiomyocytes have altered NFAT and ERK phosphorylation resulting in increased nuclear localization. A, total NFAT and phosphoserine 165 NFAT were measured by Western blotting of transfected cardiomyocytes either paced at 0.5 Hz for 8 h (P) or unpaced (UP). B, densitometric quantification of blots from paced cells indicates that expression of HCM mutations results in a significant decrease in phospho-NFAT compared with WT (n = 6). Similarly, C, ERK phosphorylated at threonine 202/tyrosine 204 and total ERK were measured by Western blotting of transfected cardiomyocytes either paced or unpaced. D, densitometric quantification of blots from paced cells show that expression of HCM mutations results in a significant increase in phospho-ERK compared with WT (n = 6). Relative changes in phosphorylation of NFAT and ERK in unpaced cardiomyocytes and the percentages of change in phosphorylation caused by pacing for each group are tabulated in Table S8. E and F show representative immunofluorescence images using anti-NFAT and anti-ERK antibodies, respectively. Individual panels compare HCM mutant to WT-infected cardiomyocytes either unpaced or paced at 0.5 Hz for 8 h.…