Frequency-dependent acceleration of relaxation in mammalian heart: a property not relying on phospholamban and SERCA2a phosphorylation

Abstract

An increase in stimulation frequency causes an acceleration of myocardial relaxation (FDAR). Several mechanisms have been postulated to explain this effect, among which is the Ca(2+)-calmodulin-dependent protein kinase (CaMKII)-dependent phosphorylation of the Thr(17) site of phospholamban (PLN). To gain further insights into the mechanisms of FDAR, we studied the FDAR and the phosphorylation of PLN residues in perfused rat hearts, cat papillary muscles and isolated cat myocytes. This allowed us to sweep over a wide range of frequencies, in species with either positive or negative force-frequency relationships, as well as to explore the FDAR under isometric (or isovolumic) and isotonic conditions. Results were compared with those produced by isoprenaline, an intervention known to accelerate relaxation (IDAR) via PLN phosphorylation. While IDAR occurs tightly associated with a significant increase in the phosphorylation of Ser(16) and Thr(17) of PLN, FDAR occurs without significant changes in the phosphorylation of PLN residues in the intact heart and cat papillary muscles. Moreover, in intact hearts, FDAR was not associated with any significant change in the CaMKII-dependent phosphorylation of sarcoplasmic/endoplasmic Ca(2+) ATPase (SERCA2a), and was not affected by the presence of the CaMKII inhibitor, KN-93. In isolated myocytes, FDAR occurred associated with an increase in Thr(17) phosphorylation. However, for a similar relaxant effect produced by isoprenaline, the phosphorylation of PLN (Ser(16) and Thr(17)) was significantly higher in the presence of the beta-agonist. Moreover, the time course of Thr(17) phosphorylation was significantly delayed with respect to the onset of FDAR. In contrast, the time course of Ser(16) phosphorylation, the first residue that becomes phosphorylated with isoprenaline, was temporally associated with IDAR. Furthermore, KN-93 significantly decreased the phosphorylation of Thr(17) that was evoked by increasing the stimulation frequency, but failed to affect FDAR. Taken together, the results provide direct evidence indicating that CaMKII phosphorylation pathways are not involved in FDAR and that FDAR and IDAR do not share a common underlying mechanism. More likely, a CaMKII-independent mechanism could be involved, whereby increasing stimulation frequency would disrupt the SERCA2a-PLN interaction, leading to an increase in SR Ca(2+) uptake and myocardial relaxation.

Figure 1. Frequency-dependent acceleration of relaxation (FDAR) depends on a functional sarcoplasmic reticulum (SR)

A, records of left ventricular developed pressure (LVDP) obtained from a perfused rat heart. The increase of the contraction frequency produced a decrease in LVDP (negative staircase). B, normalized records of LVDP showing that this force–frequency relationship is associated with an accelerated relaxation. C, inhibition of SR function with the treatment of 0.5 μm ryanodine (Ry) plus 2 μm thapsigargin (Thaps) produced a negative inotropic effect at both stimulation frequencies and suppressed both the negative staircase observed in A and FDAR. D, normalized records of LVDP to illustrate better the suppression of FDAR in the presence of Ry and Thaps. E, bars show the ratio between the time to half-relaxation (t½) at the high and low frequencies (FDAR index) obtained in the absence (Control) or the presence of an inhibited SR (Ry + Thaps). Inhibition of SR function abolished the decrease of the ratio below 1 that was produced by FDAR. Results are expressed as means ± s.e.m. of n = 4–6 experiments. *P < 0.05 with respect a ratio of 1; **P < 0.05 with respect to control.

Figure 2. Effects of increasing contraction frequency (A) and isoprenaline (B) on relaxation and phosphorylation of Ser¹⁶ and Thr¹⁷ residues of PLN in the perfused rat heart

The increase in the contraction frequency from 120 to 510 beats min⁻¹ produced a decrease in t½(▪ in A) similar to that induced by increasing isoprenaline from 0.3 to 300 nm (▪ in B). While the relaxant effect of isoprenaline was associated with the enhancement of the phosphorylation of both PLN residues, the increase in the contraction frequency did not modify PLN phosphorylation either at Ser¹⁶ (^) or Thr¹⁷ (•) residues. Results are expressed as means ± s.e.m. of n = 4–12 experiments; t½ is expressed as change (Δms) with respect to 120 beats min⁻¹ (A) or the absence of drug (B). *P < 0.05 with respect to 120 beats min⁻¹ (A) or the absence of drug (B).

Figure 3. Phosphorylation of Ser¹⁶ and Thr¹⁷ residues of PLN at a similar relaxant effect produced by either increasing contraction frequency or addition of appropriate isoprenaline concentration in cat papillary muscles

A similar decrease of t½ was obtained when cat papillary muscles stimulated at 10 beats min⁻¹ were then stimulated at 60 beats min⁻¹ or perfused with 1 μm isoprenaline. Only the relaxant effect produced by the β-agonist was associated with enhancement of PLN phosphorylation at the Ser¹⁶ and Thr¹⁷ residues. Phosphorylation of Ser¹⁶ and Thr¹⁷ residues was maximal at this concentration of isoprenaline. Results are expressed as means ± s.e.m. of n = 6–13 experiments. *P < 0.05 with respect to 1 μm isoprenaline.

Figure 4. Do FDAR and IDAR share a common mechanism?

A and B illustrate the effects of β-adrenergic stimulation on FDAR. A, rat hearts were stimulated at increasing frequencies in the absence (Control) and in the presence of 30 nm isoprenaline (Iso). A similar decrease of t½ to that induced by increasing contraction frequency from 120 to 510 beats min⁻¹ was attained in the absence and in the presence of the β-agonist. B, the lack of action of isoprenaline upon FDAR was also detected in the ratio between t½ at 510 and 120 beats min⁻¹. The significant decrease of this ratio below 1 was not different between control and isoprenaline-stimulated hearts. Results are expressed as means ± s.e.m. of n = 5–12 experiments. *P < 0.05 with respect to a ratio of 1. C and D illustrate the effects of the enhancement of contraction frequency on IDAR. C, hearts, stimulated at 120 beats min⁻¹, were perfused with increasing isoprenaline concentrations. At the maximal concentration-dependent relaxant effect of isoprenaline (300 nm), increasing the contraction frequency to 510 beats min⁻¹ produced an additional decrease in t½. D, phosphorylation levels of the Ser¹⁶ and Thr¹⁷ residues of PLN induced by 300 nm isoprenaline were not modified by a change in stimulation frequency from 120 to 510 beats min⁻¹. Results are expressed as means ± s.e.m. of n = 4 experiments.

Figure 5. Effect of stimulation frequency on CaMKII phosphorylation of SR membranes

A, SR membranes isolated from perfused rat hearts stimulated at 120 and 510 beats min⁻¹ were back phosphorylated with the endogenous CaMKII and [γ-³²P]-ATP. Left panel, autoradiogram; right panel, overall results, expressed as means ± s.e.m. of n = 9 experiments. Note that the in situ treatment (frequency change) did not alter the ability of SERCA2a and PLN to serve as substrate for the in vitro phosphorylation. B, immunoblot of SR membranes probed with SERCA PS-38 antibody. The antibody did not detect any significant phosphorylation of SERCA2a in SR membranes, although it was able to recognize the phosphoepitope peptide conjugated to a scaffolding protein (calibration standard, Cal-38) run in parallel.

Figure 6. Effects of increasing contraction frequency on relaxation and phosphorylation of Ser¹⁶ and Thr¹⁷ residues of PLN in cat myocytes

A, average data of isolated cat myocytes stimulated for 3 min at increasing frequencies. The decrease in the half-relaxation time of the Ca²⁺ transient (t½ transient, ▪) was associated with an increase in the phosphorylation of Thr¹⁷ (•) independently of that of Ser¹⁶ (^), detected as in Fig. 2. B, correlation between Thr¹⁷ phosphorylation and t_½·t_½ is expressed as change (Δms) with respect to 10 beats min⁻¹. Results are expressed as means ± s.e.m. of n = 4–9 experiments.

Figure 7. Phosphorylation of Ser¹⁶ and Thr¹⁷ residues of PLN at a similar relaxant effect produced by either increasing contraction frequency or increasing isoprenaline concentration in cat myocytes

A similar decrease in t½ transient was obtained when isolated cat myocytes were either submitted to a frequency change from 10 to 60 beats min⁻¹ or superfused with 3 nm isoprenaline. While IDAR and FDAR were associated with a significant increase in the phosphorylation of Thr¹⁷, IDAR was also associated with a significant increase in the phosphorylation of Ser¹⁶. Results are expressed as means ± s.e.m. of n = 8–10 experiments. *P < 0.05 with respect to 10 beats min⁻¹.

Figure 8. Time course of relaxation and phosphorylation of PLN residues during an increase in stimulation frequency and administration of isoprenaline in cat myocytes

A, upper panel shows a representative immunoblot of Thr¹⁷ PLN phosphorylation during an increase in stimulation frequency from 10 to 50 beasts min⁻¹. Lower panel shows the average data of Thr¹⁷ PLN phosphorylation (n = 5 for each data point, •) and of the decrease in t½ transient (n = 3) for the same frequency step. The decrease in t½ transient (▪) became significant 2.5 s after the change of frequency, while phosphorylation of Thr¹⁷ occurred more slowly. B, upper panel shows a representative immunoblot of Ser¹⁶ and Thr¹⁷ phosphorylation during administration of 30 nm isoprenaline. Lower panel shows the average data of Ser¹⁶ (^) and Thr¹⁷ (•) PLN phosphorylation (n = 5 for each data point) and of the decrease in t½ transient (n = 6,▪) for the same isoprenaline concentration. The phosphorylation of Ser¹⁶ of PLN occurred temporally associated with the onset of IDAR. Phosphorylation of Thr¹⁷ attained significant levels after 2 min of isoprenaline administration and was associated with a further enhancement of IDAR. *P < 0.05 with respect to the phosphorylation of Ser¹⁶ or Thr¹⁷ at 10 beats min⁻¹ (A) and before isoprenaline administration (B). †P < 0.05 with respect to the t½ value at 10 beats min⁻¹ (A) and before isoprenaline administration (B).

Figure 9. Effects of CaMKII inhibition on the relaxation and Thr¹⁷ phosphorylation induced by a frequency step from 10 to 50 beats min⁻¹

A, representative immunoblot of Ser¹⁶ PLN phosphorylation (PSer¹⁶-PLN) and Thr¹⁷ PLN phosphorylation (PThr¹⁷-PLN) at 10 and 50 beats min⁻¹ in the absence and the presence of 1 μm KN-93 to inhibit CaMKII. Inhibition of CaMKII reduced the phosphorylation of Thr¹⁷ attained at 50 beats min⁻¹. B, bars show the overall results (n = 6–9 experiments). C and D, the diminished Thr¹⁷ phosphorylation did not occur associated with either a reduction in t½ or in the time constant of the Ca²⁺ transient decline (τ). Results are expressed as means ± s.e.m. of n = 6–10 experiments. *P < 0.05 with respect to 10 beats min⁻¹.