Characterization of a Novel PKA Phosphorylation Site, Serine-2030, Reveals No PKA Hyperphosphorylation of the Cardiac Ryanodine Receptor in
http://www.100md.com
循环研究杂志 2005年第4期
the Cardiovascular Research Group (B.X., M.Z., S.R.W.C.), Department of Physiology and Biophysics, Smooth Muscle Research Group (C.S., M.P.W., S.R.W.C.), Department of Biochemistry and Molecular Biology, University of Calgary, Calgary, AB, Canada
Department of Anesthesiology (M.T.J., D.C.W.), Medical College of Wisconsin, Milwaukee, Wis
Laboratory of Cardiovascular Sciences (D.Y., H.C.), National Institutes of Health, Baltimore, Md
Wales Heart Research Institute (A.L.), Cardiff University, UK.
Abstract
Hyperphosphorylation of the cardiac Ca2+ release channel (ryanodine receptor, RyR2) by protein kinase A (PKA) at serine-2808 has been proposed to be a key mechanism responsible for cardiac dysfunction in heart failure (HF). However, the sites of PKA phosphorylation in RyR2 and their phosphorylation status in HF are not well defined. Here we used various approaches to investigate the phosphorylation of RyR2 by PKA. Mutating serine-2808, which was thought to be the only PKA phosphorylation site in RyR2, did not abolish the phosphorylation of RyR2 by PKA. Two-dimensional phosphopeptide mapping revealed two major PKA phosphopeptides, one of which corresponded to the known serine-2808 site. Another, novel, PKA phosphorylation site, serine 2030, was identified by Edman sequencing. Using phospho-specific antibodies, we showed that the novel serine-2030 site was phosphorylated in rat cardiac myocytes stimulated with isoproterenol, but not in unstimulated cells, whereas serine-2808 was considerably phosphorylated before and after isoproterenol treatment. We further showed that serine-2030 was stoichiometrically phosphorylated by PKA, but not by CaMKII, and that mutations of serine-2030 altered neither the FKBP12.6-RyR2 interaction nor the Ca2+ dependence of [3H]ryanodine binding. Moreover, the levels of phosphorylation of RyR2 at serine-2030 and serine-2808 in both failing and non-failing canine hearts were similar. Together, our data indicate that serine-2030 is a major PKA phosphorylation site in RyR2 responding to acute -adrenergic stimulation, and that RyR2 is not hyperphosphorylated by PKA in canine HF.
Key Words: heart failure PKA phosphorylation ryanodine receptor FKBP12.6 phosphospecific antibody
Introduction
It is well known that cardiac myocytes from failing hearts exhibit depressed Ca2+ transients and reduced sarcoplasmic reticulum (SR) Ca2+ content,1eC4 but the precise mechanisms underlying these defects in the Ca2+ homeostasis of HF remain elusive. Alterations in several cardiac Ca2+ handling proteins including the cardiac SR Ca2+ ATPase, Na+/Ca2+ exchanger, and cardiac Ca2+ release channel (ryanodine receptor, RyR2) have been implicated.4 A large body of evidence indicates that in HF the function of the SR Ca2+ ATPase is reduced, whereas the function of the Na+/Ca2+ exchanger is enhanced. On the other hand, the role of RyR2 in HF has been controversial.5 Although impaired RyR2 function is likely to be involved in the abnormal Ca2+ handling commonly observed in HF, the exact alterations in RyR2 function in HF have yet to be defined.
Enhanced phosphorylation of RyR2 by PKA has been proposed as a major impairment that contributes to the reduced SR Ca2+ content and contractility in HF.6 It has been shown that RyR2 in failing hearts is hyperphosphorylated by PKA at a single residue, serine-2808, as a result of an elevated level of circulating catecholamines. This hyperphosphorylation of RyR2 was found to cause dissociation of a 12.6-kDa FK506 binding protein (FKBP12.6) from RyR2, and, consequently, resulted in an increased sensitivity of the channel to Ca2+ activation.6 However, the status of PKA phosphorylation of RyR2 in HF, the impact of PKA phosphorylation of RyR2 on FKBP12.6 binding, and the effect of FKBP12.6 dissociation on RyR2 channel properties are highly contentious. Jiang et al demonstrated that there are no measurable differences in the PKA phosphorylation level and basal channel activity of RyR2 between failing and nonfailing canine hearts.7 Furthermore, several studies have shown no correlation between PKA phosphorylation of RyR2 at serine-2808 and FKBP12.6 dissociation.7eC9 Removal of FKBP12.6 from RyR2 did not significantly affect the single channel properties of RyR2.10
In light of the observations that -adrenergic receptor density was reduced and phosphatase activity was increased in failing hearts,11,12 the finding that RyR2 is hyperphosphorylated by PKA in HF is somewhat unexpected. It is well known that both RyR2 and phospholamban (PLB) are substantially phosphorylated by PKA on -adrenergic stimulation. If RyR2 was hyperphosphorylated by PKA as a consequence of increased -adrenergic stimulation, one would expect that PLB would also be hyperphosphorylated in HF. In contrast, hypophosphorylation, rather than hyperphosphorylation, of PLB by PKA has consistently been observed in failing hearts.13eC15
One possible reason for the discrepancies regarding the phosphorylation state of RyR2 in HF may lie in the complexity of RyR2 phosphorylation. It has been shown that RyR2 is phosphorylated at multiple sites by various kinases including PKA, the Ca2+- and calmodulin-dependent protein kinase II (CaMKII), PKC, and PKG.16 For instance, tryptic phosphopeptide mapping of RyR2 revealed one major and two minor PKA-phosphorylated peptides, and three major CaMKII-phosphorylated peptides, suggesting the existence of multiple PKA and CaMKII phosphorylation sites in RyR2.16 Furthermore, the same phosphorylation site could be phosphorylated by different kinases. For example, serine-2808 is phosphorylated by both PKA and CaMKII.17eC19
To precisely determine the state of phosphorylation of RyR2 by PKA, one needs to fully understand the biochemistry of PKA phosphorylation of RyR2. In the present study, we have focused on investigations of whether or not RyR2 is phosphorylated by PKA at a single residue, serine-2808, and if RyR2 is hyperphosphorylated by PKA in failing hearts. We identified a novel PKA phosphorylation site, serine-2030, in RyR2. We have shown that serine-2030 is markedly phosphorylated in rat cardiac myocytes on treatment with isoproterenol, but not in resting cells, whereas serine-2808 was considerably phosphorylated in the absence of -adrenergic stimulation. Importantly, we observed similar levels of phosphorylation of serine-2030 or serine-2808 in failing and nonfailing canine hearts. Our studies have shed novel insights into the biochemistry of PKA phosphorylation of RyR2 and the state of RyR2 phosphorylation in normal and failing hearts.
Materials and Methods
Mutations were introduced into the mouse RyR2 cDNA by the overlap extension method using the polymerase chain reaction. DNA transfection was performed using the Ca2+ phosphate precipitation method. The purified S2808A protein was phosphorylated in vitro by PKA in the presence of [-32P]ATP. The PKA-phosphorylated S2808A protein was isolated by SDS-PAGE and used to identify additional PKA phosphorylation sites in RyR2 at the Keck Facility of Yale University. The antieCserine-2030 phosphorylation site antiserum was custom-generated by Global Peptide Services.
An expanded Materials and Methods section can be found in the online data supplement available at http://circres.ahajournals.org.
Results
Mutation of Serine-2808 Does Not Abolish Phosphorylation of RyR2 by PKA
To determine whether PKA can phosphorylate RyR2 at sites other than serine-2808, we generated a single point mutation, S2808A, in RyR2. RyR2(wt) and the S2808A mutant were expressed in HEK293 cells and pulled down by GST-FKBP12.6. The pulled-down RyR2(wt) and S2808A proteins were phosphorylated by PKA using [-32P]ATP. Figure 1Ab shows that the RyR2(wt) protein was phosphorylated by PKA (lane 1), but not by boiled PKA (lane 2). Surprisingly, the S2808A mutant protein was also phosphorylated by PKA (lane 3), but not by boiled PKA (lane 4). These observations indicate that serine-2808 is not the sole PKA site in RyR2.
Multiple PKA Phosphorylation Sites Exist in RyR2
Purified RyR2(wt) and S2808A mutant proteins were phosphorylated by PKA in the presence of [-32P]ATP followed by extensive digestion with trypsin. The trypsin-digested 32P-labeled RyR2(wt) and S2808A mutant proteins were separated in the first dimension by electrophoresis and by thin layer chromatography in the second dimension. The autoradiograms of separated tryptic peptides are shown in Figure 1B. Two major radioactive spots (1 and 2) and one faint spot were detected in the tryptic digest of RyR2(wt). The 32P-labeling pattern of the tryptic digest of the S2808A mutant was essentially identical to that of RyR2(wt), except that one of the major radioactive spots, spot 1, was missing. These observations indicate that spot 1 contains the phosphopeptide encompassing the serine-2808 phosphorylation site and that at least one additional major PKA phosphorylation site exists in RyR2.
Identification of a Novel PKA Phosphorylation Site, Serine-2030
About 100 e of 32P-phosphorylated S2808A protein was purified and completely digested with trypsin. The tryptic peptides were then separated by HPLC. A single major peak was found in fractions 53 and 54, which were combined and further digested with chymotrypsin. The resulting peptides were separated again by HPLC. A single peak was detected in fraction 76 (Figure 2B). Liquid-phase Edman sequencing of fraction 76 revealed a partial amino acid sequence of Leu-Leu-X-Leu-Val-Glu-Lys, which matches the amino acid sequence 2028 to 2034, Leu-Leu-Ser-Leu-Val-Glu-Lys, of RyR2 (Figure 2C). Thus, serine-2030 is a previously unknown PKA phosphorylation site.
Residues Serine-2030 and Serine-2808 Are Two Major PKA Phosphorylation Sites in RyR2
RyR2(wt), and the mutants S2030A, S2808A, and S2030A/S2808A, were expressed in HEK293 cells and 32P-phosphorylated by PKA. Figure 3B shows that although single mutations, S2030A and S2808A, reduced the level of PKA phosphorylation, the double mutation S2030A/S2808A nearly abolished PKA phosphorylation of RyR2. These data demonstrate that RyR2 contains two major PKA phosphorylation sites, serine-2030 and serine-2808.
To specifically determine the stoichiometry of phosphorylation of RyR2 at the novel serine-2030 site by PKA, we quantified the level of PKA phosphorylation by measuring the incorporation of 32P and the level of RyR2 protein by determining [3H]ryanodine binding to the single point mutant S2808A and the double mutant S2808A/S2030A. We found that the double mutant S2808A/S2030A incorporated 0.20±0.03 mol Pi per mol RyR monomer (mean±SEM, n=3), whereas the single mutant S2808A incorporated 1.28±0.07 mol Pi per mol RyR monomer (n=3) (Figure 3C). These data indicate that serine-2030 can be stoichiometrically phosphorylated by PKA.
Characterization of a Phosphospecific Antibody Recognizing the Serine-2030 Phosphorylation Site
To probe the phosphorylation status of RyR2 at serine-2030, we generated a phosphospecific antibody against the serine-2030 site (Figure 4A), anti-S2030(PO3), the specificity of which was confirmed by immunoblotting. As shown in Figure 4B, the anti-S2030(PO3) antibody strongly recognized the phosphorylated peptide, S2030(PO3)-peptide (Figure 4Ba), but not the nonphosphorylated peptide, S2030-peptide (Figure 4Bb). Figure 4Bc shows that the S2030-peptide after treatment with PKA can be recognized by the anti-S2030(PO3) antibody. These data indicate that the anti-S2030(PO3) antibody is specific to the phosphorylated form of the S2030-peptide.
The specificity of the anti-S2030(PO3) antibody was further examined using intact recombinant and native RyR2 proteins. RyR2(wt) and the S2030A mutant were expressed with or without a cDNA encoding the catalytic subunit of PKA in HEK293 cells. The expressed RyR2 wt and mutant proteins were immunoprecipitated and immunoblotted with anti-RyR(34c) (Figure 4Ca) and anti-S2030(PO3) (Figure 4Cb) antibodies. As seen in Figure 4C, the RyR2 wt and S2030A mutant proteins, coexpressed with or without PKA, were expressed at equivalent levels (Figure 4Ca). The anti-S2030(PO3) antibody recognized only the RyR2(wt) coexpressed with PKA (Figure 4Cb, lane 3), but did not interact with either RyR2(wt) expressed alone or the S2030A mutant expressed with or without PKA (Figure 4Cb, lanes 1, 2, and 4, respectively).
Figure 4Db shows that the anti-S2030(PO3) antibody recognized the PKA-treated native RyR2, but not RyR2 treated with boiled PKA. Interestingly and in contrast to phosphorylation at serine-2808,9 phosphorylation of the serine-2030 site was not detected in either recombinant or native RyR2 (Figure 4Cb, lane 1, and 4Db, lane 1). The specificity of phosphorylation of serine-2030 by PKA was further examined using a PKA-specific inhibitor, PKI5eC24. Phosphorylation of serine-2030 by PKA was completely abolished by the addition of 5 eol/L PKI5eC24 (Figure 4E).
Serine-2030 Is Phosphorylated in Rat Cardiac Myocytes Stimulated With Isoproterenol
Cell homogenates were prepared from rat cardiac myocytes treated with 1 eol/L isoproterenol for various periods of time, solubilized by SDS, and immunoblotted with the anti-RyR2 (Figure 5Aa), anti-S2030(PO3) (Figure 5Ab), or anti-S2808(PO3) antibody (Figure 5Ac). Consistent with the observations with recombinant and native RyR2, no detectable level of phosphorylation of serine-2030 was observed in rat cardiac myocytes before isoproterenol stimulation (Figure 5Ab, lane 1). On the other hand, isoproterenol (1 eol/L) stimulation for 10 minutes and 1 hour dramatically increased phosphorylation at serine-2030 (Figure 5Ab, lanes 2 and 3). Intriguingly, serine-2030 phosphorylation returned to a low level after prolonged stimulation (24 hours) with isoproterenol (Figure 5Ab, lane 4). To determine the phosphorylation status of serine-2808 under the same conditions, the anti-S2030(PO3) blot was stripped and reprobed with the anti-S2808(PO3) antibody. In keeping with previous reports, serine-2808 was phosphorylated in rat cardiac myocytes before isoproterenol stimulation (Figure 5Ac, lane 1), and its phosphorylation increased on isoproterenol treatment7 (lanes 2 and 3). The level of phosphorylation at serine-2808 was also reduced after prolonged exposure to isoproterenol (24 hours). Collectively, our data demonstrate that serine-2030 is a physiologically relevant phosphorylation site.
To determine whether serine-2030 is a substrate for CaMKII, we coexpressed RyR2 and CaMKII (the -C isoform) in HEK293 cells. We then assessed the phosphorylation state of RyR2 at both serine-2030 and serine-2808 using the anti-S2030(PO3) and anti-S2808(PO3) antibodies, respectively. Coexpression of CaMKII increased the phosphorylation level of RyR2 at serine-2808 by 30% (29.2±0.3%, n=3) (Figure 5Bc and 5C). On the other hand, no phosphorylation of RyR2 at serine-2030 was detected with or without coexpression of CaMKII (Figure 5Bb). Similar studies were also performed in rat cardiac myocytes. No phosphorylation of RyR2 at serine-2030 was detected in cells overexpressing a constitutively active form of CaMKII, whereas phosphorylation of PLB at the CaMKII site, threonine-17, was clearly observed. In addition, we found that the expression of a dominant-negative CaMKII or the addition of a CaMKII inhibitor, KN-93, did not reduce the level of phosphorylation of serine-2808 at rest (unpublished data, 2004). These results indicate that serine-2030 is unlikely to be a CaMKII phosphorylation site. They also suggest that additional kinases may be involved in the phosphorylation of serine-2808 at rest.
Effect of PKA Phosphorylation and Mutations at Serine-2030 on FKBP12.6-RyR2 Interaction
HEK293 cells were transfected with RyR2(wt), S2030A, or S2030D cDNA, or cotransfected with the RyR2(wt) and PKA cDNAs. The expressed wt and mutant RyR2 proteins were then pulled down by GST-FKBP12.6. As shown in Figure 6Aa, all RyR2 variants, RyR2(wt), PKA-phosphorylated RyR2(wt), S2030A, and S2030D, were pulled down by GST-FKBP12.6 at comparable levels. Immunoblotting using the anti-S2030(PO3) antibody confirmed that RyR2(wt) was phosphorylated at serine-2030 when coexpressed with PKA (Figure 6Ab, lane 2), but the RyR2(wt) expressed alone, and the S2030A and S2030D mutants were not (Figure 6Ab, lanes 1, 3, and 4), whereas phosphorylation of serine-2808 was detected in RyR2(wt) and the S2030A and S2030D mutants with or without coexpression of PKA (Figure 6Ac). Thus, these results demonstrate that GST-FKBP12.6 is able to interact with RyR2 irrespective of the phosphorylation status of RyR2 at serine-2030.
To determine whether RyR2(wt) and the RyR2 mutants interact with FKBP12.6 in HEK293 cells, we cotransfected the cells with FKBP12.6 and wt or mutant RyR2 cDNAs. As shown in Figure 6Bb, RyR2(wt) coexpressed with PKA (lane 2) was phosphorylated at serine-2030, but RyR2(wt) expressed alone or the S2030A or S2030D mutants were not (lanes 1, 3, and 4). Importantly, all RyR2 variants, the nonphosphorylated RyR2(wt), phosphorylated RyR2(wt), and the S2030A and S2030D mutants, were able to precipitate the coexpressed FKBP12.6 (Figure 6Bc). These data further demonstrate that FKBP12.6 can interact with both the serine-2030 phosphorylated and nonphosphorylated forms of RyR2.
Effect of Mutations at Serine-2030 on the Sensitivity of RyR2 to Ca2+ Activation
In an attempt to assess the functional impact of serine-2030 phosphorylation on RyR2 channel function, we determined the sensitivities to Ca2+ activation of RyR2(wt), the S2030A mutant that completely removes phosphorylation, and the S2030D mutant that is thought to mimic phosphorylation. As shown in Figure 7, the Ca2+ dependence of [3H]ryanodine binding to the S2030A and S2030D mutants is virtually identical to that of RyR2 (wt). The EC50 values were 0.14±0.01 eol/L (mean±SEM, n=3) for RyR2 (wt), 0.17±0.01 eol/L (n=3) for S2030A, and 0.17±0.01 eol/L (n=3) for S2030D.
RyR2 Is Not Hyperphosphorylated by PKA in Failing Canine Hearts
To investigate whether phosphorylation of RyR2 is altered in HF, we performed immunoblotting analyses of tissue homogenates from failing and nonfailing canine heart samples using the anti-S2808(PO3) (Figure 8A) and anti-S2030(PO3) (Figure 8B) antibodies. Consistent with previous reports,7 we observed considerable levels of phosphorylation of RyR2 at serine-2808 in both failing and nonfailing hearts (Figure 8Ab), but no hyperphosphorylation in failing hearts (Figure 8Ac). On the other hand, low or near background levels of phosphorylation of serine-2030 were detected in failing and nonfailing hearts, but there was no hyperphosphorylation of serine-2030 in failing hearts (Figure 8Bb). It should be noted that the anti-S2030(PO3) antibody is able to recognize the phosphorylated serine-2030 of canine RyR2 (Figure 4Db) and RyR2 from intact canine hearts stimulated with isoproterenol (Figure 8Bb, right panel). Thus, the lack of a phosphorylation signal is not due to the failure of the anti-S2030(PO3) antibody, but due to the lack of significant phosphorylation of serine-2030.
Discussion
The state of PKA phosphorylation of RyR2 in HF remains highly controversial despite intensive investigation. In our opinion, this controversy is due largely to the lack of understanding of the molecular basis of phosphorylation of RyR2. It is generally believed that serine-2808 is the only PKA phosphorylation site in RyR2. Consequently, the level of phosphorylation of serine-2808 has widely been used as an indicator of phosphorylation of RyR2 by PKA.6,20 However, the results of our present study do not support this common belief. We showed that, in addition to serine-2808, RyR2 was phosphorylated by PKA at a previously unknown site, serine-2030 (Figures 2 and 3). We demonstrated that serine-2030 phosphorylation responded to acute -adrenergic stimulation, whereas serine-2808 was considerably phosphorylated in the absence of -adrenergic stimulation (Figures 4, 5, and 8). These data clearly indicate that serine-2808 is not the only PKA site in RyR2, and that the level of serine-2808 phosphorylation does not correlate closely with -adrenergic stimulation. Thus, our findings raise a concern about the appropriateness of using serine-2808 phosphorylation as an indicator of RyR2 phosphorylation by PKA, and indicate that phosphorylation of serine-2030 more accurately reflects PKA phosphorylation of RyR2.
Serine-2030 Is a Specific PKA Phosphorylation Site
Although serine-2030 was predicted to be a CaMKII phosphorylation site rather than a PKA site based on sequence analysis,17 several lines of evidence indicate the opposite. Serine-2030 was phosphorylated by an active PKA, but not by an inactive PKA (boiled PKA) (Figure 4), demonstrating that the observed phosphorylation of serine-2030 in the presence of exogenous PKA was mediated by the added PKA, but not by other kinases potentially copurified with RyR2. We also found that serine-2030 was stoichiometrically phosphorylated by PKA (Figure 3), and that phosphorylation of serine-2030 by PKA was completely abolished by the addition of the PKA-specific inhibitor, PKI5eC24 (Figure 4). Moreover, coexpression of CaMKII with RyR2 led to phosphorylation of RyR2 at serine-2808, but not at serine-2030 (Figure 5). Hence, our results support the conclusion that serine-2030 is a PKA, but not a CaMKII, phosphorylation site.
The identification of an additional PKA phosphorylation site in RyR2 is apparently inconsistent with a recent report that showed that a single point mutation, S2809A, in human RyR2 diminished PKA phosphorylation.20 This discrepancy may result from species differences in RyR2 phosphorylation. To test this possibility, we determined whether serine-2030 of RyR2 from various species could be phosphorylated. Using a phosphospecific antibody, we were able to detect phosphorylation of serine-2030 of the canine (Figure 4D), mouse (Figure 4C and 4E), rat (Figure 5A), and human RyR2 (not shown). Thus, the discrepancy is unlikely to be due to species difference; rather, it likely results from differences in experimental conditions.
Unlike serine-2030, serine-2808 is a substrate for multiple kinases. It was originally identified as a unique CaMKII phosphorylation site in RyR218,19 and later shown to be also a PKA site.6 In addition, serine-2808 was considerably phosphorylated in rat cardiac myocytes in the absence of -adrenergic stimulation and in the presence of a CaMKII inhibitor, KN-93, and a dominant-negative CaMKII (unpublished data, 2004), suggesting that serine-2808 may be phosphorylated by kinase(s) other than PKA and CaMKII at rest. Thus, the extent of phosphorylation of serine-2808 would depend not only on the activity of PKA, but also on the activity of CaMKII and probably other kinase(s). Influence of various kinases on serine-2808 phosphorylation is probably a major source of the discrepancy concerning the state of phosphorylation of RyR2 by PKA.
Phosphoryaltion State of RyR2 in HF
Our assessments of the level of serine-2030 phosphorylation revealed that RyR2 was not hyperphosphorylated by PKA in HF (Figure 8), whereas Marks and his colleagues6 showed that RyR2 was hyperphosphorylated by PKA in HF. Because serine-2030 is a substrate for PKA but not for CaMKII, whereas serine-2808 is a substrate for both PKA and CaMKII, this discrepancy could potentially be explained by the possibility that RyR2 is hyperphosphorylated by CaMKII, rather than by PKA. In line with this possibility, Wang et al21 has recently shown that sustained -adrenergic stimulation, a condition thought to occur in HF, led to a transient activation of the cAMP/PKA signaling pathway, followed by a persistent activation of CaMKII. Consistent with this observation, we found that phosphorylation of RyR2 at serine-2030 was transient, whereas phosphorylation of serine-2808 persisted during sustained - stimulation (Figure 5). Given the transient nature of PKA activation, which is better suited for fight-or-flight response, and of serine-2030 phosphorylation in response to sustained -stimulation, it is difficult to imagine that hyperphosphorylation of RyR2 by PKA could be maintained in HF. In keeping with this view, we found that phosphorylation of the serine-2030 PKA site was hardly detected in both failing and nonfailing canine (Figure 8), human, and rat hearts (unpublished data, 2004). Hence, enhanced cardiac contractility as a result of sustained -adrenergic stimulation may be mediated by activation of the CaMKII, rather than the PKA signaling pathway.21
Although the idea that RyR2 is hyperphosphorylated by CaMKII but not by PKA in HF is possible, it seems to be inconsistent with a recent report by Wehrens et al.20 They found that RyR2 was phosphorylated by CaMKII at a single site, serine-2815, but not at serine-2808. Thus, their findings suggest that increased serine-2808 phosphorylation observed in HF is unlikely to be associated with CaMKII. This idea also appears to be inconsistent with the results of our present study. In contrast to Marks’ group, we failed to detect hyperphosphorylation of RyR2 at serine-2808 in failing hearts. It is important to emphasize that the issue of RyR2 phosphorylation by CaMKII is also controversial. The exact CaMKII phosphorylation sites in RyR2 and whether RyR2 is hyperphosphorylated by CaMKII in HF are unclear.
Alternatively, the discrepancy regarding the state of PKA phosphorylation of RyR2 may result from technical differences in experimental conditions. In our studies, the level of serine-2808 phosphorylation was determined using SDS-solubilized tissue homogenates isolated from quickly frozen heart tissues. In the studies by Marks and his colleagues, the extent of phosphorylation of RyR2 was determined using RyR2 proteins immunoprecipitated from Triton X-100 solubilized postmitochondrial or SR fractions.6 During solubilization and immunoprecipitation, some kinases or phosphatases may remain active and thus may alter the phosphorylation state of serine-2808. The activities of these kinases or phosphatases might be different between failing and nonfailing hearts and under different experimental conditions, thus leading to different extents of phosphorylation of serine-2808 independent of PKA. Hence, technical differences in sample handling may account for the different phosphorylation levels of serine-2808 and thus of RyR2 reported by different groups. To accurately assess the level of phosphorylation of RyR2, it is critical to use samples in which the activities of kinases and phosphatases have quickly been inactivated (eg, by the treatment with SDS). Accordingly, we suggest that SDS-treated tissue homogenates from heart muscles, rather than immunoprecipitated RyR2 proteins, be used directly for the assessment of phosphorylation of RyR2 to minimize further RyR2 phosphorylation or dephosphorylation.
Impact of Serine-2030 Phosphorylation and Mutation
Our data indicate that PKA phosphorylation of RyR2 at the novel serine-2030 site does not abolish FKBP12.6 binding, consistent with our previous finding.9 It should be emphasized that complete PKA phosphorylation of both native RyR2 and recombinant RyR2 does not dissociate the endogenous or coexpressed FKBP12.6 from RyR2.9 It should also be noted that native RyR2 proteins isolated from canine cardiac SR, although already heavily phosphorylated at serine-2808,17 exhibited stoichiometric binding to FKBP12.6.10 Thus, our findings together with previous observations clearly demonstrate that there is no correlation between PKA phosphorylation of RyR2 and FKBP12.6 dissociation.
Our data also indicate that mutations at serine-2030 do not alter the sensitivity of RyR2 to Ca2+ activation as determined by [3H]ryanodine binding. Stange et al8 have also recently shown that mutations at the serine-2808 site have no measurable effect on RyR2 channel function. In addition, PKA phosphorylation does not appear to alter the properties of Ca2+ sparks in mouse ventricular myocytes.22 On the other hand, Marx et al6 reported that phosphorylation of RyR2 by PKA increased the sensitivity of the RyR2 channel to Ca2+ activation. Thus, the exact role of PKA phosphorylation in RyR2 channel function is unclear and requires further investigation. However, it is clear that phosphorylation of serine-2030 by PKA is physiologically relevant, as its phosphorylation is closely associated with -adrenergic stimulation. Generation and characterization of mouse models that contain serine-2030 mutations in RyR2 will provide some definitive insights into the physiological role of phosphorylation of RyR2 by PKA.
In summary, the present study reveals that serine-2030 of RyR2 is a novel PKA phosphorylation site that responds to acute -adrenergic stimulation, and that RyR2 in failing canine hearts is not hyperphosphorylated by PKA. Furthermore, we found that phosphorylation of RyR2 by PKA at serine-2030 does not dissociate FKBP12.6 from RyR2. Taken together, our data do not support previous findings that (1) RyR2 is phosphorylated by PKA at a single residue; (2) RyR2 is hyperphosphorylated by PKA in HF; and (3) phosphorylation of RyR2 by PKA dissociates FKBP12.6 from RyR2.6 Additional investigations are needed to unveil the molecular mechanisms of regulation of RyR2 by PKA and other kinases in normal and diseased hearts.
Acknowledgments
This work was supported by research grants from CIHR to M.P.W. and S.R.W.C., and from the American Heart Association Northland Affiliate and Medical College of Wisconsin to M.T.J. B.X. is a recipient of an AHFMR Studentship Award. M.P.W. is an AHFMR Scientist and Canada Research Chair (Tier I) in Biochemistry. S.R.W.C. is an AHFMR Senior Scholar. The authors would like to thank Drs Hector Valdivia, John Tyberg, and Masakazu Obayashi for providing the canine cardiac tissues, Dr Andrew Braun for providing the cDNAs encoding PKA and CaMKII, Jeff Bolstad for critical reading of the manuscript, and John Tessmer and David Schwabe for technical assistance.
References
Nuss HB, Kaab S, Kass DA, Tomaselli GF, Marban E. Cellular basis of ventricular arrhythmias and abnormal automaticity in heart failure. Am J Physiol Heart Circ Physiol. 1999; 277: H80eCH91.
Zaugg C, Buser P. When calcium turns arrhythmogenic: intracellular calcium handling during the development of hypertrophy and heart failure. Croat Med J. 2001; 42: 24eC32.
Janse M. Electrophysiological changes in heart failure and their relationship to arrhythmogenesis. Cardiovasc Res. 2004; 61: 208eC217.
Pogwizd SM, Bers DM. Cellular Basis of Triggered Arrhythmias in Heart Failure. Trends in Cardiovascular Medicine. 2004; 14: 61eC66.
Bers DM, Eisner DA, Valdivia HH. Sarcoplasmic Reticulum Ca2+ and Heart Failure: Roles of Diastolic Leak and Ca2+ Transport. Circ Res. 2003; 93: 487eC490.
Marx SO, Reiken S, Hisamatsu Y, Jayaraman T, Burkhoff D, Rosemblit N, Marks AR. PKA phosphorylation dissociates FKBP12.6 from the calcium release channel (ryanodine receptor): defective regulation in failing hearts. Cell. 2000; 101: 365eC376.
Jiang MT, Lokuta AJ, Farrell EF, Wolff MR, Haworth RA, Valdivia HH. Abnormal Ca2+ release, but normal ryanodine receptors, in canine and human heart failure. Circ Res. 2002; 91: 1015eC1022.
Stange M, Xu L, Balshaw D, Yamaguchi N, Meissner G. Characterization of recombinant skeletal muscle (Ser-2843) and cardiac muscle (Ser-2809) ryanodine receptor phosphorylation mutants. J Biol Chem. 2003; 278: 51693eC51702.
Xiao B, Sutherland C, Walsh MP, Chen SRW. Protein kinase A phosphorylation at serine-2808 of the cardiac Ca2+-release channel (ryanodine receptor) does not dissociate 12.6-kDa FK506-binding protein (FKBP12.6). Circ Res. 2004; 94: 487eC495.
Timerman AP, Onoue H, Xin HB, Barg S, Copello J, Wiederrecht G, Fleischer S. Selective binding of FKBP12.6 by the cardiac ryanodine receptor. J Biol Chem. 1996; 271: 20385eC20391.
Bristow M, Ginsburg R, Minobe W, Cubicciotti R, Sageman W, Lurie K, Billingham M, Harrison D, Stinson E. Decreased catecholamine sensitivity and beta-adrenergic-receptor density in failing human hearts. N Engl J Med. 1982; 307: 205eC211.
Neumann J. Altered phosphatase activity in heart failure, influence on Ca2+ movement. Basic Res Cardiol. 2002; 97 (suppl 1): I91eCI95.
Schmidt U, Hajjar RJ, Kim CS, Lebeche D, Doye AA, Gwathmey JK. Human heart failure: cAMP stimulation of SR Ca2+-ATPase activity and phosphorylation level of phospholamban. Am J Physiol Heart Circ Physiol. 1999; 277: H474eCH480.
Schwinger RHG, Munch G, Bolck B, Karczewski P, Krause E-G, Erdmann E. Reduced Ca2+-sensitivity of SERCA 2a in failing human myocardium due to reduced serin-16 phospholamban phoshorylation. J Mol Cell Cardiol. 1999; 31: 479eC491.
Sande JB, Sjaastad I, Hoen IB, Bokenes J, Tonnessen T, Holt E, Lunde PK, Christensen G. Reduced level of serine16 phosphorylated phospholamban in the failing rat myocardium: a major contributor to reduced SERCA2 activity. Cardiovasc Res. 2002; 53: 382eC391.
Takasago T, Imagawa T, Furukawa K, Ogurusu T, Shigekawa M. Regulation of the cardiac ryanodine receptor by protein kinase-dependent phosphorylation. J Biochem (Tokyo). 1991; 109: 163eC170.
Rodriguez P, Bhogal MS, Colyer J. Stoichiometric phosphorylation of cardiac ryanodine receptor on serine-2809 by calmodulin-dependent kinase II and protein kinase A. J Biol Chem. 2003; 278: 38593eC38600.
Witcher DR, Kovacs RJ, Schulman H, Cefali DC, Jones LR. Unique phosphorylation site on the cardiac ryanodine receptor regulates calcium channel activity. J Biol Chem. 1991; 266: 11144eC11152.
Witcher DR, Strifler BA, Jones LR. Cardiac-specific phosphorylation site for multifunctional Ca2+/calmodulin-dependent protein kinase is conserved in the brain ryanodine receptor. J Biol Chem. 1992; 267: 4963eC4967.
Wehrens XHT, Lehnart SE, Reiken SR, Marks AR. Ca2+/calmodulin-dependent protein kinase II phosphorylation regulates the cardiac ryanodine receptor. Circ Res. 2004; 94: e61eCe70.
Wang W, Zhu W, Wang S, Yang D, Crow MT, Xiao R-P, Cheng H. Sustained 1-adrenergic stimulation modulates cardiac contractility by Ca2+/calmodulin kinase signaling pathway. Circ Res. 2004; 95: 798eC806.
Li Y, Kranias EG, Mignery GA, Bers DM. Protein kinase A phosphorylation of the ryanodine receptor does not affect calcium sparks in mouse ventricular myocytes. Circ Res. 2002; 90: 309eC316.(Bailong Xiao, Ming Tao Ji)
Department of Anesthesiology (M.T.J., D.C.W.), Medical College of Wisconsin, Milwaukee, Wis
Laboratory of Cardiovascular Sciences (D.Y., H.C.), National Institutes of Health, Baltimore, Md
Wales Heart Research Institute (A.L.), Cardiff University, UK.
Abstract
Hyperphosphorylation of the cardiac Ca2+ release channel (ryanodine receptor, RyR2) by protein kinase A (PKA) at serine-2808 has been proposed to be a key mechanism responsible for cardiac dysfunction in heart failure (HF). However, the sites of PKA phosphorylation in RyR2 and their phosphorylation status in HF are not well defined. Here we used various approaches to investigate the phosphorylation of RyR2 by PKA. Mutating serine-2808, which was thought to be the only PKA phosphorylation site in RyR2, did not abolish the phosphorylation of RyR2 by PKA. Two-dimensional phosphopeptide mapping revealed two major PKA phosphopeptides, one of which corresponded to the known serine-2808 site. Another, novel, PKA phosphorylation site, serine 2030, was identified by Edman sequencing. Using phospho-specific antibodies, we showed that the novel serine-2030 site was phosphorylated in rat cardiac myocytes stimulated with isoproterenol, but not in unstimulated cells, whereas serine-2808 was considerably phosphorylated before and after isoproterenol treatment. We further showed that serine-2030 was stoichiometrically phosphorylated by PKA, but not by CaMKII, and that mutations of serine-2030 altered neither the FKBP12.6-RyR2 interaction nor the Ca2+ dependence of [3H]ryanodine binding. Moreover, the levels of phosphorylation of RyR2 at serine-2030 and serine-2808 in both failing and non-failing canine hearts were similar. Together, our data indicate that serine-2030 is a major PKA phosphorylation site in RyR2 responding to acute -adrenergic stimulation, and that RyR2 is not hyperphosphorylated by PKA in canine HF.
Key Words: heart failure PKA phosphorylation ryanodine receptor FKBP12.6 phosphospecific antibody
Introduction
It is well known that cardiac myocytes from failing hearts exhibit depressed Ca2+ transients and reduced sarcoplasmic reticulum (SR) Ca2+ content,1eC4 but the precise mechanisms underlying these defects in the Ca2+ homeostasis of HF remain elusive. Alterations in several cardiac Ca2+ handling proteins including the cardiac SR Ca2+ ATPase, Na+/Ca2+ exchanger, and cardiac Ca2+ release channel (ryanodine receptor, RyR2) have been implicated.4 A large body of evidence indicates that in HF the function of the SR Ca2+ ATPase is reduced, whereas the function of the Na+/Ca2+ exchanger is enhanced. On the other hand, the role of RyR2 in HF has been controversial.5 Although impaired RyR2 function is likely to be involved in the abnormal Ca2+ handling commonly observed in HF, the exact alterations in RyR2 function in HF have yet to be defined.
Enhanced phosphorylation of RyR2 by PKA has been proposed as a major impairment that contributes to the reduced SR Ca2+ content and contractility in HF.6 It has been shown that RyR2 in failing hearts is hyperphosphorylated by PKA at a single residue, serine-2808, as a result of an elevated level of circulating catecholamines. This hyperphosphorylation of RyR2 was found to cause dissociation of a 12.6-kDa FK506 binding protein (FKBP12.6) from RyR2, and, consequently, resulted in an increased sensitivity of the channel to Ca2+ activation.6 However, the status of PKA phosphorylation of RyR2 in HF, the impact of PKA phosphorylation of RyR2 on FKBP12.6 binding, and the effect of FKBP12.6 dissociation on RyR2 channel properties are highly contentious. Jiang et al demonstrated that there are no measurable differences in the PKA phosphorylation level and basal channel activity of RyR2 between failing and nonfailing canine hearts.7 Furthermore, several studies have shown no correlation between PKA phosphorylation of RyR2 at serine-2808 and FKBP12.6 dissociation.7eC9 Removal of FKBP12.6 from RyR2 did not significantly affect the single channel properties of RyR2.10
In light of the observations that -adrenergic receptor density was reduced and phosphatase activity was increased in failing hearts,11,12 the finding that RyR2 is hyperphosphorylated by PKA in HF is somewhat unexpected. It is well known that both RyR2 and phospholamban (PLB) are substantially phosphorylated by PKA on -adrenergic stimulation. If RyR2 was hyperphosphorylated by PKA as a consequence of increased -adrenergic stimulation, one would expect that PLB would also be hyperphosphorylated in HF. In contrast, hypophosphorylation, rather than hyperphosphorylation, of PLB by PKA has consistently been observed in failing hearts.13eC15
One possible reason for the discrepancies regarding the phosphorylation state of RyR2 in HF may lie in the complexity of RyR2 phosphorylation. It has been shown that RyR2 is phosphorylated at multiple sites by various kinases including PKA, the Ca2+- and calmodulin-dependent protein kinase II (CaMKII), PKC, and PKG.16 For instance, tryptic phosphopeptide mapping of RyR2 revealed one major and two minor PKA-phosphorylated peptides, and three major CaMKII-phosphorylated peptides, suggesting the existence of multiple PKA and CaMKII phosphorylation sites in RyR2.16 Furthermore, the same phosphorylation site could be phosphorylated by different kinases. For example, serine-2808 is phosphorylated by both PKA and CaMKII.17eC19
To precisely determine the state of phosphorylation of RyR2 by PKA, one needs to fully understand the biochemistry of PKA phosphorylation of RyR2. In the present study, we have focused on investigations of whether or not RyR2 is phosphorylated by PKA at a single residue, serine-2808, and if RyR2 is hyperphosphorylated by PKA in failing hearts. We identified a novel PKA phosphorylation site, serine-2030, in RyR2. We have shown that serine-2030 is markedly phosphorylated in rat cardiac myocytes on treatment with isoproterenol, but not in resting cells, whereas serine-2808 was considerably phosphorylated in the absence of -adrenergic stimulation. Importantly, we observed similar levels of phosphorylation of serine-2030 or serine-2808 in failing and nonfailing canine hearts. Our studies have shed novel insights into the biochemistry of PKA phosphorylation of RyR2 and the state of RyR2 phosphorylation in normal and failing hearts.
Materials and Methods
Mutations were introduced into the mouse RyR2 cDNA by the overlap extension method using the polymerase chain reaction. DNA transfection was performed using the Ca2+ phosphate precipitation method. The purified S2808A protein was phosphorylated in vitro by PKA in the presence of [-32P]ATP. The PKA-phosphorylated S2808A protein was isolated by SDS-PAGE and used to identify additional PKA phosphorylation sites in RyR2 at the Keck Facility of Yale University. The antieCserine-2030 phosphorylation site antiserum was custom-generated by Global Peptide Services.
An expanded Materials and Methods section can be found in the online data supplement available at http://circres.ahajournals.org.
Results
Mutation of Serine-2808 Does Not Abolish Phosphorylation of RyR2 by PKA
To determine whether PKA can phosphorylate RyR2 at sites other than serine-2808, we generated a single point mutation, S2808A, in RyR2. RyR2(wt) and the S2808A mutant were expressed in HEK293 cells and pulled down by GST-FKBP12.6. The pulled-down RyR2(wt) and S2808A proteins were phosphorylated by PKA using [-32P]ATP. Figure 1Ab shows that the RyR2(wt) protein was phosphorylated by PKA (lane 1), but not by boiled PKA (lane 2). Surprisingly, the S2808A mutant protein was also phosphorylated by PKA (lane 3), but not by boiled PKA (lane 4). These observations indicate that serine-2808 is not the sole PKA site in RyR2.
Multiple PKA Phosphorylation Sites Exist in RyR2
Purified RyR2(wt) and S2808A mutant proteins were phosphorylated by PKA in the presence of [-32P]ATP followed by extensive digestion with trypsin. The trypsin-digested 32P-labeled RyR2(wt) and S2808A mutant proteins were separated in the first dimension by electrophoresis and by thin layer chromatography in the second dimension. The autoradiograms of separated tryptic peptides are shown in Figure 1B. Two major radioactive spots (1 and 2) and one faint spot were detected in the tryptic digest of RyR2(wt). The 32P-labeling pattern of the tryptic digest of the S2808A mutant was essentially identical to that of RyR2(wt), except that one of the major radioactive spots, spot 1, was missing. These observations indicate that spot 1 contains the phosphopeptide encompassing the serine-2808 phosphorylation site and that at least one additional major PKA phosphorylation site exists in RyR2.
Identification of a Novel PKA Phosphorylation Site, Serine-2030
About 100 e of 32P-phosphorylated S2808A protein was purified and completely digested with trypsin. The tryptic peptides were then separated by HPLC. A single major peak was found in fractions 53 and 54, which were combined and further digested with chymotrypsin. The resulting peptides were separated again by HPLC. A single peak was detected in fraction 76 (Figure 2B). Liquid-phase Edman sequencing of fraction 76 revealed a partial amino acid sequence of Leu-Leu-X-Leu-Val-Glu-Lys, which matches the amino acid sequence 2028 to 2034, Leu-Leu-Ser-Leu-Val-Glu-Lys, of RyR2 (Figure 2C). Thus, serine-2030 is a previously unknown PKA phosphorylation site.
Residues Serine-2030 and Serine-2808 Are Two Major PKA Phosphorylation Sites in RyR2
RyR2(wt), and the mutants S2030A, S2808A, and S2030A/S2808A, were expressed in HEK293 cells and 32P-phosphorylated by PKA. Figure 3B shows that although single mutations, S2030A and S2808A, reduced the level of PKA phosphorylation, the double mutation S2030A/S2808A nearly abolished PKA phosphorylation of RyR2. These data demonstrate that RyR2 contains two major PKA phosphorylation sites, serine-2030 and serine-2808.
To specifically determine the stoichiometry of phosphorylation of RyR2 at the novel serine-2030 site by PKA, we quantified the level of PKA phosphorylation by measuring the incorporation of 32P and the level of RyR2 protein by determining [3H]ryanodine binding to the single point mutant S2808A and the double mutant S2808A/S2030A. We found that the double mutant S2808A/S2030A incorporated 0.20±0.03 mol Pi per mol RyR monomer (mean±SEM, n=3), whereas the single mutant S2808A incorporated 1.28±0.07 mol Pi per mol RyR monomer (n=3) (Figure 3C). These data indicate that serine-2030 can be stoichiometrically phosphorylated by PKA.
Characterization of a Phosphospecific Antibody Recognizing the Serine-2030 Phosphorylation Site
To probe the phosphorylation status of RyR2 at serine-2030, we generated a phosphospecific antibody against the serine-2030 site (Figure 4A), anti-S2030(PO3), the specificity of which was confirmed by immunoblotting. As shown in Figure 4B, the anti-S2030(PO3) antibody strongly recognized the phosphorylated peptide, S2030(PO3)-peptide (Figure 4Ba), but not the nonphosphorylated peptide, S2030-peptide (Figure 4Bb). Figure 4Bc shows that the S2030-peptide after treatment with PKA can be recognized by the anti-S2030(PO3) antibody. These data indicate that the anti-S2030(PO3) antibody is specific to the phosphorylated form of the S2030-peptide.
The specificity of the anti-S2030(PO3) antibody was further examined using intact recombinant and native RyR2 proteins. RyR2(wt) and the S2030A mutant were expressed with or without a cDNA encoding the catalytic subunit of PKA in HEK293 cells. The expressed RyR2 wt and mutant proteins were immunoprecipitated and immunoblotted with anti-RyR(34c) (Figure 4Ca) and anti-S2030(PO3) (Figure 4Cb) antibodies. As seen in Figure 4C, the RyR2 wt and S2030A mutant proteins, coexpressed with or without PKA, were expressed at equivalent levels (Figure 4Ca). The anti-S2030(PO3) antibody recognized only the RyR2(wt) coexpressed with PKA (Figure 4Cb, lane 3), but did not interact with either RyR2(wt) expressed alone or the S2030A mutant expressed with or without PKA (Figure 4Cb, lanes 1, 2, and 4, respectively).
Figure 4Db shows that the anti-S2030(PO3) antibody recognized the PKA-treated native RyR2, but not RyR2 treated with boiled PKA. Interestingly and in contrast to phosphorylation at serine-2808,9 phosphorylation of the serine-2030 site was not detected in either recombinant or native RyR2 (Figure 4Cb, lane 1, and 4Db, lane 1). The specificity of phosphorylation of serine-2030 by PKA was further examined using a PKA-specific inhibitor, PKI5eC24. Phosphorylation of serine-2030 by PKA was completely abolished by the addition of 5 eol/L PKI5eC24 (Figure 4E).
Serine-2030 Is Phosphorylated in Rat Cardiac Myocytes Stimulated With Isoproterenol
Cell homogenates were prepared from rat cardiac myocytes treated with 1 eol/L isoproterenol for various periods of time, solubilized by SDS, and immunoblotted with the anti-RyR2 (Figure 5Aa), anti-S2030(PO3) (Figure 5Ab), or anti-S2808(PO3) antibody (Figure 5Ac). Consistent with the observations with recombinant and native RyR2, no detectable level of phosphorylation of serine-2030 was observed in rat cardiac myocytes before isoproterenol stimulation (Figure 5Ab, lane 1). On the other hand, isoproterenol (1 eol/L) stimulation for 10 minutes and 1 hour dramatically increased phosphorylation at serine-2030 (Figure 5Ab, lanes 2 and 3). Intriguingly, serine-2030 phosphorylation returned to a low level after prolonged stimulation (24 hours) with isoproterenol (Figure 5Ab, lane 4). To determine the phosphorylation status of serine-2808 under the same conditions, the anti-S2030(PO3) blot was stripped and reprobed with the anti-S2808(PO3) antibody. In keeping with previous reports, serine-2808 was phosphorylated in rat cardiac myocytes before isoproterenol stimulation (Figure 5Ac, lane 1), and its phosphorylation increased on isoproterenol treatment7 (lanes 2 and 3). The level of phosphorylation at serine-2808 was also reduced after prolonged exposure to isoproterenol (24 hours). Collectively, our data demonstrate that serine-2030 is a physiologically relevant phosphorylation site.
To determine whether serine-2030 is a substrate for CaMKII, we coexpressed RyR2 and CaMKII (the -C isoform) in HEK293 cells. We then assessed the phosphorylation state of RyR2 at both serine-2030 and serine-2808 using the anti-S2030(PO3) and anti-S2808(PO3) antibodies, respectively. Coexpression of CaMKII increased the phosphorylation level of RyR2 at serine-2808 by 30% (29.2±0.3%, n=3) (Figure 5Bc and 5C). On the other hand, no phosphorylation of RyR2 at serine-2030 was detected with or without coexpression of CaMKII (Figure 5Bb). Similar studies were also performed in rat cardiac myocytes. No phosphorylation of RyR2 at serine-2030 was detected in cells overexpressing a constitutively active form of CaMKII, whereas phosphorylation of PLB at the CaMKII site, threonine-17, was clearly observed. In addition, we found that the expression of a dominant-negative CaMKII or the addition of a CaMKII inhibitor, KN-93, did not reduce the level of phosphorylation of serine-2808 at rest (unpublished data, 2004). These results indicate that serine-2030 is unlikely to be a CaMKII phosphorylation site. They also suggest that additional kinases may be involved in the phosphorylation of serine-2808 at rest.
Effect of PKA Phosphorylation and Mutations at Serine-2030 on FKBP12.6-RyR2 Interaction
HEK293 cells were transfected with RyR2(wt), S2030A, or S2030D cDNA, or cotransfected with the RyR2(wt) and PKA cDNAs. The expressed wt and mutant RyR2 proteins were then pulled down by GST-FKBP12.6. As shown in Figure 6Aa, all RyR2 variants, RyR2(wt), PKA-phosphorylated RyR2(wt), S2030A, and S2030D, were pulled down by GST-FKBP12.6 at comparable levels. Immunoblotting using the anti-S2030(PO3) antibody confirmed that RyR2(wt) was phosphorylated at serine-2030 when coexpressed with PKA (Figure 6Ab, lane 2), but the RyR2(wt) expressed alone, and the S2030A and S2030D mutants were not (Figure 6Ab, lanes 1, 3, and 4), whereas phosphorylation of serine-2808 was detected in RyR2(wt) and the S2030A and S2030D mutants with or without coexpression of PKA (Figure 6Ac). Thus, these results demonstrate that GST-FKBP12.6 is able to interact with RyR2 irrespective of the phosphorylation status of RyR2 at serine-2030.
To determine whether RyR2(wt) and the RyR2 mutants interact with FKBP12.6 in HEK293 cells, we cotransfected the cells with FKBP12.6 and wt or mutant RyR2 cDNAs. As shown in Figure 6Bb, RyR2(wt) coexpressed with PKA (lane 2) was phosphorylated at serine-2030, but RyR2(wt) expressed alone or the S2030A or S2030D mutants were not (lanes 1, 3, and 4). Importantly, all RyR2 variants, the nonphosphorylated RyR2(wt), phosphorylated RyR2(wt), and the S2030A and S2030D mutants, were able to precipitate the coexpressed FKBP12.6 (Figure 6Bc). These data further demonstrate that FKBP12.6 can interact with both the serine-2030 phosphorylated and nonphosphorylated forms of RyR2.
Effect of Mutations at Serine-2030 on the Sensitivity of RyR2 to Ca2+ Activation
In an attempt to assess the functional impact of serine-2030 phosphorylation on RyR2 channel function, we determined the sensitivities to Ca2+ activation of RyR2(wt), the S2030A mutant that completely removes phosphorylation, and the S2030D mutant that is thought to mimic phosphorylation. As shown in Figure 7, the Ca2+ dependence of [3H]ryanodine binding to the S2030A and S2030D mutants is virtually identical to that of RyR2 (wt). The EC50 values were 0.14±0.01 eol/L (mean±SEM, n=3) for RyR2 (wt), 0.17±0.01 eol/L (n=3) for S2030A, and 0.17±0.01 eol/L (n=3) for S2030D.
RyR2 Is Not Hyperphosphorylated by PKA in Failing Canine Hearts
To investigate whether phosphorylation of RyR2 is altered in HF, we performed immunoblotting analyses of tissue homogenates from failing and nonfailing canine heart samples using the anti-S2808(PO3) (Figure 8A) and anti-S2030(PO3) (Figure 8B) antibodies. Consistent with previous reports,7 we observed considerable levels of phosphorylation of RyR2 at serine-2808 in both failing and nonfailing hearts (Figure 8Ab), but no hyperphosphorylation in failing hearts (Figure 8Ac). On the other hand, low or near background levels of phosphorylation of serine-2030 were detected in failing and nonfailing hearts, but there was no hyperphosphorylation of serine-2030 in failing hearts (Figure 8Bb). It should be noted that the anti-S2030(PO3) antibody is able to recognize the phosphorylated serine-2030 of canine RyR2 (Figure 4Db) and RyR2 from intact canine hearts stimulated with isoproterenol (Figure 8Bb, right panel). Thus, the lack of a phosphorylation signal is not due to the failure of the anti-S2030(PO3) antibody, but due to the lack of significant phosphorylation of serine-2030.
Discussion
The state of PKA phosphorylation of RyR2 in HF remains highly controversial despite intensive investigation. In our opinion, this controversy is due largely to the lack of understanding of the molecular basis of phosphorylation of RyR2. It is generally believed that serine-2808 is the only PKA phosphorylation site in RyR2. Consequently, the level of phosphorylation of serine-2808 has widely been used as an indicator of phosphorylation of RyR2 by PKA.6,20 However, the results of our present study do not support this common belief. We showed that, in addition to serine-2808, RyR2 was phosphorylated by PKA at a previously unknown site, serine-2030 (Figures 2 and 3). We demonstrated that serine-2030 phosphorylation responded to acute -adrenergic stimulation, whereas serine-2808 was considerably phosphorylated in the absence of -adrenergic stimulation (Figures 4, 5, and 8). These data clearly indicate that serine-2808 is not the only PKA site in RyR2, and that the level of serine-2808 phosphorylation does not correlate closely with -adrenergic stimulation. Thus, our findings raise a concern about the appropriateness of using serine-2808 phosphorylation as an indicator of RyR2 phosphorylation by PKA, and indicate that phosphorylation of serine-2030 more accurately reflects PKA phosphorylation of RyR2.
Serine-2030 Is a Specific PKA Phosphorylation Site
Although serine-2030 was predicted to be a CaMKII phosphorylation site rather than a PKA site based on sequence analysis,17 several lines of evidence indicate the opposite. Serine-2030 was phosphorylated by an active PKA, but not by an inactive PKA (boiled PKA) (Figure 4), demonstrating that the observed phosphorylation of serine-2030 in the presence of exogenous PKA was mediated by the added PKA, but not by other kinases potentially copurified with RyR2. We also found that serine-2030 was stoichiometrically phosphorylated by PKA (Figure 3), and that phosphorylation of serine-2030 by PKA was completely abolished by the addition of the PKA-specific inhibitor, PKI5eC24 (Figure 4). Moreover, coexpression of CaMKII with RyR2 led to phosphorylation of RyR2 at serine-2808, but not at serine-2030 (Figure 5). Hence, our results support the conclusion that serine-2030 is a PKA, but not a CaMKII, phosphorylation site.
The identification of an additional PKA phosphorylation site in RyR2 is apparently inconsistent with a recent report that showed that a single point mutation, S2809A, in human RyR2 diminished PKA phosphorylation.20 This discrepancy may result from species differences in RyR2 phosphorylation. To test this possibility, we determined whether serine-2030 of RyR2 from various species could be phosphorylated. Using a phosphospecific antibody, we were able to detect phosphorylation of serine-2030 of the canine (Figure 4D), mouse (Figure 4C and 4E), rat (Figure 5A), and human RyR2 (not shown). Thus, the discrepancy is unlikely to be due to species difference; rather, it likely results from differences in experimental conditions.
Unlike serine-2030, serine-2808 is a substrate for multiple kinases. It was originally identified as a unique CaMKII phosphorylation site in RyR218,19 and later shown to be also a PKA site.6 In addition, serine-2808 was considerably phosphorylated in rat cardiac myocytes in the absence of -adrenergic stimulation and in the presence of a CaMKII inhibitor, KN-93, and a dominant-negative CaMKII (unpublished data, 2004), suggesting that serine-2808 may be phosphorylated by kinase(s) other than PKA and CaMKII at rest. Thus, the extent of phosphorylation of serine-2808 would depend not only on the activity of PKA, but also on the activity of CaMKII and probably other kinase(s). Influence of various kinases on serine-2808 phosphorylation is probably a major source of the discrepancy concerning the state of phosphorylation of RyR2 by PKA.
Phosphoryaltion State of RyR2 in HF
Our assessments of the level of serine-2030 phosphorylation revealed that RyR2 was not hyperphosphorylated by PKA in HF (Figure 8), whereas Marks and his colleagues6 showed that RyR2 was hyperphosphorylated by PKA in HF. Because serine-2030 is a substrate for PKA but not for CaMKII, whereas serine-2808 is a substrate for both PKA and CaMKII, this discrepancy could potentially be explained by the possibility that RyR2 is hyperphosphorylated by CaMKII, rather than by PKA. In line with this possibility, Wang et al21 has recently shown that sustained -adrenergic stimulation, a condition thought to occur in HF, led to a transient activation of the cAMP/PKA signaling pathway, followed by a persistent activation of CaMKII. Consistent with this observation, we found that phosphorylation of RyR2 at serine-2030 was transient, whereas phosphorylation of serine-2808 persisted during sustained - stimulation (Figure 5). Given the transient nature of PKA activation, which is better suited for fight-or-flight response, and of serine-2030 phosphorylation in response to sustained -stimulation, it is difficult to imagine that hyperphosphorylation of RyR2 by PKA could be maintained in HF. In keeping with this view, we found that phosphorylation of the serine-2030 PKA site was hardly detected in both failing and nonfailing canine (Figure 8), human, and rat hearts (unpublished data, 2004). Hence, enhanced cardiac contractility as a result of sustained -adrenergic stimulation may be mediated by activation of the CaMKII, rather than the PKA signaling pathway.21
Although the idea that RyR2 is hyperphosphorylated by CaMKII but not by PKA in HF is possible, it seems to be inconsistent with a recent report by Wehrens et al.20 They found that RyR2 was phosphorylated by CaMKII at a single site, serine-2815, but not at serine-2808. Thus, their findings suggest that increased serine-2808 phosphorylation observed in HF is unlikely to be associated with CaMKII. This idea also appears to be inconsistent with the results of our present study. In contrast to Marks’ group, we failed to detect hyperphosphorylation of RyR2 at serine-2808 in failing hearts. It is important to emphasize that the issue of RyR2 phosphorylation by CaMKII is also controversial. The exact CaMKII phosphorylation sites in RyR2 and whether RyR2 is hyperphosphorylated by CaMKII in HF are unclear.
Alternatively, the discrepancy regarding the state of PKA phosphorylation of RyR2 may result from technical differences in experimental conditions. In our studies, the level of serine-2808 phosphorylation was determined using SDS-solubilized tissue homogenates isolated from quickly frozen heart tissues. In the studies by Marks and his colleagues, the extent of phosphorylation of RyR2 was determined using RyR2 proteins immunoprecipitated from Triton X-100 solubilized postmitochondrial or SR fractions.6 During solubilization and immunoprecipitation, some kinases or phosphatases may remain active and thus may alter the phosphorylation state of serine-2808. The activities of these kinases or phosphatases might be different between failing and nonfailing hearts and under different experimental conditions, thus leading to different extents of phosphorylation of serine-2808 independent of PKA. Hence, technical differences in sample handling may account for the different phosphorylation levels of serine-2808 and thus of RyR2 reported by different groups. To accurately assess the level of phosphorylation of RyR2, it is critical to use samples in which the activities of kinases and phosphatases have quickly been inactivated (eg, by the treatment with SDS). Accordingly, we suggest that SDS-treated tissue homogenates from heart muscles, rather than immunoprecipitated RyR2 proteins, be used directly for the assessment of phosphorylation of RyR2 to minimize further RyR2 phosphorylation or dephosphorylation.
Impact of Serine-2030 Phosphorylation and Mutation
Our data indicate that PKA phosphorylation of RyR2 at the novel serine-2030 site does not abolish FKBP12.6 binding, consistent with our previous finding.9 It should be emphasized that complete PKA phosphorylation of both native RyR2 and recombinant RyR2 does not dissociate the endogenous or coexpressed FKBP12.6 from RyR2.9 It should also be noted that native RyR2 proteins isolated from canine cardiac SR, although already heavily phosphorylated at serine-2808,17 exhibited stoichiometric binding to FKBP12.6.10 Thus, our findings together with previous observations clearly demonstrate that there is no correlation between PKA phosphorylation of RyR2 and FKBP12.6 dissociation.
Our data also indicate that mutations at serine-2030 do not alter the sensitivity of RyR2 to Ca2+ activation as determined by [3H]ryanodine binding. Stange et al8 have also recently shown that mutations at the serine-2808 site have no measurable effect on RyR2 channel function. In addition, PKA phosphorylation does not appear to alter the properties of Ca2+ sparks in mouse ventricular myocytes.22 On the other hand, Marx et al6 reported that phosphorylation of RyR2 by PKA increased the sensitivity of the RyR2 channel to Ca2+ activation. Thus, the exact role of PKA phosphorylation in RyR2 channel function is unclear and requires further investigation. However, it is clear that phosphorylation of serine-2030 by PKA is physiologically relevant, as its phosphorylation is closely associated with -adrenergic stimulation. Generation and characterization of mouse models that contain serine-2030 mutations in RyR2 will provide some definitive insights into the physiological role of phosphorylation of RyR2 by PKA.
In summary, the present study reveals that serine-2030 of RyR2 is a novel PKA phosphorylation site that responds to acute -adrenergic stimulation, and that RyR2 in failing canine hearts is not hyperphosphorylated by PKA. Furthermore, we found that phosphorylation of RyR2 by PKA at serine-2030 does not dissociate FKBP12.6 from RyR2. Taken together, our data do not support previous findings that (1) RyR2 is phosphorylated by PKA at a single residue; (2) RyR2 is hyperphosphorylated by PKA in HF; and (3) phosphorylation of RyR2 by PKA dissociates FKBP12.6 from RyR2.6 Additional investigations are needed to unveil the molecular mechanisms of regulation of RyR2 by PKA and other kinases in normal and diseased hearts.
Acknowledgments
This work was supported by research grants from CIHR to M.P.W. and S.R.W.C., and from the American Heart Association Northland Affiliate and Medical College of Wisconsin to M.T.J. B.X. is a recipient of an AHFMR Studentship Award. M.P.W. is an AHFMR Scientist and Canada Research Chair (Tier I) in Biochemistry. S.R.W.C. is an AHFMR Senior Scholar. The authors would like to thank Drs Hector Valdivia, John Tyberg, and Masakazu Obayashi for providing the canine cardiac tissues, Dr Andrew Braun for providing the cDNAs encoding PKA and CaMKII, Jeff Bolstad for critical reading of the manuscript, and John Tessmer and David Schwabe for technical assistance.
References
Nuss HB, Kaab S, Kass DA, Tomaselli GF, Marban E. Cellular basis of ventricular arrhythmias and abnormal automaticity in heart failure. Am J Physiol Heart Circ Physiol. 1999; 277: H80eCH91.
Zaugg C, Buser P. When calcium turns arrhythmogenic: intracellular calcium handling during the development of hypertrophy and heart failure. Croat Med J. 2001; 42: 24eC32.
Janse M. Electrophysiological changes in heart failure and their relationship to arrhythmogenesis. Cardiovasc Res. 2004; 61: 208eC217.
Pogwizd SM, Bers DM. Cellular Basis of Triggered Arrhythmias in Heart Failure. Trends in Cardiovascular Medicine. 2004; 14: 61eC66.
Bers DM, Eisner DA, Valdivia HH. Sarcoplasmic Reticulum Ca2+ and Heart Failure: Roles of Diastolic Leak and Ca2+ Transport. Circ Res. 2003; 93: 487eC490.
Marx SO, Reiken S, Hisamatsu Y, Jayaraman T, Burkhoff D, Rosemblit N, Marks AR. PKA phosphorylation dissociates FKBP12.6 from the calcium release channel (ryanodine receptor): defective regulation in failing hearts. Cell. 2000; 101: 365eC376.
Jiang MT, Lokuta AJ, Farrell EF, Wolff MR, Haworth RA, Valdivia HH. Abnormal Ca2+ release, but normal ryanodine receptors, in canine and human heart failure. Circ Res. 2002; 91: 1015eC1022.
Stange M, Xu L, Balshaw D, Yamaguchi N, Meissner G. Characterization of recombinant skeletal muscle (Ser-2843) and cardiac muscle (Ser-2809) ryanodine receptor phosphorylation mutants. J Biol Chem. 2003; 278: 51693eC51702.
Xiao B, Sutherland C, Walsh MP, Chen SRW. Protein kinase A phosphorylation at serine-2808 of the cardiac Ca2+-release channel (ryanodine receptor) does not dissociate 12.6-kDa FK506-binding protein (FKBP12.6). Circ Res. 2004; 94: 487eC495.
Timerman AP, Onoue H, Xin HB, Barg S, Copello J, Wiederrecht G, Fleischer S. Selective binding of FKBP12.6 by the cardiac ryanodine receptor. J Biol Chem. 1996; 271: 20385eC20391.
Bristow M, Ginsburg R, Minobe W, Cubicciotti R, Sageman W, Lurie K, Billingham M, Harrison D, Stinson E. Decreased catecholamine sensitivity and beta-adrenergic-receptor density in failing human hearts. N Engl J Med. 1982; 307: 205eC211.
Neumann J. Altered phosphatase activity in heart failure, influence on Ca2+ movement. Basic Res Cardiol. 2002; 97 (suppl 1): I91eCI95.
Schmidt U, Hajjar RJ, Kim CS, Lebeche D, Doye AA, Gwathmey JK. Human heart failure: cAMP stimulation of SR Ca2+-ATPase activity and phosphorylation level of phospholamban. Am J Physiol Heart Circ Physiol. 1999; 277: H474eCH480.
Schwinger RHG, Munch G, Bolck B, Karczewski P, Krause E-G, Erdmann E. Reduced Ca2+-sensitivity of SERCA 2a in failing human myocardium due to reduced serin-16 phospholamban phoshorylation. J Mol Cell Cardiol. 1999; 31: 479eC491.
Sande JB, Sjaastad I, Hoen IB, Bokenes J, Tonnessen T, Holt E, Lunde PK, Christensen G. Reduced level of serine16 phosphorylated phospholamban in the failing rat myocardium: a major contributor to reduced SERCA2 activity. Cardiovasc Res. 2002; 53: 382eC391.
Takasago T, Imagawa T, Furukawa K, Ogurusu T, Shigekawa M. Regulation of the cardiac ryanodine receptor by protein kinase-dependent phosphorylation. J Biochem (Tokyo). 1991; 109: 163eC170.
Rodriguez P, Bhogal MS, Colyer J. Stoichiometric phosphorylation of cardiac ryanodine receptor on serine-2809 by calmodulin-dependent kinase II and protein kinase A. J Biol Chem. 2003; 278: 38593eC38600.
Witcher DR, Kovacs RJ, Schulman H, Cefali DC, Jones LR. Unique phosphorylation site on the cardiac ryanodine receptor regulates calcium channel activity. J Biol Chem. 1991; 266: 11144eC11152.
Witcher DR, Strifler BA, Jones LR. Cardiac-specific phosphorylation site for multifunctional Ca2+/calmodulin-dependent protein kinase is conserved in the brain ryanodine receptor. J Biol Chem. 1992; 267: 4963eC4967.
Wehrens XHT, Lehnart SE, Reiken SR, Marks AR. Ca2+/calmodulin-dependent protein kinase II phosphorylation regulates the cardiac ryanodine receptor. Circ Res. 2004; 94: e61eCe70.
Wang W, Zhu W, Wang S, Yang D, Crow MT, Xiao R-P, Cheng H. Sustained 1-adrenergic stimulation modulates cardiac contractility by Ca2+/calmodulin kinase signaling pathway. Circ Res. 2004; 95: 798eC806.
Li Y, Kranias EG, Mignery GA, Bers DM. Protein kinase A phosphorylation of the ryanodine receptor does not affect calcium sparks in mouse ventricular myocytes. Circ Res. 2002; 90: 309eC316.(Bailong Xiao, Ming Tao Ji)