Frequency-encoding Thr17 Phospholamban Phosphorylation Is Independent of Ser16 Phosphorylation in Cardiac Myocytes*

Both Ser16 and Thr17 of phospholamban (PLB) are phosphorylated, respectively, by cAMP-dependent protein kinase (PKA) and Ca2+/calmodulin-dependent protein kinase II (CaMKII). PLB phosphorylation relieves cardiac sarcoplasmic reticulum Ca2+ pump from inhibition by PLB. Previous studies have suggested that phosphorylation of Ser16 by PKA is a prerequisite for Thr17 phosphorylation by CaMKII and is essential to the relaxant effect of β-adrenergic stimulation. To determine the role of Thr17 PLB phosphorylation, we investigated the dual-site phosphorylation of PLB in isolated adult rat cardiac myocytes in response to β1-adrenergic stimulation or electrical field stimulation (0.1–3 Hz) or both. A β1-adrenergic agonist, norepinephrine (10−9–10−6 m), in the presence of an α1-adrenergic antagonist, prazosin (10−6 m), selectively increases the PKA-dependent phosphorylation of PLB at Ser16in quiescent myocytes. In contrast, electrical pacing induces an opposite phosphorylation pattern, selectively enhancing the CaMKII-mediated Thr17 PLB phosphorylation in a frequency-dependent manner. When combined, electric stimulation (2 Hz) and β1-adrenergic stimulation lead to dual phosphorylation of PLB and exert a synergistic effect on phosphorylation of Thr17 but not Ser16. Frequency-dependent Thr17 phosphorylation is closely correlated with a decrease in 50% relaxation time (t 50) of cell contraction, which is independent of, but additive to, the relaxant effect of Ser16phosphorylation, resulting in hastened contractile relaxation at high stimulation frequencies. Thus, we conclude that in intact cardiac myocytes, phosphorylation of PLB at Thr17 occurs in the absence of prior Ser16 phosphorylation, and that frequencydependent Thr17 PLB phosphorylation may provide an intrinsic mechanism for cardiac myocytes to adapt to a sudden change of heart rate.

Phospholamban (PLB) 1 is the major regulator of cardiac sarcoplasmic reticulum (SR) Ca 2ϩ -ATPase and Ca 2ϩ transport across the SR membrane, thereby modulating myocardial relaxation (for review see Ref. 1). Specifically, PLB in its dephos-phorylated state is an inhibitor of the SR Ca 2ϩ pump (2). Phosphorylation of PLB at Ser 16 and Thr 17 by PKA and CaMKII removes this inhibition by increasing the affinity of the pump for Ca 2ϩ (1)(2)(3). In the beating mammalian heart, ␤-adrenergic stimulation increases both PKA-and CaMKIImediated phosphorylation of Ser 16 and Thr 17 (4 -6). More recent studies have shown that the effects of ␤-adrenergic stimulation on PLB phosphorylation are mostly attributable to ␤ 1but not ␤ 2 -adrenergic receptor subtype (7)(8)(9)(10).
Over the last two decades, intensive studies have been focused on the physiological significance of the dual site phosphorylation of PLB. These previous studies in perfused hearts or in vivo have provided several lines of evidence leading to the concept that Ser 16 phosphorylation is a prerequisite for phosphorylation of Thr 17 and that Ser 16 phosphorylation is largely responsible for ␤-adrenergic modulation of cardiac relaxation (4, 5, 11, 12, 14 -18). Recent studies in transgenic mice overexpressing the Ser 16 3 Ala 16 PLB mutant have further demonstrated that prevention of Ser 16 phosphorylation abolishes Thr 17 phosphorylation and attenuates ␤-adrenergic responses (19). Thus, it is widely accepted that phosphorylation of PLB at Ser 16 is obligatory for Thr 17 phosphorylation and that Ser 16 phosphorylation is the dominant molecular event responsible for accelerated cardiac relaxation. However, in vitro studies in the isolated SR membranes have consistently indicated that Ser 16 and Thr 17 can be readily and independently phosphorylated by PKA and CaMKII, respectively, and that when both are phosphorylated, there is an additive interaction (20,21). The apparent discrepancy between in vivo and in vitro PLB phosphorylation is yet to be reconciled.
To resolve this paradox and to further address the relative contribution of PKA-and CaMKII-mediated PLB phosphorylation in beat-to-beat cardiac functional modulation, individually we manipulated PKA activity, using ␤-adrenergic stimulation in quiescent rat ventricular myocytes, and CaMKII activity, by electrically pacing the myocytes at different stimulation frequency (0.1-3 Hz) in the absence of ␤-adrenergic stimulation. Both stimuli were also combined to explore possible interactions between PKA-and CaMKII-mediated signaling. Under those experimental conditions, we measured PLB phosphorylation at Ser 16 and Thr 17 as well as relaxation time of cell contraction. Here, we report our surprising findings that electrical stimulation alone increases CaMKII-dependent phosphorylation of PLB at Thr 17 in a frequency-dependent manner without altering PKA-mediated Ser 16 phosphorylation, that phosphorylation of Thr 17 is markedly enhanced by ␤-adrenergic stimulation in the electrically paced but not in quiescent myocytes, and that Thr 17 phosphorylation is associated with a significant relaxant effect, regardless of ␤-adrenergic stimulation.
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

EXPERIMENTAL PROCEDURES
Measurements of Cell Contraction and Relaxation-Single ventricular myocytes were isolated from adult rat (2-4-month old) hearts by a standard enzymatic technique (22). The cells were suspended in HEPES buffer, pH 7.4, containing (in mmol/liter): 20 HEPES, 1 CaCl 2 , 137 NaCl, 5 KCl, 15 dextrose, 1.3 MgSO 4 , and 1.2 NaH 2 PO 4 . Cells were kept at rest or stimulated at different frequencies ranging from 0.1 to 3 Hz at 23°C, and cell length was monitored from the bright-field image by an optical edge tracking method using a photodiode array (model 1024 SAQ, Reticon) with a 3-ms time resolution (22). The half-time of contractile relaxation (t 50 ) was measured as the time from cell peak shortening to 50% relaxation.
Site-specific PLB Phosphorylation-The detection of site-specific PLB phosphorylation was performed as described previously (9). Briefly, a 500-l suspension of isolated rat ventricular myocytes was stimulated over a wide range of frequencies (0.1-3 Hz) at 23°C. Following a 5-min stimulation, 4ϫ sample buffer was added, and the samples were frozen in liquid nitrogen. For ␤ 1 -adrenergic stimulation, myocytes were incubated with norepinephrine (NE, 10 Ϫ9 -10 Ϫ6 M) and prazosin (10 Ϫ6 M) for 10 min. In another subset of experiments, myocytes were first incubated with NE (10 Ϫ7 M) plus prazosin (10 Ϫ6 M) for 5 min and then were stimulated electrically (at 2 Hz) for another 5 min in the continued presence of ␤ 1 -adrenergic stimulation. Samples were solubilized prior to electrophoresis at 95°C for 5 min to fully dissociate PLB into its monomers. Following electrophoresis, proteins were transferred to a polyvinylidene difluoride membrane (PVDF, Sigma), which was probed with the phosphorylation site-specific PSer 16 and PThr 17 PLB antibodies (PhosphoProtein Research). Protein concentration was determined by the method of Lowry et al. (23) using ovalbumin as standard. Following incubation with a peroxidase-conjugated antibody (Dianova), films were exposed to the chemiluminescence (ECL, Amersham Pharmacia Biotech) reaction and quantified with a video documentation system (Bio-Rad).
Statistics and Other Assays-Results are presented as the mean Ϯ S.E. Statistical significance was determined by Student's t test, the Mann-Whitney test (when variances were significantly different), or one-way analysis of variance, when appropriate. Values with p Ͻ 0.05 were considered statistically significant.

Dose-Response of NE-induced Ser 16 PLB Phosphorylation-
Freshly isolated single rat quiescent ventricular myocytes were challenged by NE (10 Ϫ9 -10 Ϫ6 M), a ␤ 1 -adrenergic receptor agonist, in the presence of an ␣ 1 -adrenergic antagonist, prazosin (10 Ϫ6 M). Phosphorylation of Ser 16 or Thr 17 in response to ␤ 1 -adrenergic stimulation was determined by immunoblots with anti-PSer 16 PLB and anti-PThr 17 PLB antibodies. As shown in Fig. 1, NE increased PKA-mediated PLB phosphorylation at Ser 16 in a dose-dependent manner, with an EC 50 of ϳ10 Ϫ7 M. In contrast, NE, even at the maximal concentration (10 Ϫ6 M), had only a very minor effect on CaMKII-dependent phosphorylation of Thr 17 (Fig. 1).
Frequency-dependent Thr 17 PLB Phosphorylation in the Absence of ␤-Adrenergic Stimulation-To examine PLB phosphorylation at the CaMKII site Thr 17 , myocytes were electrically paced for 5 min at frequencies ranging from 0.1 to 3.0 Hz. Fig.  2 shows the response of Ser 16 or Thr 17 PLB phosphorylation to stimulation frequency in the absence of ␤-adrenergic stimulation. There was a near linear increase in Thr 17 PLB phosphorylation with increasing stimulation frequency, and the phosphorylation was not yet saturated at the highest stimulation frequency (147.70 Ϯ 13.98 at 3 Hz versus 44.07 Ϯ 11.35 at rest, p Ͻ 0.01). In the same cells, however, no significant PKAmediated PLB phosphorylation at Ser 16 was observed at any stimulation frequency. The pattern of Ser 16 and Thr 17 phosphorylation induced by electrical stimulation alone was thus opposite to that in quiescent cells induced by ␤-adrenergic stimulation, suggesting that CaMKII-mediated Thr 17 PLB phosphorylation can occur independent of any prior Ser 16 PLB phosphorylation in intact cardiac myocytes.
Ser 16 and Thr 17 PLB Phosphorylation in Response to Combined Electrical and ␤-Adrenergic Stimulation-To further in-vestigate possible interactions between the PKA and CaMKII phosphorylation sites, we measured site-specific PLB phosphorylation using electrical stimulation in conjunction with ␤ 1adrenergic stimulation. Following a 5-min incubation with 10 Ϫ7 M NE plus 10 Ϫ6 M prazosin, myocytes were either kept at rest or electrically paced at 2 Hz for another 5 min in the continued presence of NE. Electrical stimulation had no significant effect on NE-induced Ser 16 PLB phosphorylation (Fig.  3A). In contrast, ␤ 1 -adrenergic stimulation markedly augmented the frequency-dependent PLB phosphorylation at Thr 17 , resulting in a 3-fold augmentation compared with that in the absence of ␤-adrenergic stimulation (Fig. 3B). Thus, there is a synergistic interaction between electrical stimulation and ␤-adrenergic stimulation in conferring Thr 17 PLB phosphorylation, whereas the effect of ␤-adrenergic stimulation on Ser 16 phosphorylation is independent of electrical stimulation.
Modulatory Effect of Stimulation Frequency on Cardiac Relaxation-To determine the functional role of Thr 17 PLB phosphorylation, we examined the t 50 of contraction in the absence as well as in the presence of ␤ 1 -adrenergic stimulation in rat cardiomyocytes. We found that in the absence of ␤ 1 -adrenergic stimulation, t 50 after a 5-min pacing period was abbreviated relative to t 50 of the first post-rest beat. The higher the pacing frequency, the briefer the  4C shows that the frequency-dependent abbreviation of 50% relaxation time was linearly correlated to the frequency-dependent increase in Thr 17 PLB phosphorylation but was totally dissociated from Ser 16 phosphorylation (which remained un- changed during electrical pacing) (Fig. 2). In the presence of ␤ 1 -adrenergic stimulation, the first post-rest beat relaxed 14.1% faster relative to NE-untreated cells (Fig. 4B), and this was accompanied by an enhanced PLB phosphorylation occurring exclusively at the PKA site ( Figs. 1 and 3A). The relaxant effect of NE was almost doubled by the application of 2 Hz electrical stimulation (Fig. 4B). Concomitantly, the enhanced relaxant effect was accompanied by a markedly potentiated Thr 17 phosphorylation (Fig. 4A) in the absence of any change in Ser 16 PLB phosphorylation (Fig. 3A), suggesting that the increased Thr 17 PLB phosphorylation has an additive effect to hasten relaxation during combined ␤ 1 -adrenergic and electrical stimulation.

DISCUSSION
There are four major findings in the present study. First, ␤-adrenergic receptor stimulation in quiescent ventricular myocytes increases PKA-dependent phosphorylation of PLB at Ser 16 and accelerates the relaxation rate of the first post-rest beat, with little effect on the CaMKII-dependent phosphorylation of Thr 17 (Fig. 1). This observation is consistent with the previous notion that Ser 16 is the direct and dominant site for PKA-mediated phosphorylation underlying ␤-adrenergic relaxant effect. Second, and more importantly, in the absence of ␤-adrenergic stimulation, electrical stimulation enhances phosphorylation of PLB at Thr 17 in a frequency-dependent manner without altering the phosphorylation status of Ser 16 (Fig. 2). These results suggest that phosphorylation states of Ser 16 and Thr 17 can be regulated independently in intact cardiac myocytes. Third, a combination of the ␤ 1 -adrenergic agonist, NE (10 Ϫ7 M), and electrical stimulation (at 2 Hz) induces dual phosphorylation of PLB at Ser 16 and Thr 17 , exhibiting a synergistic effect on Thr 17 but not Ser 16 phosphorylation (Fig.  3). Finally, we provided direct evidence that the frequency-de-pendent Thr 17 PLB phosphorylation is associated with a significant relaxant effect, independent of ␤-adrenergic-receptor stimulation (Fig. 4).
Independent Modulation of PLB Phosphorylation at Ser 16 and Thr 17 -The observations described above indicate that CaMKII-dependent Thr 17 phosphorylation can occur independent of PKA-mediated Ser 16 phosphorylation and is likely involved in the modulation of cardiac relaxation. This conclusion is in contrast to the well accepted sequential model of PLB phosphorylation at Ser 16 and Thr 17 (4, 5, 11, 12, 14 -19). Several differences between the present and previous studies may account for the different outcomes. First, our experimental setting permits either a selective PKA activation (␤ 1 -adrenergic stimulation in quiescent cardiac myocytes) or a selective CaMKII activation (electrical stimulation in the absence of ␤-adrenergic stimulation), avoiding possible cross-talk between PKA and CaMKII signaling pathways. In contrast, most previous studies were performed in vivo, or in isolated beating hearts, and could not distinguish the primary effects of PKA from its secondary effects via interaction with the CaMKII signaling pathway (see below). Additionally, using site-specific antibodies for PSer 16 or PThr 17 PLB with an appropriate exposure time for the chemiluminescence reaction (see "Experimental Procedures"), we have greatly improved the sensitivity to detect a low level phosphorylation of PLB in the present study. Our results indicate that the increase in Thr 17 PLB phosphorylation induced by electrical stimulation (2 Hz) is ϳ30% of that induced by a combined ␤-adrenergic and electrical stimulation (ϳ15% of the total phosphorylation of PLB). This relatively low level of PLB phosphorylation could have been overlooked in previous studies, particularly when it was measured by 32 P incorporation into PLB. It is noteworthy that although previous studies have demonstrated the reliability and specificity of the site-specific antibodies for PSer 16 and PThr 17 PLB (12, 16, 24), a recent study (25) shows poor reactivities of those antibodies with dual-phosphorylated PLB. If this were the case, we might underestimate the degree of the sequential PLB phosphorylation. However, this is unlikely, because the present results show that in the presence of both electrical and ␤-adrenergic stimuli anti-PSer 16 and anti-PThr 17 detected a clear increase in the phosphorylation of both sites ( Fig. 3 and 4). Finally, it has been proposed that in vivo or in beating hearts, the dependence of Thr 17 phosphorylation on prior Ser 16 phosphorylation is due to a counteraction of protein phosphatases on phosphoproteins (16). In principle, a low level of protein phosphatase activity in freshly isolated myocytes relative to that found in vivo might also explain the discrepancy between this and previous reports (4, 12, 14 -18). However, this explanation is unlikely because in quiescent cardiac myocytes ␤-adrenergic stimulation increases Thr 17 PLB phosphorylation only in the presence of protein phosphatase inhib-itors (either okadaic acid or calyculin A), 2 suggesting that protein phosphatases are rather active in intact isolated cardiac myocytes.
Synergistic Effect of ␤-Adrenergic and Electrical Stimulation on Thr 17 Phosphorylation-It is noteworthy that the increase in phosphorylation at the CaMKII site Thr 17 in response to a combination of NE and 2 Hz electrical stimulation is much greater than the summation of its separate responses to these two stimuli (Fig. 3), suggesting a synergistic interaction between ␤-adrenergic and electrical stimuli. The most plausible explanation is that ␤-adrenergic stimulation indirectly enhances Thr 17 phosphorylation through cross-talking with CaMKII signaling pathway. For instance, ␤-adrenergic stimulation elevates CaMKII activity by enhancing sarcolemmal Ltype Ca 2ϩ currents (I Ca ), SR Ca 2ϩ cycling, and intracellular Ca 2ϩ transients (26 -28). This mechanism appears to be minimized in the absence of intracellular Ca 2ϩ transients, as evidenced by the insensitivity of Thr 17 phosphorylation to ␤-adrenergic stimulation in quiescent myocytes (Fig. 1). In addition to the Ca 2ϩ -dependent cross-talk, PKA may also enhance CaMKII signaling by inhibiting protein phosphatase 1 (PP1), which is the major phosphatase dephosphorylating PLB (29), either by direct inhibition of PP1 activity via phosphorylation of its regulatory subunit (29) or by indirect inhibition of the phosphatase via phosphorylation and subsequent activation of endogenous phosphatase inhibitor 1 (30,31). Both mechanisms prevent PP1 from dephosphorylating PLB. In this regard, it has been shown that in beating hearts, a protein phosphatase inhibitor, okadaic acid, permits Thr 17 PLB phosphorylation in response to high extracellular [Ca 2ϩ ], in the absence of ␤-adrenergic stimulation (16). Regardless of the specific mechanism, the potent effect of ␤-adrenergic stimulation to enhance CaMKII signaling may masquerade as a sequential phosphorylation of Ser 16 and Thr 17 under certain experimental conditions.
Unique Property of Thr 17 Phosphorylation: A Frequency Detector-The present results indicate that CaMKII-mediated phosphorylation of PLB at Thr 17 is linearly correlated to the increase in stimulation frequency. The strong frequency dependence of Thr 17 phosphorylation may be attributed largely to an intrinsic "memory" ability of CaMKII. Recent studies have demonstrated that CaMKII undergoes autophosphorylation during activation (13). The autophosphorylated CaMKII retains some kinase activity even in the interval of intracellular Ca 2ϩ transients, leading to an accumulative CaMKII activation during high frequency stimulation (13).
The intrinsic memory property of CaMKII makes Thr 17 PLB phosphorylation unique as compared with PKA-mediated Ser 16 phosphorylation. As a result of frequency-dependent CaMKII activation, phosphorylation of Thr 17 PLB may function as a frequency detector of heart rate; the higher the heart rate and the more frequent intracellular Ca 2ϩ transients, the greater is CaMKII-mediated Thr 17 PLB phosphorylation. Because frequency-dependent Thr 17 PLB phosphorylation is closely associated with the frequency-dependent acceleration of relaxation (Fig. 4), CaMKII-mediated Thr 17 PLB phosphorylation may provide a constant beat-to-beat intrinsic regulation of cardiac contractile kinetics.
In summary, the present results provide the first documentation that in intact cardiac myocytes, frequency-dependent Thr 17 PLB phosphorylation by CaMKII occurs independently of PKA-mediated Ser 16  A, average data of Thr 17 PLB phosphorylation in response to ␤ 1 -adrenergic stimulation by NE, electrical stimulation at 2 Hz, or a combination of ␤ 1 -adrenergic and electrical stimulation (mean Ϯ S.E., n ϭ 6 -10 for each data point, * p Ͻ 0.01). B, corresponding % of decrease in t 50 of relaxation relative to the t 50 value of the first post-rest (P.rest) beat (n ϭ 13-30 for each data point, * p Ͻ 0.05). C, correlation between Thr 17 PLB phosphorylation (n ϭ 5-10) and abbreviation of the t 50 of relaxation (n ϭ 13-30) in the absence of ␤-adrenergic stimulation. sequential model of PLB phosphorylation but also suggest a novel autoregulatory mechanism in cardiac beat-to-beat functional modulation.