Ca Controls Functional Expression of the Cardiac K (cid:1) Transient Outward Current via the Calcineurin Pathway*

The transient outward K (cid:1) current ( I to ) modulates transmembrane Ca 2 (cid:1) influx into cardiomyocytes, which, in turn, might act on I to . Here, we investigated whether Ca 2 (cid:1) modifies functional expression of I to . Whole-cell I to were recorded using the patch clamp tech- nique in single right ventricular myocytes isolated from adult rats and incubated for 24 h at 37 °C in a serum-free medium containing various Ca 2 (cid:1) concentrations ([Ca 2 (cid:1) ] o ). Increasing the [Ca 2 (cid:1) ] o from 0.5 to 1.0 and 2.5 m M produced a gradual decrease in I to density without change in current kinetics. Quantitative reverse tran-scriptase-PCR showed that a decrease of the Kv4.2 mRNA could account for this decrease. In the acetoxymethyl ester form of 1,2-bis(2-aminophenoxy)ethane-N , N , N (cid:1) , N (cid:1) -tetraacetic acid (BAPTA-AM)-loaded myocytes (a permeant Ca 2 (cid:1) chelator), I to density increased significantly when cells were exposed for 24 h to either 1 or 2.5 m M [Ca 2 (cid:1) ] o . Moreover, 24-h exposure to the Ca 2 (cid:1) channel agonist, Bay K8644, in 1 m M [Ca 2 (cid:1) ] o induced a decrease in I to density, whereas the Ca 2 (cid:1) channel antagonist,

K ϩ channels play critical roles in a wide variety of physiological processes including regulation of heart rate and contraction. Among other cardiac K ϩ currents, the transient outward K ϩ current (I to ) is crucial because it controls the amplitude of the plateau phase and the duration of the action potential. As a consequence, I to strongly modulates transmembrane Ca 2ϩ entry and, thereby, the excitation-contraction coupling (1). Thus, any change in I to has profound pathophysiological consequences, often leading to the generation of life-threatening arrhythmias. For example, changes in the density and/or the properties of I to occur in conjunction with myocardial damage or disease, including acute or chronic diabetes mellitus, hypertrophy induced by pressure or volume overload, and cardiac failure (2). Unidentified common pathway(s) may underlie a down-regulation of the expression of I to channels in all these pathological conditions. A possible candidate for triggering such channel remodeling could be the altered level of intracellular Ca 2ϩ ([Ca 2ϩ ] i ).
Even if defined as Ca 2ϩ -independent, by contrast to the 4-aminopyridine-resistant Ca 2ϩ -activated transient outward chloride current, I to might be regulated by Ca 2ϩ . A modulation of I to by extracellular Ca 2ϩ ([Ca 2ϩ ] o ) induced by depolarizing shifts in the gating parameters has been documented (3). Moreover, I to inactivation in human atrial myocytes might be controlled by [Ca 2ϩ ] i -dependent processes, involving Ca 2ϩ /calmodulin-dependent protein kinase II (CaMKII) 1 (4). At the molecular level, shal-type voltage-gated K ϩ channels (in rat predominantly Kv4.2 (5)) are the pore-forming subunits of the I to channel that are regulated by auxiliary subunits (5,6). Interestingly, the regulatory subunits KCHiPs and the neuronal calcium sensor (NCS) that modulate I to expression and kinetics are Ca 2ϩ -binding proteins. Thus, Ca 2ϩ controls the gating of Kv4-KChiP (7) or Kv4-NCS-1 complexes (8). In addition to these acute effects, Ca 2ϩ also affects cardiac gene expression via excitation-transcription coupling. It is now well recognized that Ca 2ϩ entry into neuronal cells through voltage or ligand-gated channels triggers neuronal activity-dependent gene expression critical for neurobiological adaptive changes (9). In cardiac cells, Ca 2ϩsensitive transcription factors may play a key role in maintaining the contractile phenotype and contribute to adaptive or pathological changes in the structural or functional properties of the heart (10). Notably, long term changes in intracellular Ca 2ϩ level have been linked to altered functional expression of the ion channel. In cultured neonatal cardiomyocytes, increases in [Ca 2ϩ ] i cause a fall in density of the Na ϩ current (11). Exposure of adult rat ventricular myocytes in culture to high Ca 2ϩ increases Ca 2ϩ channel mRNA and protein abundance, producing a corresponding change in the L-type Ca 2ϩ current (I Ca ) (12).
In the present study, we hypothesized that Ca 2ϩ regulates the functional expression of I to . Experimental manipulations of Ca 2ϩ entry via I Ca , in adult rat ventricular myocytes incubated for 24 h, showed that increased cytosolic Ca 2ϩ decreases I to and down-regulates Kv4.2 mRNA expression, which involves Ca 2ϩ / calmodulin-regulated protein phosphatase calcineurin cascade.

EXPERIMENTAL PROCEDURES
Cell Isolation and Incubation-Cardiac ventricular myocytes were isolated from adult male Wistar rats (250 -280 g) using an enzymatic * 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  perfusion method described previously (13). Taking into account the electrical heterogeneity, only the right ventricle was selected (14). Myocytes were incubated for 24 h at 37°C in Tyrode's solution supplemented with 100 IU/ml penicillin and 0.1 g/ml streptomycin as described previously (13). Before electrophysiological recordings, myocytes were washed out for 10 min to minimize any acute effects of the drugs used during the incubation.
Electrophysiology-Whole-cell currents were monitored at 0.1 Hz with an Axopatch 1D amplifier and recorded with a pCLAMP-7 (Axon Instruments) at 21-23°C. Series resistance was electronically compensated (40 -60%). The I to , defined as the 3 mM 4-aminopyridine-sensitive current, was recorded as described previously (13). Because cell size variation might account for differences, current amplitudes were normalized to cell capacitance (C m ) and expressed as densities (picoampere/picofarad). No difference in the C m of myocytes in the different incubation conditions was observed (Table I).
Reverse Transcription and Real-time PCR-Total RNA was extracted using the TRIzol method (InVitrogen) from whole dish of cardiomyocytes (samples), and its integrity was analyzed by electrophoresis with a chip-based RNA analysis system (Agilent Technologies). To obtain cDNA, 200 ng of RNA was reverse-transcripted using the Taqman Gold RT-PCR kit (Applied Biosystems). Real-time PCR analysis was done with an iCycler iQ detection system (Bio-Rad) using a specific mix containing specific primers and probe and AmpliTaq Gold DNA polymerase, as described elsewhere (15,16). Each real-time quantitative PCR assay was performed twice using n samples.
Calcineurin Activity-Cellular calcineurin phosphatase activity was measured on cell extract using the Calbiochem Calcineurin Cellular Activity Assay Kit according to the manufacturer's instructions. The fraction of total phosphatase activity due to calcineurin was determined by detection of free phosphate released in absence or presence of EGTA buffer. Colorimetric measure (assay with Malachite Green) was done at 620 nm on a plate reader (Dynatech MR 5000).
[Ca 2ϩ ] i Measurement-After 24-h incubation, cells were loaded with 2.5 M Fura-2 AM (Molecular Probes) for 30 min at 37°C. Cells were then rinsed in respective storage solution (1 or 2.5 mM [Ca 2ϩ ] o ), and fluorescence measurements were made using the MetaFluor imaging system (Universal Imaging Corp.). [Ca 2ϩ ] i was calculated as reported previously (11); the used dissociation constant of Fura-2 for Ca 2ϩ was 0.141 M.
Statistical Analysis-Data were presented as mean Ϯ S.E. Unpaired Student's t test was used for comparisons. A value of p Ͻ 0.05 was accepted as statistically significant. Fig. 1A (Fig. 1A, right). Detailed analysis revealed that these variations were not related to changes in voltage-or time-dependent properties of I to during the 24-h incubation (Table I).   24-h

incubation in different conditions
The voltage-dependent availability were fit to a Boltzmann distribution with E 50 inac , the potential of half-inactivation, and k inac , the slope factor. The time to peak values, T peak ϩ40 mV , were determined as the time from the onset of depolarization at ϩ40 mV to the time of maximal current amplitude and reflect activation rates. The inactivation kinetics were determined by fitting the decay phase of the current traces to a monoexponential function with inac ϩ40 mV , time constant at ϩ40 mV. pF, picofarads.  (n ϭ 24)). Then we repeated our electrophysiological experiments on cells incubated 24 h in either 1 or 2.5 mM [Ca 2ϩ ] o in the presence of the membranepermeant calcium chelator, BAPTA-AM. As shown in Fig. 2, 10 M BAPTA-AM not only increased I to in cells exposed to 1 mM [Ca 2ϩ ] o but also blunted the decrease expected at 2.5 mM [Ca 2ϩ ] o (Fig. 1). Thus, decreased [Ca 2ϩ ] i resulted in a significant increase of I to slope conductance.
Ca 2ϩ Channels Mediate Ca 2ϩ -dependent Down-regulation of I to -The mechanisms by which [Ca 2ϩ ] o promotes an increase in [Ca 2ϩ ] i may involve greater leak across the sarcolemma through L-type Ca 2ϩ channels or stimulation of the Na ϩ /Ca 2ϩ exchanger working in the "reverse mode" (17). Even if small, the open probability of the L-type Ca 2ϩ channel at rest is not null as assessed by single channel recording (18). Moreover, increasing the driving force across the membrane in presence of higher [Ca 2ϩ ] o will enhance channel conductance (19). We measured I to in myocytes incubated for 24 h with either a Ca 2ϩ channel agonist (Bay K8644) or a Ca 2ϩ channel antagonist (Fig. 2, ϩNifedipine) (Fig. 2). It is worth noting that none of those treatments altered I to kinetics and voltage dependence (Table I).
These experiments suggested, therefore, that down-regulation of the functional expression of I to occurs mostly by modulation of [Ca 2ϩ ] i through L-type calcium current,.
Ca 2ϩ -induced Down-regulation of I to Involves the Calcineurin Pathway-Many of the actions of Ca 2ϩ are mediated through its interaction with CaM. CaM serves as an intracellular sensor for Ca 2ϩ and selectively activates specific downstream signaling pathways in response to local changes in [Ca 2ϩ ] i (20). To focus on the Ca 2ϩ -triggered pathway involved in the down-regulation of I to , one group of cells was treated with the Ca 2ϩ /CaM inhibitor W7 (21). The effect of 1 M W7 in the presence of 2.5 mM [Ca 2ϩ ] o is shown in Fig. 3. W7 prevented the down-regulation of I to induced by high Ca 2ϩ , with no change in the kinetics and in the voltage dependence of I to (Table I). This finding suggested that Ca 2ϩ modulation of I to involves CaM.
Two Ca 2ϩ /CaM-dependent enzymes that have major effects on cardiac muscle function are the CaMKII (22) and the phosphatase calcineurin (23). These enzymes have distinct Ca 2ϩ sensitivities, partner proteins, and subcellular localizations that enable them to discriminate between different types of Ca 2ϩ signals and regulate different functions (9). To characterize further the Ca 2ϩ pathway for long term regulation of I to , we used the CaMKII inhibitor KN-62 that has been reported to specifically inhibit Ca 2ϩ /CaM protein kinase isoforms (24). Exposure of cells to 1 M KN-62 in 2.5 mM [Ca 2ϩ ] o for 24 h did not prevent the down-regulation of I to (Fig 3A). In contrast, incubation with tacrolimus (FK506, 1 M) or CsA (25 M), known to block calcineurin (25), prevented I to down-regulation (Fig. 3A). Neither the time-nor the voltage-dependent properties of I to were changed after these treatments (Table I).
Exposure of cardiomyocytes to higher doses of W7 (26), FK506, or CsA (27) might exert direct acute effects on I Ca and/or I to . To confirm calcineurin involvement, we checked its activity in myocytes incubated for 24 h in either 1 or 2.5 mM [Ca 2ϩ ] o , using a colorimetric assay. Calcineurin activity determined by the amount of phosphate release was increased significantly in 2.5 mM [Ca 2ϩ ] o -treated cells (Fig. 3B), suggesting that enhancement of calcineurin activity after Ca 2ϩ treatment mediates I to down-regulation. We concluded, therefore, that the Ca 2ϩ /CaM-regulated protein phosphatase, calcineurin, is involved in the long term Ca 2ϩ -dependent down-regulation of I to .

DISCUSSION
The present study shows that increased Ca 2ϩ influx through L-type Ca 2ϩ channels causes an increase in [Ca 2ϩ ] i , which up-regulates the Ca 2ϩ /calcineurin pathway and leads to a down-regulation of the expression of I to (Kv4.2 gene) in isolated adult rat ventricular myocytes. Fig. 4 summarizes our data.
Long term changes in cardiac cellular excitability can be generated by regulating the expression of I to channel genes (1,5,6,28). This was well demonstrated by use of transgenic and targeted deletion strategies in mice (29). On the other hand, electrical activity might affect K ϩ channel gene expression. It has been reported that membrane depolarization or Bay K8644 (30) enhances expression of the rapid inactivating Shaker K ϩ channel Kv1.4 mRNA related to I to in newborn rat cardiomyo-cytes (31). Chronic membrane depolarization of cultured neonatal myocytes reduces I to density without affecting either current kinetics or voltage dependence (32). Our data suggest that Ca 2ϩ influx through voltage-dependent Ca 2ϩ channels is a route for modulation of functional expression of I to . Indeed, the Ca 2ϩ channel agonist Bay K8644 mimics the effect of high [Ca 2ϩ ] o on I to (decrease), whereas the Ca 2ϩ channel antagonist nifedipine has the opposite effect (increase) analogous to the effect of reducing [Ca 2ϩ ] o .
We demonstrate here that the calmodulin inhibitor, W7, blunts the decrease in I to at high [Ca 2ϩ ] o . An emerging body of work indicates that the Ca 2ϩ -calmodulin complex acts as an important second messenger for various signals, including angiotensin II, endothelin-1, ␣-adrenergic agents, and a mechanical stretch that triggers hypertrophic growth of the myocardium (33,34). Interestingly, these modulators regulate both Kv4 channel current densities and mRNA levels. For example, angiotensin II has been implicated in the modulation of I to (35). An autocrine/paracrine release of ET-1 modulates Kv4.2 and Kv1.2 (35). Thyroid status also regulated I to density and Kv4.2 mRNA levels (36). Thus, some of the effects on I to in response to a variety of endogenous modulators and hormones known, or postulated, to modulate cardiac function might involve changes in Ca 2ϩ signaling, in particular, related to I Ca regulation. In support of this view, we recently demonstrated that aldosterone-induced downregulation of I to functional expression occurs secondary to modulation of Ca 2ϩ signaling (increased I Ca ) (13).
Members of the diverse superfamily of voltage-activated K ϩ channels are modulated by phosphorylation. Such modulations are catalyzed by a variety of protein kinases including CaMKII. In atrial myocytes, phosphorylation of the K ϩ channel by CaMKII affects the fast inactivation kinetics (4 the different treatments or effects of CamKII inhibitor on the Ca 2ϩ down-regulation of I to . However, our experiments were conducted in the presence of millimolar levels of EGTA in the pipette that will attenuate acute intracellular Ca 2ϩ variations. More recently, CaMKII phosphorylation of neuronal Kv4.2 at the C terminus has been shown to affect cell-surface expression of A-type K ϩ channels (37). Our results did not rule out this effect in cardiomyocytes but show that increased diastolic [Ca 2ϩ ] i effect on I to seems CamKII-independent.
Calcineurin inhibition by CsA has been shown to prevent the decreases in mRNA levels of Kv4.2 and Kv4.3 and I to density after myocardial infarction (38). Moreover, in contrast to a previous report (39), overexpression of calcineurin decreases the density and function of the depolarization-activated K ϩ currents, and the density of I to is restored by CsA treatment in transgenic mice (40). Consistent with these observations, we report here that calcineurin inhibition with either FK506 or CsA at low concentrations prevents Ca 2ϩ down-regulation of I to at high [Ca 2ϩ ] o .
Changes in the cytoplasmic free Ca 2ϩ concentration constitute one of the main routes by which information is transferred from extracellular signals received by animal cells to intracellular sites. In heart, these changes have been implicated in regulating diverse physiological and pathological processes.
Numerous studies indicate that alterations in intracellular Ca 2ϩ signaling are a primary stimulus for the hypertrophic response (41,42). As reported, the Ca 2ϩ /calmodulin-activated phosphatase calcineurin and its downstream transcriptional effector calcineurin-nuclear factor of activated T-cells (NFAT) have been implicated as transducers of the pathological hypertrophic response (43). Importantly, decreased I to and downregulated expression of ventricular Kv4 mRNAs are consistent findings in the hypertrophic myocardium, which can lead to increases in Ca 2ϩ influx through L-type Ca 2ϩ channels during longer action potential (44). On the other hand, we suggest that, in this pathological condition, increased diastolic [Ca 2ϩ ] i might induce down-regulation of I to . Thus, Ca 2ϩ influx is under control I to through action potential duration influence, resulting in elevated intracellular Ca 2ϩ levels, which in turn regulates I to and then reshapes action potential waveforms favoring Ca 2ϩ influx. We summarize this mechanism as Ca 2ϩ -induced Ca 2ϩ entry. More generally, our finding, if extended to other cell types, might help to understand how cells coordinate the expression of K ϩ channel for regulation of excitability.