Urotensin-II Receptor Stimulation of Cardiac L-type Ca2+ Channels Requires the βγ Subunits of Gi/o-protein and Phosphatidylinositol 3-Kinase-dependent Protein Kinase C β1 Isoform*

Background: Regulation of L-type Ca2+ channels has important roles in determining the electrical properties of cardiomyocytes. Results: U-II potentiates ICa,L via U-IIR that couples to the PI3K-dependent PKCβ1 isoform. Conclusion: U-IIR stimulation of ICa,L contributes to the increase in the amplitude of sarcomere shortening. Significance: Regulation of ICa,L by U-IIR plays important roles in cardiovascular actions including cardiac positive inotropic effects and increasing cardiac output. Recent studies have demonstrated that urotensin-II (U-II) plays important roles in cardiovascular actions including cardiac positive inotropic effects and increasing cardiac output. However, the mechanisms underlying these effects of U-II in cardiomyocytes still remain unknown. We show by electrophysiological studies that U-II dose-dependently potentiates L-type Ca2+ currents (ICa,L) in adult rat ventricular myocytes. This effect was U-II receptor (U-IIR)-dependent and was associated with a depolarizing shift in the voltage dependence of inactivation. Intracellular application of guanosine-5′-O-(2-thiodiphosphate) and pertussis toxin pretreatment both abolished the stimulatory effects of U-II. Dialysis of cells with the QEHA peptide, but not scrambled peptide SKEE, blocked the U-II-induced response. The phosphatidylinositol 3-kinase (PI3K) inhibitor wortmannin as well as the class I PI3K antagonist CH132799 blocked the U-II-induced ICa,L response. Protein kinase C antagonists calphostin C and chelerythrine chloride as well as dialysis of cells with 1,2bis(2aminophenoxy)ethaneN,N,N′,N′-tetraacetic acid abolished the U-II-induced responses, whereas PKCα inhibition or PKA blockade had no effect. Exposure of ventricular myocytes to U-II markedly increased membrane PKCβ1 expression, whereas inhibition of PKCβ1 pharmacologically or by shRNA targeting abolished the U-II-induced ICa,L response. Functionally, we observed a significant increase in the amplitude of sarcomere shortening induced by U-II; blockade of U-IIR as well as PKCβ inhibition abolished this effect, whereas Bay K8644 mimicked the U-II response. Taken together, our results indicate that U-II potentiates ICa,L through the βγ subunits of Gi/o-protein and downstream activation of the class I PI3K-dependent PKCβ1 isoform. This occurred via the activation of U-IIR and contributes to the positive inotropic effect on cardiomyocytes.

Isolation of Ventricular Myocytes-All the procedures and protocols conformed to the Guide for Care and Use of Laboratory Animals published by the National Institutes of Health, and the local institutional ethical committee approved the study. Left ventricular myocytes were enzymatically dissociated from rat hearts using a modified method described previously (26,27). Briefly, male Sprague-Dawley rats (200 -300 g) were injected with heparin (1000 IU intraperitoneally) and then euthanized in a CO 2 chamber. The heart was dissected out and transferred to ice-cold Tyrode's solution. The aorta was cannulated, and the heart was mounted on a Langendorff apparatus and perfused with prewarmed (37°C) and oxygenated Tyrode's solution containing protease type XIV and collagenase type I for 12 min until the heart was flaccid. The left ventricles were dissected out, cut into small pieces, and gently bathed in Tyrode's solution. Isolated cells were filtered and maintained in oxygenated Kraft-Brühe (KB) solution. Cells with a rod shape and clear cross-striation were used for experiments. For short hairpin RNA (shRNA) knockdown experiments, primary cultures of adult rat ventricular myocytes were cultured in serumfree Medium 199 (Invitrogen) containing taurine (5 mM), creatine (5 mM), L-carnitine (5 mM), and sodium bicarbonate (26 mM) and plated in 24-well plates onto 12-mm glass coverslips coated with matrigel (BD Biosciences). All the media used for the cell culture contained 50 IU penicillin and 50 g/ml streptomycin, and the cells were incubated under sterile conditions at 5% CO 2 and 37°C.
Protein Kinase A (PKA) Activity Assay-PKA activity in homogenates from isolated cardiomyocytes was determined by enzyme-linked immunosorbent assay (ELISA; Promega, Madison, WI) according to the manufacturer's instructions. Briefly, the cells were pretreated with either vehicle or KT-5720 for 30 min followed by treatment with either vehicle (0.1% DMSO) or forskolin for 15 min. The cells were washed with ice-cold PBS, placed on ice, and incubated with 200 l of lysis buffer. After a 10-min incubation on ice, the cells were transferred to microcentrifuge tubes. Cell lysates were centrifuged for 15 min, and aliquots of the supernatants containing 0.2 g of protein were assayed for PKA activity. The activity is expressed as relative light units Ϫ1 /amount of protein.
Determination of PKC Activity-The cells were pretreated with either vehicle or calphostin C for 30 min at 37°C followed by treatment with either vehicle or U-II for 15 min. Cells were resuspended in 50 l of buffer A (20 M Tris, 2 M EDTA, 0.5 M EGTA, and 1 M PMSF, pH 7.4) containing 50 M 2-mercaptoethanol and 25 l of 10% Nonidet P-40 followed by sonication for 45 s. The cell lysates were centrifuged at 100,000 ϫ g for 1 h and stored at Ϫ20°C. PKC activity was determined using the PepTag PKC assay kit (Promega) according to the manufacturer's instructions.
Phosphatidylinositol 3-Kinase (PI3K) Activity Assay-Cells were stimulated with or without U-II (0.1 M) for 15 min. After stimulation, PI3K activity was determined using a phosphatidylinositol 3-kinase activity ELISA kit according to the manufacturer's instructions (Tian-Ao Biotechnology). The absorbance of samples was measured at 450 nm with an EL 340 Bio Kinetic Reader (Bio-Tek Instruments).
Adenovirus Transduction-Three shRNAs targeting the sequence of PKC␤ 1 (GenBank TM accession number NM_ 012713.3) or G q (GenBank accession number Y17161.1) were designed, and the best knockdown effects of shRNA for G q (5Ј-CGCGGAGAUCGAGAAACAG-3Ј) and PKC␤ 1 (5Ј-CAC-AUCCAACAAGUUGGCCGUUUCA-3Ј) were selected for subsequent experiments. Additional scrambled sequences also were designed as negative controls (NCs) (G q , AGAUG-AGAGAGCACACCGG; PKC␤ 1 , GUACUCACCUGACUCAU-CAUGCAAG). The recombinant adenoviruses containing G q shRNA (Ad-G q -shRNA) or PKC␤ 1 shRNA (Ad-PKC␤ 1 -shRNA) were packaged by GeneChem (Shanghai, China) using the pAdeasy system with pAdtrack-CMV-GFP as a shuttle vector and pAdeasy-1 as an adenoviral backbone. The virus was purified on two consecutive cesium chloride gradients, dialyzed, and titered. The titer of viral preparations used in the studies ranged from 10 9 to 10 10 pfu/ml. Forty-eight hours after infection, cells showing green fluorescence under an inverted fluorescence microscope (Ti-2000, Nikon) were subjected to whole-cell patch clamp analysis. RNAi efficacy on G q or PKC␤ 1 expression was analyzed by Western blot analysis.
Measurement of Myocyte Contraction-Cell shortening of ventricular myocytes was assessed by a video-based edge detection system (IonOptix Corp.). The cells were placed in a perfusion chamber mounted on the stage of an inverted microscope (Zeiss IM). The cells were field-stimulated with 20% suprathreshold voltage at a frequency of 0.5 Hz with a pair of platinum electrodes. The myocyte being studied was displayed on the computer monitor with the help of an IonOptix MyoCam charge-coupled device camera, which was attached to the sidearm of the microscope. SoftEdge acquisition software (IonOptix Corp.) captured and converted the changes in sarcomere length to digital signals. Only rod-shaped myocytes with clear edges were selected for experiments. Signals were recorded when steady-state contraction was reached in each experimental medium.
Electrophysiology and Data Analysis-Whole cell patch clamp electrophysiology was used to measure the L-type Ca 2ϩ currents (I Ca,L ) in rat ventricular myocytes. Cells were placed in a recording dish and perfused with a bath solution containing 140 mM tetraethylammonium chloride, 2 mM CaCl 2 , 0.5 mM MgCl 2 , 5.5 mM glucose, 5 mM CsCl, and 10 mM HEPES, pH 7. 35 with CsOH. Recordings were performed at room temperature (22-24°C) using a MultiClamp 700B amplifier (Molecular Devices). Recording pipettes (World Precision Instruments) had 2-3-megaohm resistance when filled with internal solution containing 110 mM CsCl, 4 mM Mg-ATP, 0.3 mM Na 2 -GTP, 10 mM EGTA, and 25 mM HEPES, pH 7.3 with CsOH. pClamp 10.2 was used to acquire and analyze all data. I Ca,L were recorded in voltage clamp mode, and signals were filtered at 1 kHz using a low pass Bessel filter and digitized at 5 kHz. Series resistance and capacitance readings were taken directly from the amplifier after electronic subtraction of capacitive transients. Series resistance was compensated to the maximal extent (at least 75%). Current traces were corrected using on-line P/6 trace subtraction. In experiments in which cells were dialyzed with compounds or peptides, current measurements were started at least 5 min after breaking the patch. I Ca,L were measured at their peak amplitude within the 200-ms test pulse and used to determine the percentage of I Ca,L increase. Concentration-response curves were fitted by the following sigmoidal Hill equation: I/I control ϭ 1/(1 ϩ 10 (log(IC50) Ϫ X) n H ) where X is the decadic logarithm of the concentration used, IC 50 is the concentration at which the half-maximum effect occurs, and n H is the Hill coefficient. Activation data were fitted by the following modified Boltzmann equation: where G max is the fitted maximal conductance, V half is the membrane potential for half-activation, and k is the slope factor. Steady-state inactivation of I A was fitted with the following negative Boltzmann equation: where I max is maximal current, V half is the membrane potential for half-inactivation, and k is the slope factor. Summary data are expressed as mean Ϯ S.E. GraphPad Prism 5.0 was used to analyze the data. Statistical significance was determined using Student's t test when two groups were compared and one-way analysis of variance with a post hoc Bonferroni test when three or more groups were compared. Results were considered statistically significant at a p value of Ͻ0.05.

RESULTS
U-II Enhances I Ca,L -Whole-cell currents were elicited by a 200-ms stepping voltage from a holding potential of Ϫ60 to 0 mV. These currents can be blocked by nicardipine (5 M), a specific L-type Ca 2ϩ channel blocker (data not shown). Addition of U-II (0.1 M) to the bath caused a significant increase in peak I Ca,L to 28.3% of the basal level (n ϭ 8; Fig. 1, A and B). Following removal of U-II, the amplitude of I Ca,L recovered partially within 5 min (Fig. 1, B and C). Further examination of the U-II effect demonstrated that U-II increased I Ca,L in a concen-tration-dependent manner (Fig. 1D). The relationship between U-II concentration and the degree of increase observed is described by a logistic equation where the concentration of U-II producing half-maximal increase (IC 50 ) is 62.7 nM, the apparent Hill coefficient is 0.86, and the maximal stimulatory effect is 45.2 Ϯ 1.8% (n ϭ 9; Fig. 1D).
Next, we determined whether the biophysical properties of I Ca,L were affected by U-II. A current-voltage curve was evoked by a series of depolarizing pulses from a holding potential of Ϫ60 mV to test potentials between Ϫ60 and ϩ50 mV. Population data showed that 0.1 M U-II significantly down-shifted the current-voltage curve (n ϭ 12; Fig. 1E), and at 0 mV, the current density increased from 6.8 Ϯ 0.7 to 9.1 Ϯ 0.6 pA/picofarad (p Ͻ 0.01, n ϭ 12; Fig. 1F). Further effects mediated by U-II including the voltage dependences of activation and inactivation were examined. We observed a significant shift of the steady-state inactivation potentials of I Ca,L by 8.5 mV in the depolarized direction (V half from Ϫ32.6 Ϯ 0.6 to Ϫ24.1 Ϯ 0.8 mV, n ϭ 11; Fig. 1, G and I), whereas the activation potential did not change significantly (V half from Ϫ12.6 Ϯ 0.5 to Ϫ12.8 Ϯ 0.8 mV, n ϭ 9; Fig. 1, H and I). These results suggest that the increase in I Ca,L observed after the application of U-II may be due to the retention of a decreased proportion of inactivated channels.
U-IIR Mediates U-II-induced Increase in I Ca,L -U-IIR is the functional receptor for U-II in vivo (2). We examined U-IIR participation in I Ca,L responses to U-II by examining the expression profile in rat ventricular myocytes. RT-PCR analysis demonstrated that the transcripts for U-IIR (predicted size of amplicon is 537 bp) were present in adult rat ventricular myocytes. Negative control reactions in which reverse transcriptase was excluded from the RT step did not yield any PCR products . F, summary of results showing the current density of I Ca,L at 0 mV as indicated in E. G-I, U-II did not significantly alter the steady-state activation curve of L-type Ca 2ϩ channels (H; n ϭ 9) but caused a rightward shift of the steady-state inactivation curve (G; n ϭ 11). To determine the steady-state inactivation, I Ca,L was evoked by a 100-ms test pulse to 0 mV after the 3-s conditioning pulses ranging from Ϫ80 to ϩ30 mV with 10-mV increments.
To determine the voltage dependence of activation, tail currents were elicited by repolarization to Ϫ60 mV after 40-ms test pulses from Ϫ50 to ϩ40 mV in increments of 10 mV. ( Fig. 2A). Protein levels of U-IIR were assessed by immunoblotting with subunit-specific antibodies. Immunoblotting analysis revealed that U-IIR is endogenously expressed in adult rat ventricular myocytes. The rat testis expresses U-IIR and was therefore used as a positive control (Fig. 2B). We next determined the involvement of U-IIR in U-II-induced changes in I Ca,L . Palosuran (1 M), a specific U-IIR antagonist, alone had no significant effect on I Ca,L in adult rat ventricular myocytes (increase of 1.5 Ϯ 1.2%, n ϭ 7), whereas pretreatment of cells with palosuran (1 M) completely abolished the U-II-induced I Ca,L response (increase of 2.6 Ϯ 1.3%, n ϭ 9, p Ͻ 0.001; Fig. 2, C and D), suggesting that the U-II-induced increase in I Ca,L was dependent on U-IIR.
U-II-induced Changes in I Ca,L Require the ␤␥ Subunits of G i/o -U-IIR is a G-protein-coupled receptor and has been shown to couple to G q/11 in the cardiomyocytes (2,32). To investigate the potential involvement of heterotrimeric G-proteins, cells were dialyzed with the non-hydrolyzable GDP analog GDP-␤-S (1 mM). GDP-␤-S completely abolished the increase in I Ca,L induced by U-II (increase of 2.7 Ϯ 1.1%, n ϭ 8; Fig. 3A), indicating the requirement for G-protein activation. We next determined the involvement of G q/11 in the response mediated by U-IIR. Because of a lack of commercially available specific G q/11 inhibitors, an adenovirus-based shRNA knockdown approach was used to examine the effect of U-II on I Ca,L in G q -silenced ventricular myocytes. Western blot analysis showed that the expression of G q was significantly reduced in cells transduced with G q shRNA (Ad-G q -shRNA) compared with the cells transduced with control shRNA (Ad-NC-shRNA; Fig. 3B). Knockdown of G q did not affect the U-II-induced I Ca,L increase (increase of 27.9 Ϯ 1.7%, n ϭ 12; Fig. 3C), and there was no significant difference in the I Ca,L density between control shRNA and non-transduced control cells (Figs. 1F and 3C). These results support the hypothesis that the U-II-induced increase in I Ca,L is independent of G q/11 in adult rat ventricular myocytes. In cultured ventricular myocytes (48 h), the current density of I Ca,L slightly, but not significantly, decreased at 0 mV (6.6 Ϯ 0.4 pA/picofarad for control, n ϭ 9; 5.9 Ϯ 1.1 pA/picofarad for cultured cells, n ϭ 11; Fig. 3D). In addition, the voltage dependence of I Ca,L activation and inactivation was also measured in control and in cultured cells. Similar values for halfmaximal activation and inactivation voltage (V half ) (V half from Ϫ13.1 Ϯ 0.4 to Ϫ12.3 Ϯ 0.6 mV for activation curve, n ϭ 9; V half from Ϫ29.7 Ϯ 0.7 to Ϫ28.6 Ϯ 0.9 mV for inactivation curve, n ϭ 12; Fig. 3E) and slope factor (k) (k from 7.7 Ϯ 0.6 to 7.5 Ϯ 0.9 mV for activation curve, n ϭ 9; k from 8.9 Ϯ 0.6 to 9.1 Ϯ 0.8 mV for inactivation curve, n ϭ 12; Fig. 3E) were obtained in both groups. We next examined the involvement of different G-protein subtypes in the U-II-mediated modulation of I Ca,L . Inactivation of G s by pretreating ventricular myocytes with cholera toxin (0.5 g/ml for 24 h) did not affect the ability of U-II to increase I Ca,L (increase of 32.2 Ϯ 3.1%, n ϭ 10; Fig. 3F). Conversely, inhibition of G i/o by pretreating ventricular myocytes with PTX (0.2 g/ml for 24 h) abolished the stimulatory effect of U-II (increase of 1.3 Ϯ 2.1%, n ϭ 9; Fig. 3F). These results indicate that G i/o is involved in the transduction pathways leading to the increase in I Ca,L in response to U-IIR stimulation.
The potential role of the native ␤␥ subunits (G␤␥) of G i/oprotein in the U-IIR-mediated increase of I Ca,L was examined further. Intracellular infusion of synthetic peptide QEHA was used to bind G␤␥ subunits released after receptor activation, thus preventing the activation of the upstream effectors. Application of QEHA (10 M) through the recording pipette blocked the U-II-induced response (increase of 2.8 Ϯ 3.2%, n ϭ 9; Fig. 3, G and H), whereas similar dialysis of a scrambled peptide, SKEE (10 M), did not alter the ability of U-II to increase I Ca,L (increase of 27.5 Ϯ 1.5%, n ϭ 10; Fig. 3H). Together, these findings suggest that the G␤␥ subunit of the G i/o -protein complex mediates the U-II-induced increase in I Ca,L .
U-IIR-mediated Stimulation of I Ca,L Involves the Class I PI3K but Not Akt-Next, we investigated in detail the mechanism underlying the U-IIR-mediated I Ca,L increase. Examination of the potential involvement of intracellular signaling pathways revealed no evidence for the involvement of PKA. Preincubation of cells with the PKA inhibitor KT-5720 (1 M) did not alter the ability of U-II to increase I Ca,L (increase of 27.6 Ϯ 1.8%, n ϭ 10; Fig. 4, A and C). Similar results were obtained following addition of another PKA inhibitor, protein kinase inhibitor 5-24 (increase of 28.2 Ϯ 2.1%, n ϭ 9; Fig. 4, B and C). Pretreatment of ventricular myocytes with 1 M KT-5720 abolished the ability of forskolin (20 M) to increase PKA activity (Fig. 4D), thus confirming the inhibitory effect of KT-5720. Previous studies have shown that the immediate downstream mediator of G␤␥ is PI3K; therefore, to examine the role of PI3K in the response mediated by U-II, we first determined PI3K activity in ventricular myocytes. U-II significantly induced phosphatidylinositol 3,4,5-trisphosphate accumulation, and this effect was blocked by pretreatment of the cells with the PI3K inhibitor wortmannin (0.5 M) (Fig. 4E). Furthermore, wortmannin pretreatment completely abolished the increase of I Ca,L induced by U-II (increase of 0.9 Ϯ 2.5%, n ϭ 9; Fig. 4, F and H). It should be noted that pretreatment of cells with CH5132799 (1 M), which selectively inhibits class I PI3Ks, also blocked the U-II-induced I Ca,L response (increase of 2.3 Ϯ 1.1%, n ϭ 9; Fig. 4H). Akt is a common downstream target of PI3K (33)(34)(35). Therefore, we examined whether U-II action is mediated by Akt activation. We assayed the activity of Akt in ventricular myocytes treated with 0.1 M U-II. Fig. 4I illustrates that the level of phosphorylated Akt increased following treatment with U-II (0.1 M), whereas total Akt did not change. This effect was abolished by Akt inhibitor III (10 M ; Fig. 4I). To further explore the role of Akt in the modulation of I Ca,L by U-II, cells were pretreated with Akt inhibitor III prior to 0.1 M U-II stimulation. Interestingly, in the presence of Akt inhibitor III (10 M), U-II still caused a significant increase in I Ca,L (increase of 27.7 Ϯ 4.1%, n ϭ 11; Fig. 4, G and H). Taken together, these results suggest that U-II increases I Ca,L through PI3K activation (likely via the class I PI3K) but independently of PKA or Akt signaling.
U-IIR-induced Stimulation of I Ca,L Involves PKC-PKC activation has been shown to modulate L-type Ca 2ϩ channels (22,36) and can act as a downstream effector of G␤␥ activation (37). We further investigated the role of PKC in U-II-induced changes in I Ca,L . U-II (0.1 M) significantly increased PKC activity (ϳ2.3-fold) in rat ventricular myocytes (Fig. 5A). This response was abolished in cells pretreated with the class I PI3K inhibitor CH5132799 (1 M) (Fig. 5A). Calphostin C (50 nM), an inhibitor of both novel and classic PKC isoforms, abolished the ability of U-II to increase I Ca,L (increase of 1.5 Ϯ 1.6%, n ϭ 10; Fig. 5B). Similar results were obtained with another classic and novel PKC antagonist, chelerythrine chloride (1 M) (increase of 1.3 Ϯ 2.9%, n ϭ 9; Fig. 5B). Interestingly, the inhibition of PKC on U-II-induced responses by these antagonists was reproduced by Ro 31-8220 (2 M), which blocks only classic PKC isoforms (increase of 1.9 Ϯ 1.8%, n ϭ 10; Fig. 5B). Application of calphostin C (50 nM), chelerythrine chloride (1 M), or Ro 31-8220 (2 M) alone had no significant effect on I Ca,L in adult rat ventricular myocytes (increase of Ϫ1.7 Ϯ 1.9% for calphostin C, n ϭ 6; increase of 2.1 Ϯ 2.3% for chelerythrine, n ϭ 5; increase of Ϫ1.0 Ϯ 2.1% for Ro 31-8220, n ϭ 6; Fig. 5B). In contrast to novel PKC isoforms, activation of classic PKC isoforms requires cytoplasmic Ca 2ϩ . Intracellular dialysis of the fast Ca 2ϩ chelator BAPTA (20 mM), but not the inactive analog dn-BAPTA, completely abolished the U-II-mediated response (increase of 1.3 Ϯ 2.1% for BAPTA, n ϭ 10; increase of 27.5 Ϯ 3.6% for dn-BAPTA, n ϭ 8; Fig. 5, C and D), thus supporting the involvement of classic PKC isoforms in this process. The Classic PKC␤1 Isoform Is Involved in U-II Response-Next, we aimed to determine the exact PKC isoform involved in the U-II-induced I Ca,L increase. To achieve this, we first investigated the protein expression profiles of classic PKC isoforms in rat ventricular myocytes. Western blot analysis revealed that PKC␣, PKC␤ 1 , and PKC␤ 2 are expressed in adult rat ventricular myocytes, whereas PKC␥ could not be detected (Fig. 6A). Rat brain expresses all four classic PKC isoforms and was used as a positive control for the various PKC isoform antisera (Fig.  6A). Pretreatment of cells with HBDDE (1 M), a PKC␣ and PKC␥ inhibitor, did not affect the stimulatory effects of U-II on I Ca,L (increase of 26.9 Ϯ 3.3%, n ϭ 10; Fig. 6, B and D). In contrast, pretreating ventricular myocytes with the PKC␤ antagonist LY333531 (0.2 M) abolished the U-II-mediated change in I Ca,L (increase of 1.8 Ϯ 2.1%, n ϭ 9; Fig. 6, C and D). In addition, dialysis of cells with ␤IV5-3 (0.1 M), a PKC␤ 1 -specific inhibitor, completely blocked the U-II-induced I Ca,L increase (increase of 0.9 Ϯ 1.2%, n ϭ 9; Fig. 6E). Application of scrambled ␤IV5-3 (0.1 M) (increase of 28.1 Ϯ 1.9%, n ϭ 9; Fig.  6E) or the PKC␤ 2 -specific inhibitor ␤IIV5-3 (0.1 M) (increase of 27.3 Ϯ 3.1%, n ϭ 10; Fig. 6E) did not elicit this effect. Together, these results suggest that PKC␤ 1 may be involved in the U-II-induced increase in I Ca,L . To confirm this, we used an adenovirus-based shRNA approach to knock down PKC␤ 1 in  ventricular myocytes. Western blot analysis revealed that the expression of PKC␤ 1 was substantially reduced in cells transduced with PKC␤ 1 -specific shRNA (Ad-PKC␤ 1 -shRNA) compared with the cells transduced with control shRNA (Ad-NC-shRNA) (Fig. 6F), whereas the expression of PKC␤ 2 was not affected. Knockdown of PKC␤ 1 in ventricular myocytes almost completely eliminated the U-II-mediated increase in I Ca,L (increase of 1.3 Ϯ 0.7%, n ϭ 9; Fig. 6G). Cellular activation of PKC is linked to its translocation and binding to the plasma membrane. Therefore, to support the results from the previous experiments, we examined the translocation of PKC␤ 1 from the cytosolic to the membrane fractions of the cell following treatment with U-II. Western blot analysis indicated an increase of membrane-bound PKC␤ 1 and a decrease in the cytosolic fraction following stimulation with U-II (0.1 M; Fig. 6H). Collectively, these results suggest that the U-II-mediated increase in I Ca,L occurs through the classic PKC␤ 1 pathway.
U-II Increased the Amplitude of Sarcomere Shortening-L-type Ca 2ϩ channels are at the top of a cascade of events that initiate excitation-contraction coupling and thus regulate the strength of cardiac contraction (18 -20). L-type Ca 2ϩ channel blockers can cause a negative inotropism; conversely, any agent that increases I Ca,L might, in theory, serve as a positive inotrope (20). To further examine the functional implications of the I Ca,L increase induced by U-II, we tested the effects of U-II on cardiomyocyte contractility in rat ventricular myocytes. Our results showed that application of U-II (0.1 M) resulted in a rapid (usually within 5 min after administration) increase in the amplitude of sarcomere shortening (increase of 30.8 Ϯ 3.1%, n ϭ 12 from five hearts; Fig. 7, A and B). The effects of U-II on cardiomyocyte shortening in response to depolarizing pulses were reversible after washout of U-II from the bath solution (Fig. 7B). We next determined the involvement of U-IIR in U-II-induced changes in cardiomyocyte shortening. Our results showed that pretreatment of cells with palosuran (1 M), a specific U-IIR antagonist, abolished the U-II-induced response (increase of 2.1 Ϯ 1.7%, n ϭ 6 from three hearts; Fig.  7C), suggesting the involvement of U-IIR in the U-II-induced increase in the amplitude of sarcomere shortening. Furthermore, pretreating ventricular myocytes with the PKC␤ antagonist LY333531 (0.2 M) completely abolished the U-II-mediated change in sarcomere shortening (increase of 3.5 Ϯ 2.8%, n ϭ 6 from three hearts; Fig. 7C), whereas the PKC␣ inhibitor HBDDE (1 M) elicited no such effect (increase of 27.6 Ϯ 1.9%, n ϭ 10 from six hearts; Fig. 7C). To further verify that this U-II-induced response was mediated by the I Ca,L increase, we investigated whether Bay K8644, an L-type Ca 2ϩ channel agonist, would occlude the U-IIR-mediated increase in the amplitude of sarcomere shortening. Indeed, Bay K8644 at 0.5 M induced a significant increase in I Ca,L (increase of 37.6 Ϯ 3.1%, n ϭ 7; Fig. 7D). Application of Bay K8644 (0.5 M) to ventricular myocytes mimicked the U-II-induced increase in the amplitude of sarcomere shortening (Fig. 7, E and F). Notably, application of U-II (0.1 M) after the maximum Bay K8644-induced response failed to produce any further increase either in I Ca,L (increase of 35.9 Ϯ 6.5%, n ϭ 7; Fig. 7D) or in the amplitude of sarcomere shortening (Fig. 7, E and F). These results together suggest that U-II increased the amplitude of sarcomere shortening through a U-IIR-dependent PKC␤ and L-type Ca 2ϩ channel pathway.

DISCUSSION
The present study provides mechanistic data describing a novel functional role of U-II in modulating L-type Ca 2ϩ channels as well as sarcomere shortening in adult rat ventricular myocytes. These results suggest that this response is mediated by U-IIR coupled to the ␤␥ subunits of G i/o -proteins and subsequent activation of the class I PI3K-dependent PKC␤ 1 isoform. A schematic diagram of this proposed pathway is shown in Fig. 8.
U-IIR activation is detected through PTX-insensitive G-protein G q/11 coupling sequentially to phospholipase C (2,32). Interestingly, in adult rat ventricular myocytes, we found that the ␤␥ subunits of PTX-sensitive G i/o -protein are involved in the U-IIR-mediated L-type Ca 2ϩ channel stimulation because 1) the effect was blocked by the nonselective G-protein inhibitor GDP-␤-S and 2) pretreatment with PTX, but not cholera toxin, abolished the response to U-II, indicating the involvement of G i/o in rat ventricular myocytes. It has been suggested previously that G i/o can interact directly with L-type Ca 2ϩ channels (38); however, such a mechanism is unlikely to be involved in our system. We did find the involvement of G␤␥ subunits in the U-II-induced response because pipette application of the peptide QEHA, which disrupts interactions between G␤␥ subunits and Ca 2ϩ channels, but not its scrambled peptide SKEE abolished U-II-induced stimulation. However, G␤␥ did not seem to interact with the L-type Ca 2ϩ channels because the U-II-induced response was further blocked by the inhibition of downstream protein kinases. Importantly, the outcome of G␤␥ regulation of Cav1.2 was a reduction in I Ca,L due to direct interaction with the N terminus of ␣1C subunits (39). The apparent independence of G␤␥ subunits in the modulation of L-type calcium channels may indicate that rat ventricular myocytes express a Ca 2ϩ channel that is insensitive to the G␤␥ subunits of G i/o . An alternative hypothesis is that different Cav1.2 channel splice variants are able to generate different L-type Ca 2ϩ channel functions (40). For example, alternative splicing of the Cav1.2 channel changes the sensitivity of the L-type Ca 2ϩ channel to dihydropyridines (41). In addition, it is possible that U-IIR-mediated G␤␥ activation does not result in a functional increase in L-type Ca 2ϩ channel current in adult rat ventricular myocytes.
Presently, it is unclear how G␤␥ stimulates L-type Ca 2ϩ channel activity. G␤␥ subunits can activate PKA to modulate various targets including calcium channels (42,43). For example, I Ca,L recorded from isolated cardiomyocytes were shown to be increased by ␤-adrenergic receptors via the cAMP/PKA-dependent pathway (44). In contrast, I Ca,L inhibition by CB1 cannabinoid receptor activation was prevented by the application of PKA inhibitors (45). Similarly, Cav1.2 current inhibition by integrin receptor activation was blocked by the addition of the PKA inhibitor H89 (46). However, in the present study, we found that the stimulatory effect of U-II on I Ca,L was PKAindependent, which suggests that other non-cAMP/PKA-dependent mechanisms are involved in the stimulatory effect of U-II. PI3K (particularly PI3K␥) is a known downstream target of G␤␥ signaling, and there is evidence that G␤␥ stimulates Cav1.2 via PI3K (47). In this study, we found that the response to U-II was abolished by the selective PI3K inhibitors, suggesting that PI3K also participates in the U-IIR/G␤␥ pathway in rat ventricular myocytes. In contrast, the stimulatory effect of U-II on PKC activity was abolished by PI3K blockade, suggesting that PKC is downstream of PI3K rather than the reverse.
Next, we wished to determine how PI3K activates PKC. Phosphatidylinositol 3,4,5-trisphosphate, the lipid product of PI3K, targets several different second messengers (48,49) including PKC. Furthermore, PI3K␥ itself has serine kinase activity, and this could lead to PKC activation (50 -53). The G␤␥/PI3K pathway has been reported to regulate chloride channels in oocytes and has been linked to phosphatidylinositol 3,4,5-trisphosphate-dependent activation of PKC, an atypical PKC that is insensitive to both Ca 2ϩ and diacylglycerol (54,55). In addition, the G␤␥/PI3K pathway has been shown to activate a novel PKC isoform in rabbit portal vein myocytes (50). Conversely, the U-IIR-induced response is blocked with calphostin C. Calphostin C inhibits both classic and novel PKCs but not atypical PKCs. Furthermore, the effects of U-II in the present study were also inhibited when selective antagonists of classic PKCs and a fast intracellular calcium chelator were used, leading us to the conclusion that classic PKC isoforms are involved.
The following data suggest that PKC␤ 1 is involved in the U-II-induced increase in I Ca,L . 1) Pharmacological inhibition of PKC ␤ , but not PKC␣, completely abolished the increase of U-II on I Ca,L . 2) PKC␤ 1 has been identified in rat ventricular myocytes, and U-II increases the membrane expression of PKC␤ 1 .
3) Pharmacological inhibition of PKC␤ 1 or knockdown with shRNA blocked the U-II-induced I Ca,L response. These results are supported by previous studies showing that PKC activation increases I Ca,L in neonatal mouse ventricular myocytes (56). It has also been shown that PKC phosphorylates the Cav1.2 ␣1C calcium channel subunit, resulting in the up-regulation of L-type Ca 2ϩ channel activity (57). Similar results have been reported in rat portal vein myocytes (50). In contrast, a PKCinduced I Ca,L decrease has been described in the heart (58,59) and cerebral artery smooth muscle cells (60). Biphasic effects of PKC and no effect of PKC activation on L-type Ca 2ϩ channels have also been reported (61,62). Although the regulation of L-type Ca 2ϩ channels by PKC remains controversial, the differential modulation of L-type Ca 2ϩ channel activity by PKC may also involve different parameters. First, the expression and/or activation of endogenous PKC isoforms is tissue/cell-specific, and remarkable heterogeneity across PKC-dependent signal transduction pathways exists including that for ion channel modulation (63,64). As described in the present study, we suggest that the PKC␤ 1 isoform is involved in the U-II-induced I Ca,L response. Second, PKC modulation of L-type Ca 2ϩ channels may involve PKC-interacting proteins; this is the case for the Cav2.2 N-type channel (65). PKC-interacting proteins confer specificity on individual PKC isoforms by regulating their activity and cellular location, endowing isoforms with the ability to mediate specific cellular functions (66,67). Finally, cellspecific splice variants of Cav1.2 ␣1C (68) or ␤ subunits (69) might modulate the pharmacological properties of L-type Ca 2ϩ channels in different ways. Therefore, we cannot exclude the possibility that an intermediate protein, phosphorylated by a different PKC isoform, may be involved in the observed U-IIRmediated response.
In conclusion, the present study provides evidence of new mechanisms involved in the modulation of L-type Ca 2ϩ channels by U-II in adult rat ventricular myocytes. We propose that the marked increase in I Ca,L induced by U-II is mediated through U-IIR and involves the G␤␥ subunits of G i/o and downstream class I PI3K-dependent activation of the PKC␤ 1 pathway. We found no evidence of a role for PKA and Akt signaling. This novel mechanism may contribute to the physiological functions of U-II including ventricular contraction in the mammalian cardiovascular system.