β-Adrenergic Stimulation Increases Cav3.1 Activity in Cardiac Myocytes through Protein Kinase A

The T-type Ca2+ channel (TTCC) plays important roles in cellular excitability and Ca2+ regulation. In the heart, TTCC is found in the sinoatrial nodal (SAN) and conduction cells. Cav3.1 encodes one of the three types of TTCCs. To date, there is no report regarding the regulation of Cav3.1 by β-adrenergic agonists, which is the topic of this study. Ventricular myocytes (VMs) from Cav3.1 double transgenic (TG) mice and SAN cells from wild type, Cav3.1 knockout, or Cav3.2 knockout mice were used to study β-adrenergic regulation of overexpressed or native Cav3.1-mediated T-type Ca2+ current (ICa-T(3.1)). ICa-T(3.1) was not found in control VMs but was robust in all examined TG-VMs. A β-adrenergic agonist (isoproterenol, ISO) and a cyclic AMP analog (dibutyryl-cAMP) significantly increased ICa-T(3.1) as well as ICa-L in TG-VMs at both physiological and room temperatures. The ISO effect on ICa-L and ICa-T in TG myocytes was blocked by H89, a PKA inhibitor. ICa-T was detected in control wildtype SAN cells but not in Cav3.1 knockout SAN cells, indicating the identity of ICa-T in normal SAN cells is mediated by Cav3.1. Real-time PCR confirmed the presence of Cav3.1 mRNA but not mRNAs of Cav3.2 and Cav3.3 in the SAN. ICa-T in SAN cells from wild type or Cav3.2 knockout mice was significantly increased by ISO, suggesting native Cav3.1 channels can be upregulated by the β-adrenergic (β-AR) system. In conclusion, β-adrenergic stimulation increases ICa-T(3.1) in cardiomyocytes, which is mediated by the cAMP/PKA pathway. The upregulation of ICa-T(3.1) by the β-adrenergic system could play important roles in cellular functions involving Cav3.1.

Introduction T-type Ca 2+ channels (TTCCs or Cav3) belong to one of the families of voltage-dependent Ca 2+ channels. These channels are activated and inactivated at low membrane potentials (the threshold is about 260 mV) with rapid time-dependent decay (transient) and tiny single channel currents and thus termed Ttype. They are encoded by three genes, Cav3.1 (a1G), Cav3.2 (a1H) and Cav3.3 (a1I) [1,2,3,4,5]. The identification of the genes encoding TTCCs [2,3,5] allows the examination of the properties, distribution and function of each subtype of TTCCs and offers the potential to design isoform-specific TTCC antagonists to treat related channelopathies.
TTCCs are present in a wide variety of tissues including the heart, brain, skeletal muscle, testis and spermatids, indicating multiple functions of these channels such as cardiac rhythm generation, neuronal excitability, hormone secretion, neurotransmitter release, vascular tone regulation, muscle contraction, gene expression, cell metabolism, differentiation, and proliferation [2,3,5,6]. Therefore, abnormal expression and function of TTCCs are associated with many diseases including cardiac hypertrophy and arrhythmia, hypertension, epilepsy, autism, and cancer [6].
TTCCs are expressed in the whole heart during the embryonic stage but their expression in the ventricle decreases rapidly after birth [7]. Cav3.1 and Cav3.2 expression is retained in the sinoatrial node (SAN), atrioventricular node (AVN) and Purkinje fibers of the adult heart, indicating a role in cardiac automaticity and conduction [7]. Mice deficient of Cav3.2 showed normal sinoatrial rhythm [8], but mice lacking Cav3.1 had prolonged SAN recovery time, slowed pacemaker activity of SAN cells and heart rate, and delayed atrioventricular conduction. These results indicate Cav3.1, rather than Cav3.2, is the major TTCC participant in cardiac rhythm generation in the mouse heart [9].
Since b-adrenergic system is critical for heart rate regulation and Cav3.1 is involved in cardiac rhythm generation, it is important to examine the regulation of the TTCC by the badrenergic/PKA system. The regulation of TTCCs by cAMPdependent protein kinase A (PKA) has been controversial probably due to the differences in experimental conditions, cell types and the existence of specific isoforms [10]. In general it is believed that PKA has little effects on TTCCs [11,12,13]. Phosphorylation of Cav3.2 by PKA has been shown to permit the inhibitory effect of Gbc dimmers [14]. In contrast, T-type Ca 2+ current (I Ca-T , probably through Cav3.2 because it was sensitive to low concentration of Ni 2+ ) in frog atrial myocytes was reported to be increased by isoproterenol via a cAMP/PKA independent mechanism [15]. The same group showed that cAMP/PKA downstream to b-adrenergic receptor might phosphorylate a protein to enhance high-voltage prepulse-induced facilitation of TTCCs [16]. In addition, Lenglet et al. also reported that Cav3.2 TTCC activity recorded in rat glomerulosa cells was augmented by PKA after the stimulation of 5HT7 receptors [17]. To date, there is no report of the regulation of Cav3.1 by the b-adrenergic receptor/cAMP/PKA cascade in cardiac or other native mammalian cells.
In this study, we sought to determine whether Cav3.1 is regulated by b-adrenergic receptor/PKA signaling pathway using ventricular myocytes from Cav3.1 transgenic mice and sinoatrial node cells from wildtype or Cav3.2 knockout mice. We have found that the activity of both overexpressed and native Cav3.1 channel is enhanced by a b-adrenergic agonist, isoproterenol. This effect was mediated by the adenylyl cyclase/cAMP/PKA system because cAMP recapitulated the effect of isoproterenol while a PKA inhibitor (H89) abolished the effect of ISO on I Ca-T(3.1) . The upregulation of Cav3.1 by PKA may contribute to the regulation of the heart rate by the b-adrenergic system.

Ethical Approval
This study conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and was approved by the Institutional Animal Use and Care Committee at Temple University.

Ventricular Myocyte and Sinoatrial Node Cell Isolation
Mouse ventricular myocytes (VMs) were isolated using a constant-pressure Langendorff apparatus as described [20]. Animals were anesthetized with sodium pentobarbital (120 mg/ kg body weight). The heart was excised and digested via retrograde perfusion of the heart with normal Tyrode solution containing type II collagenase (290 U/mL) and (in mM): CaCl 2 0.02, glucose 10, HEPES 5, KCl 5.4, MgCl 2 1.2, NaCl 150, sodium pyruvate 2 (pH 7.4 with NaOH). After 8-10 min, the ventricles were minced and isolated ventricular myocytes were dissociated by gentle aspiration of minced tissue. Isolated VMs then were filtered with a 200 mm-diameter mesh. VMs were maintained in normal Tyrode solution containing 0.5% bovine serum albumin and the extracellular Ca 2+ was titrated up to 1 mM. Myocytes were used within 8 hours after isolation.
SAN cells were isolated as described previously [21,22]. Animals were anesthetized with sodium pentobarbital (120 mg/kg BW, intraperitoneal injection) and heparinized intravenously. SAN cells were isolated with a ''chunk'' technique as follow: After the heart was excised, it was placed into Tyrode's solution (35uC) containing (in mM) 140 NaCl, 5.0 HEPES, 5.5 Glucose, 5.4 KCl, 1.8 CaCl 2 , 1.0 MgCl 2 (pH = 7.4). The SAN region was the one with spontaneous activity and where the contraction of the heart initiates. The SAN tissue was dissected out with a dissecting scope according to the landmarks of the heart (delimited by the orifice of superior vena cava, crista terminalis and atrial septum) as described [21,22]. The SAN tissue was cut into smaller pieces, which were transferred and rinsed in a ''low Ca 2+ '' digestion containing (mM) 140 NaCl, 5.0 Hepes, 5.5 Glucose, 5.4 KCl, 0.2 CaCl 2 , 0.5 MgCl 2 , 1.2 KH 2 PO 4 , 50 Taurine, and 1 mg/mL BSA (pH = 6.9). Then SAN tissue pieces were digested in 5 mL of ''low Ca 2+ '' solution containing collagenase type I (Worthington, 225 U/mL), elastase (Worthington, 1.8 U/mL), and protease type XIV (0.8 U/mL, Sigma) for 30 min at 37uC. After digestion, the tissue was washed with 10 mL of Kraft-Bruhe medium containing (mM) 100 potassium glutamate, 5.0 Hepes, 20 Glucose, 25 KCl, 10 potassium aspartate, 2.0 MgSO 4 , 10 KH 2 PO 4 , 20 taurine, 5 creatine, 0.5 EGTA, and 1 mg/mL BSA (pH = 7.2) for 3 times and then the cells were dissociated with a transfer pipette by pipetting up and down the tissue chunks. Dissociated cells were put at room temperature for 5 min. The cells were stored at 4uC and studied within 5 hours.

Electrophysiology
Ca 2+ currents (I Ca ) were measured using whole-cell voltageclamp techniques with an Axopatch 2B voltage-clamp amplifier and pClamp8 software and 1-3 MV pipettes. Ca 2+ currents were measured under discontinuous voltage-clamp mode. The real clamping voltage and Ca 2+ currents were measured simultaneously. To achieve good voltage control, the gain was set between 8 to 50. To ensure the quality of our data, only data with the loss of voltage control ,10 mV were included for our report. VMs were placed in a heated chamber (3562uC or room temperature) on the stage of an inverted microscope (Nikon Diaphot, Japan) and initially perfused with normal Tyrode solution. The pipette contained a Na + -free and K + -free solution consisting of (in mM): Cs-aspartic acid 130, EGTA 10, MgCl 2 1, NMDG 10, HEPES 10, TEA-Cl 20, Tris-ATP 5, Tris-GTP 0.25, pH 7.2. Once a gigaohm seal was obtained, the patched membrane was ruptured to allow 10 minutes of dialysis of the myocyte. The perfusate was switched to a Na + -free, K + -free solution containing (in mM): CaCl 2 2, 4aminopyridine (4-AP) 2, CsCl 5.4, Glucose 10, HEPES 5, MgCl 2 1.2, NMDG 150, pH 7.4. Myocyte capacitance was obtained with the membrane test function in Clampex 8.0, which hyperpolarizes the cell for 25 mV from the holding potential to measure cell capacitance. The total calcium current (I Ca )-voltage relationship was determined by measuring I Ca from a holding potential of 290 mV using square wave pulses in a 10 mV-increment. I Ca-L was measured at the same test voltages from a holding potential of 250 mV. To minimize run-down of Ca 2+ currents, these two Ca 2+ currents measured at the same test potential were separated by a 1000 ms holding potential of 250 mV ( Figure 1B) and 20 s was set between sweeps. I Ca-T at each test membrane potential was determined by subtracting raw I Ca-L from raw total I Ca at this test potential, which is a traditional way for measuring I Ca-T [1]. Mibefradil (1 mM) or BayK 8644 (1 mM) was applied via perfusate to the cell while the Ca 2+ currents measured at the test potential 240 mV from the Vh of 290 mV and at 10 mV from the Vh of 250 mV to 10 mV were continuously monitored. BayK 8644, a DHP agonist, was selected because it could readily and affirmatively determine if BayK 8644 had any stimulatory effect on I Ca-T even in the presence of a rundown of I Ca-T .
To determine ISO effect on the current-voltage (I-V) relationships of I Ca-L and I Ca-T , I-V relationships of total I Ca and I Ca-L were recorded at baseline. Subsequently, 1 mM isoproterenol (ISO, Sigma) were applied through perfusate while the currents elicited from the Vh = 290 mV to 240 mV (mostly I Ca-T ) and from the V h = 250 mV to 10 mV (mostly I Ca-L ) were monitored until the maximum effect of ISO was observed. Then, the I-V relationships of total I Ca , I Ca-L and I Ca-T were determined again as described above. To determine if direct activation of PKA was able to regulate I Ca-T , a nonhydrolyzable cAMP analog (dibutyryl-cAMP, Sigma, 10 mM) was dialyzed into the cell for 10 minutes and the I Ca-L and I Ca-T were determined afterwards [23]. To determine if ISO effects on I Ca-L and I Ca-T is mediated by PKA, a PKA inhibitor (H89, Sigma, 5 mM), was included in the pipette filling solution and patched VMs were dialyzed for 10 minutes. Thereafter, the I-V relationships of total I Ca , I Ca-L and I Ca-T were measured before and after the application of ISO.
To investigate the effects of ISO and db-cAMP on voltagedependent inactivation of I Ca-L and I Ca-T(3.1) , double-pulse protocols were applied with or without these drugs. For I Ca-T(3.1) , the first pulse was the prepulse to different test voltages (2110 mV to 210 mV in a10 mV-increment) from the holding potential 290 mV. The second pulse was separated by a 5 ms repolarizing period to 290 mV from the prepulses and depolarized to 240 mV, a voltage almost only I Ca-T and minimal I Ca-L contamination could be recorded. For I Ca-L , the prepulses were evoked from the holding potential of 250 mV to different membrane potentials and the second pulse was evoked from a short return to 250 mV for 5 ms to 0 mV (for 400 ms).
Since in small SAN cells, rundown of I Ca-L and I Ca-T was faster (often I Ca-T was decreased by .30% within 10 minutes), it became a significant issue to study ISO effects if a long recording time was needed. Full I-V curves before and after ISO could not obtained. Therefore, to study ISO effects on I Ca-T in SAN cells, we only studied ISO effect at a single voltage (from the holding potential of 290 mV to 240 mV) in the presence of nifedipine (an I Ca-L blocker, 10 mM). The use of nifedipine ensured only I Ca-T was recorded at this voltage even ISO could shift the activation of I Ca-L to more negative voltage and thus avoided the use of double pulses to minimize recording time to minimize I Ca-T rundown.
The conductance of L-and T-type Ca 2+ channels (G Ca ) was calculated by dividing the current by the driving force (V t -E Ca ', where E Ca ' stands for the apparent reversal potential of Ca 2+ ). The activation-voltage (G-V) relationships were plotted using normalized L-or T-type Ca 2+ channel conductance (G Ca /G Ca,max ) as the y-axis and test potentials as the x-axis. The G-V curve is fitted with a Boltzmann function G where V t is the test voltage and V 0.5, d' is the voltage at which half of the G max can be elicited, k is the slope factor. The voltagedependent inactivation curve (f ' ) was fitted with a Boltzmann function G/G max = 1/[1+exp[2(V t 2V 0.5, f' )/k]], where V t is the test voltage and V 0.5, f' is the voltage at which half of the channels were inactivated, k is the slope factor. Real-time PCR SAN tissue was pooled from 4-8 mice for wild type (n = 24) and transgenic (n = 19) animals. Total mRNA was extracted from snap-frozen SAN tissue, ventricular tissue, or brain (as a positive control for T-type Ca 2+ channel expression) using Trizol reagent and quantitated by a UV spectrometer. Real-time PCR was done with the SYBR Green Real Time PCR kit (Applied Biosystems, Carlsbad, CA) according to the instruction of the kit and an Eppendorff Mastercycler RT-PCR machine. GAPDH was used as the internal control. The DDCt-method was used to determine the abundance of Cav3 mRNAs relative to GAPDH. The primers were (59 to 39): Cav3.1: forward: TGTGGAAATGGTGGT-GAAGA and reverse: ACTGCGGAGAAGCTGACATT; Cav3.2: forward: GCTGTTTGGGAGGCTAGAAT and reverse: CGAAGGTGACGAAGTAGACG; Cav3.3: forward: TGGGCATTTTTGGCAAGAA and reverse: CAGTGCG-GATGGCTGACA; GAPDH: forward: TGCACCAC-CAACTGCTTAG and reverse: GATGCAGGGATGATGTTC.

Data Analysis
Obtained data were analyzed offline with Clampfit 8 (Molecular Device, CA) as described previously [23], managed with Microsoft Excel and presented with GraphPad Prizm 5.0 (La Jolla, CA, USA). In short, I Ca-T was obtained by subtracting the raw I Ca-L from the raw total I Ca as described previously [24]. The currentvoltage (I-V) relationships were constructed by plotting the peak amplitudes of total I Ca , I Ca-L and derived I Ca-T against the test voltages.

Statistics
Data in the text are reported as mean6SEM. When appropriate, paired and unpaired T-test, ANOVA or ANOVA for repeated measures were used to detect significance with SAS 9.0 (SAS Institute Inc.). P values of #0.05 were considered significant. In this paper, n is the number of cells examined from at least 3 animals.

Results
Cav3.1 Mediated T-type Ca 2+ Current (I Ca-T(3.1) ) Was Observed Only in VMs from Cav3.1 Double Transgenic Mice To examine whether Cav3.1 was expressed in our transgenic system, I Ca-T was measured in VMs from transgenic (TG) and control hearts. Typical examples of total Ca 2+ current (I Ca ), L-type Ca 2+ current (I Ca-L ) and T-type Ca 2+ current (I Ca-T ) in control and TG VMs are shown in Figure 1B. The current-voltage relationships (I-V curves) of total I Ca , I Ca-L and I Ca-T in both control (n = 12) and TG myocytes (n = 5) are shown in Figure 1C and D. There was no detectable I Ca-T observed in all 12 tested control cells ( Figure 1B and C) but a great density of I Ca-T (maximum I Ca-T amplitude: 212.161.3pA/pF) was found in Cav3.1 TG VMs ( Figure 1B and D). I Ca-T had a threshold of ,260 mV and peaked at ,240 mV. The time-dependent decay rates (kinetics) of the I Ca-T were significantly faster than those of the I Ca-L recorded in the same cell ( Figure 1B), as reported previously [1].
To confirm the identity of putative I Ca-T , mibefradil (1 mM) was used to determine its sensitivity to this TTCC antagonist. Mibefradil suppressed I Ca-T at 240 mV by 66.868.7% (Figure 2A and C) but had minimal effect on I Ca-L at 10 mV (decreased by 11.162.8%; Figure 2B and D). To further confirm that I Ca-T could be effectively separated by holding the cell at different membrane potentials with minimal contamination of I Ca-L , BayK 8644, a dihydropyridine agonist specific for I Ca-L , was used to test dihydropyridine effect on the two components of I Ca . As suggested in Figure 1C and D, I Ca recorded at 240 mV from V h of 290 mV was almost completely consisting of I Ca-T while I Ca recorded at 10 mV from V h of 250 mV should be only I Ca-L . As predicted, I Ca at 240 mV from the V h of 290 mV was not changed by BayK ( Figure 2E, F and H) while I Ca at 10 mV from the V h of 250 mV was significantly increased 49.760.8% by BayK ( Figure 2E, G and H). When the I-V curves of presumable I Ca-L and I Ca-T , separated by the strategy of holding the cell at different membrane potentials, were examined, BayK evidently increased the presumable I Ca-L but not the presumable I Ca-T ( Figure 2I, J and K). These results suggest that the separation of I Ca-L and I Ca-T by holding the cell at different membrane potentials is effective, and that I Ca-T is not sensitive to dihydropyridines but sensitive to mibefradil.

ISO Significantly Increased I Ca-T(3.1) in TG VMs
To study whether I Ca-T(3.1) can be regulated by the b-adrenergic system in cardiac myocytes, isoproterenol, a b-adrenergic agonist, was applied to TG VMs. Since I Ca at 240 mV from the V h of 290 mV was almost completely consisting of the T-type Ca 2+ current, ISO effect on I Ca-T(3.1) was monitored by recording I Ca at 240 mV. Figure 3A showed I Ca at 240 mV before and after ISO and Figure 3B showed the time course of the ISO effect on a myocyte, suggesting a significant upregulation of I Ca-T(3.1) . I-V curves of total I Ca ( Figure 3C), I Ca-L ( Figure 3D), and I Ca-T ( Figure 3E) before and after ISO showed that b-adrenergic stimulation (ISO) significantly increased both I Ca-L and I Ca-T at most test voltages. The maximal I Ca-L was increased by 81.6% (from 27.661.2pA/pF to 213.861.3pA/pF) and the maximal I Ca-T(3.1) was increased by 55.5% (from 212.861.9pA/pF to 219.962.4pA/pF). It is well known that b-adrenergic stimulation shifts the activation of I Ca-L to more negative voltages and accelerates the decay of I Ca-L via a Ca 2+ -dependent inactivation mechanism [24]. In this study, we found that ISO shifted the peak I Ca voltage for I Ca-L but not for I Ca-T(3.1) (Figure 3D and E). It seems that there is no voltage dependent effect of ISO because ISO did not change the voltage-dependence of I Ca-T(3.1) activation and inactivation although it shifted the activation and inactivation of I Ca-L to more negative voltages ( Figure 3F and 3G).

Nonhydrolyzable Dibutyryl-cAMP (db-cAMP) Significantly Increased I Ca-T(3.1) at Both Physiological and Room Temperatures
The stimulatory effect of b-adrenergic agonists on calcium channels in cardiac myocytes can be mediated by mulitple mechanisms [24]: 1, PKA activation through the b-adrenergic receptor/Gas/adenyl cyclase/cAMP/PKA pathway; 2, direct modulation by Gas or Gbc after the binding of adrenergic agonists to the adrenergic receptor and subsequent dissociation of Gas from Gbc dimer; 3. The newly found cAMP sensor, EPAC (exchange protein directly activated by cAMP), regulates I Ca-T . We tested if direct activation of PKA by a cAMP analog (db-cAMP) was able to stimulate I Ca-T(3.1) . Db-cAMP significantly increased I Ca-L by 154.9% (with db-cAMP 215.563.3pA/pF versus without db-cAMP 26.160.8pA/pF, Figure 4B) and increased I Ca-T(3.1) by 102.8% (with db-cAMP 22.762.6pA/pF vs. without db-cAMP 211.260.9pA/pF, Figure 4C). The voltage-dependence of I Ca-T activation and inactivation was not altered by db-cAMP ( Figure 4F) although db-cAMP shifted the activation curve and inactivation curve to the left ( Figure 4E). These findings clearly Albeit our results with db-cAMP imply that the stimulatory effect of ISO might be mediated by PKA, recently another cAMP sensor, an exchange protein directly activated by cAMP (EPAC), has been found in cells from various tissues including the heart [25]. To rule out the involvement of EPAC and confirm the role of PKA in the stimulatory effects of ISO, a PKA-specific inhibitor, H89, was included in the pipette filling solution. In the presence of H89, ISO did not increase the current recording at the Vt = 240 mV from Vh = 290 mV (mainly I Ca-T(3.1) ) before and after the application of ISO (raw currents in Figure 5A and time course in Figure 5B). I-V curves of I Ca-L and I Ca-T were not changed by ISO with H89 in the pipette filling solution ( Figure 5C & D). These results support the idea that PKA activated by b-adrenergic agonists causes the upregulation of I Ca-T(3.1) .

I Ca-T(3.1) is the T-type Ca 2+ Current in Mouse SAN Cells and is Upregulated by Isoproterenol
At last, we examined if native I Ca-T in SAN cells is regulated by the b-adrenergic system. Total I Ca , I Ca-T and I Ca-L were recorded courses of amplitude changes of presumable I Ca-L and I Ca-T in response to BayK. H, Normalized increases in presumable I Ca-L and I Ca-T by BayK. I, J and K, I-V curves of total I Ca , I Ca-L and I Ca-T before and after BayK. doi:10.1371/journal.pone.0039965.g002 from wild type and Cav3.1 KO SAN cells. In wild type SAN cells, I Ca-T was clearly recorded with a maximum current of 22.0860.64pA/pF (Figure 6 A-C). In contrast, in the SAN cells from Cav3.1 knockout mice, T-type Ca 2+ current is completely absent, indicating that Cav3.1 is the major mediator of I Ca-T in mouse SAN cells (Figure 6 D-F) as previous study suggests [9]. Real-time PCR further confirmed that the major TTCC in the SAN is Cav3.1. Cav3.2 mRNA abundance was very low and Cav3.3 mRNA was not detectable in SAN tissue although all three types of TTCCs were detected in the brain (Figure 6 G). I Ca ( presumably I Ca-T ) measured at 240 mV in wild-type SAN cells bathed in 10 mM nifedipine (to block I Ca-L ) was increased by 115.7621.1% with 1 mM ISO ( Figure 6 H and I). To further make sure that I Ca-T in SAN cells was Cav3.1-mediated and could be stimulated by 1 mM ISO, I Ca-T was recorded in Cav3.2 knockout SAN cells and ISO did increase I Ca-T in these cells by 161.0620.5% ( Figure 6 J and K).

Discussion
What are the New Findings in this Study?
First, we found that b-adrenergic stimulation increased I Ca-T(3.1) in cardiac myocytes expressing exogenous or endogenous Cav3.1 channels. To our knowledge, this is the first detailed The Regulation of TTCCs by the b-adrenergic/PKA Pathway TTCCs are distributed in various cell types serving a wide range of functional roles [3,26]. Abnormal expression of T-type Ca 2+ channels is involved in pathological conditions including epilepsy [27], neurogenic pain [28], cancer [29,30], and cardiac hypertrophy [31]. Therefore, the regulation of TTCCs has been studied since their discoveries [10]. However, early studies often show conflicting results possibly due to unknown molecular identities and TTCC heterogeneity in cells [10]. Interestingly, it has been shown that the same neurotransmitter, hormone, or protein kinase exerts different effects on subtypes of TTCCs or on the same TTCC subtype in different tissues [10]. In addition, in our study, we have observed a run-down phenomenon of I Ca-T at baseline or during b-adrenergic stimulation in some myocytes, which could mask the stimulatory effects of b-adrenergic.
Specifically for b-adrenergic/PKA regulation of TTCCs, it remains controversial [32]. Most previous studies showed little effect of PKA on TTCCs [11,12,13]. Phosphorylation of Cav3.2 by PKA permits the inhibition of I Ca-T by Gbc dimmers [14]. On the other hand, T-type Ca 2+ current has been reported be increase by ISO in frog atrial cells through a cAMP/PKA-independent mechanism [15] while in rat glomerulosa cells by 5-HT via a PKA-dependent mechanism [17]. Cav3.2 channel activity can be increased by db-cAMP in heterologous systems as well [32,33]. To date, there has been no report about Cav3.1 regulation by badrenergic system, which is not present in heterologous systems. It has been shown that in the mouse heart, Cav3.1 is the primary TTCC found in SAN cells as our study also suggests [9]. Here, for the first time, we have shown that in cardiac myocytes only expressing Cav3.1 (TG myocytes or Cav3.2 knockout SAN cells), I Ca-T(3.1) was significantly increased by ISO. Db-cAMP reproduces the effect of ISO on I Ca-L(3.1) at both room and physiological temperatures while H89 (a PKA inhibitor) blocked ISO effect on I Ca-L and I Ca-T in TG myocytes, confirming that the stimulatory effect of ISO on I Ca-L and I Ca-T is mediated by PKA.

Potential Mechanisms for PKA-dependent Upregulation of I Ca-T(3.1)
In the current study, we did not determine how I Ca-T(3.1) is regulated by PKA activated by ISO. PKA might phosphorylate Cav3.1 directly as it does on L-type (Cav1) Ca 2+ channels. Since it is generally believed that Cav3 channels do not have accessory subunits, the PKA site(s) could be on the Cav3.1a subunit. It is possible that PKA-dependent phosphorylation of Cav3.1 augments the open probability of the channel, as it does to the L-type Ca 2+ channel. A second possibility is that PKA might phosphorylate another molecule to indirectly augment the open probability of the channel. Further studies will be needed to define the mechanism. The effect of b-adrenergic stimulation on I Ca-T could be reversed by dephosphorylation of Cav3.1a or a related protein if PKA phosphorylation is the mechanism for the upregulation of Cav3.1 activity. Multiple protein serine/threonine phosphatases including PP1, PP2A, and PP2B, are expressed in cardiac myocytes. Which phosphatase is involved in Cav3.1 by ISO warrants further investigation.

Relevance of b-adrenergic Mediated Upregulation of I Ca-T(3.1) in the Heart
In the heart, T-type calcium channels contribute to the heart rate generation [34], and it is possible that b-adrenergic/PKA regulation of Cav3.1 may participate in the positive chronotropic effects of b-adrenergic agonists. I Ca-T is also reported to be reemerged and a participator in excitation-contraction coupling in stressed hearts; it is likely that b-adrenergic mediated upregulation of I Ca-T might also contribute to cardiac contraction in stressed hearts although our previous studies have shown that overexpressed Cav3.1 is not effective to load the SR and trigger SR Ca 2+ release [35]. Furthermore, if there is abnormally high I Ca-T (e.g., overstimulation of the Cav3 channels by the sympathetic nervous system) in the cardiac pacemaking tissues and the conduction system, there could be tachycardia or atrial fibrillation or ectopic ventricular contraction. We have observed ectopic ventricular contraction in some Cav3.1 TG mice. In line with this, blocking Cav3 channels is able to reduce arrhythmic events and sudden cardiac death in a mouse heart failure model [36].

Conclusion
In cardiac myocytes, Cav3.1 current is increased by badrenergic agonists. This effect is mediated by protein kinase A. The regulation of Cav3.1 by b-adrenergic/PKA signaling pathway could play a role in heart rate regulation, arrhythmias and regulating other cellular functions involving Cav3.1.