Facilitated Hyperpolarization Signaling in Vascular Smooth Muscle-overexpressing TRIC-A Channels*

Background: TRIC channels likely mediate counter-K+ movements during store Ca2+ release, but their roles are unknown in many cell types. Results: Smooth muscle-specific Tric-a-transgenic mice developed hypotension due to facilitated hyperpolarization signaling in vascular muscle. Conclusion: TRIC-A channels facilitate Ca2+ spark generation in vascular muscle. Significance: TRIC-A channels may have unexplained roles in store Ca2+ distribution but provide a potential pharmaceutical target for vasomodulation. The TRIC channel subtypes, namely TRIC-A and TRIC-B, are intracellular monovalent cation-specific channels and likely mediate counterion movements to support efficient Ca2+ release from the sarco/endoplasmic reticulum. Vascular smooth muscle cells (VSMCs) contain both TRIC subtypes and two Ca2+ release mechanisms; incidental opening of ryanodine receptors (RyRs) generates local Ca2+ sparks to induce hyperpolarization and relaxation, whereas agonist-induced activation of inositol trisphosphate receptors produces global Ca2+ transients causing contraction. Tric-a knock-out mice develop hypertension due to insufficient RyR-mediated Ca2+ sparks in VSMCs. Here we describe transgenic mice overexpressing TRIC-A channels under the control of a smooth muscle cell-specific promoter. The transgenic mice developed congenital hypotension. In Tric-a-overexpressing VSMCs from the transgenic mice, the resting membrane potential decreased because RyR-mediated Ca2+ sparks were facilitated and cell surface Ca2+-dependent K+ channels were hyperactivated. Under such hyperpolarized conditions, L-type Ca2+ channels were inactivated, and thus, the resting intracellular Ca2+ levels were reduced in Tric-a-overexpressing VSMCs. Moreover, Tric-a overexpression impaired inositol trisphosphate-sensitive stores to diminish agonist-induced Ca2+ signaling in VSMCs. These altered features likely reduced vascular tonus leading to the hypotensive phenotype. Our Tric-a-transgenic mice together with Tric-a knock-out mice indicate that TRIC-A channel density in VSMCs is responsible for controlling basal blood pressure at the whole-animal level.


The TRIC channel subtypes, namely TRIC-A and TRIC-B, are intracellular monovalent cation-specific channels and likely mediate counterion movements to support efficient Ca 2؉ release from the sarco/endoplasmic reticulum. Vascular smooth muscle cells (VSMCs) contain both TRIC subtypes and two Ca 2؉ release mechanisms; incidental opening of ryanodine receptors (RyRs) generates local Ca 2؉ sparks to induce hyperpolarization and relaxation, whereas agonist-induced activation of inositol trisphosphate receptors produces global Ca 2؉ transients causing contraction.
Tric-a knock-out mice develop hypertension due to insufficient RyR-mediated Ca 2؉ sparks in VSMCs. Here we describe transgenic mice overexpressing TRIC-A channels under the control of a smooth muscle cell-specific promoter. The transgenic mice developed congenital hypotension. In Trica-overexpressing VSMCs from the transgenic mice, the resting membrane potential decreased because RyR-mediated Ca 2؉ sparks were facilitated and cell surface Ca 2؉ -dependent K ؉ channels were hyperactivated. Under such hyperpolarized conditions, L-type Ca 2؉ channels were inactivated, and thus, the resting intracellular Ca 2؉ levels were reduced in Tric-a-overexpressing VSMCs. Moreover, Tric-a overexpression impaired inositol trisphosphate-sensitive stores to diminish agonist-induced Ca 2؉ signaling in VSMCs. These altered features likely reduced vascular tonus leading to the hypotensive phenotype. Our Tric-a-transgenic mice together with Tric-a knock-out mice indicate that TRIC-A channel density in VSMCs is responsible for controlling basal blood pressure at the whole-animal level.
Two major classes of intracellular Ca 2ϩ channels, inositol 1,4,5-trisphosphate receptors (IP 3 Rs) 3 and ryanodine receptors (RyRs), provide the fundamental pathway for Ca 2ϩ release from internal stores into the cytoplasm (1)(2)(3)(4). IP 3 R-mediated Ca 2ϩ release from IP 3 -sensitive stores in the sarco/endoplasmic reticulum (SR/ER) plays a central role in cellular signaling systems. RyR-mediated Ca 2ϩ release from caffeine-sensitive stores also regulates important cellular functions such as excitationcontraction coupling, secretion/exocytosis, and gene expression, primarily in excitable cells. When the divalent cation Ca 2ϩ is released from the SR/ER, a negative potential is likely generated on the luminal side and may inhibit subsequent Ca 2ϩ release. Therefore, counterion movements may balance SR/ER membrane potential to establish efficient Ca 2ϩ release mechanisms in various cell types (5,6). Indeed, previous electrophysiological and fluorometric analyses detected several ionic currents, such as K ϩ , Cl Ϫ , and H ϩ fluxes, across the SR/ER membrane (7)(8)(9)(10). However, the channel/transporter components mediating such ionic flows have not been identified, and thus, their contributions to the proposed counterion movements are largely unknown.
Recently, we identified two TRIC (trimeric intracellular cation) channel subtypes, TRIC-A and TRIC-B, that assemble into bullet-shaped homotrimers to form monovalent cation-selective channels in the SR/ER and nuclear membranes (11)(12)(13). Double-knock-out mice lacking both the Tric-a and Tric-b genes show embryonic cardiac failure, and the mutant car-diomyocytes display weak RyR-mediated Ca 2ϩ release from the SR (14). Tric-a knock-out mice develop hypertension associated with vascular hypertonicity, and the mutant vascular smooth muscle cells (VSMCs) show insufficient Ca 2ϩ sparks for inducing hyperpolarization (15). Moreover, mutant skeletal muscle from Tric-a knock-out mice occasionally exhibits alternan contraction responses, likely due to destabilized RyR-mediated Ca 2ϩ release (16). On the other hand, Tric-b knock-out mice develop respiratory failure at birth, and the mutant alveolar epithelial cells exhibit compromised IP 3 R-mediated Ca 2ϩ release and insufficient handling of surfactant lipids (17). These defects observed in the knock-out mice support our hypothesis that TRIC channel subtypes partly mediate counterion movements to facilitate SR/ER Ca 2ϩ release.
VSMCs possess both caffeine-and IP 3 -sensitive stores and also contain both TRIC-A and TRIC-B channels. Therefore, VSMCs are an ideal model system to examine the physiological functions of TRIC channel subtypes. We recently detected not only insufficient RyR-mediated Ca 2ϩ sparks but also facilitated IP 3 R-mediated Ca 2ϩ transients in Tric-a knock-out VSMCs (15). These observations led to the hypothesis that TRIC-A channels preferentially support RyR-mediated Ca 2ϩ release and also regulate Ca 2ϩ distribution between the caffeine-and IP 3 -sensitive stores. To further verify this hypothesis, we planned to generate mutant mice carrying the SMC-specific Tric-a transgene and to examine altered features in Tric-aoverexpressing VSMCs.

EXPERIMENTAL PROCEDURES
Transgenic Mice and Blood Pressure Monitoring-Transgenic mice overexpressing TRIC-A channels under the control of the human ␣-smooth muscle actin promoter were generated and genotyped as previously described (15). The transgenic and wild-type littermates (8 -12 weeks old) were examined physiologically and biochemically in this study. All experiments were conducted with the approval of the Animal Research Committee at Kyoto University according to the regulations on animal experimentation at Kyoto University. To telemetrically monitor arterial blood pressure, PA-C10 transmitters (Data Sciences International, New Brighton, MN) were implanted into mice, and the pressure-sensing catheters were inserted into the aortic arch under anesthesia. Radio signals from the implanted transmitter were captured using the Physiotel RPC-1 receiver (Data Sciences International), and the data were stored online using the Dataquest ART data acquisition system (Data Sciences International). Blood pressure and heart rate in the conscious state were also monitored by tail-cuff plethysmography (Model BP-98A-L, Softron) during the daytime.
Anatomical Analysis-Histological and ultrastructural analyses were carried out as described previously (18). Briefly, mesenteric arteries from adult mice were fixed in 3% paraformaldehyde, 2.5% glutaraldehyde, and 0.1 M sodium cacodylate, pH 7.4. After the tissues were dehydrated and embedded in Epon, ultrathin sections (ϳ80 nm thick) were prepared and stained with toluidine blue for histological observation or uranyl acetate and lead citrate for electron microscopic analysis (JEM-200CX, JEOL).
Imaging Analysis-Single VSMCs were enzymatically isolated from mesenteric arteries and subjected to Ca 2ϩ spark and membrane potential monitoring as described previously (15,19). Briefly, single VSMCs were seeded onto glass-bottomed dishes and incubated with 5 M Fluo-4AM (Dojindo) in HEPES-buffered saline: 137 mM NaCl, 5.9 mM KCl, 2.2 mM CaCl 2 , 1.2 mM MgCl 2 , 14 mM glucose, and 10 mM HEPES, pH 7.4, with NaOH). The Ca 2ϩ spark images with excitation at 488 nm and emission at Ͼ520 nm were captured using a total internal reflection fluorescence microscopy system (TE-2000U, Nikon), and the data were analyzed using the custom software Aquacosmos (Hamamatsu Photonics). For membrane potential monitoring, single VSMCs on glass-bottomed dishes were perfused with HEPES-buffered saline containing 200 nM oxonol VI (Fluka). Fluorescence images with excitation at 559 nm and emission at Ͼ606 nm were captured at a sampling rate of ϳ1.6 s using a confocal microscope system (FV1000, Olympus).
For monitoring intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) in VSMCs, the inner surface of the artery was gently wiped with gauze to remove endothelial cells as described previously (20). The resulting arterial muscle segments were incubated with 5 M Fura-PE3AM (Santa Cruz) and 0.02% cremophor EL (Sigma) and tightly fixed with fine steel pins onto a silicone rubber sheet, which was placed on a glass-bottomed dish. A CCD camera (ImagEM, Hamamatsu Photonics) mounted on the microscope (DMI 4000B, Leica) equipped with a polychromatic illumination system (MetaFluor ver7.7r3, Molecular Device) was used to capture fluorescence images with excitation at 340 and 380 nm and emission at Ͼ510 nm at room temperature (ϳ23°C).
Patch Clamp and Membrane Potential Recording-Wholecell voltage clamp recording was carried out using the Axopatch 200B amplifier (Molecular Device) essentially as described previously (21). For spontaneous transient outward current (STOC) monitoring in single VSMCs, the STOC pipette solution (140 mM KCl, 4 mM MgCl 2 , 5 mM Na 2 ATP, 0.05 mM EGTA, and 10 mM HEPES, pH 7.2, with KOH) and HEPESbuffered saline as a bathing solution were utilized. For the recording of total Ca 2ϩ -dependent K ϩ currents, the BK pipette solution (140 mM KCl, 1 mM MgCl 2 , 6.1 mM CaCl 2 , 1 mM Na 2 ATP, 10 mM EGTA, and 10 mM HEPES, pH 7.2, with KOH, pCa 6.5) and HEPES-buffered saline containing 100 M CdCl 2 as a bathing solution were utilized. All physiological measurements were carried out at room temperature.
The arterial muscle segments were subjected to transmembrane potential monitoring with high resistance glass microelectrodes (30 -50 megaohms) filled with 3 M KCl solution as described previously (22). Membrane potential was determined as an average of the potentials obtained from stable impalements longer than 1 min from different areas in the muscle segment. The transmembrane potential was amplified by a high-input impedance amplifier with capacitance neutralization (MEZ-7200, Nihon Kohden), and the electrical signals were digitized and analyzed using a Power Lab system (AD Instruments). The bath solution used was modified Krebs solution babbled with 95% O 2 , 5% CO 2 and maintained at 37°C at pH ϳ 7.4 (112 mM NaCl, 4.7 mM KCl, 2.2 mM CaCl 2 , 1.2 mM MgSO 4 , 25 mM NaHCO 3 , 1.2 mM KH 2 PO 4 , and 14 mM glucose).

RT-PCR and Immunochemical
Analysis-For quantitative PCR analysis, total RNA preparations from various tissues (at least three mice of each genotype) were used as templates for cDNA synthesis (ReverTra ACE qPCR-RT kit, Toyobo) and analyzed using a real-time PCR system according to the manufacturer's instructions (Thermal Cycler TP800, Takara). The cycle threshold was determined from the cDNA amplification curve as an index for relative mRNA content in each reaction. The PCR primer sets used in this study were listed previously (15). For immunochemical analysis, tissue lysates and fixed VSMCs were examined using primary antibodies to TRIC channel subtypes and calnexin (Santa Cruz) as described previously (14,18).

Hypotension in SMC-specific Tric-a-Transgenic Mice-We
previously generated transgenic mice with SMC-specific overexpression of TRIC-A channels using the ␣-smooth muscle actin promoter (23). Although the constructed Tric-a-transgene rescues the hypertensive phenotype developed in Tric-a knock-out mice (15), the effects of the transgene have not been examined in a wild-type genetic background. Therefore, we generated two Tric-a-transgenic mouse lines in wild-type backgrounds, which we designated as TgA3 and TgA20 mice. Both transgenic mouse lines grew and reproduced normally (supplemental Fig. S1A). RT-PCR analysis demonstrated that the transgene was highly expressed in blood vessels and other tissues containing SMCs, and its expression was faintly detected in the liver and skeletal muscle (supplemental Fig. S1B). In accordance with our previous observation that the transgene dosage in the TgA20 mouse line is higher than in the TgA3 line (15), TgA20 mice showed higher transgene expression in SMCcontaining tissues than TgA3 mice in RT-PCR analysis (supplemental Fig. S1C).
By telemetric blood pressure monitoring, both TgA20 and TgA3 mice, even at a young-adult age, showed low blood pressure levels throughout the entire day (Fig. 1A). This hypotensive phenotype was further confirmed by tail-cuff monitoring (Fig. 1B). Most likely reflecting the difference in transgene expression between the mouse lines, TgA20 mice developed slightly lower blood pressure than TgA3 mice. Elevated blood pressure stimulates the baroreceptor reflex and effectively decreases heart rate in normal animals (24). However, in both transgenic mice, a slow heart rate was clearly observed under hypotensive conditions. The blockage of cardiac autonomic controls by coapplication of the muscarinic blocker atropine and the ␤-adrenoreceptor blocker metoprolol ameliorated the bradycardic phenotype, whereas low blood pressure was clearly maintained in the transgenic mice (Fig. 1B). Therefore, ectopic Tric-a expression in baroreceptors and/or autonomic neurons likely produced vagal-dominant states in both transgenic mice. Moreover, the ␣1-adrenoreceptor blocker prazosin induced similar vasodilating effects in both wild-type and transgenic mice (supplemental Fig.  S1E). These observations suggest that impaired sympathetic stimuli to the heart and vascular smooth muscle were not significant contributors to hypotension in the transgenic mice. It may be reasonable that altered vascular functions underlie the hypotensive phenotype in both transgenic mice because Tric-a knock-out mice develop hypertension due to increased vascular tonus (15).
Tric-a-Overexpressing VSMCs from the Transgenic Mice-Resistance arteries distributed to peripheral tissues are primarily responsible for blood pressure maintenance. The mesenteric artery is a typical resistance vessel and was examined in our studies below. By RT-PCR analysis of mesenteric arteries, the Tric-a mRNA levels in TgA20 and TgA3 mice were at least 400-fold greater than those of wild-type controls ( Fig. 2A). Western blot analysis confirmed Tric-a overexpression, and enhanced immunoreactivities provided estimates that the TRIC-A protein levels in the arteries from the transgenic mice were increased more than 500-fold compared with wild-type arteries (Fig. 2B). However, neither the heart nor skeletal muscle from the transgenic mice showed an obvious enhancement of TRIC-A immunoreactivity (supplemental Fig. S1D). Such SMC-specific overexpression of TRIC-A channels may not cause serious damage to the fundamental functions of resistance arteries because the transgenic mice retained circadian blood pressure variation (Fig. 1A) and regular blood pressure fluctuation in response to typical vasodilators targeting VSMCs (supplemental Fig. S1E). A, telemetric blood pressure monitoring is shown. Circadian fluctuations in systolic blood pressure (SBP) were monitored, and the data were averaged over each 2-h interval during a 24-h period. B, systolic blood pressure and heart rate (HR) monitoring by tail-cuff plethysmography during the daytime. After the base-line monitoring (Ctrl), autonomic controls were blocked using an intraperitoneal injection of the muscarinic antagonist atropine (Atr, 4 mg/kg) and the ␤-antagonist metoprolol (Met, 4 mg/kg). Upon drug application, HR was quickly attenuated, but systolic blood pressure changed in a time-dependent manner (see supplemental Fig. S1E). The data represent the mean Ϯ S.E., and the numbers of mice examined are shown in parentheses. Significant differences between the genotypes are marked with asterisks (*, p Ͻ 0.05; **, p Ͻ 0.01 by Student's t test). MAY 31, 2013 • VOLUME 288 • NUMBER 22

JOURNAL OF BIOLOGICAL CHEMISTRY 15583
Histological observations revealed no obvious abnormalities in mesenteric arteries from TgA20 and TgA3 mice (Fig. 2C). For example, under relaxation conditions after incubation in a Ca 2ϩ -free bathing solution, similar passive diameters and VSMC layer thicknesses of the arteries were observed in the transgenic and wild-type mice (supplemental Fig. S2A). These observations suggest that the hypotensive phenotype was mainly caused by insufficient myogenic tonus of resistance arteries under in vivo conditions in the transgenic mice.
Electron microscopic observation detected irregular membranous ultrastructures in mesenteric arteries from the transgenic mice. In Tric-a-overexpressing VSMCs, stacked rough ER elements accumulated around cell-interior regions (Fig. 2D,  left panels). The generation of stacked rough ER has been repeatedly reported in various cell types overexpressing integral SR/ER membrane proteins (25,26), and such irregular ER elements were not detected in wild-type VSMCs. Indeed, dense immunofluorescence signals of ectopic TRIC-A channels were predominantly observed in the cell-interior regions of VSMCs from transgenic mice (supplemental Fig. S2B), suggesting that the ER stacks were generated by nonspecific effects of excess TRIC-A protein. In SMCs, IP 3 Rs are distributed uniformly over the nuclear and ER/SR membranes (27,28), whereas RyRs are predominantly localized in peripheral SR elements (29,30). Therefore, it is possible that the formation of the stacked rough ER may have a functional impact on IP 3 -sensitive stores. However, such a morphological abnormality may minimally affect caffeine-sensitive stores.
In VSMCs, free-floating small vesicles, originally referred to as "surface vesicles" (31), were abundantly arranged underneath the cell membrane (Fig. 2D, right panels). Accompanied by the surface vesicles, larger-sized vacuoles with myelin figures (multilamellar lipid deposits) or without electron-dense material in the lumen were frequently observed in Tric-a-overexpressing VSMCs. Such vacuoles were rarely detected in wildtype VSMCs. To resolve the stacked ER elements, the large vacuoles may be aberrantly generated in Tric-a-overexpressing VSMCs. Alternatively, the excess amounts of TRIC-A protein may have promoted the swelling and fusion of normal small vesicles, generating the vacuoles. Swelling of SR elements is frequently accompanied by stored Ca 2ϩ overloading in skeletal and cardiac muscle (16,32,33), and RyR-mediated Ca 2ϩ sparks are predominantly generated underneath the cell membrane in VSMCs (3,4). The facilitated formation of the vacuoles may imply altered Ca 2ϩ -handling characteristics in caffeine-sensitive stores of Tric-a-overexpressing VSMCs.
Facilitated Hyperpolarization Signaling in Tric-a-Overexpressing VSMCs-In VSMCs, incidental activation of RyRs generates Ca 2ϩ sparks and then activates big-conductance Ca 2ϩdependent K ϩ (BK) channels to evoke STOCs, leading to hyperpolarization (3,4). To identify the direct cause of the hypotensive phenotype, we investigated Ca 2ϩ spark and STOC generation in VSMCs from the transgenic mice because hyperpolarization signaling is impaired in Tric-a knock-out VSMCs (15). Our Ca 2ϩ spark imaging captured hyperactivated sparks at specified intracellular hotspots in single VSMCs prepared from the transgenic mice (Fig. 3A). Statistical analysis of the imaging data detected no difference in Ca 2ϩ spark amplitude between the genotypes. However, Ca 2ϩ spark frequency was substantially elevated, and the number of spark hotspots was significantly increased in Tric-a-overexpressing VSMCs from both TgA20 and TgA3 mice (Fig. 3, B-D). RT-PCR analysis detected no altered gene expression of the key components in A, shown is quantitative detection of Tric-a mRNA in mesenteric artery (MA) and aorta. Total tissue RNA preparations from transgenic (#3 and #20) and wild-type mice (n ϭ 3) were examined by quantitative RT-PCR using a primer set for amplifying Tric-a mRNA from both the endogenous gene and the transgene. The cDNA fragments amplified by 33 PCR cycles were analyzed by agarose gel electrophoresis. RT-PCR data obtained are summarized in the right graph. The cycle threshold (Ct) indicates the cycle number at which the amount of amplified cDNA reached a fixed threshold in each reaction. B, Western blot analysis of TRIC-A protein in MA and aorta is shown. The vessels dissected from the Tric-a-transgenic (#3 and #20), Tric-a knock-out (AϪ/Ϫ), and wild-type mice were homogenized to prepare total cell lysates (postnuclear fractions). Wild-type lysates (10 g of protein) with or without lysates from the transgenic mice (0.1 g) were analyzed with an antibody against the TRIC-A protein. Lysates from the Tric-a knock-out mice (10 g) served as negative controls. Arrowheads indicate TRIC-A protein bands. Immunoreactive signals were digitalized and statistically analyzed to estimate the relative contents of TRIC-A protein in the different genotypes as shown in the right graph. C, normal histology in MA from Tric-a-transgenic mice is shown. Mesenteric artery preparations were incubated in a Ca 2ϩ -free solution for ϳ20 min and then fixed for anatomical analysis. Thin sections were stained with toluidine blue for photomicroscopic observation. IVL, intravascular lumen. Scale bar, 10 m. D, shown is formation of stacked ER elements and vacuoles in Tric-aoverexpressing VSMCs. Rough ER (rER) elements at the cell-interior portion (left panels: scale bar, 500 nm) are shown. Stacks of rough ER elements were frequently observed in VSMCs from transgenic mice (ϳ27% of TgA3 cells and 80% of TgA20 cells), whereas such stacked ER was not detected in wild-type controls (n ϭ 45-97 cells from 3-6 mice in each genotype). Surface vesicles and vacuoles were located at the cell periphery (right panels: scale bar, 200 nm). TRIC-A overexpression appeared to promote the formation of largesized vacuoles containing myelin figures (arrows) or no electron-dense materials (arrowheads). Such vacuoles were frequently observed in Ͼ50% of VSMCs from TgA3 and TgA20 mice but only detected in 9.5% of wild-type VSMCs (n ϭ 53-89 cells from 6 mice in each genotype). intracellular Ca 2ϩ signaling and G protein-coupled receptor signaling, such as RyR, IP 3 R, Ca 2ϩ -pump protein, ␣-adrenoreceptor, and phospholipase C, in mesenteric arteries and thoracic aorta from the transgenic mice (supplemental Fig. S3A). Therefore, Tric-a overexpression appears to hyperactivate Ca 2ϩ spark generation without affecting the cellular levels of major Ca 2ϩ -handling proteins in VSMCs.
For monitoring BK channel-mediated STOCs, single VSMCs were examined by patch clamp recording using the nominally Ca 2ϩ -free STOC pipette solution. VSMCs from the transgenic mice showed highly activated STOCs (Fig. 4A). In particular, both STOC frequency and amplitude were markedly increased at a holding potential of Ϫ60 mV near the resting potential in the Tric-a-overexpressing cells (Fig. 4B). We also monitored total K ϩ currents using the Ca 2ϩ -containing BK pipette solution and examined the effects of the BK channel inhibitor iberiotoxin on VSMCs. VSMCs from the transgenic mice showed normal levels of total K ϩ currents both sensitive and insensitive to iberiotoxin (supplemental Fig. S3B). Therefore, despite normal cell-surface densities of BK channels, STOC generation was likely facilitated under physiological conditions in Tric-aoverexpressing VSMCs. The enhanced STOCs were fully consistent with the hyperactivated Ca 2ϩ sparks observed (Fig. 3), suggesting that hyperpolarization signaling generated by functional cross-talk between the RyR and BK channels was stimulated in VSMCs from the transgenic mice.
Deepened Resting Potential in Tric-a-Overexpressing VSMCs-In SMCs, BK channel-mediated STOCs contribute to the main-tenance of resting membrane potential and the fine-tuning of cellular excitability (3,4). Based on the facilitated STOCs observed, we next focused on membrane potential in Tric-aoverexpressing VSMCs and leveraged confocal microscopic imaging using the voltage-dependent dye oxonol VI. Imaging analysis showed that fractional fluorescence intensity was clearly decreased in VSMCs from both TgA20 and TgA3 mice at basal conditions (Fig. 5A). The decreased intensities suggested that Tric-a-overexpressing VSMCs were deeply hyperpolarized during the resting state (Fig. 5B). We previously prepared the calibration plot representing the relationship between the fluorescence intensity and membrane potential using the reported resting potential of Ϫ59.9 mV in wild-type VSMCs (15,34). The calibration plot estimated that the resting membrane potentials of TgA3, TgA20, and Tric-a knock-out VSMCs were Ϫ63.6, Ϫ63.9, and Ϫ54.6 mV, respectively. The estimated potentials agreed closely with actual measurement values from direct monitoring using high resistance microcapillary electrodes; the detected resting potentials of wild-type, TgA20, and Tric-a knock-out VSMCs were Ϫ59.4 Ϯ 0.6, Ϫ64.8 Ϯ 1.0, and Ϫ53.7 Ϯ 0.7 mV, respectively (supplemental Fig. S3C). Moreover, from oxonol VI imaging, enlarged intensity shifts were evoked by iberiotoxin in VSMCs from the transgenic mice (Fig. 5C). The enhanced responses directly reflected the hyperactivation of BK channels at resting conditions in Tric-a-overexpressing VSMCs. However, while being exposed to iberiotoxin, similar fractional intensities were observed in wild-type and Tric-a-overexpressing VSMCs. Because many channels and transporters, in addition to BK channels, contribute to generating resting potential, the iberiotoxin-induced realignment of resting intensities suggested that the components other than BK channels functioned normally on the cell surface of Tric-a-overexpressing VSMCs. In summary, these results suggest that specific hyperactivation of BK channels lowered resting membrane potential in VSMCs from the transgenic mice.
Altered Ca 2ϩ Handling in Tric-a-Overexpressing VSMCs-Of the excitable cell types, SMCs have relatively shallow resting membrane potentials. In VSMCs at basal conditions, L-type voltage-gated Ca 2ϩ channels generate persistent Ca 2ϩ influx  The membrane potential of single VSMCs was controlled by the whole-cell patch clamp technique using the STOC pipette to monitor membrane currents. Representative STOC recording data are illustrated in A. The data for STOC frequency and amplitude are summarized in B. The data represent the mean Ϯ S.E., and the numbers of cells examined from at least three mice are shown in parentheses. Significant differences between the genotypes are indicated by asterisks (*, p Ͻ 0.05; **, p Ͻ 0.01 by t test). and maintain both resting [Ca 2ϩ ] i and intrinsic tonus (3,4). We examined cellular Ca 2ϩ handling in VSMCs from the transgenic mice and found several abnormal features. First, our ratiometric Fura-PE3 imaging detected that resting [Ca 2ϩ ] i was markedly decreased in VSMCs from the transgenic mice (Fig. 6,  A and B). In accordance with the detected membrane potentials, TgA20 cells exhibited slightly lower [Ca 2ϩ ] i than TgA3 cells. In a Ca 2ϩ -free extracellular solution, similar resting [Ca 2ϩ ] i levels were observed in wild-type and Tric-a-overexpressing VSMCs. Moreover, the downward-adjusted resting [Ca 2ϩ ] i was clearly rescued by the L-type Ca 2ϩ channel inhibitor verapamil and the activator BayK8644 in Tric-a-overexpressing VSMCs (Fig. 6C). Therefore, L-type Ca 2ϩ channels were likely inactivated at basal conditions, and resting [Ca 2ϩ ] i was thus decreased in Tric-a-overexpressing VSMCs.
VSMCs contain both IP 3 -and caffeine-sensitive stores, and they are functionally interconnected (3,4). The SR/ER Ca 2ϩpump inhibitor cyclopiazonic acid (CPA) causes Ca 2ϩ leakage from both stores, and the total stored Ca 2ϩ contents can be estimated from CPA-induced responses. CPA-induced Ca 2ϩ depletion from intracellular stores generally triggers store-operated Ca 2ϩ entry (35). VSMCs from the transgenic mice exhibited normal CPA-induced and store-operated Ca 2ϩ entry responses (Fig. 6, D and E). However, Tric-a-overexpressing VSMCs showed weak IP 3 R-mediated Ca 2ϩ release in response to the ␣-adrenoreceptor agonist phenylephrine (PE) (Fig. 6, F  and G). We also observed that PE-induced responses were diminished after normal caffeine-evoked transients in Tric-aoverexpressing VSMCs (supplemental Fig. S4A). Furthermore, in response to sequential applications of PE and caffeine under Ca 2ϩ -free conditions, Tric-a-overexpressing VSMCs exhibited poor IP 3 R-mediated transients immediately followed by enhanced RyR-mediated transients (Fig. 6, H and I). It is unlikely that poor PE-induced transients were due to compromised phosphoinositide turnover signaling because RT-PCR analysis suggested normal expression of major components of adrenoreceptor-mediated signaling in Trica-overexpressing VSMCs (supplemental Fig. S3A). Therefore, in addition to facilitating Ca 2ϩ spark generation as described above, Tric-a overexpression appeared to affect IP 3 -sensitive stores in VSMCs.

DISCUSSION
Single-channel recording in planar lipid bilayers demonstrated that purified TRIC-A and TRIC-B proteins form monovalent cation-specific channels (12)(13)(14). Therefore, TRIC channel subtypes likely function as SR/ER K ϩ channels under intracellular conditions. Although TRIC channels exhibit marked voltage dependence, they are likely to contribute to persistent K ϩ flux when membrane potential and osmotic pressure fluctuate in the SR/ER (11). In our previous study, Tric-a knock-out mice displayed high vascular tonus and developed hypertension because the hyperpolarization signaling triggered by Ca 2ϩ sparks was compromised in the knock-out VSMCs (15). In the present study, as a polar opposite of the TRIC-Anull model, SMC-specific Tric-a-transgenic mice developed hypotension, and Ca 2ϩ spark generation was facilitated in Trica-overexpressing VSMCs. Therefore, we reasonably conclude that TRIC-A channels support RyR-mediated Ca 2ϩ spark generation and contribute to setting the basal tonus in VSMCs. From a biophysical point of view, the density-dependent contribution of TRIC-A channels to Ca 2ϩ spark generation can be explained by an oversimplified scheme (supplemental Fig. S4B). Several lines of evidence indicate that RyR subtypes are nonselective cation channels equipped with large diameter pores and can bidirectionally conduct Ca 2ϩ and K ϩ (36). Our model tentatively assumes that only K ϩ provided by TRIC-A or RyR channels acts as a counter ion for stabilizing SR membrane potential during Ca 2ϩ release. In wild-type VSMCs, both TRIC-A and RyR channels may conduct counter-K ϩ to support Ca 2ϩ release. However, the SR K ϩ currents are likely insufficient to stabilize membrane potential during vital Ca 2ϩ release, and many events of incidental RyR opening may fail to grow to the Ca 2ϩ sparks visualized in Ca 2ϩ -imaging analysis. Even lower SR K ϩ currents are proposed for Tric-a knock-out VSMCs, and thus, the growth to detectable Ca 2ϩ sparks is further attenuated by the TRIC-A-null conditions. In contrast, in Tric-a-overexpressing VSMCs, enhanced SR K ϩ currents may no longer attenuate the growth of Ca 2ϩ sparks. Additionally, the excess K ϩ currents mediated by overexpressed TRIC-A channels may persistently prevent negative-shifting of the membrane potential toward the Ca 2ϩ reversal potential and could effectively maintain Ca 2ϩ release from partially depleted stores. Such biophysical mechanisms likely underlie the facilitated Ca 2ϩ sparks observed in Tric-a-overexpressing VSMCs. In the present study, several altered functions linked to the hypotensive phenotype were characterized in Tric-a-overexpressing VSMCs (Fig. 7). Excess TRIC-A channels facilitate Ca 2ϩ spark (Fig. 3) and STOC generation (Fig. 4) and decrease resting membrane potential in VSMCs (Fig. 5). Under such FIGURE 5. Decreased resting membrane potential in Tric-a-overexpressing VSMCs. Single VSMCs were examined by confocal microscopic imaging using the voltage-dependent dye oxonol VI. Cellular fluorescence intensities were normalized to the maximum value to yield the fractional intensity (F/F 140K ). A, shown are representative imaging data. B, shown are the summarized data of the resting intensity. C, shown are the summarized data of the intensity shift by iberiotoxin (IBTX). The data represent the mean Ϯ S.E., and the numbers of cells examined from at least three mice are shown in parentheses. Significant differences between the genotypes are indicated by asterisks (**, p Ͻ 0.01 by t test).
hyperpolarized conditions, L-type Ca 2ϩ channels are relatively inactivated, and resting [Ca 2ϩ ] i is suppressed in Tric-a-overexpressing VSMCs (Fig. 6). Decreased [Ca 2ϩ ] i in VSMCs likely reduces the intrinsic tonus of resistance arteries, leading to hypotension in the Tric-a-transgenic mice (Fig. 1).
VSMCs possess both caffeine-and IP 3 -sensitive stores and contain both TRIC-A and TRIC-B channels with similar estimated expression levels (15). In accordance with previous studies (37)(38)(39), our Ca 2ϩ imaging also detected that caffeine-induced responses were larger than agonist-induced responses in wild-type VSMCs from mesenteric arteries (Fig. 6H). In addition to the altered features directly linked to hypotension, Trica-overexpressing VSMCs exhibited an unexplainable abnormality in SR Ca 2ϩ handling; PE-induced Ca 2ϩ release was attenuated despite unchanged total Ca 2ϩ content in the SR (Fig.  6D). In contrast, Tric-a knock-out VSMCs exhibit facilitated PE-induced Ca 2ϩ release from the Ca 2ϩ -overloaded SR (15). The alterations in PE-induced response appear to display a mirror image between Tric-a-overexpressing and knock-out VSMCs (supplemental Fig. S4A). Low Ca 2ϩ contents of IP 3 -sensitive stores may contribute to impaired PE-induced Ca 2ϩ release in Tric-a-overexpressing cells because the sequential application of PE and caffeine implied poor loading in IP 3 -sensitive stores (Fig. 6H). In this case, overexpressed TRIC-A channels may selectively gather Ca 2ϩ into caffeine-sensitive stores to relatively deplete IP 3 -sensitive stores in VSMCs. Because IP 3 R channels are activated by both high luminal and cytoplasmic Ca 2ϩ (1,38), lowered luminal and cytoplasmic Ca 2ϩ levels may synergistically inhibit IP 3 R channels in Tric-a-overexpressing VSMCs, whereas elevated Ca 2ϩ levels on both sides could facilitate IP 3 Rs in Tric-a knock-out VSMCs. Alternatively, altered PE-induced responses could be explained by differentially modulated IP 3 Rs in Tric-a-overexpressing and knock-out VSMCs. In addition to generating countercurrents favorable to RyR activation, the TRIC-A protein might inhibit IP 3 R channels in VSMCs. These postulated inhibitory effects might be enhanced to remarkably depress PE-induced Ca 2ϩ release in Tric-a-overexpressing VSMCs but might be removed to facilitate PE-evoked responses in Tric-a knock-out VSMCs. Furthermore, other molecular mechanisms can also be hypoth- esized for the altered PE-induced Ca 2ϩ release depending upon TRIC-A contents. To further investigate the mysterious role of TRIC-A channels in IP 3 -sensitive stores, physiological measurements of the mutant VSMCs under membrane-permeable "skinned" conditions together with comprehensive proteomic survey for its binding partners in Ca 2ϩ stores would be required in future studies. From another point of view, it can be presumed that TRIC-A and TRIC-B channels may have distinct subtype-specific roles in SR Ca 2ϩ handling. To verify this possibility, Tric-b-overexpressing VSMCs, producible with the use of similar transgenic technology, will provide an important model system.
In our clinical study the TRIC-A gene polymorphisms in the Japanese population were associated with both hypertension susceptibility and sensitivity to antihypertensive medications (15). As discussed above, TRIC-A channel density in VSMCs is an important determinant of basal blood pressure at the whole animal level. Therefore, the human hypertensive risk allele is likely associated with relatively low Tric-a expression in VSMCs. Hypertension represents a major health problem with an appalling annual toll, and various antihypertensive drugs have been developed. However, hypertension still remains resistant to multiple antihypertensive medications in a considerable number of patients. Although renal sympathetic denervation has recently come into use in patients with malignant hypertension, novel target proteins are still required for creating new medication strategies (40). Our transgenic mice together with the knock-out mice suggest that TRIC-A channel openers may potentially be drugs with benefits for malignant hypertension.