Downregulation of Ca2+-Activated Cl− Channel TMEM16A Mediated by Angiotensin II in Cirrhotic Portal Hypertensive Mice

Portal hypertension is defined as an increased pressure in the portal venous system and occurs as a major complication in chronic liver diseases. The pathological mechanism underlying the pathogenesis and development of portal hypertension has been extensively investigated. Vascular tone of portal vein smooth muscles (PVSMs) is regulated by the activities of several ion channels, including Ca2+-activated Cl− (ClCa) channels. TMEM16A is mainly responsible for ClCa channel conductance in vascular smooth muscle cells, including portal vein smooth muscle cells (PVSMCs). In the present study, the functional roles of TMEM16A channels were examined using two experimental portal hypertensive models, bile duct ligation (BDL) mice with cirrhotic portal hypertension and partial portal vein ligation (PPVL) mice with non-cirrhotic portal hypertension. Expression analyses revealed that the expression of TMEM16A was downregulated in BDL-PVSMs, but not in PPVL-PVSMs. Whole-cell ClCa currents were smaller in BDL-PVSMCs than in sham- and PPVL-PVSMCs. The amplitude of spontaneous contractions was smaller and the frequency was higher in BDL-PVSMs than in sham- and PPVL-PVSMs. Spontaneous contractions sensitive to a specific inhibitor of TMEM16A channels, T16Ainh-A01, were reduced in BDL-PVSMs. Furthermore, in normal PVSMs, the downregulation of TMEM16A expression was mimicked by the exposure to angiotensin II, but not to bilirubin. This study suggests that the activity of ClCa channels is attenuated by the downregulation of TMEM16A expression in PVSMCs associated with cirrhotic portal hypertension, which is partly mediated by increased angiotensin II in cirrhosis.


INTRODUCTION
Portal hypertension is characterized by a pathological increase in pressure in the portal venous system (Bandali et al., 2017). It is a common and severe complication of chronic liver diseases. The most frequent cause is cirrhosis. Hepatic fibrosis due to cirrhosis causes an increase in hepatic vascular tone and morphological changes occurring during vascular remodeling. Increased hepatic resistance leads to portal hypertension and subsequent disturbances in the splanchnic and systemic circulations, resulting in serious symptoms such as gastroesophageal varices, variceal hemorrhage, splenomegaly, ascites, and hepatic encephalopathy (Bosch et al., 2009;García-Pagán et al., 2012). The normal portal venous pressure is 5-10 mmHg (Bandali et al., 2017) and the hepatic venous pressure gradient (HVPG), which is a normal pressure gradient between the portal vein and the inferior vena cava, is typically 1-5 mmHg (Bosch et al., 2009). Currently, HVPG measurement is the gold-standard method to evaluate the presence and severity of portal hypertension. Clinically significant portal hypertension is defined as an increase in HVPG to ≥10 mmHg (Bosch et al., 2009). The molecular mechanism underlying the pathogenesis and development of portal hypertension has been discussed from many functional aspects.
Vascular tone and pressure are mainly regulated by the cytosolic Ca 2+ concentration ([Ca 2+ ] cyt ) under the physiological and pathological conditions. Following vasoconstrictor stimulation and membrane depolarization, an increase in [Ca 2+ ] cyt is caused by two major pathways: 1) Ca 2+ influx through the voltage-dependent Ca 2+ channels (VDCCs) and receptor-operated Ca 2+ (ROC) channels in the plasma membrane, and 2) Ca 2+ release from the intracellular Ca 2+ store sites such as the sarcoplasmic reticulum (SR) and mitochondria (Chalmers et al., 2007). Increased [Ca 2+ ] cyt in vascular smooth muscles facilitates excitation-contraction coupling. In parallel, it stimulates Ca 2+ -activated ion channels in the plasma membrane and regulates vascular tone. For example, the activation of Ca 2+ -activated K + (K Ca ) channels causes a membrane hyperpolarization and reduces the activity of VDCCs (Latorre et al., 2017). On the other hand, the opening of Ca 2+ -activated Cl − (Cl Ca ) channels causes membrane depolarization and increases the VDCC activity (Kitamura and Yamazaki, 2001;Bulley and Jaggar, 2014). Therefore, the balance between K Ca and Cl Ca channel conductance determines the membrane excitability, myogenic tone, contraction/relaxation, and pressure in vascular smooth muscles.
Cl Ca channels are involved in many physiological processes such as smooth muscle contraction, epithelial secretion, cell volume regulation, neuronal signaling, and sensory transduction (Pedemonte and Galietta, 2014). In vascular smooth muscles, Cl Ca channels play significant roles in the regulation of membrane excitability, myogenic tone, and muscle contraction (Pedemonte and Galietta, 2014). So far, TMEM16A has been identified as a molecular entity responsible for Cl Ca channels in vascular smooth muscles (Pedemonte and Galietta, 2014), including portal veins (Davis et al., 2010;Ohshiro et al., 2014a;Ohshiro et al., 2014b). Under physiological conditions, the activation of TMEM16A Cl Ca channels shifts the resting membrane potential to the positive direction and promotes Ca 2+ influx through VDCCs, resulting in an increase in vascular tone and pressure (Kitamura and Yamazaki, 2001;Bulley and Jaggar, 2014). On the other hand, there is limited information on the pathological role of TMEM16A Cl Ca channels in cardiovascular diseases. A recent report showed that TMEM16A channels regulate the proliferation of portal vein smooth muscle cells (PVSMCs) in portal hypertension (Zeng et al., 2018).
The present study was performed to elucidate the involvement of TMEM16A Cl Ca channels in the pathological mechanism of portal hypertension, using expression analyses, electrophysiological recordings, and contractility measurements. Two different portal hypertensive animal models, bile duct ligation (BDL) mice as a portal hypertensive model resulting from cirrhosis and partial portal vein ligation (PPVL) mice as an idiopathic portal hypertensive model without hepatic dysfunction (Abraldes et al., 2006;Geerts et al., 2008;Bosch and Iwakiri, 2018), were prepared to analyze the functional expression of TMEM16A channels in PVSMCs. The expression and function of TMEM16A channels were downregulated in BDL-PVSMCs compared with sham-and PPVL-PVSMCs, which was partly mediated by angiotensin II, the serum concentration of which increases in patients with cirrhosis.

Ethical Approval
All experiments were approved by the Ethics Committee of Nagoya City University (H30-P-1) and were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the Japanese Pharmacological Society.

BDL Operation
Male mice (C57BL/6, 8-week-old; Japan SLC, Hamamatsu, Japan) were anesthetized under isoflurane (Wako Pure Chemical Industries, Osaka, Japan) inhalation. After the midline abdominal incision, the common bile duct was occluded with a double ligature using a suture (nylon 8-0; Alfresa Pharma, Osaka, Japan). The first ligature was made below the junction of the hepatic ducts and the second ligature was made above the entrance of the pancreatic ducts. The portion of the bile duct between the two ligatures was cut to avoid re-permeabilization. In sham-operated mice, the abdominal cavity was opened and the common bile duct was isolated, but there was no ligation. Experiments were performed at 4-5 weeks after sham or BDL operation (Abraldes et al., 2006;Geerts et al., 2008;Bosch and Iwakiri, 2018).

PPVL Operation
Under isoflurane inhalation anesthesia, a midline abdominal incision was made and the portal vein was separated from the surrounding tissues. A ligature (nylon 6-0) was tied around both portal vein and a blunt-tipped 27-gauge needle. Subsequent removal of the needle yielded calibrated stenosis of the portal vein. In sham-operated mice, the abdominal cavity was opened and the portal vein was separated, but there was no ligation. Experiments were performed at 2-3 weeks after sham or PPVL operation (Abraldes et al., 2006;Geerts et al., 2008;Bosch and Iwakiri, 2018).

Contractility Measurements
Dissected portal veins were placed in aerated Krebs' solution. A small piece of portal vein (5 mm long) was prepared and placed in an organ bath containing aerated Krebs' solution at 37°C. One end of the segment was pinned to a rubber plate at the bottom of the bath and the other end was connected to an isometric transducer. The strips were stretched to 1 mN of tension. The Krebs' solution had an ionic composition of 112 mM NaCl, 4.7 mM KCl, 2.2 mM CaCl 2 , 1.2 mM MgCl 2 , 25 mM NaHCO 3 , 1.2 mM KH 2 PO 4 , and 14 mM glucose. The pH was adjusted to 7.4 by gassing with a mixture of 95% O 2 and 5% CO 2 .

Statistical Analysis
Pooled data are shown as the mean ± S.E. The significance of differences between two groups was assessed by the Student's t-test using BellCurve for Excel software (version 3.10; Social Survey Research Information, Tokyo, Japan). A p value of <0.05 was considered to be significant.

Expression of TMEM16 Family in PVSMs From Portal Hypertensive Mice
The expression profiles of the TMEM16 family in PVSMs from portal hypertensive mice were analyzed using real-time PCR and immunocytochemical staining. Among the TMEM16 family, TMEM16A mRNA was expressed (0.090 ± 0.006 of GAPDH, n = 4) and TMEM16F and TMEM16K mRNAs were also detected (0.072 ± 0.002 and 0.066 ± 0.010, respectively, n = 4; Figure 2A) in sham-PVSMs, as reported previously (Ohshiro et al., 2014a). In BDL-PVSMs, the mRNA expression of TMEM16A was clearly reduced (0.051 ± 0.005, n = 6, p = 0.001). TMEM16F mRNA was slightly attenuated in BDL-PVSMs (0.059 ± 0.004, n = 6, p = 0.042), whereas TMEM16K mRNA was identical to that in sham-PVSMs (0.051 ± 0.002, n = 6). On the other hand, there were no significant differences in the expression of TMEM16 genes between sham-and PPVL-PVSMs (n = 4 to 6; Figure 2B). Because the reduced expression of TMEM16A proteins in BDL-PVSMs was shown by Western blotting (Zeng et al., 2018), immunocytochemical staining was performed for protein expression analysis in the present study. Immunocytochemical staining can detect the localization of the protein in cells. Based on immunocytochemical staining, TMEM16A-coding proteins were abundantly expressed at the plasma membrane of sham-PVSMCs (n = 38), but it was clearly attenuated in BDL-PVSMCs (n = 39; Figure 2C). On the other hand, the distribution of TMEM16A proteins in PPVL-PVSMCs (n = 19) was almost the same as that in sham-PVSMCs (n = 32; Figure 2D). Taken together, TMEM16A expression was downregulated in BDL-PVSMCs at the mRNA and protein levels compared with that in sham-and PPVL-PVSMCs.

Cl Ca Currents in PVSMCs From Portal Hypertensive Mice
Cl Ca currents were measured in freshly isolated murine PVSMCs using K + deficient and Cl − rich solutions containing TEA-Cl under whole-cell voltage-clamp conditions (see Materials and Methods). In addition, to activate Cl Ca currents, the Ca 2+ concentration in the pipette solution was fixed to pCa 6.0 (1 μM). Single PVSMCs were stimulated from the holding potential of −60 mV to selected test potentials (from −100 to +100 mV by 20-mV increments, every 15 s) for 1 s, and then repolarized to −80 mV for 500 ms. The cell capacitance of sham-PVSMCs was 34.9 ± 2.3 pF (n = 12). The depolarizing pulse elicited time-dependent outward currents at membrane potentials positive to +40 mV in sham-PVSMCs (16.1 ± 2.3 pA/pF at +100 mV, n = 12; Figures 3A,B). Upon repolarization, characteristic inward tail currents were recorded (−12.0 ± 1.8 pA/pF at −80 mV following +100 mV depolarization, n = 12). The current-voltage relationship demonstrated that the reversal potential was approximately 0 mV (n = 12). Our previous report showed that these currents were inhibited by 100 μM niflumic acid (a conventional Cl Ca channel blocker) and 10-30 μM T16A inh -A01 (a specific TMEM16A channel blocker) (Ohshiro et al., 2014a).
Next, Cl Ca currents were measured in BDL-PVSMCs. The cell capacitance was similar to that in sham-PVSMCs (40.5 ± 4.6 pF, n = 7). The outward currents were smaller in BDL-PVSMCs than in sham-PVSMCs (9.8 ± 0.9 pA/pF, n = 7, p = 0.022; Figures  3A,B). The inward currents were also reduced in BDL-PVSMCs (−6.9 ± 0.8 pA/pF, n = 7, p = 0.023). Similar to sham-PVSMCs, the current-voltage relationship demonstrated that the reversal potential was approximately 0 mV (n = 7). On the other hand, the amplitude of outward and inward Cl Ca currents in PPVL-PVSMCs was similar to that in sham-PVSMCs (n = 5 to 7; Figures 3C,D). This suggested that, in BDL-PVSMCs, the activity of Cl Ca currents was reduced by the downregulation of TMEM16A expression.

Effects of Bilirubin on the Expression and Activity of TMEM16A Channels in Murine PVSMCs
As the plasma bilirubin level is increased in chronic liver diseases (Geerts et al., 2008), the effects of bilirubin on the expression and activity of TMEM16A channels were examined in murine PVSMCs. Under pCa 6.0 conditions in the pipette solution, whole-cell Cl Ca currents were not affected by the application of 3 μM bilirubin to PVSMCs (26.2 ± 6.6 pA/pF at +100 mV, n = 7, p = 0.419 versus control, 19.4 ± 4.8 pA/pF, n = 7; Figures  6A,B). The expression of TMEM16A mRNA was unaltered by treatment with 30 μM bilirubin for 24 h in PVSMs (0.97 ± 0.02 of GAPDH, n = 9, p = 0.918 versus vehicle control, 1.00 ± 0.18, n = 9; Figure 6C). Therefore, bilirubin did not alter the expression or activity of TMEM16A channels in PVSMCs.

Effects of Angiotensin II on Expression and Activity of TMEM16A Channels in Murine PVSMCs
As the plasma angiotensin II level was reported to be increased in chronic liver diseases (Grace et al., 2012), the effects of angiotensin II on the expression and activity of TMEM16A channels were examined in murine PVSMCs. Under pCa 6.0 conditions in the pipette solution, whole-cell Cl Ca currents were not affected by the application of 1 μM angiotensin II to PVSMCs (15.6 ± 4.7 pA/pF at +100 mV, n = 4, p = 0.946 versus control, 16.1 ± 4.4 pA/pF, n = 4; Figures 7A,B). Real-time PCR analysis revealed that the expression of TMEM16A mRNA was downregulated by the treatment with 1 μM angiotensin II for 24 h in PVSMs (0.68 ± 0.07 of GAPDH, n = 9, p = 0.032 versus vehicle control, 1.00 ± 0.11, n = 12; Figure 7C). This suggested that the expression of TMEM16A was downregulated by angiotensin II in PVSMCs.

DISCUSSION
Portal hypertension is a frequent major complication of cirrhosis. As the development of portal hypertension causes serious clinical symptoms such as gastroesophageal varices, variceal hemorrhage, splenomegaly, ascites, and hepatic encephalopathy, the underlying mechanism has been extensively investigated (Bosch et al., 2009;García-Pagán et al., 2012). In the present study, the pathological role of TMEM16A Cl Ca channels in portal hypertension was focused on using cirrhotic BDL and noncirrhotic PPVL mice. We found that 1) the expression of TMEM16A was downregulated in BDL-PVSMCs, but not in PPVL-PVSMCs, 2) the activity of Cl Ca currents was reduced in BDL-PVSMCs, 3) the involvement of TMEM16A channels in spontaneous contractions was attenuated in BDL-PVSMs, and 4) the downregulation of TMEM16A expression in BDL-PVSMCs was partly induced by angiotensin II. TMEM16A channels are ubiquitously expressed in epithelia, interstitial cells of Cajal (ICC), neurons, endothelium, and smooth muscles to regulate their functions (Pedemonte and Galietta, 2014). In vascular smooth muscles, TMEM16A Cl Ca channels function in the regulation of membrane excitability, myogenic tone, and muscle contraction (Pedemonte and Galietta, 2014). In the present study, expression analyses of the TMEM16 family revealed that TMEM16A, TMEM16F, and TMEM16K genes were abundant in PVSMs, as reported previously (Ohshiro et al., 2014a). The most important finding was that the expression level of TMEM16A was lower in BDL-PVSMs (43%) than in sham-PVSMs, whereas the expression level in PPVL-PVSMCs was almost the same as that in sham-PVSMs. TMEM16F transcripts slightly decreased in BDL-PVSMs (18%). Although TMEM16F is necessary for the Ca 2+ -dependent phospholipid scramblase, it is hypothesized to form Cl − channels. However, the activation of TMEM16F channels requires a higher concentration of [Ca 2+ ] cyt (EC 50 of >10 μM) than TMEM16A channels (EC 50 of 0.3 μM) (Pedemonte and Galietta, 2014). Therefore, TMEM16F channels may not function as Cl Ca channels under the physiological conditions in PVSMCs. TMEM16K is not considered an ion channel (Pedemonte and Galietta, 2014). Taken together, the present study mainly focused on the pathophysiological functions of TMEM16A Cl Ca channels in PVSMCs from cirrhotic portal hypertensive mice.
The amplitudes of peak and tail Cl Ca currents were smaller in BDL-PVSMCs than in sham-PVSMCs. Our previous report showed that the Cl Ca currents were inhibited by niflumic acid and T16A inh -A01 (Ohshiro et al., 2014a). The spontaneous contractions in sham-PVSMs were also blocked by T16A inh -A01. T16A inh -A01 may be somewhat less selectivity, but this suggests that TMEM16A Cl Ca channel conductance is involved in the formation of spontaneous contractions, which play a significant role in the blood flow from mesenteric vascular beds to the liver. The spontaneous contractions in PVSMs depend on the activities of VDCCs and Cl Ca channels, and are inhibited by niflumic acid (Kirkup et al., 1996;Saleh and Greenwood, 2005). In the present study, there was no significant difference in the expression level of VDCC subunits (α 1C , β 2 , and β 3 ) between sham-and BDL-PVSMs (data not shown). On the other hand, the sensitivity to T16A inh -A01 on spontaneous contractions in BDL-PVSMs was markedly lower (sham, 93% decrease versus BDL, 52% decrease in amplitude; and sham, 79% decrease versus BDL, 30% decrease in frequency). Taken together, these decreases in Cl Ca channel components of currents and spontaneous contractions are consistent with the downregulation of TMEM16A expression in BDL-PVSMCs. In vascular smooth muscles, a decrease in Cl Ca conductance shifts the resting membrane potential to the hyperpolarizing direction. This blocks Ca 2+ influx through VDCCs, resulting in the attenuation of cell excitability in the form of Ca 2+ -dependent action potentials (Kitamura and Yamazaki, 2001;Bulley and Jaggar, 2014). Therefore, this may be a mechanism for protecting against increased portal pressure and related symptoms. However, the spontaneous contractions in sham-PVSMs were mostly inhibited by T16A inh -A01, whereas these in BDL-PVSMs were often observed even in the presence of T16A inh -A01, suggesting that the expression of other Cl Ca channels was upregulated as a compensatory mechanism. Further experiments are necessary to elucidate the molecular switching of Cl Ca channels in BDL-PVSMCs.
It has been reported that BDL-induced cirrhosis alters the expression level of ion channels such as voltage-dependent Na + channels (Lee et al., 2016), epithelial Na + channels (Kim et al., 2006), large-conductance K Ca (BK Ca ) channels (Yuan et al., 2016;Jadeja et al., 2017), ATP-sensitive K + channels (Yuan et al., 2016), transient receptor potential canonical subfamily (TRPC) channels (Nedungadi and Cunningham, 2014;Jadeja et al., 2017), and transient receptor potential vanilloid subfamily (TRPV) channels (Nedungadi et al., 2012;Belghiti et al., 2013;Hong-Qian Wang et al., 2019). In addition, we noted the downregulation of TMEM16A (and also TMEM16F) expression in cirrhotic BDL mice, but not in non-cirrhotic PPVL mice. As both mouse models exhibited portal hypertrophy, the downregulation of TMEM16A expression in BDL-PVSMCs was not due to an increase in portal venous pressure. Rather, it may be caused by liver failure because the difference between BDL and PPVL mice is the presence and absence of fibrosis of the liver, respectively. In addition to PVSMCs, the downregulation of TMEM16A expression was observed in aortic smooth muscles from BDL mice (unpublished observation). Taken together, the downregulation may be mediated by endogenous compounds associated with cirrhosis, but not by portal hypertension.
Some endogenous substances are released into the blood from damaged liver cells. We focused on bilirubin (Geerts et al., 2008) and angiotensin II (Grace et al., 2012), which are known to be increased in the blood in cirrhosis. Bilirubin regulates the expression and activity of voltage-dependent Na + channels (Shi et al., 2019), epithelial Na + channels (Xue-Ping , acid-sensing ion channels (ASICs) (Lai et al., 2020), and VDCCs (Liang et al., 2017;Albanna et al., 2019). Although we examined the effects of bilirubin on the expression and activity of TMEM16A channels, it had no effects in PVSMCs. On the other hand, angiotensin II down-/up-regulates vascular ion channels such as VDCCs (Kharade et al., 2013), voltagedependent K + channels (Barrese et al., 2018), BK Ca channels (Nieves-Cintrón et al., 2007), and transient receptor potential melastatin subfamily (TRPM) channels (He et al., 2005;   ). In addition, angiotensin II downregulates TMEM16A functions in rat basilar smooth muscle cells from 2K2C renohypertensive rats (Wang et al., 2012) and TMEM16A expression in human aortic smooth muscle cells (Zhang et al., 2015), whereas it upregulates TMEM16A functions in aorta and mesenteric arterial smooth muscle cells from spontaneous hypertensive rats (SHR) (Wang et al., 2015). Our study demonstrated that angiotensin II, the plasma concentration of which increases in patients with cirrhotic portal hypertension from 20 to 200 pg/ml (Grace et al., 2012), downregulates the TMEM16A expression in murine PVSMCs. As angiotensin II did not affect Cl Ca currents, it did not interact directly with TMEM16A channels in PVSMCs. Thus, the underlying mechanism is likely to through indirect action such as genomic effects. The expression level of TMEM16A channels is altered in cardiovascular diseases such as systemic hypertension (Wang et al., 2012;Wang et al., 2015), pulmonary hypertension (Forrest et al., 2012;Sun et al., 2012;Papp et al., 2019), and diabetic gastroparesis (Mazzone et al., 2011). In addition, cirrhosis induces the downregulation of TMEM16A expression in vascular smooth muscle cells. We demonstrated that TMEM16A Cl Ca channels play a role in vascular tone and contraction in PVSMCs from the cirrhotic portal hypertensive model. More recently, similar downregulation was observed in the same preparation, but it was involved in cell proliferation (Zeng et al., 2018). The activity of TMEM16A channels was suggested to be involved in cell survival and death, in addition to smooth muscle contraction (Pedemonte and Galietta, 2014). Although TMEM16A channels can activate Ca 2+ /calmodulin-dependent protein kinase II (CaMKII) and mitogen-activated protein kinase (MAPK) pathways (Wang et al., 2017), the signaling mechanism of TMEM16A channels for the proliferation and migration of smooth muscle cells remains unclear. Taken together, TMEM16A Cl Ca channels play an important role in both the regulation of vascular tone and remodeling in cirrhotic portal hypertension. On the other hand, the pathological mechanism of non-cirrhotic portal hypertension remains unclear in the present study. Further studies using PPVL mice are necessary because there are no specific drugs for the treatment of idiopathic portal hypertension currently.
In conclusion, the expression and activity of TMEM16A channels were downregulated in PVSMCs from BDL mice associated with cirrhosis. This downregulation is partly mediated by increased angiotensin II in cirrhosis. Currently, non-selective β blockers, vasopressin analogues, and somatostatin analogues are commonly used to prevent portal hypertensive bleeding in cirrhosis. Novel therapeutic targets have been discussed as treatment options for portal hypertension; however, these strategies are far from satisfactory (Schwabl and Laleman, 2017). Our study revealed the involvement of TMEM16A Cl Ca channels in the pathological mechanism underlying cirrhotic portal hypertension and provides a novel therapeutic target.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material.

ETHICS STATEMENT
The animal study was reviewed and approved by the Ethics Committee of Nagoya City University (H30-P-1).

AUTHOR CONTRIBUTIONS
RK and HY contributed to conception and design of experiments; RK, NF, AD, and NK performed experiments; RK, NF, AD, NK, YS, and YI analyzed data; RK and HY wrote manuscript.

FUNDING
This study was supported by a Grant-in-Aid for Scientific Research on Innovative Areas (17H05537 to HY), Grantin-Aid for Promotion of Joint International Research (Fostering Joint International Research (B)) (18KK0218 to YS, YI, and HY), Grant-in-Aid for Scientific Research (B) (19H03381 to YS), and Grants-in-Aid for Scientific Research (C) (25460104, 16K08278, and 19K07125 to HY) from the Japan Society for the Promotion of Science.