Genetic analysis of the regulation of the voltage-gated calcium channel homolog Cch1 by the γ subunit homolog Ecm7 and cortical ER protein Scs2 in yeast

The yeast Cch1/Mid1 Ca2+ channel is equivalent to animal voltage-gated Ca2+ channels and activated in cells incubated in low Ca2+ medium. We herein investigated the third subunit, Ecm7, under the same cell culture conditions. The deletion of ECM7 slightly lowered Ca2+ influx activity in the CNB1+ background, in which calcineurin potentially dephosphorylates Cch1, but markedly lowered this activity in the cnb1Δ background. The deletion of the C-terminal cytoplasmic region of Ecm7 also reduced Ca2+ influx activity. We identified a novel Cch1-interacting protein, Scs2, which is known as a cortical endoplasmic reticulum membrane protein. The deletion of SCS2 did not affect Ca2+ influx activity when calcineurin was inhibited by FK506, but enhanced this activity by 35% when the enzyme was not inhibited. However, this enhancement was canceled by the deletion of ECM7. These results suggest that Cch1/Mid1 is regulated differentially by Ecm7 and Scs2 in a manner that is dependent on the phosphorylation status of Cch1.


Introduction
Ca 2+ is a second messenger that induces changes in many fundamental cellular events, including enzyme activation, secretion, and gene expression [1]. Therefore, the permeation of Ca 2+ through cellular membranes is crucial and strictly regulated by a number of transporters and channels, one of which belongs to the voltage-gated Ca 2+ channel (VGCC) superfamily. PLOS  catalytic subunit Cna1 or Cna2 and the regulatory subunit Cnb1. When the calcineurin inhibitor, FK506, was added to yeast cells exposed to α-factor or endoplasmic reticulum (ER) stress, Ca 2+ influx through the Cch1/Mid1 channel was found to markedly increase [7,9]. In addition, it is shown that Cch1 is phosphorylated in vivo in cells treated with either FK506 or the ERstressor tunicamycin or both and is dephosphorylated by calcineurin in vitro [14]. Thus, the phosphorylation status could be important for the activity of this channel. Scs2 is an integral ER membrane protein homologous to VAP (vesicle-associated membrane protein-associated protein) and connects the ER to the plasma membrane to regulate the level of phosphatidylinositol 4-phosphate (PI4P) in the plasma membrane by controlling the access of another integral ER membrane protein, Sac1 phosphatase, to its substrate PI4P [15,16]. At the ER/plasma membrane contact sites, interactions between Scs2 and the oxysterol-binding homology protein Osh3 activates ER-localized Sac1. Scs2 also interacts with the septin Shs1 to create the ER diffusion barrier, by which diffusion of integral ER proteins is restricted between the ER in the mother cell and the ER in the daughter cell [17,18]. Based on a genome-wide screen for mutants with altered Ca 2+ uptake, Scs2 has been identified as a member of proteins whose defect results in a deficiency in a low affinity Ca 2+ influx system (LACS) that is stimulated by α-factor in YPD medium [8].
In the present study, we show genetically that Ecm7 and Scs2, which we identified as a new Cch1-binding protein, differentially regulate the Ca 2+ influx activity of the Cch1/Mid1 channel in a manner that depends potentially on the phosphorylation status of this channel or its regulators.

Ca 2+ accumulation assay
The accumulation of Ca 2+ was measured according to a previously described method [4]. Briefly, cells grown in SD.Ca100 medium at 30˚C to the exponential phase were incubated for 2 h with 45 CaCl 2 (185 kBq/ml; 1.85 kBq/nmol; PerkinElmer, Waltham, MA, USA) and 6 μM α-factor. Aliquots (duplicate) of the cultures were collected, filtered through Millipore filters (type HA; 0.45 μm, Merck, Tokyo, Japan) that had been presoaked in 5 mM CaCl 2 , and washed five times with the same solution. The radioactivity retained on the filters was counted with the scintillation cocktail, Emulsifier-Safe (PerkinElmer) in a liquid scintillation counter (Beckman Coulter LS 6500, Tokyo, Japan).

Viability assay
Cells were grown in SD.Ca100 medium and exposed to 6 μM α-factor as described above.
After an 8-h exposure, the viability of cells was assessed according to the methylene blue method described by Iida et al. [21]. This method is based on the ability of viable cells to reduce methylene blue to the colorless leukomethylene blue. Nonviable cells cannot reduce the dye and stain blue. Therefore, methylene blue-negative cells are viable and methylene bluepositive cells are nonviable. Viability was expressed as the percentage of the number of methylene blue-negative cells in the total number of cells. The number of methylene blue-negative cells is comparable to colony-forming units [21]. pBCT-CCH1H P TDH3 -CCH1H-T ADH1 LEU2 ARS1 CEN4 amp r ColE1-ori-rop [22] pBCT-CCH1H-EGFP P TDH3 -CCH1H-EGFP-T ADH1 LEU2 ARS1 CEN4 amp r ColE1-ori-rop [22] pBCY-BTC Western blotting Western blotting was performed according to a previously described method [4]. To determine the content of Cch1 in cells lacking ECM7 and/or CNB1, crude extracts were subjected to SDS-PAGE followed by Western blotting. For co-immunoprecipitation experiments, proteins immunoprecipitated with an anti-Cch1 antibody was analyzed by SDS-PAGE and Western blotting. The antibodies used were a rabbit polyclonal anti-Cch1 antibody [22], mouse monoclonal anti-FLAG M2 antibody (Sigma-Aldrich Japan, Cat. No. F3165, Tokyo, Japan), mouse monoclonal anti-Pma1 antibody (GenScript Japan, Cat. No. A00101, Tokyo, Japan) and a rabbit polyclonal anti-enolase [23].

Fluorescence microscopy
The subcellular localization of Cch1-EGFP in cells treated for 2 h with 6 μM α-factor was visualized using a confocal laser scanning microscope (LSM 780, Carl Zeiss Microscopy, Shinjuku, Tokyo, Japan) using excitation wavelengths of 488 nm.

Co-immunoprecipitation
Cells (total of~1 x 10 8 cells) growing in SD medium were harvested, washed once with Milli-Q water, resuspended in BB-buffer (10 mM Tris-HCl [pH 7.5], 300 mM sorbitol, 100 mM NaCl, 5 mM MgCl 2 , 1 mM PMSF, and 1 tablet/10 ml of protease inhibitors [Roche Diagnostics, Basel, Switzerland]), and then lysed with glass beads by vortexing. Crude extracts were ultracentrifuged at 55,000 rpm (137,000 x g) at 4˚C for 30 min. The pellet (membrane fraction) was washed once with BB-buffer by ultracentrifugation and resuspended in 1 ml of IP-buffer (50 mM Tris-HCl [pH 8.0], 1.0% Triton X-100, 150 mM NaCl, 2 mM EDTA, 1 mM PMSF, and 1 tablet/10 ml of protease inhibitors) and left to stand at 4˚C for 60 min to be solubilized. The solution was ultracentrifuged at 55,000 rpm at 4˚C for 30 min and the supernatant was mixed and incubated with 30 μl of nProtein A Sepharose 4 Fast Flow (GE Healthcare, CT, USA) at 4˚C for 60 min with rotation. The mixture was centrifuged at 12,000 rpm for 1 min and the supernatant was mixed with rabbit anti-Cch1 (3.4 μl; 2.5 μg) and incubated at 4˚C for 60 min with rotation. nProtein A Sepharose 4 Fast Flow (50 μl) was supplemented to the mixture, which was incubated again at 4˚C for 60 min with rotation. The mixture was centrifuged at 12,000 rpm for 1 min, and the pellet was washed twice with centrifugation in IP-buffer and then washed once with wash-buffer (50 mM Tris-HCl [pH 8.0]). The final pellet was dissolved in SDS-sample buffer (50 mM Tris-HCl [pH 6.8], 4% SDS, 10% glycerol, 0.05% bromophenol blue, and 5% ß-mercaptoethanol). After heating at 95˚C for 5 min followed by centrifugation at 15,000 rpm at 25˚C, the supernatant was subjected to SDS-PAGE and a Western blot analysis or stocked at -20˚C until used. Western blots were used to detect Cch1, FLAG-tagged Scs2, and Pma1.

Statistical analysis
The significance of differences was evaluated using an unpaired Student's t-test, with a maximum p value of < 0.05 being required for significance.

Ecm7 is required for HACS in calcineurin-deficient cells
In order to study the role of Ecm7, we employed low Ca 2+ medium containing 100 μM CaCl 2 , named SD.Ca100, in which HACS becomes active in α-factor-treated cells [4,7]. We incubated the ecm7Δ mutant as well as its control, the cch1Δ mutant and their parental ECM7 + CCH1 + strain (designated here as the wild type) with α-factor for 2 h for Ca 2+ accumulation assays and 8 h for viability assays. The results obtained showed that only slightly less Ca 2+ accumulated in the ecm7Δ mutant (by 17%) than in the wild-type strain, while Ca 2+ accumulation was reduced by 77% in the cch1Δ mutant (Fig 2A). The ecm7Δ cch1Δ double mutant quantitatively showed the same accumulation of Ca 2+ as the cch1Δ mutant. Viability assays showed similar results between the ecm7Δ and cch1Δ mutants and wild-type strain (Fig 2B). These results suggest that Ecm7 only negligibly contributes to HACS in cells exposed to α-factor, even in low Ca 2+ medium.
We then examined the contribution of Ecm7 to HACS using the cnb1Δ mutation, which produces no calcineurin activity, because a previous study reported that α-factor-induced Ca 2+ accumulation in the ecm7Δ mutant was approximately 50% less than that in the wild-type strain, even in YPD medium, when the calcineurin inhibitor, FK506 was added [8]. Our results showed that, in SD.Ca100 medium, α-factor-induced Ca 2+ accumulation was 7.4-fold higher in the cnb1Δ mutant than in the wild-type strain and also that the high level of accumulation in the cnb1Δ mutant was markedly decreased to one-fourth when ECM7 was deleted (Fig 3A). Viability assays showed that the viability of the cnb1Δ mutant was further lowered in the ecm7Δ cnb1Δ mutant (Fig 3B). Since HACS or its regulator is not dephosphorylated by calcineurin in the cnb1Δ mutant, the above results suggest that Ecm7 is selectively required for a phosphorylated form of HACS or its regulator. Alternatively, it is possible that a phosphorylated form of Ecm7 is important for the activity of HACS or its regulator.
In the Western blot analysis and confocal laser scanning microscopy, we found that the deletion of ECM7 and/or CNB1 did not affect the content of the Cch1 protein (Fig 4) or the subcellular localization of Cch1-EGFP expressed from a strong TDH3 promoter (S1 Fig). It should be noted that expression from a strong promoter could sometimes lead to mislocalization of a protein of interest. In fact, while the localization of Cch1-Myc expressed from the CCH1 promoter is reported to be localized exclusively in the plasma membrane [24], Cch1-EGFP expressed from the TDH3 promoter was localized in not only the plasma membrane but also the ER membrane (S1 Fig). Nevertheless, we would like to point out that the fluorescence images of Cch1-EGFP were very similar between cnb1Δ, ecm7Δ cnb1Δ and wildtype cells.

The cytoplasmic C-terminal region of Ecm7 is partially necessary
The cytoplasmic C-termini of human γ subunit isoforms are highly divergent, except for the terminal seven amino acid residues, which are conserved [25]. Similarly, the amino acid alignment of Ecm7 and its orthologs from several species of the genus Saccharomyces revealed that regions near the cytoplasmic C-termini were divergent, whereas the terminal 37 amino acid residues were well conserved. In addition, mass spectrometry-based proteomics identified phosphorylation sites at seven Ser residues in the C-terminal region ranging from Ser 360 to the C terminus (Phe 448 ) [26] (Fig 1).
In order to examine the importance of these regions and some of the phosphorylation sites, we performed truncation analysis. The C-terminally truncated forms of Ecm7, Ecm7 1-322 lacking all of the seven phosphorylation sites and Ecm7 1-412 lacking three C-terminal phosphorylation sites (Fig 1), were expressed in the ecm7Δ cnb1Δ mutant and Ca 2+ accumulation was measured (Fig 5). After the subtraction of background Ca 2+ accumulation in cells carrying an empty vector from that in cells expressing each mutant Ecm7 (see the dotted line in Fig 5), we found that the Ca 2+ accumulation activities of Ecm7 1-322 and Ecm7 1-412 were significantly reduced, but were still maintained at approximately one-fourth that of the wild-type Ecm7 (Fig 5). This result indicates that the C-terminally truncated regions, which contain the seven phosphorylation sites, are required for the full activity of Ecm7, but are not necessarily essential. In order to examine the validity of this suggestion, we constructed the following three mutant proteins: (1) Ecm7 ΔS429-S432 , a mutant Ecm7 protein with the deletion of the region (Ser 429 to Ser 432 ) containing the two phosphorylation sites and (2) Ecm7 1-432 , a mutant Ecm7 protein with the deletion of the C-terminal region ranging from Glu 433 to the C terminus. We then measured their Ca 2+ accumulation activities. The results obtained showed that the two mutant Ecm7 proteins exhibited normal Ca 2+ accumulation activities. Since Ecm7 1-412 showed reduced activity, the region between 413 and 428 amino acid residues appears to be important for the activity of Ecm7, although its contribution is restrictive.

Scs2 positively regulates HACS
In the search for membrane proteins that potentially interact with HACS, we screened candidates with the split-ubiquitin membrane-based yeast two-hybrid system [27,28], using Cch1 as bait and S. cerevisiae cDNA libraries as prey. We obtained 12 candidates, among which Scs2 was the most frequently detected. Details on this screening will be published elsewhere; we herein focused on Scs2, particularly its genetic study in terms of HACS regulation because this protein had been reported as a member of LACS [8]. We confirmed that N-terminally tagged FLAG-Scs2 co-immunoprecipitated with Cch1 when anti-Cch1 antibodies were used as the primary antibody followed by Western blotting with the anti-FLAG antibody (Fig 6A-6C).
Scs2 is a type II integral membrane protein with a single C-terminal transmembrane segment that localizes at the ER membrane, and is a homolog of VAP [15]. In order to investigate the role of Scs2 in Ca 2+ influx through HACS, the scs2Δ mutant was constructed and examined for Ca 2+ accumulation 2 h after exposure to α-factor in SD.Ca100 medium. The results obtained showed that 35% more Ca 2+ accumulated in the scs2Δ mutant than in the wild-type strain (Fig 7A). This enhancement was not observed in the scs2Δ ecm7Δ mutant, suggesting that Scs2 negatively regulates HACS activity and that Ecm7 has a positive effect on the function of a system regulated negatively by Scs2 although the molecular basis of this effect is unknown. In contrast, when FK506 was added to the medium, Ca 2+ accumulation was not enhanced in the scs2Δ mutant (Fig 7B). Therefore, the role of Scs2 in the regulation of HACS activity appears to potentially depend on the phosphorylation status of HACS or its regulator.
SCS2 has a paralog, SCS22 [29]. The scs22Δ mutant showed the wild-type levels of Ca 2+ accumulation and cell viability after exposure to α-factor and the scs2Δ scs22Δ double mutant showed the same phenotypes as the scs2Δ mutant (data not shown), suggesting that Scs22 is not involved in the regulation of the Cch1/Mid1 channel under our assay conditions.

Discussion
In the present study, we identified two regulators of HACS, Ecm7 and Scs2, under the experimental conditions that activate HACS, i.e. under low extracellular Ca 2+ conditions: the former is a positive regulator, while the latter is a negative regulator. We showed that Ecm7, a yeast homolog of animal VGCC γ subunits, was hardly necessary for α-factor-induced HACS activity when calcineurin was functional in cells incubated in low Ca 2+ medium. However, when the CNB1 gene encoding a subunit of this protein phosphatase was deleted or this enzyme was inhibited by FK506, Ecm7 switched to a strong positive regulator. Although Ecm7 possesses seven phosphorylation sites in the intracellular C-terminal region [26], these were suggested to be nonessential for the function of Ecm7 (Fig 5; see the next paragraph). Thus, a phosphorylated form of Ecm7 would not be required for HACS activation in calcineurin-deficient cells. Therefore, Ecm7 appears to selectively activate the phosphorylated form of HACS or its regulator. One of the Ecm7 targets may be Cch1 because the entire molecule of Mid1 is localized extracellularly and, thus, cannot be dephosphorylated by cytoplasmic calcineurin (Iida K et al., manuscript in preparation). The calcineurin deficiency-dependent activation of HACS was also reported for S. cerevisiae cells exposed to tunicamycin, an ER stressor; however, the involvement of Ecm7 was negligible, and this may have been because YPD medium, which is rich in nutrients and Ca 2+ , was used to cultivate cells [8]. A previous study demonstrated that HACS hardly functions in this rich medium [7].
A truncation analysis showed that the large cytoplasmic C-terminal region of Ecm7 was partially responsible for the full Ca 2+ influx activity of HACS, suggesting that the seven phosphorylation sites present in this region are less important. Particularly, two C-terminally p < 0.05 (WT vs. cnb1Δ); *5, p < 0.05 (cch1Δ vs. cnb1Δ ecm7Δ). Data are the mean ± SD of three independent experiments. https://doi.org/10.1371/journal.pone.0181436.g003 truncated forms, Ecm7 1-322 and Ecm7 1-412 , had the same activity each other (Fig 5), suggesting that the amino acid residue 323-412 segment containing four phosphorylation sites (Ser 360 , Ser 381 , Ser 387 , and Ser 394 ) is not involved in Ecm7 function. In addition, Ecm7 Δ429-432 lacking two phosphorylation sites (Ser 429 and Ser 432 ) and Ecm7 1-432 lacking the phosphorylation site Ser 435 did not lose the activity at all (Fig 5). Therefore, it is suggested that all of the seven phosphorylation sites do not contribute to the full activity of Ecm7 at least under the experimental conditions we employed. Among the eight members of mammalian γ subunits (γ 1 to γ 8 ), only one member, γ 2 , was found to be phosphorylated in the cytoplasmic C-terminal region [30,31]. Since Ecm7 has four transmembrane segments that are highly conserved among the VGCC γ subunit family, the four transmembrane segments with a large extracellular loop between segments 1 and 2 may play a positive regulatory role.
Another regulator, Scs2, was newly discovered by the screening of Cch1-interacting membrane proteins with the split-ubiquitin membrane-based yeast two-hybrid system. The deletion of the SCS2 gene enhanced α-factor-induced HACS activity by 35% and this enhancement was canceled in the absence of Ecm7 or when cells were incubated with the calcineurin inhibitor FK506. Since Scs2 is shown to be phosphorylated at Ser 106 , Ser 153 , Thr 200 , Ser 201 , Ser 203 , and  Thr 204 [26,32], it is possible that HACS activity is dually regulated by Ecm7 and Scs2 in a calcineurin-dependent manner. The following hypothesis could be possible: when calcineurin is non-functional due to a low cytosolic Ca 2+ concentration, Ecm7 may activate a phosphorylated form of HACS or its regulator. However, this activation may not be applied to its dephosphorylated form. When calcineurin is functional due to a high cytosolic Ca 2+ concentration, Scs2 may repress the activity of a dephosphorylated form of HACS or its regulator, but may not affect its phosphorylated form. This hypothesis warrants further studies looking at HACS phosphorylation and dephosphorylation.
Scs2 is localized to the cortical ER membrane, at which it acts as an ER-to-plasma membrane-tethering protein [18]. Cch1 is located in the plasma membrane [24], while Mid1 is localized to the ER and plasma membranes [33][34][35][36]. Since the Cch1/Mid1 channel or HACS is activated by ER stress [9], it is tempting to speculate that Scs2 transduces an ER signal to the Cch1/Mid1 channel in order to regulate Ca 2+ influx through the plasma membrane.
In conclusion, we demonstrated that the VGCC γ subunit homolog Ecm7 functioned as a strong positive regulator of Cch1 in calcineurin-deficient cnb1Δ cells exposed to α-factor in low-Ca 2+ medium. However, this positive regulation is weak in CNB1 + cells. We identified the cortical ER resident protein Scs2 as a novel interactor of Cch1 and demonstrated that Scs2 negatively regulated Cch1 in calcineurin-functional cells.
Supporting information S1 Fig. The deletion of ECM7 and CNB1 does not affect the subcellular localization of Cch1-EGFP. Cch1-EGFP images of cells treated for 2 h with 6 μM α-factor were obtained by confocal scanning fluorescence microscopy. Cch1-EGFP was expressed from the TDH3 promoter and appeared to be localized in the ER and plasma membranes in the cnb1Δ and ecm7Δ cnb1Δ mutants as well as in the wild-type strain. (TIFF)