Renal cell carcinoma (RCC) encompasses diverse histological and molecular subtypes and is a prevalent form of kidney cancer with significant clinical implications [1]. Clear cell RCC (ccRCC), the most common subtype, accounts for approximately 75% of all RCC cases and is characterized by its clear cytoplasm resulting from lipid and glycogen accumulation [2]. A hallmark of ccRCC is the frequent inactivation of the von Hippel-Lindau (VHL) gene, leading to increased hypoxia-inducible factor (HIF) activity and subsequent angiogenesis [3]. Additionally, ccRCC is associated with mutations in the MET oncogene [4] and chromatin-remodeling genes, such as those in the SWI/SNF complex [5]. Papillary RCC (pRCC), comprising approximately 15% of RCC cases, is further classified into type 1 and type 2, each with distinct histological features [2]. Type 1 pRCC often involves MET proto-oncogene mutations, whereas type 2 is linked to alterations in tricarboxylic acid cycle enzyme genes, such as fumarate hydratase and succinate dehydrogenase [6]. These genetic variations underlie the variable behavior and therapeutic responses observed in pRCC. Chromophobe RCC (chRCC), accounting for approximately 5% of RCC cases, is identified by its large polygonal cells with eosinophilic cytoplasm [2]. Frequent genetic alterations in chRCC involve the TP53 and PTEN genes, leading to deregulated cell cycle control and activation of the Akt/mTOR pathway [7]. The unique genetic profiles and relatively indolent nature of chRCC set it apart from other RCC subtypes. Despite the distinct genetic and clinical features defining RCC subtypes, an incomplete understanding of their molecular mechanisms underscores the need for further study toward tailored therapies.

The deregulation of second messenger Ca²⁺ signaling has been implicated in the pathogenesis of various diseases, including tumor progression [8]. Store-operated Ca²⁺ entry (SOCE), facilitated by a pore-forming Orai1 channel at the plasma membrane and endoplasmic reticulum Ca2+ sensor STIM1, serves as the primary Ca²⁺ influx mechanism in epithelial cells [9]. The roles of Orai1 and STIM1 in SOCE and cancer progression are well-documented across multiple malignancies, including breast [10-14], liver [15], lung [16], colon [17], ccRCC [18,19], ovarian [20], and gastric cancers [21]. However, few cancer studies have focused on STIM2, Orai2, or Orai3.

In breast cancer, Orai1 and Orai3 have distinct expression patterns and functional roles. Orai1 is implicated in the development and progression of estrogen receptor (ER) negative (ER(–)) breast cancers, such as MDA-MB231 [13,14], while Orai3 has shown anti-proliferative effects in ER-positive (ER(+)) MCF7 cells, indicating its potential in modulating cancer growth and metastasis [13,14]. Furthermore, although STIM2 knockout appears to have minimal effects on SOCE [22], a high STIM1 to low STIM2 ratio has been associated with poor prognosis in breast cancer [23]. This association is not consistently observed across other cancer types, such as glioblastoma [24] and prostate cancer [25].

This study aims to unveil the distinctive regulatory mechanisms and functional significance of Orai1, Orai3, and STIM1/2 in ccRCC and pRCC subtypes. Through a comparative analysis of SOCE-associated gene expression, migration, and functional Ca²⁺ influx in metastatic Caki-1 and primary pRCC Caki-2 cells, we seek to clarify the role of Orai channels in RCC pathogenesis and contribute to the development of novel therapeutic strategies targeting RCC.

TCGA RNASeq data analysis

Gene expression profiling data and clinical annotations were gathered from UCSC Xena (https://xena.ucsc.edu/). The Cancer Genome Atlas Kidney Clear Cell Carcinoma (TCGA-KIRC) project encompasses 606 patients, comprising 72 standard and 534 cancer cases. Similarly, the Kidney Papillary Cell Carcinoma (TCGA-KIRP) project includes 323 patients, with 32 normal and 291 cancer cases. Both projects provide normalized count data transformed to log (expr + 1) following genome-wide transcriptome profiling of kidney tissues using the Illumina HiSeq 2000 platform. Kaplan–Meier survival analysis assessed survival rates based on Orai1, Orai3, and STIM2 expression. The cut-off values were determined by selecting the FPKM values that maximized the survival difference between groups, as indicated by the lowest log-rank p-value.

Cell culture and small interfering RNA knockdown

Two renal cancer cell lines, Caki-1 and Caki-2, were obtained from the American Type Culture Collection (ATCC) and the Korean Cell Line Bank. The cells were cultured according to previously established protocols [18].

Oligonucleotides for non-targeting control siRNA were procured from Santa Cruz Biotechnology, and human Orai1-3 siRNA was obtained from Dharmacone (M-014998-01-0005, Orai1; M-015012-01-0005, Orai2; M-015896-00-0005, Orai3; M-011785-00-0020, STIM1; M-013166-01-0020, STIM2). Following the manufacturer's instructions, target gene silencing was induced using DharmaFect reagent (Thermo Scientific). Knockdown genes were analyzed 96 h after transfection. The wound healing assay was performed 72 h after siRNA transfection.

Wound-healing assay (In vitro scratch assay)

The wound-healing migration assay was conducted according to previously established protocols in the presence of the anti-tumor drug mitomycin C (Sigma-Aldrich) to exclude proliferative effects [18]. Cells were cultured with or without fetal bovine serum (FBS) (10%) and IGF-1 (100 nM, Sigma-Aldrich). Microscopic images were captured at 0, 24, and 48 h after drug treatment (time 0, initial time point) and analyzed using the Image J program (ver. 1.41; National Institutes of Health).

Colony formation assay

The colony formation assay was performed as previously described [18]. Caki-1 and Caki-2 cells were transfected with siRNA targeting Orai1, Orai3, or a control oligo. After 96 h, cells were trypsinized, re-seeded at 100 cells per well in 6-well plates, and incubated for 14–20 days. Colonies were stained with 1% methylene blue for visualization.

Live cell Ca²⁺ imaging

Measurement of intracellular Ca²⁺ concentration ([Ca²⁺]i) was previously described [18]. 2-aminoethoxydiphenyl borinate (2-APB, Sigma-Aldrich) was used as a pharmacological Orai3 activator and Orai1 inhibitor. Standard physiological salt solution (PSS) was used for bath perfusion, comprising (in mM) 135 NaCl, 5 KCl, 1 MgCl₂, 2 CaCl₂, 10 HEPES, and 5.5 Glucose (pH 7.4). Ca²⁺-free PSS was also perfused in a chamber composed of 135 NaCl, 5 KCl, 1 MgCl₂, 10 HEPES, 5.5 Glucose, and 1 EGTA (pH 7.4, in mM). Cyclopiazonic acid (CPA, Tocris) used at a concentration of 20 μM under 0 mM Ca²⁺ conditions. An Evident IX73 microscope (Evident Corporation) equipped with a pE-340 (CoolLED) and a 16-bit CMOS cooling camera IRIS9 sCMOS (Teledyne Photometrics) recorded fluorescence. This setup allowed rapid changes in the excitation filters (ET340x and ET380x, Chroma). Fura-2 ratiometric fluorescence was collected through the T400lp–ET510/80m beam splitter–emission cube from Chroma. Cells were exposed to 340 and 380 nm light for 200 and 100 ms, respectively, with acquisitions every 2–5 sec using VisiView software (Visitron Systems). Fura-2 Ca²⁺ imaging was analyzed using the Visiview software (Visitron Systems) or MetaFluor (Sutter Instruments).

RT- and real-time PCR

Total RNA was purified using the Hybrid-RTM total RNA purification kit (GeneAll) according to the manufacturer's instructions. Complementary DNA (cDNA) was synthesized using ReverTraAce qPCR RT Master Mix with gDNA Remover (Toyobo). RT- and real-time PCR procedures were previously described [26]. Primer sequences are in Supplementary Table 1.

Western blot

Western blotting (WB) was performed as described previously [27]. The primary antibodies used for immunoblotting were: β-actin (1:3,000, #sc-47778) and Cyclin D1 (1:200, #sc-246) from Santa Cruz Biotechnology; STIM1 (1:1,000, #S6072) from Sigma Aldrich; p-FAK (Tyr397) (1:1,000, #8556) and t-FAK (1:1,000, #13009) from Cell Signaling Technology Inc.; and STIM2 (1:500, #21192-1-AP) from ProteinTech. Bands were detected and quantified using the Vilber Lourmat (Fusion Solo 6S Edge V.070) with its analysis software.

Data analysis

Data were analyzed using the GraphPad Prism Software (version 10.1.0; GraphPad Software). Statistical comparisons between two data groups were conducted using an unpaired Student's t-test, while multiple comparisons were performed using one-way ANOVA followed by Tukey's multiple comparison tests. Significant levels were set at p-values of less than 0.05, 0.01, 0.001, and 0.0001 for single and multiple comparisons. The results were presented as mean ± S.E.M.

Statement of ethics

This study does not require research ethics approval, as both the use of TCGA data and the commercially purchased human RCC cell lines (Caki-1 and Caki-2) from ATCC do not fall under the jurisdiction of IRB and follow established ethical guidelines.

Expression profiling of SOCE-associated genes in RCC subtypes

To assess the expression of SOCE-associated genes in ccRCC and pRCC, we analyzed mRNA levels using RNA sequencing (RNAseq) data from the TCGA KIRC and KIRP databases. Initial analysis through UCSC Xena data hubs, including both normal and cancer tissues, revealed increased Orai1, Orai3, and STIM1 expression in both RCC types. While Orai2 did not show significant expression changes in ccRCC, it was upregulated in pRCC. Conversely, STIM2 expression was elevated in ccRCC compared to normal tissues but decreased in pRCC (Fig. 1A-D).

Figure 1. Expression profiling of store-operated calcium entry (SOCE)-associated genes in renal cell carcinoma (RCC) subtypes. (A, C) Heatmaps depict the expression patterns of five SOCE-related genes and all samples in clear cell RCC (ccRCC) and papillary RCC (pRCC), respectively. Cells represent the z-transformed (across samples) expression value of each gene. (B, D) Violin plots represent the differential patterns of the five genes between cancer and normal. Dotted lines, points, and whiskers represent the median value, interquartile range (IQR), and extreme values (maximum and minimum) no further than 1.5 × IQR. (E) The z-scores of log RNA-seq V2 RSEM were calculated based on all normal samples of ccRCC and pRCC expressing relative mean mRNA expression values for Orai1-3 and STIM1, 2. The z-score threshold ± 3. Box whisker plots express the minimum, median, and maximum values. (F) Relative frequency of genetic alterations of Orais and STIMs in ccRCC (“C”) and pRCC (“P”) using cBioPortal web. Alteration types, mRNA low (or Shallow deletion) and high (or Gain), were analyzed by copy number alteration (CNA) using GISTIC (Genomic Identification of Significant Targets In Cancer). Data were compared using Student’s t-test (B, D, and E). ns., not significant. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Further comparison of mRNA expression of Orai and STIM proteins between ccRCC and pRCC utilized z-scores from the TCGA PanCancer Atlas project data (ccRCC, n = 510; pRCC, n = 283) [28]. These scores standardized expression levels, with positive values indicating higher than mean expression. The average z-scores of Orai1-3 and STIM1 exhibited upregulation in both RCCs; however, STIM2 displayed low expression in pRCC (Fig. 1E).

Analysis of mRNA level alterations (high or low) and copy number alterations (CNA) using GISTIC2.0 on the cBioPortal identified consistent trends with the z-score analysis. Orai1, Orai3, and STIM1 demonstrated increased expression in both RCC types, while STIM2 was notably less expressed in pRCC (Fig. 1F). These findings highlight the high expression of Orai1 and Orai3 across RCC subtypes and the differential expression of Orai2 and STIM2 between ccRCC and pRCC.

Gender-specific prognostic impact of Orai3 expression in RCC subtypes

To investigate the correlation of Orai1 and Orai3 with cancer grades and survival probability, we conducted next-generation RNAseq analysis. Orai1 mRNA expression was significantly elevated in both ccRCC and pRCC patients, consistent with our previous findings in ccRCC (Supplementary Fig. 1A, C) [19]. Orai3 mRNA expression increased with advancing cancer stages in both RCCs, especially in pRCC across all grades (Fig. 2A, C).

Figure 2. Prognostic impact of Orai3 and STIM2 expression in renal cell carcinoma (RCC) subtypes. (A, C) FPKM for Orai3 according to normal and RCC grades (Stage I, II, III, and IV) in clear cell RCC (ccRCC) and papillary RCC (pRCC), respectively. Box whisker plots express the minimum, median, and maximum values of FPKM (expressed as log(expr + 1)), where FPKM was adjusted using a scaling factor instead of total mapped reads to remove a million units. ***p ≤ 0.001 of one-way ANOVA (normal vs. tumor stages). (B, D) Kaplan–Meler survival analysis comparing females (left) and males (right) in the low and high mRNA levels of Orai3 in ccRCC and pRCC, respectively. The cut-off value for Orai3 in ccRCC is 13.5, and in pRCC is 16.87. (E, G) Correlation analysis between gene copy number (log2-transformed values) of STIM2 and Orai3 in ccRCC (total n = 355) and pRCC (total n = 241), respectively. The linear regression line with a 95% confidence interval and the Spearman r correlation coefficient. (F, H) Kaplan–Meler survival analysis comparing males and females with high Orai3 and low STIM2 mRNA levels in ccRCC and pRCC, respectively. The cut-off value for Orai3 is 13, and for STIM2 is 4.5. p-values were determined by a log-rank test (B, D, F, and H).

Next, given the essential role of Orai3 in ER(+) breast cancer [14,29], we analyzed data by gender due to the lack of ER information in the RCC TCGA database. Elevated Orai3 expression correlated with poorer survival in both genders for ccRCC, with a more pronounced effect in females, who had half the survival duration of males (~10 years) at the 45% survival rate (Fig. 2B). Similar to our previous findings that Orai1 is associated with reduced survival in ccRCC patients [18,19], high Orai1 expression was significantly associated with decreased survival, specifically in male ccRCC patients, but not in females (Supplementary Fig. 1B). These results suggest that Orai1 may serve as a poor prognostic marker in male ccRCC patients, while Orai3 might play a more critical role in reducing survival in females, potentially due to its regulation by the ER. In pRCC, the gender-specific impact was even more significant in pRCC female patients (Fig. 2D). Similar to Orai3, high Orai1 expression was associated with significantly reduced survival in females but not in males (Supplementary Fig. 1D). This result indicates that both Orai1 and Orai3 are critical prognostic factors in pRCC, particularly for female patients. Hence, these data demonstrate that elevated Orai3 expression levels significantly influence cancer progression and grading, leading to poor survival, particularly in female patients in both ccRCC and pRCC.

Reverse-correlation between STIM2 reduction and Orai3 increase in pRCC affects gender-specific survival

To unravel the role of STIM2 in RCC, the STIM2 mRNA expression levels across different cancer grades were analyzed. The STIM2 expression in pRCC consistently exhibited lower levels than in ccRCC, with the lowest expression observed in stage III (Supplementary Fig. 2). A reverse correlation between STIM2 and Orai3 expression was observed in both RCCs (Fig. 2E, G), indicating that a decrease in STIM2 expression is associated with increased Orai3 expression in these cancers. Categorizing the distribution of Orai3 and STIM2 into four quadrants based on their expression levels revealed a significant number of pRCC patients in the quadrant with high Orai3 and low STIM2 (140 out of 241 in total pRCC and 77 out of 355 in total ccRCC) (Fig. 2E, G). Further prognosis analysis by gender within this group unveiled notable differences between ccRCC (Fig. 2F) and pRCC (Fig. 2H). Specifically, in pRCC, only females showed a significantly worse prognosis than other groups. These findings suggest that the expression of SOCE components, varying by RCC subtype and gender, may affect survival prognosis.

Expression and functional modulation of SOCE-associated genes in Caki-1 and Caki-2 cells

Functional SOCE regulation was explored in vitro using ccRCC (Caki-1) and primary pRCC (Caki-2) cell lines; both are vhl wild-type (Fig. 3A) [30,31]. mRNA expression levels of Orai and STIM proteins were assessed via RT-PCR and q-PCR, revealing higher SOCE-related gene expression in Caki-1 compared to in Caki-2 (Fig. 3B). Although Orai3 expression in Caki-1 was slightly lower than the other Orai channels, this difference was not significant (Fig. 3C), similar to the non-significant difference between STIM1 and STIM2 expression in Caki-1 (Fig. 3C). In contrast, Caki-2 showed significantly higher Orai3 expression compared to other Orai channels with markedly lower STIM2 than STIM1 expression (Fig. 3D).

Figure 3. Expression profiling of store-operated Ca2+ entry (SOCE)-associated genes in Caki-1 and Caki-2 cells, with differential modulation of Orai1 and -3 by 2-aminoethoxydiphenyl borinate (2-APB). (A) Human renal cell carcinoma (RCC) cell line models were employed for clear cell RCC (ccRCC) and papillary RCC (pRCC). (B) RT-PCR, (C, D) q-PCR for the mRNA expression levels of Orai1-3 and STIM1, 2 in Caki-1 and Caki-2, respectively. (E) Representative images displaying SOCE were recorded in cells after a one-hour pre-incubation with 2-APB (50 μM) or vehicle (control, DMSO) in Caki-1 and Caki-2, respectively. (F, G) Summary of the SOCE (F) and basal calcium (G), highlighted by the blue square in panel (E). SOCE is the difference between the peak and basal Fura-2 ratio (F340/F380), and basal Ca²⁺ is the average Fura-2 ratio from 0 to 60 sec. n = 60–96 cells per each group. (H, J) Representative SOCE images were recorded in Caki-1 and Caki-2 transfected with control oligo or targeting siRNAs against STIMs, respectively. (I, K, and L, M) Summary of the SOCE (I, K) and basal Ca²⁺ (L, M) in panels (H) and (J), respectively. n = 59–369 cells per each group. Bar graphs expressed as mean ± S.E.M. analyzed with Student’s t-test (C, D, F, and G) or one-way ANOVA (I, K, L, and M). ns., not significant; DMSO, dimethyl sulfoxide; CPA, cyclopiazonic acid. *p ≤ 0.05, ***p ≤ 0.001, ****p ≤ 0.0001. All experiments were performed three times with similar results.

Further, SOCE experiments using 2-APB revealed distinct responses in Caki-1 and Caki-2 (Fig. 3E-G). 2-APB is a pharmacological inhibitor of SOCE via the Orai1/STIM1 pathway but an activator of Orai3 [32]. Both cell lines exhibited a reduction in ER Ca²⁺ following one hour of pretreatment with 2-APB, aligning with previous studies indicating that 2-APB induces ER Ca²⁺ leakage [33]. In Caki-1, 2-APB reduced SOCE, consistent with inhibiting the Orai1/STIM1 pathway, whereas 2-APB increased SOCE in Caki-2 (Fig. 3E, F). Notably, basal Ca²⁺ influx at steady state was much higher in Caki-2 cells compared to Caki-1 cells (Fig. 3G). These findings indicate distinct Ca²⁺ influx mechanisms in Caki-1 and Caki-2 cells, with SOCE and basal Ca²⁺ leak being differently mediated by Orai1 and Orai3 channels.

After pharmacological assessing the effects of Orai1 and Orai3 channels on SOCE in Caki-1 and Caki-2 cells, we evaluated the impact of genetically knocking down these channels on SOCE and basal Ca²⁺ levels. The efficiency of siRNA-mediated knockdown of Orai1-3 was confirmed by real-time PCR, showing a reduction in mRNA levels compared to the control oligo (Supplementary Fig. 3). Notably, the siRNA knockdown of Orai1 significantly reduced SOCE in both cell lines, while the knockdown of Orai3 did not result in a noticeable decrease in SOCE (Fig. 3H-K). However, when both Orai1 and Orai3 were knocked down simultaneously, SOCE was reduced even further compared to the knockdown of Orai1 alone. This double knockdown effect indicates that while Orai3 alone does not significantly affect SOCE, it has an additive effect when knocked down with Orai1. As described earlier, Caki-2 cells exhibit higher basal Ca²⁺ influx than Caki-1 cells, with Orai3 being the dominant channel. When assessing the effects of Orai1 and Orai3 knockdown on basal Ca²⁺ influx at steady state, lowering Orai3 by siRNA effectively reduced basal Ca²⁺ entry in Caki-2 cells, not Caki-1 cells. However, this reduction was not further enhanced by the double knockdown of Orai1 and Orai3 (Fig. 3L, M). These findings suggest that Orai3 regulates basal Ca²⁺ levels independently of SOCE in Caki-2 cells, predominantly expressing Orai3. Orai3 may have a role in maintaining elevated basal Ca²⁺ influx, distinct from its contribution to SOCE.

Both Orai1 and Orai3 are critical for cell migration and proliferation in Caki-1 and Caki-2

Serum growth factors and IGF-1 regulate cell migration and proliferation in RCC [34]. Our study used a wound healing assay to compare cell migration (Fig. 4A, B). In Caki-1 cells, migration was significantly enhanced by treatment with serum (10% FBS) and IGF-1 (100 nM), as illustrated in Fig. 4A. Conversely, Caki-2 cells exhibited comparatively slower migration rates (Fig. 4B). These findings underscore the crucial role of serum- and IGF-1-stimulated RCC cell migration.

Figure 4. Regulation of migration by Orai1 and Orai3 in Caki-1 and Caki-2. (A, B) Representative photos of wound-healing assays show that either 10% fetal bovine serum (FBS) or IGF-1 (100 nM) increases the migration of Caki-1 and Caki-2, respectively. (C, D) Representative images and the number of migrated cells demonstrating the impact of Orai1-3 silencing on 10% FBS-induced cell migration in Caki-1 and Caki-2, respectively. (E) Colony formation assay showing the impact of Orai1 or Orai3 knockdown on the proliferative capability of Caki-1 and Caki-2. Data are expressed as mean ± S.E.M. and analyzed with one-way ANOVA (A–D). **p ≤ 0.01, ***p ≤ 0.001. All experiments were performed three times with similar results.

To elucidate the mechanism behind migration regulation by Orai channels, we employed siRNA in Caki-1 and Caki-2 cells under standard culture conditions (with serum) to identify the primary Orai channel involved in this process. Migration was significantly reduced in both cell lines following siRNA targeting Orai1 and Orai3 (Fig. 4C, D) and STIM1 (Supplementary Fig. 4), but not by targeting Orai2 (Fig. 4C, D) or STIM2 (Supplementary Fig. 4). To understand how Orai1 and Orai3 are involved in cellular processes related to cell migration, we examined p-FAK and its reduction after siRNA-mediated knockdown of Orai1 and Orai3, indicating that these channels in both cell lines could influence FAK signaling (Supplementary Fig. 5A, B).

Furthermore, knockdown of both Orai1 and Orai3 significantly suppressed cell proliferation (Fig. 4E). However, Cyclin D1 levels remained unchanged with Orai1 knockdown but reduced considerably after Orai3 knockdown (Supplementary Fig. 5C, D). This result suggests that Orai3 has a more direct role in cell proliferation by regulating Cyclin D1. Despite the unchanged Cyclin D1 levels with Orai1 knockdown in both Caki-1 and Caki-2 cells, cell proliferation was still reduced (Fig. 4E). This reduction is likely due to the decrease in p-FAK caused by Orai1 knockdown in both cell lines (Supplementary Fig. 5A, B), as FAK is a crucial regulator of both cell migration and proliferation [35]. Thus, the reduction in p-FAK likely contributed to the decreased cell proliferation despite stable Cyclin D1 levels following the siRNA knockdown of Orai1 in both RCC cells. These findings indicate that Ca²⁺ influx through Orai1 and Orai3 channels is essential for cell migration and proliferation.

This study comprehensively analyzed SOCE-associated genes in ccRCC and pRCC using clinical samples from the TCGA database. Our findings reveal a significant upregulation of Orai1, Orai3, and STIM1 in both RCC subtypes. Notably, pRCC exhibited higher Orai2 and lower STIM2 expression compared to ccRCC. Furthermore, elevated Orai3 expression was associated with poorer survival outcomes in both genders for ccRCC, with a particularly pronounced impact in females with pRCC. Functional studies demonstrated that 2-APB inhibits SOCE in the ccRCC cell line Caki-1 by targeting Orai1, whereas it increases SOCE in the pRCC cell line Caki-2 by activating Orai3. These findings provide valuable insights into the distinct roles of Orai1 and Orai3 in cell migration across different RCC cell lines. Additionally, our combined analysis of TCGA data and q-PCR in RCC cell lines highlighted reduced STIM2 expression in both RCCs, suggesting that its potentially compromised inhibitory function on SOCE might inadvertently promote metastasis and growth.

Building on our previous study, which proposed Orai1 as an early diagnostic marker due to its association with lower Fuhrman nuclear grade, pathologic T stage, TNM stage, and favorable prognosis [19], a recent study has shed new light on the oncogenic role of Orai3 in ccRCC [36]. Our current results reveal that Orai3 mRNA is prominently expressed among SOCE genes in both RCCs and increases with tumor grades. Drawing parallels with colorectal cancer, where a higher Orai3:Orai1 ratio correlated with poor prognosis [37], our findings suggest that increasing Orai3 expression in RCC might elevate the Orai3:Orai1 ratio, potentially impacting survival probability. Moreover, previous studies in prostate cancer have demonstrated that the formation of Orai1/Orai3 heteromeric channels, activated by arachidonic acid (AA), promotes cell proliferation via NFAT nuclear translocation and Cyclin D1 expression [38]. AA activated Orai1/3 heteromeric channel-mediated SOCE in a membrane-delimited and store-independent fashion [39].

Furthermore, an increased Orai1:Orai3 ratio favors the formation of these heteromeric channels over homomeric Orai1 channels, which are involved in apoptosis [38]. In contrast, Orai3 overexpression leads to apoptosis resistance and promotes cell proliferation [38]. In this study, we demonstrated that both Orai1 and Orai3 play pivotal roles in regulating migration and cell proliferation in both RCCs through distinct Ca²⁺ signaling pathways. Knockdown of Orai1 and Orai3 significantly reduced cell proliferation, with Orai3 knockdown notably decreasing Cyclin D1 expression (see Supplementary Fig. 5C, D), suggesting a more direct role for Orai3 in the regulation of Cyclin D1 and cell proliferation. Additionally, Orai1 knockdown decreased SOCE in both cell lines, while lowering Orai3 did not affect SOCE, indicating the formation of Orai1/Orai3 heteromeric channels. These findings support the importance of Orai1/3 in Ca²⁺-mediated signaling, particularly in AA-induced store-independent Ca²⁺ entry in prostate cancer [38]. The inhibition of cell migration through knockdown of Orai3 in both RCC subtypes further substantiates the significant role of Orai3 in RCC metastasis. Additionally, the higher expression of SOCE-related genes in Caki-1 compared to Caki-2, correlating with differences in migration, implies a potential link between increased SOCE activity and cellular aggressiveness. While considering the possible involvement of store-independent AA-regulated Ca²⁺-selective (ARC) channels through heteromeric Orai1/3 channels [39], our study underscores the imperative need for advancing study on Orai3 for prospective therapeutic developments in RCC.

The FDA approval of Belzutifan (MK-6482) in 2021 for treating RCC in patients with VHL disease, acting as a hypoxia-inducible factor 2-α (HIF-2α) inhibitor [40], highlights the critical role of HIF pathways in RCC [41]. HIF-1/2α can activate Orai1 and Orai3 in the hypoxic cancer microenvironment [42]. Even though HIF-1/2α levels can increase under hypoxic conditions even in wild-type VHL human RCC cell lines, Caki-1 and Caki-2 [43], our findings, coupled with previous reports, suggest that Orai1 and Orai3 may be upregulated in RCC through mechanisms independent of VHL or due to VHL mutations, potentially driven by increased levels of HIF-1/2α. This hypothesis warrants further investigation to confirm the link and to determine whether Belzutifan could also target Orai1 and Orai3.

Notably, the gender-specific effects of high Orai3 levels on prognosis, especially in pRCC females, align with findings from breast cancer studies indicating that miRNA-18a, regulated by the ER, can upregulate Orai3 [11]. Given that increased miRNA-18a expression correlates with poorer overall survival in ccRCC [44], this suggests that ER(+) miRNA-18a-mediated Orai3 upregulation may significantly influence prognosis in pRCC, meriting further investigation.

The observed expression pattern of elevated Orai3 and reduced STIM2 in pRCC compared to ccRCC is associated with a survival trend mirroring the impact of Orai3 expression, where females exhibit significantly worse outcomes. This pattern underscores the emerging important role of STIM2 across various cancer types, emphasizing its influence on tumor growth, metastasis, and migration [12,25,45-47]. While most studies report overexpression of STIM2 in malignancies such as breast [12,46], cervical [47], and prostate cancer [25], its depletion in colorectal cancer is linked to increased tumor growth and metastasis [45]. This variation suggests a complex role for STIM2 that warrants further investigation, particularly its effects on growth and metastasis in pRCC. Additionally, the higher expression of Orai2 in pRCC compared to ccRCC patients points to an area requiring additional research to understand its implications fully. Overall, our study proposes the importance of SOCE as a crucial pathway governing intracellular Ca²⁺ levels in RCCs, uncovering the pivotal roles of overexpressed Orai3 and Orai1 in the metastasis of RCC. Specifically, the augmented expression of Orai3 alongside reduced STIM2 levels in pRCC provides new perspectives on molecular subtypes, guiding potential diagnostic and therapeutic approaches. This investigation illuminates the potential of targeting Orai channels and STIM proteins as a novel strategy to hinder the progression and metastasis of RCC.

This study was supported by the Medical Research Center Program (2017R1A5A2015369) and the Basic Science Research Program (NRF-2019R1A2C1084880, 2022R1A2C2011079, and 2022R1C1C2009853) through the National Research Foundation of Korea (NRF). This research was supported by Korea Basic Science Institute (National Research Facilities and Equipment Center) grant funded by the Ministry of Science and ICT (No. RS-2024-00404529). This work was supported by the Soonchunhyang University Research Fund (No. 20231294). This research was also supported by the BK21 FOUR program through the NRF under the Ministry of Education, supporting KHH and JYO.

We thank Seoyun Jun, Ho Gyeong Lee, Rahyun Won, Hayeon Oh, So Jeong Park, and Sunhee Park at Soonchunhyang University for the revised manuscript's technical support in cell culture and Western blot experiments. We also express our gratitude to Boyeong (Kathy) An at UC Berkeley for critical proofreading and comments on the manuscript.

The authors declare no conflicts of interest.

Supplementary data including one table and five figures can be found with this article online at https://doi.org/10.4196/kjpp.24.126