Targeting ER Stress with Saikosaponin A to Overcome Resistance under Radiation in Gastric Cancer Cells

Saikosaponin A is a triterpene saponin and a potentially bioactive compound derived from Bupleurum falcatum L. However, the molecular mechanisms and effects of saikosaponin A in gastric cancer remain unknown. In the present study, I evaluated the effects of saikosaponin A on cell death and endoplasmic reticulum stress via calcium and reactive oxygen species release. Targeting reactive oxygen species with diphenyleneiodonium and N-acetylcysteine inhibited cell death and protein kinase RNA-like ER kinase signaling pathway by down-regulating Nox4 and inducing glucose-regulated protein 78 exosomes. Furthermore, saikosaponin A caused a synergistic inhibitory effect of the epithelial mesenchymal transition phenomenon, indicating the reversible phenotype modulation by epithelial cells under radiation exposure in radiation-resistant gastric cancer cells. These results suggest that saikosaponin A-mediated calcium and reactive oxygen species-induced endoplasmic reticulum stress overcome radio-resistance and induce cell death under radiation in gastric cancer cells. Therefore, saikosaponin A in combination with radiation may be a potential strategy for gastric cancer therapy.


Introduction
Each year, more than one million cases of gastric cancer (GC) are diagnosed worldwide, and GC is the fifth most common cancer type around the world [1]. Anti-cancer therapy aims still to kill or slow the growth of tumor cells with surgical therapy, and many therapeutic strategies, including chemotherapy, radiotherapy, cryotherapy, and immunotherapy, have been developed and challenged [2]. Chemotherapy and radiotherapy are the two most representative types of cancer treatment; however, chemo-and radio-resistance and the reduction in efficacy in GC remain serious obstacles [3,4]. Moreover, both chemotherapy and radiotherapy cause many side effects, such as the death of normal cells, fatigue, weight change, pain, gastrointestinal problems, and hair loss [5]. Therefore, herbal or natural medicine, specifically the use of natural or herbal compounds extracted from plants, may be a potential strategy to reduce the side effects of these therapies [6,7]. Natural medicines extracted from plants, food, and fruit have low toxicity and are normally safe, designating them potential candidates for enhancing the efficacy of anticancer activity [8]. Some traditional herbal medicines are known as potential anti-cancer drugs such as saikosaponin A (SSA), saikosaponin B (SSB), and saikosaponin D (SSD) extracted from Bupleurum falcatum L. Unlike other saikosaponin compounds, SSA and SSD are the major active compounds that exert anti-inflammatory, anti-viral, and anti-cancer effects [9]. In addition, SSA and SSD regulate cell cycle arrest, anti-proliferation, and apoptotic cell death in various cancer cell types, including colon, liver, breast, pancreatic, and lung [10][11][12][13][14]. SSA exerts antioxidant and anti-tumor effects through the induction of apoptotic cell death in liver and breast cancer [15]. SSA in combination with cisplatin treatment induces apoptotic cell death by inducing reactive oxygen species (ROS) release and inhibits the proliferation and tumorigenesis [16]. These findings suggest that SSA is a potential anti-cancer drug. The endoplasmic reticulum (ER) is a vital intracellular organelle in the secretory pathway and exerts many functions, including homeostasis, protein synthesis, calcium (Ca 2+ ) storage, transport, degradation, folding, modification, trafficking, and lipid metabolism [17]. ER stress triggers the unfolded protein response (UPR) and is classified by three ER stress sensors, including protein kinase RNA (PKR)-like ER kinase (PERK), inositol-requiring protein-1α (IRE1α), and activating transcription factor 6 (ATF6) [18]. Under physiological stress, PERK is phosphorylated, and this signaling is prolonged by the phosphorylation of eukaryotic translation initiation factor 2 (eIF2α). The phosphorylation of eIF2α induces the activation of activating transcriptional factor 4 (ATF4) and transcriptional factor C/EBP homologous protein (CHOP) [19]. Recent reports suggest that the PERK-eIF2α-ATF4-CHOP axis overcomes radiation resistance and induces cell death in cancer cells [20].
Therefore, the present study aimed to evaluate the efficacy of SSA treatment in overcoming radio-resistance via the PERK-ATF4-CHOP pathway and the induction of apoptotic cell death under radiation exposure in GC cells.

SSA Treatment Induces Apoptosis in GC Cells
To determine the cytotoxic effect of SSA on various GC cells, I performed a cell viability assay. As shown in Figure 1A,B, SSA decreased the cell viability in AGS, SNU-638, SNU-216, MKN-74, MKN-7, and NCI-N87 cells in a dose-and time-dependent manner (0, 1, 2.5, 5, 10, and 20 µM, 24 h; 0, 8, 16, and 24 h, 10 µM). To further evaluate the effects of SSA in vivo, a GC xenograft mouse model was constructed with AGS cells. Mice in the 5 mg/kg and 10 mg/kg SSA groups showed lower tumor volumes than the control group and exhibited dose-dependent efficacy ( Figure 1C). Body weight did not significantly differ among the groups ( Figure 1D).

SSA Mediates Apoptotic Cell Death through the ER Stress Pathway in GC Cells
Emerging studies indicate that ER stress leads to cell death and overcomes radioresistance in cancer [21]. To confirm whether SSA induces ER stress signaling in GC cells, I measured SSA-triggered Ca 2+ release. When intracellular Ca 2+ release assay was performed, SSA induced intracellular Ca 2+ release in a concentration-and time-dependent manner ( Figure 3A). To affirm the levels of SSA-treated ER stress markers, such as GRP78, p-PERK, PERK, p-eIF2α, eIF2α, ATF4, and CHOP in a time-dependent manner, Western blot analyses were performed. SSA treatment induced the upregulation of GRP78, p-PERK, p-eIF2α, ATF4, and CHOP levels ( Figure 3B). Emerging reports suggest that GRP78 is located in the exosome fractions in cancer cells and is up-regulated during the ER stress response [22]. To study the function of GRP78 on SSA-mediated exosome release, I treated AGS and MKN-74 cells with SSA and extracted the secreted exosomes from their culture supernatants. SSA increased the levels of the exosome marker CD63 in a time-dependent manner, and GRP78 was notably upregulated in exosomes extracted from SSA-treated cell culture medium compared to that from the control ( Figure 3C). Our results suggest that GRP78 + exosomes are an important for SSA-triggered ER stress and cell death. I also assessed the effect of SSA in combination with the ER stress inducer thapsigargin (TG) on GC cells. Combination experiments showed that both TG and SSA reduced cell viability and enhanced LDH release, caspase-3 activity, and intracellular Ca 2+ release compared with the control (Figure 3D-G). In addition, SSA, in combination with TG, increased GRP78, p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 levels ( Figure 3H).
PERK-eIF2α-ATF4-CHOP axis overcomes radiation resistance and induces cell death in cancer cells [20]. Therefore, the present study aimed to evaluate the efficacy of SSA treatment in overcoming radio-resistance via the PERK-ATF4-CHOP pathway and the induction of apoptotic cell death under radiation exposure in GC cells.

The Inhibition of ER Stress Signaling Blocks SSA-Induced Apoptotic Cell Death
To confirm whether GRP78 from the cell lysate and exosomes regulates SSA-mediated apoptotic cell death in GC cells, a GRP78 knockdown experiment was performed. GRP78 knockdown triggered the increase in cell viability and the reduction in LDH release in SSA-treated GC cells compared to control cells ( Figure 4A,B). Compared to controls with SSA treatment, GRP78 knockdown with SSA treatment inhibited GRP78, p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 ( Figure 4C). The PERK-CHOP axis is a potential factor for cell death in cancer and chemoresistance [23]. I carried out a PERK knockdown study using a PERK-specific siRNA in SSA-treated GC cells. These cells were transfected with PERK siRNA (30 nM, 24 h) and treated with SSA. In the control cells, SSA decreased cell viability and enhanced LDH release, whereas SSA-treated PERK knockdown cells showed higher cell viability and lower LDH release than control cells ( Figure 4D,E). Western blot analyses indicated that control cells had increased p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 with SSA treatment, whereas PERK-knockdown cells showed decreased p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 with SSA treatment ( Figure 4F). In a CHOP-knockdown experiment, SSA treatment reduced cell viability and increased LDH release in control cells, whereas SSA-treated CHOP knockdown cells showed higher cell viability and lower LDH release than control cells ( Figure 4G). Western blot analyses showed that control cells had increased levels of CHOP and cleaved caspase-3 with SSA treatment, whereas PERK-knockdown cells had decreased levels of CHOP and cleaved caspase-3 with SSA treatment ( Figure 4H).
culture medium compared to that from the control ( Figure 3C). Our results suggest that GRP78 + exosomes are an important for SSA-triggered ER stress and cell death. I also assessed the effect of SSA in combination with the ER stress inducer thapsigargin (TG) on GC cells. Combination experiments showed that both TG and SSA reduced cell viability and enhanced LDH release, caspase-3 activity, and intracellular Ca 2+ release compared with the control ( Figure 3D-G). In addition, SSA, in combination with TG, increased GRP78, p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 levels ( Figure 3H).

SSA Induces ER Stress and Cell Death via ROS and Ca 2+ Generation in GC Cells
Many reports suggest that the accumulation of reactive oxygen species (ROS) induces ER stress and cell death in cancer cells [24]. To identify whether SSA induces ROS production, I performed a cellular ROS assay after staining with the fluorescent probe, DCFDA. After a 16-hour treatment with SSA, DCFDA fluorescence was enhanced by 3.5-4-fold in AGS and MKN-74 cells when compared to control cells ( Figure 5A). To assess whether SSA-induced ROS release was related to ER stress and cell death in AGS and MKN-74 cells, ROS inhibitors, including NAC and DPI, were co-treated with SSA. As shown in Figure 5B-F, both NAC and DPI blocked SSA's ability to decrease cell viability and enhance LDH cytotoxicity, ROS release, and Ca 2+ production. Western blot analyses revealed that NAC and DPI blocked the SSA-mediated increase in p-PERK, p-eIF2α, ATF4, and CHOP in AGS and MKN-74 cells. Emerging reports suggest that NADPH oxidase Nox4 is a powerful regulator for ROS release, and the Nox4-ROS pathway regulates cell survival and cell death in cancer [25]. To identify whether SSA modulates ROS and ER stress by activating Nox4, I performed knockdown experiments using Nox4-specific siRNA in AGS and MKN-74 cells, cell viability and LDH assays, and Western blot analyses. My results indicate that SSA did not reduce cell viability, and it did not enhance LDH cytotoxicity in Nox4 knockdown cells when compared to control cells ( Figure 5G,H). After quantifying the Nox4 knockdown in AGS and MKN-74 cells, Western blot analyses showed that SSA treatment up-regulated Nox4 and CHOP expression in control cells. However, in Nox4 knockdown cells, SSA treatment decreased the expression of Nox4 and CHOP ( Figure 5I).
down cells showed higher cell viability and lower LDH release than control cells ( Figure  4D,E). Western blot analyses indicated that control cells had increased p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 with SSA treatment, whereas PERK-knockdown cells showed decreased p-PERK, p-eIF2α, ATF4, CHOP, and cleaved caspase-3 with SSA treatment ( Figure 4F). In a CHOP-knockdown experiment, SSA treatment reduced cell viability and increased LDH release in control cells, whereas SSA-treated CHOP knockdown cells showed higher cell viability and lower LDH release than control cells ( Figure 4G). Western blot analyses showed that control cells had increased levels of CHOP and cleaved caspase-3 with SSA treatment, whereas PERK-knockdown cells had decreased levels of CHOP and cleaved caspase-3 with SSA treatment ( Figure 4H).

SSA Treatment Overcomes Radio-Resistance in GC Cells
To investigate whether SSA has an anti-cancer effect in radio-resistant GC cells, I performed colony formation assays, cell viability assays, and Western blot analyses. These results indicated that SSA treatment induces a synergistic reduction in the surviving fraction at the indicated radiation exposure (2, 4, or 6 Gray (Gy)) in GC (AGS and MKN-74) and radio-resistant GC cells (AGSR and MKN-74R) when compared to control cells ( Figure 6A). Surviving fraction levels were lower in SSA-treated AGS and MKN-74 cells compared to SSA-treated AGSR and MKN-74R cells ( Figure 6A). In AGS and MKN-74 cells, SSA reduced cell viability and increased LDH cytotoxicity, and SSA/2 Gy declined cell viability and enhanced LDH cytotoxicity even further, while 2 Gy alone had no effect ( Figure 6B,C). In AGSR and MKN-74R cells, SSA treatment decreased cell viability and increased LDH cytotoxicity, and SSA/2 Gy reduced cell viability and increased LDH cytotoxicity even further, while 2 Gy alone also had no effect ( Figure 6B,C). In addition, SSA or SSA/2 Gy triggered greater cell viability and lower LDH cytotoxicity in AGSR and MKN-74R cells compared to AGS and MKN-74 cells ( Figure 6B,C). I next constructed a luciferase reporter driven by the E-cadherin promoter. A luciferase assay indicated increased activity of the E-cadherin promoter in 2 Gy and SSA-exposed AGSR and MKN74R cells ( Figure 6D). To examine whether SSA treatment affects the EMT phenomenon in radiation-exposed GC and radio-resistant GC cells, I treated AGS, MKN-74, AGSR, and MKN-74R cells with a combination of SSA and 2 Gy radiation. Western blot analyses showed that SSA and 2Gy/SSA treatments blocked N-cadherin, vimentin, Snail, and Slug and enhanced Ecadherin in AGSR cells, whereas they remained largely unchanged in AGS cells ( Figure 6E). Real-time RT-PCR indicated that 2Gy/SSA down-regulated Slug and Snail mRNA levels and up-regulated E-cadherin in AGSR and MKN-74R cells compared to control cells ( Figure 6F). These findings showed that SSA induced cell death by inhibiting the EMT phenomenon in AGSR and MKN-74R cells.
AGS and MKN-74 cells when compared to control cells ( Figure 5A). To assess whether SSA-induced ROS release was related to ER stress and cell death in AGS and MKN-74 cells, ROS inhibitors, including NAC and DPI, were co-treated with SSA. As shown in Figure 5B-F, both NAC and DPI blocked SSA's ability to decrease cell viability and enhance LDH cytotoxicity, ROS release, and Ca 2+ production. Western blot analyses revealed that NAC and DPI blocked the SSA-mediated increase in p-PERK, p-eIF2α, ATF4, and CHOP in AGS and MKN-74 cells. Emerging reports suggest that NADPH oxidase Nox4 is a powerful regulator for ROS release, and the Nox4-ROS pathway regulates cell survival and cell death in cancer [25]. To identify whether SSA modulates ROS and ER stress by activating Nox4, I performed knockdown experiments using Nox4-specific siRNA in AGS and MKN-74 cells, cell viability and LDH assays, and Western blot analyses. My results indicate that SSA did not reduce cell viability, and it did not enhance LDH cytotoxicity in Nox4 knockdown cells when compared to control cells ( Figure 5G,H). After quantifying the Nox4 knockdown in AGS and MKN-74 cells, Western blot analyses showed that SSA treatment up-regulated Nox4 and CHOP expression in control cells. However, in Nox4 knockdown cells, SSA treatment decreased the expression of Nox4 and CHOP ( Figure 5I).

Targeting PERK Blocks SSA-Induced Cell Death under Radiation Exposure in GC Cells
To further determine whether PERK signaling regulates SSA-induced cell death under radiation exposure in radio-resistant GC cells, I established stable PERK-knockdown cells. I treated these cells with 2 Gy, SSA, and SSA/2 Gy and then performed cell viability, LDH cytotoxicity, caspase-3 activity, ROS release, Ca 2+ release, and Western blot assays. These results indicated that SSA decreased cell viability and increased LDH release, caspase-3 activity, ROS release, and Ca 2+ release in control cells, whereas 2 Gy had no effect. SSA/2 Gy decreased cell viability and increased LDH release, caspase-3 activity, ROS release, and Ca 2+ release more effectively than SSA alone; conversely, in stable PERKknockdown AGSR cells, SSA, 2 Gy, and SSA/2 Gy had no effect on cell viability, LDH cytotoxicity, caspase-3 activity, ROS release, and Ca 2+ release ( Figure 7A-E). In control cells, Western blot analyses showed that SSA treatment up-regulated the expression of p-PERK, CHOP, and cleaved caspase-3, and SSA/2 Gy up-regulated the expression of p-PERK, CHOP, and cleaved caspase-3 ( Figure 7F). In PERK-knockdown AGSR cells, SSA and SSA/2 Gy did not affect the expression of p-PERK, CHOP, and cleaved caspase-3 (Figure 7F). Together, our findings suggest that PERK activation by SSA regulates ER stressmediated cell death under radiation exposure in GC cells and radio-resistant GC cells. , and were exposed to 2 Gy and SSA (10 µM, 24 hours). The luciferase activity was normalized to the activity of Renilla luciferase; *, p < 0.05. (E,F) Western blot analyses for Ecadherin, N-cadherin, vimentin, Slug, and Snail were performed and then E-cadherin, N-cadherin, and vimentin mRNA levels were measured using real-time RT-PCR.

Targeting PERK Blocks SSA-Induced Cell Death under Radiation Exposure in GC Cells
To further determine whether PERK signaling regulates SSA-induced cell death under radiation exposure in radio-resistant GC cells, I established stable PERK-knockdown cells. I treated these cells with 2 Gy, SSA, and SSA/2 Gy and then performed cell viability, LDH cytotoxicity, caspase-3 activity, ROS release, Ca 2+ release, and Western blot assays. These results indicated that SSA decreased cell viability and increased LDH release, caspase-3 activity, ROS release, and Ca 2+ release in control cells, whereas 2 Gy had no effect. SSA/2 Gy decreased cell viability and increased LDH release, caspase-3 activity, ROS release, and Ca 2+ release more effectively than SSA alone; conversely, in stable PERK-knockdown AGSR cells, SSA, 2 Gy, and SSA/2 Gy had no effect on cell viability, LDH cytotoxicity, caspase-3 activity, ROS release, and Ca 2+ release ( Figure 7A-E). In control cells, Western blot analyses showed that SSA treatment up-regulated the expression of p-PERK, CHOP, and cleaved caspase-3, and SSA/2 Gy up-regulated the expression of p-PERK, CHOP, and cleaved caspase-3 ( Figure 7F). In PERK-knockdown AGSR cells, SSA and SSA/2 Gy did not affect the expression of p-PERK, CHOP, and cleaved caspase-3 ( Figure 7F). Together, our findings suggest that PERK activation by SSA regulates ER stress-mediated cell death under radiation exposure in GC cells and radio-resistant GC cells.

Figure 7.
Targeting PERK inhibits apoptotic cell death in radio-resistant GC cells subjected to SSA in combination with 2 Gy. (A-F) PERK shRNA stable AGSR cell lines were established after AGSR cells were transfected with PERK shRNA. These cells were exposed to SSA (10 μΜ), 2 Gy, and SSA/2 Gy, and cell viability, cell cytotoxicity, caspase-3 activity, intracellular ROS release, and intracellular Ca 2+ release were determined along with Western blot analyses examining the levels of p-PERK, PERK, CHOP, and cleaved caspase-3; *, p < 0.05. β-actin was used as a protein loading control.

Discussion
In the present study, I identified apoptotic cell death mechanism via ROS and ER stress in SSA-treated GC cells. In addition, SSA/2 Gy exerted synergistic anti-tumor effects by activating ER stress and caspase-dependent cell death and by inhibiting the EMT phenomenon in radio-resistant GC cells.
ER stress via the UPR signaling pathway mediates cellular protein homeostasis and induces cell survival of damaged and stressed cells [26]. However, prolonged and excessive ER stress induces cellular dysfunction and apoptotic cell death by releasing ROS. Overwhelming the UPR and ER stress disturbs the general roles of the ER, such as homeostasis [27][28][29]. The main UPR-PERK signaling pathway (PERK-eIF2α-ATF4-CHOP axis) phosphorylates eIF2α under ER stress [30]. The phosphorylation of eIF2α induces translation and the nucleus translocation of ATF4, which then mediates apoptotic cell death via the activation of CHOP [31]. Recent studies reported that excessive ER stress induces caspase-dependent apoptotic cell death via Ca 2+ release from the ER [32]. Increased Ca 2+ production causes the accumulation of unfolded proteins that trigger ER stress by disturbing ER function [33]. Apoptotic cell death by Ca 2+ release regulates Bcl-2, and the suppression of Bcl-2 induces apoptotic cell death by regulating the Ca 2+ -dependent serinethreonine phosphatase calcineurin [34]. The interaction between the Ca 2+ -binding chaperone, calcineurin, and the UPR sensor, PERK, is increased by releasing Ca 2+ , calcineurin knockdown inhibits ER stress-induced cell death [35].
Bioactive compound saikosponins, such as SSA, SSB, SSC, and SSD extracted from Radix Bupleuri, have various in vitro and in vivo effects, including anti-cancer, anti-inflammatory, and anti-viral functions [36]. Many studies reported that SSD exerts powerful anti-tumor effects and induces apoptotic cell death via diverse molecular mechanisms Targeting PERK inhibits apoptotic cell death in radio-resistant GC cells subjected to SSA in combination with 2 Gy. (A-F) PERK shRNA stable AGSR cell lines were established after AGSR cells were transfected with PERK shRNA. These cells were exposed to SSA (10 µM), 2 Gy, and SSA/2 Gy, and cell viability, cell cytotoxicity, caspase-3 activity, intracellular ROS release, and intracellular Ca 2+ release were determined along with Western blot analyses examining the levels of p-PERK, PERK, CHOP, and cleaved caspase-3; *, p < 0.05. β-actin was used as a protein loading control.

Discussion
In the present study, I identified apoptotic cell death mechanism via ROS and ER stress in SSA-treated GC cells. In addition, SSA/2 Gy exerted synergistic anti-tumor effects by activating ER stress and caspase-dependent cell death and by inhibiting the EMT phenomenon in radio-resistant GC cells.
ER stress via the UPR signaling pathway mediates cellular protein homeostasis and induces cell survival of damaged and stressed cells [26]. However, prolonged and excessive ER stress induces cellular dysfunction and apoptotic cell death by releasing ROS. Overwhelming the UPR and ER stress disturbs the general roles of the ER, such as homeostasis [27][28][29]. The main UPR-PERK signaling pathway (PERK-eIF2α-ATF4-CHOP axis) phosphorylates eIF2α under ER stress [30]. The phosphorylation of eIF2α induces translation and the nucleus translocation of ATF4, which then mediates apoptotic cell death via the activation of CHOP [31]. Recent studies reported that excessive ER stress induces caspasedependent apoptotic cell death via Ca 2+ release from the ER [32]. Increased Ca 2+ production causes the accumulation of unfolded proteins that trigger ER stress by disturbing ER function [33]. Apoptotic cell death by Ca 2+ release regulates Bcl-2, and the suppression of Bcl-2 induces apoptotic cell death by regulating the Ca 2+ -dependent serine-threonine phosphatase calcineurin [34]. The interaction between the Ca 2+ -binding chaperone, calcineurin, and the UPR sensor, PERK, is increased by releasing Ca 2+ , calcineurin knockdown inhibits ER stress-induced cell death [35].
Bioactive compound saikosponins, such as SSA, SSB, SSC, and SSD extracted from Radix Bupleuri, have various in vitro and in vivo effects, including anti-cancer, antiinflammatory, and anti-viral functions [36]. Many studies reported that SSD exerts powerful anti-tumor effects and induces apoptotic cell death via diverse molecular mecha-nisms [37,38]. SSD induces apoptotic cell death through cell cycle arrest at G-1 phase in prostate cancer, thyroid carcinoma, non-small cell lung cancer, and renal cell carcinoma [39][40][41][42]. SSD, in combination with radiation (2 Gy), mediates powerful anti-cancer effects by inducing G-0/G-1 arrest, by upregulating p53 and Bax expression, and by downregulating Bcl2 expression under hypoxia exposure in SMMC-7721 hepatocellular carcinoma cells [43]. A Ca 2+ agent, in combination with SSD, overcame chemoresistance via p53 overexpression and induced G2/M arrest, mitochondrial fission, and apoptotic cell death via Ca 2+ release in ovarian cancer cells [16]. In addition, a novel sarcoplasmic/endoplasmic reticulum Ca 2+ -ATPase (SERCA) inhibitor, in combination with SSD, induced ER stress and autophagic cell death by releasing Ca 2+ [44]. SSA mediated apoptotic cell death via cell cycle arrest at sub-G1 phase, the caspase-9/-3 axis, the upregulation of pro-apoptotic proteins, such as Bak, Bad, and PUMA, and the downregulation of Bcl-2 expression in the rat HSC cell line, HSC-T6, and the human HSC cell line, LX-2 [45]. SSA, in combination with doxorubicin, vincristine, or paclitaxel overcomes chemoresistance by downregulating P-glycoprotein (P-gp) in multidrug resistance (MDR) overexpressing MCF-7/ADR and HepG2/ADM cells [46]. SERCA is a Ca 2+ transporter located on the sarcoplasmic reticulum (SR) and ER, and the disruption of Ca 2+ homeostasis by SERCA inhibition induces apoptotic cell death by activating ER stress, Ca 2+ accumulation, cytochrome C release, and caspase cleavage [47]. In the present study, I identified that SSA is a novel SERCA inhibitor along with SSD and thapsigargin, and that SSA mediates ER stress and apoptotic cell death via ROS release, Ca 2+ release, LDH release, and the upregulation of Nox4, and caspase activity in GC cells. Furthermore, combined therapy with SSA and radiation (2 Gy) overcomes radio resistance by inhibiting the EMT phenotype; however, in PERK knockdown radioresistant AGSR cells, SSA/2 Gy inhibited apoptotic cell death by inhibiting the PERK-ATF4-CHOP axis.
In conclusion, my findings indicate that SSA induces ER stress-mediated cell death by releasing ROS and Ca 2+ in GC and radio-resistant GC cells. Moreover, these results suggested that SSA may be a novel anti-tumor therapeutic strategy for radiotherapy.

Cell Viability
Cell viability was measured using the WST-1 assay (Roche Applied Science, Indianapolis, IN, USA). The absorbance of each well was measured at 450 nm using a microplate reader (Molecular Devices, CA, USA).

LDH Assay
Cells were seeded onto a 96-well plate in growth medium, and cell cytotoxicity was analyzed with the Pierce lactate dehydrogenase (LDH) cytotoxicity assay kit according to the manufacturer's protocol (Thermo Scientific, Waltham, MA, USA).

Caspase Activity Assay
Cells were seeded onto a 96-well plate in growth medium, and caspase-3 activity was analyzed using the caspase-3 activity assay kit (colorimetric) (Abcam, Milpitas, CA, USA) according to the manufacturer's instructions.

Development of Acquired Radio-Resistant GC Cell Lines
AGS and MKN-74 cells were seeded and cultured to approximately 50~60% confluence and exposed to 2 Gy radiation daily for three months. Then, radio-resistant GC cell lines (AGSR and MKN-74R) were established from the respective parental cell lines (AGS and MKN-74).

Colony Formation Assay
GC Cells (AGS, AGSR, MKN-74, and MKN-74R) were plated and incubated onto 60 mm dishes at a density of 1000 cells/dish. After incubation, colonies were fixed and stained with 1% methylene blue in 50% ethanol. The survival fraction was calculated using the following formula: surviving fraction = number of colonies formed/number of cells seeded x plating efficiency of the control group.

Isolation of Total RNA
Total RNA from GC cells in a 100 mm cell culture dish was isolated using Trizol reagent according to the manufacturer's protocols (Invitrogen, Carlsbad, CA, USA).

Isolation of Protein
Protein cell lysates from GC cells (2 × 10 6 cell/well) were collected in RIPA buffer containing a protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA). The supernatant was analyzed for protein content using the BCA method (Thermo Scientific, CA, USA).

Measuring of Reactive Oxygen Species (ROS)
ROS generation was measured after staining for 45 min with the DCFDA Cellular ROS Assay Kit (Abcam, Cambridge, MA, USA; ab113851). DCF fluorescence was immediately analyzed using a microplate reader (Molecular Devices, CA, USA).

Exosomes Isolation
Exosomes were obtained from the supernatant of SSA (10 µM)-treated GC cells according to the manufacturer's protocol (Total Exosome Isolation Reagent for cell culture media, Thermo Scientific, CA, USA).

Intracellular Ca 2+ Assays
AGS and MKN-74 cells were plated into a 96-well plate with growth medium at 1 × 10 4 cells/well. After 24 h, the cells were treated with SSA for 24 h. An intracellular Ca 2+ activity assay [Abcam, Ca 2+ Assay Kit (Colorimetric)] was performed and analyzed as described in the supplier's manual (Abcam, Cambridge, MA, USA).

Animals
For the animal study, five-week-old, female, athymic BALB/c nude mice (nu/nu) were purchased from OrientBio, Inc. (Daejeon, Republic of Korea), and maintained for 1 week with free access to sterile standard mouse chow (NIH-7 open formula) and water before use. Mice were housed randomly at 50 ± 20% humidity and approximately 21 ± 2 • C on a 12 h light-dark cycle (n = 10 mice/group). All animal experimental procedures were performed according to the National Institutes of Health guidelines and a protocol approved by the Institutional Animal Care and Use Committee of Kyung Hee University (KHSASP-20-250).

Tumor Xenograft Mouse Models
For the mice xenograft experiment, mice, aged six weeks, were inoculated with an AGS human gastric cancer cell line by subcutaneously (sc) implanting 1 × 10 7 cultured cells into the right thigh. Six days later, mice were grouped randomly (n = 10 per group) and SSA (5 or 10 mg/kg) was administered intraperitoneally (ip) every other day. Tumor sizes on two axes (L, longest axis; W, shortest axis) were measured three times per week using Vernier calipers. Tumor volume was calculated as (L × W 2 )/2 (mm 3 ).

Statistical Analysis
Data were expressed as the mean ± standard error (SE). Statistical analyses of the experimental data were performed using a two-sided Student's t-test. p values < 0.05 were determined to indicate statistical significance.
Author Contributions: T.W.K. designed and performed most experiments. T.W.K. discussed the data and wrote the manuscript. All authors have read and agreed to the published version of the manuscript.