Restricting extracellular Ca2+ on gefitinib-resistant non-small cell lung cancer cells reverses altered epidermal growth factor-mediated Ca2+ response, which consequently enhances gefitinib sensitivity

Non-small cell lung cancer (NSCLC), one of the leading causes of cancer-related death, has a low 5-year survival rate owing to the inevitable acquired resistance toward antitumor drugs, platinum-based chemotherapy, and targeted therapy. Epidermal growth factor (EGF)-EGF receptor (EGFR) signaling activates downstream events leading to phospholipase C/inositol trisphosphate (IP3)/Ca2+ release from IP3-sensitive Ca2+ stores to modulate cell proliferation, motility, and invasion. However, the role of EGFR-mediated Ca2+ signaling in acquired drug resistance is not fully understood. Here, we analyzed alterations of intracellular Ca2+ ([Ca2+]i) responses between gefitinib-sensitive NSCLC PC-9 cells and gefitinib-resistant NSCLC PC-9/GR cells, and we found that acute EGF treatment elicited intracellular Ca2+ ([Ca2+]i) oscillations in PC-9 cells but not in PC-9/GR cells. PC-9/GR cells presented a more sustained basal [Ca2+]i level, lower endoplasmic reticulum Ca2+ level, and higher spontaneous extracellular Ca2+ ([Ca2+]e) influx than PC-9 cells. Notably, restricting [Ca2+]e in both cell types induced identical [Ca2+]i oscillations, dependent on phospholipase C and EGFR activation. Consequently, restricting [Ca2+]e in PC-9/GR cells upregulated gefitinib-mediated poly (ADP-ribose) polymerase cleavage, an increase in Bax/Bcl-2 ratio, cytotoxicity, and apoptosis. In addition, nuclear factor of activated T cell (NFAT1) induction in response to EGF was inhibited by gefitinib in PC-9 cells, whereas EGF-mediated NFAT1 induction in PC-9/GR cells was sustained regardless of gefitinib treatment. Restricting [Ca2+]e in PC-9/GR cells significantly reduced EGF-mediated NFAT1 induction. These findings indicate that spontaneous [Ca2+]e influx in NSCLC cells plays a pivotal role in developing acquired drug resistance and suggest that restricting [Ca2+]e may be a potential strategy for modulating drug-sensitivity.


Introduction
The incidence of non-small cell lung cancer (NSCLC) is steadily increasing and accounts for 85% of lung cancer subtypes. Owing to its recurrence, NSCLC has a low 5-year survival rate of <15% [1]. Since the development of first-generation epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs), such as gefitinib, erlotinib, and ecotinib, diverse targeted therapies operating at the molecular and genetic levels have emerged rapidly. Despite these developments, intrinsic or acquired drug resistance to chemotherapeutic agents allows cancer cells to bypass cell death. Overexpression and over-activity of EGFR are observed in >60% of NSCLC cells [2]. Moreover, prolonged treatment with EGFR-TKIs frequently causes EGFR mutations and interferes with its underlying signaling pathways; thus, prolonged EGFR-TKI use limits its clinical efficacy [3].
In the 2019-guideline v3, the National Comprehensive Cancer Network indicates that genes, including EGFR, ALK, BRAF, KRAS, HER2, ROS1, RET, and MET, are therapeutic targets for treating NSCLC. TKIs targeting these fundamentally inhibit signaling cascades related to cell proliferation, by which cancer cells frequently show acquired drug resistance owing to gene mutations, including rearrangement, amplification, and point mutation [4,5]. According to the Kyoto Encyclopedia of Genes and Genomes database, EGFR and its underlying mechanisms appear dependent on the intracellular Ca 2+ ([Ca 2+ ]i) signaling pathway in NSCLC, in which EGFR activation sequentially elicits phospholipase Cγ (PLCγ) phosphorylation, inositol trisphosphate (IP 3 ) production, Ca 2+ release from the endoplasmic reticulum (ER), and protein kinase C activation. Acute human glioma cell stimulation with EGF evokes intracellular Ca 2+ responses as oscillations, which are blocked by EGFR inhibitors [6]. Thus, the correlation between intracellular Ca 2+ signaling and acquired drug resistance appears to be interactive; however, the role of Ca 2+ signaling in drug resistance is not fully understood.
Free intracellular Ca 2+ acts as a second messenger to regulate the proliferation, migration, and apoptosis of cancer cells [7]. Ca 2+ depletion in the ER mediates Stromal interaction molecule 1 (Stim1)-dependent Ca 2+ influx through store-operated Ca 2+ channels (SOCCs), such as ORAI calcium release-activated calcium modulator 1 (Orai1) and (transient receptor potential channels) TRPCs [8]. Altered Ca 2+ signaling in cancer development and progression has been frequently observed. Increased Ca 2+ influx through TRPC5 upregulates autophagic flux to prevent cancer-cell death and promotes drug resistance [9]. Anticancer drugs, including cisplatin, 5-fluorouracil, and gemcitabine in osteosarcomas and pancreatic adenocarcinomas, appear to enhance Orai1 and Stim1 expression, which prevents drug-mediated cell death [10,11]. Notably, treatment using EGFR-targeting afatinib causes Ca 2+ signaling-related gene expression in PC-9 cells. Reduced extracellular Ca 2+ ([Ca 2+ ]e) levels increase the sensitivity of PC-9 cells to afatinib [12]. These reports strongly suggest that Ca 2+ signaling in cancer cells is significantly associated with the development of acquired drug resistance. However, the role of EGFRmediated Ca 2+ signaling in acquired drug resistance is not fully understood.
In this study, we demonstrated that the EGF-mediated Ca 2+ response in NSCLC cells was altered depending on gefitinib resistance and [Ca 2+ ]e restriction on gefitinib-sensitive and gefitinib-resistant cells elicited identical [Ca 2+ ]i oscillations, which were associated with the modulation of gefitinib efficacy.

Measurement of [Ca 2+ ]i
The [Ca 2+ ]i was measured using a Ca 2+ -sensitive fluorescence dye, Fura-2/AM (Sigma Aldrich) as described previously [14]. Briefly, cells were plated on coverslips at 80% confluence and loaded with Fura-2/AM (5 μM) for 1 h at 37˚C in a 5% CO 2 incubator. Cells were perfused with HEPES buffer containing 140 mmol/L NaCl, 5 mmol/L KCl, 1 mmol/L MgCl 2 , 1 mmol/L CaCl 2 , 10 mmol/L HEPES, and 10 mmol/L glucose. The pH and osmolarity were adjusted to 7.4 and 310 mOsm. Ca 2+ -free HEPES buffer was replaced with 1 mmol/L EGTA. Following brief washing with HEPES buffer, trapped intracellular Fura-2 was excited at 340 and 380 nm. Emitted fluorescence at 510 nm was captured using a charge-coupled device camera. Images were analyzed using MetaFluor software (Molecular Devices, San Jose, CA, USA) and presented as F340/F380 ratio.

Cell viability assay
Cell viability was determined using EZ-CYTOX (Daeil Lab Service Co. Ltd., Seoul, South Korea), following the manufacturer's procedure. In brief, cells were treated with EZ-Cytox (10 μL) in each well and incubated for 30 min. Cell viability was measured at OD 450 nm using the iMAX Microplate Reader (Bio-Rad, Hercules, CA, USA).

Cytotoxicity assay
Cytotoxicity was determined by measuring extracellular glucose-6-phosphate dehydrogenase (G6PD) levels using a Vybrant cytotoxicity assay kit (Invitrogen, Carlsbad, CA, USA) following the manufacturer's protocols. Briefly, cells were seeded onto 96-well plates and incubated under indicated conditions. Cell culture medium without cells was collected and G6PD activity was determined at excitation and emission wavelengths of 544 and 590 nm, respectively. The result is expressed as the percentage of total G6PD detected in cell lysates from parallel wells.

Flow cytometry
The early apoptosis rate was determined using the FITC Annexin V Apoptosis Detection Kit with PI (BioLegend, San Diego, CA, USA) following the manufacturer's protocol. In brief, after a 24-h incubation, cells were washed with Cell Staining Buffer and resuspended with the Annexin V-FITC and PI mixture for 15 min at room temperature in the dark. After adding Annexin V Binding buffer, a minimum of 10,000 cells were analyzed using FACScan analyzer (Becton Dickinson, Franklin Lakes, NJ, USA).

Statistical analysis
Statistical analysis was conducted using Origin 2020 software (OriginLab Corporation, MA, USA). Data are presented as the mean ± standard deviation of observations obtained from more than three independent experiments. Statistical differences were analyzed using one-way ANOVA followed by Tukey's post hoc test and T-test. Values of p < 0.05 were considered statistically significant.

Basal level of [Ca 2+ ]i in PC-9/GR cells is sustained by spontaneous extracellular Ca 2+ influx, resulting in abolishment of EGF-mediated [Ca 2+ ]i oscillations
To determine altered EGF-mediated [Ca 2+ ]i responses between PC-9 and PC-9/GR cells, we performed a ratiometric assay using Fura-2/AM. Acute treatment with 200 ng/mL of EGF induced [Ca 2+ ]i oscillations in PC-9 cells but not in PC-9/GR cells ( Fig 1A). PC-9/GR cells showed a more highly sustained basal level of [Ca 2+ ]i than PC-9 cells; thus, we examined the spontaneous Ca 2+ influx in both cell types. To evaluate the spontaneous [Ca 2+ ]e influx and determine ER Ca 2+ content, cells were exposed to Ca 2+ -free HEPES buffer and HEPES buffer (1 mM Ca 2+ ). Cells were treated with cyclopiazonic acid to deplete ER Ca 2+ . The spontaneous Ca 2+ influx (indicated as F1) was greater in PC-9/GR cells than in PC-9 cells, and ER Ca 2+ content in PC-9/GR cells was 15% lower than that in PC-9 cells (Fig 1B). We additionally characterized the expression of 3 different genes, Orai1, STIM1, and SERCA2, which are essential for mediating store-operated Ca 2+ entry (SOCE). Expression of dimeric Orai1 and STIM1 showed no significant difference between PC-9 and PC-9/GR cells, whereas SERCA2 in PC-9/GR cells was significantly reduced by about 35% compared to PC-9 (S1 Fig). These results suggest that drug-resistant tumor cells somehow develop altered intracellular Ca 2+ signaling, which may act as a bypass to activate downstream signals for cell survival.

Altered EGF-mediated [Ca 2+ ]i response in PC-9/GR cells is reversed in an identical manner to that of PC-9 cells by restricting [Ca 2+ ]e
Consequently, we examined the effects of [Ca 2+ ]e restriction on EGF-stimulated PC-9 and PC-9/GR cells. EGF-mediated [Ca 2+ ]i oscillations in PC-9 cells lasted even in the absence of extracellular Ca 2+ (Fig 2A) 2 μM), an inhibitor of PLC, and gefitinib (0.1 μM) in PC-9 and PC-9/GR cells (Fig 2B and  2C). These results suggest that restricting [Ca 2+ ]e may revert the cytotoxic effects of gefitinib on PC-9/GR cells.

[Ca 2+ ]e restriction and PLC inhibition synergistically exacerbate gefitinibinduced cytotoxicity and cell viability in PC-9/GR cells
Gefitinib reduces EGFR signaling by inhibiting the ATP-binding pocket in the EGFR kinase domain, which sequentially results in cell cycle arrest and apoptosis [17]. As PC-9 and PC-9/ GR cells showed identical [Ca 2+ ]i oscillations after the removal of [Ca 2+ ]e, we examined whether the low [Ca 2+ ]e-induced [Ca 2+ ]i oscillations were associated with gefitinib-induced cytotoxicity and cell viability. PC-9 and PC-9/GR cells were treated with gefitinib in a dosedependent manner (0.1, 1, and 5 μM) and incubated in a culture medium supplemented with  or without Ca 2+ . After 24-and 48-h incubation periods, cytotoxicity and cell viability, respectively, were evaluated using enzymatic assays. We first showed that gefitinib-treated PC-9 cells exhibited a significant increase in cytotoxicity and decrease in cell viability in a dose-dependent manner regardless of [Ca 2+ ]e (left panel, Fig 3A and 3B). However, gefitinib in 1-mM [Ca 2+ ]e had no effect on cytotoxicity and cell viability in PC-9/GR cells. Thus, [Ca 2+ ]e restriction significantly exacerbated gefitinib-induced cytotoxicity and cell viability compared with the non-gefitinib-treated group and the group treated with gefitinib in 1-mM [Ca 2+ ]e (mid panel, Fig 3A and 3B). We investigated the synergistic effects of PLC inhibition and [Ca 2+ ]e restriction on cytotoxicity and cell viability. PC-9/GR cells were pretreated with U73122 and U73343 (2 μM), followed by treatment with gefitinib, either in the absence or presence of [Ca 2 + ]e. PLC inhibition by U73122 and [Ca 2+ ]e restriction synergistically exacerbated cytotoxicity and cell viability (Right panel, Fig 3A and 3B).

Restricting [Ca 2+ ]e in PC-9/GR cells enhances gefitinib-mediated early apoptosis
Biochemical assays indicated that gefitinib-mediated cytotoxicity was significantly increased by [Ca 2+ ]e restriction in PC-9/GR cells. We investigated the effects of [Ca 2+ ]e restriction on the activation of early apoptotic markers. To determine the apoptotic induction, whole-cell lysates were subjected to detect PARP cleavage and Bax/Bcl-2 expression. Gefitinib treatment caused the induction of apoptotic markers in PC-9 cells, irrespective of [Ca 2+ ]e (Fig 4A).

Restriction of [Ca 2+ ]e reduces EGF-mediated NFAT1 induction in gefitinib-resistant cells
Among the isoforms of NFATs, NFAT1 expression is higher in NSCLC cells and this overexpression is related to the poor survival of patients with NSCLC [18]. This led us to examine whether EGF-stimulated tumor cells elicited NFAT1 induction and whether this was affected by [Ca 2+ ]e restriction on gefitinib-resistant cells. Following FBS starvation for 1 h, EGF-treated (200 ng/mL) cells were incubated. EGF-mediated NFAT1 induction was significantly reduced by gefitinib (5 μM) in PC-9 cells (Fig 5A), whereas it was unaffected in gefitinib-treated PC-9/ GR cells in the presence of [Ca 2+ ]e (Fig 5B). [Ca 2+ ]e restriction on gefitinib-treated PC-9/GR cells resulted in the reduction of EGF-mediated NFAT1 induction (Fig 5B). These results indicate that NFAT1 is regulated by EGFR activation and spontaneous [Ca 2+ ]e influx is critical for inducing NFAT1 in gefitinib-resistant cells.

Discussion
Diverse signaling pathways underlying EGFR activation have been implicated in various cellular functions [16]. Increasing evidence shows that altered [Ca 2+ ]i signaling plays a key role in tumorigenesis [19]. However, its role in acquired drug resistance is not fully understood. A study has shown that the binding of EGF to EGFR in human glioma cells induces tyrosine kinase-dependent Ca 2+ oscillations [6]. The expression of sarco/endoplasmic reticulum Ca 2+ -ATPase and IP 3 R in hepatocellular carcinoma cells appears to be lower and higher, respectively, than that in human bronchial epithelial cells, leading to reduced Ca 2+ content in the ER [20]. Orai3, which mediates SOCE and allows the proliferation and metastasis of abnormal cells, has been reported to be overexpressed in NSCLCs [21]. Moreover, inhibiting Stim1 and Orai1 results in the modulation of tumor migration, metastasis, and proliferation in other cancer cells [22]. Based on these reports, we assumed that EGF-mediated Ca 2+ signaling is altered depending on acquired drug resistance. Considering that [Ca 2+ ]e influx (indicated as F2 in Fig 1B)   oscillations, which is identical to the [Ca 2+ ]i response in gefitinib-sensitive cells. Low [Ca 2+ ]emediated [Ca 2+ ]i oscillations were dependent on the blockade of EGFR activation by gefitinib, even in gefitinib-resistant cells. Considering the lower ER Ca 2+ content in PC-9/GR cells than in PC-9 cells, cellular organelles, such as lysosomes and mitochondria, might partake in maintaining Ca 2+ oscillations. However, identifying Ca 2+ stores that are involved in maintaining [Ca 2+ ]i oscillations requires further investigation. We used PC-9/GR cells with secondary mutations [13], which enhanced the affinity of EGFR toward ATP and rendered it gefitinibresistant. We observed that low [Ca 2+ ]e-mediated [Ca 2+ ]i oscillations were abolished by gefitinib and U73122 in PC-9/GR and PC-9 cells, suggesting that the absence of [Ca 2+ ]e led to enhanced gefitinib sensitivity and reversed gefitinib-mediated cytotoxicity and apoptosis. Mulder et al. demonstrated that initial targeted therapy on PC-9 cells increases the activity of Ca 2+ signaling-related proteins and deprives the cell of extracellular Ca 2+ , which results in the marked enhancement of afatinib efficacy [12]. Our data indicated that increased [Ca 2+ ]e influx in PC-9/GR cells contributed to gefitinib resistance; therefore, restricting [Ca 2+ ]e could be key in enhancing drug sensitivity. The induction of NFATs 1-4 is regulated by [Ca 2+ ]i increase, which is mediated by the receptor-activated PLC pathway or extracellular Ca 2+ influx in immune cells [26]. Inactive NFATs exist in hyper-phosphorylated states in the cytoplasm. Following a receptor-mediated [Ca 2+ ]i increase, NFATs are dephosphorylated and activated by Ca 2+ -dependent phosphatases, such as calcineurin [27]. Although still controversial, several studies have demonstrated that NFAT1 exhibits anti-apoptotic properties and promotes tumor progression [28,29]. Our study indicated that gefitinib-mediated apoptosis and EGF-mediated NFAT1 induction are significantly decreased depending on [Ca 2+ ]e restriction in gefitinib-resistant cells. Thus, sustained NFAT1 induction in gefitinib-resistant cells might play a role in acquired drug resistance.
Our study demonstrated that NSCLC cells altered EGF-mediated Ca 2+ signaling depending on gefitinib resistance. Importantly, restricting [Ca 2+ ]e in gefitinib-sensitive and gefitinibresistant cells elicited identical [Ca 2+ ]i oscillations and significantly enhanced gefitinib sensitivity in gefitinib-resistant cells. Moreover, we showed the regulatory effects of [Ca 2+ ]e on NFAT1 induction and concluded that restricting [Ca 2+ ]e should be used to treat drug-resistant NSCLCs.
Supporting information S1 Fig. Expression of Orai1, STIM1, and SERCA2 in PC-9 and PC-9/GR cells. PC-9 and PC-9/GR cells were respectively seeded on 60 mm culture dish and cultured in normal condition for overnight. Cells were lysed with RIPA buffer, and collected whole cell lysates were used for Western blot to determine the endogenous expression level of Orai1, STIM1, and SERCA2. Columns present the mean ± S.D. from 3 independent experiments. � P < 0.05. (TIF) S1 Raw images. (PPTX)