Impacts of hyperthermic chemotherapeutic agent on cytotoxicity, chemoresistance-related proteins and PD-L1 expression in human gastric cancer cells

Abstract

Objective: Gastric cancer with peritoneal metastasis is considered to be final stage gastric cancer. One current treatment approach for this condition is combined cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (HIPEC). However, the therapeutic mechanisms of HIPEC remain largely undescribed. Method: In order to assess the cellular effects of HIPEC in vitro, we treated AGS human gastric adenocarcinoma cells with or without 5-fluorouracil (5-Fu) at 37 °C or at 43 °C (hyperthermic temperature) for 1 h followed by incubation at 37 °C for 23 h. The impacts of hyperthermia/5-Fu on apoptosis, cell survival signals, oxidative stress, chemoresistance-related proteins and programmed death-ligand 1 (PD-L1) expression were measured. Results: Our results showed that hyperthermia potentiates 5-Fu-mediated cytotoxicity in AGS cells. Furthermore, the combination of 5-Fu and hyperthermia reduces levels of both phosphorylated STAT3 and STAT3, while increasing the levels of phosphorylated Akt and ERK. In addition, 5-Fu/hyperthermia enhances reactive oxygen species and suppresses superoxide dismutase 1. Chemoresistance-related proteins, such as multidrug resistance 1 and thymidylate synthase, are also suppressed by 5-Fu/hyperthermia. Interestingly, hyperthermia enhances 5-Fu-mediated induction of glycosylated PD-L1, but 5-Fu-mediated upregulation of PD-L1 surface expression is prevented by hyperthermia. Conclusion: Taken together, our findings provide insights that may aid in the development of novel therapeutic strategies and enhanced therapeutic efficacy of HIPEC.

Keywords:

• Hyperthermia
• stomach cancer
• chemosensitivity
• immune checkpoints

1. Introduction

Gastric cancer is the fifth most common malignancy and the third leading cause of cancer death worldwide [1,2]. Nearly 20–30% of gastric cancer patients are diagnosed with concurrent peritoneal dissemination perioperatively [3]. This progression to peritoneal carcinomatosis is considered to be a terminal stage of disease, and it carries a very poor prognosis of only 3–4 months expected survival [4]. Even with modern regimens of systemic chemotherapy, the median survival time can only be improved to 8.0–13.2 months [5]. In the past two decades, a more aggressive multimodal treatment protocol, cytoreductive surgery (CRS) with hyperthermic intraperitoneal chemotherapy (HIPEC), has been implemented for peritoneal carcinomatosis [6]. This treatment strategy can be successfully applied for various intraabdominal malignancies, including but not limited to gastric cancer, ovarian cancer, colon cancer, peritoneal mesothelioma and appendix low-grade mucinous neoplasms. A recent observational analysis of a prospective database (CYTO-CHIP study) has demonstrated that the addition of HIPEC to CRS can improve median survival to 18.8 months, as compared to 12.1 months for CRS alone. The 3- and 5-year recurrent survival rates were respectively 20.40% and 17.05% in the CRS-HIPEC group and 5.87% and 3.76% in the CRS alone group (p = 0.001, comparing between treatments) [7]. Although some improvements can be seen with the addition of hyperthermia to chemotherapeutic agents, the prognosis remains extremely poor and recurrence rates remain high after the treatment. Thus, there remains an urgent need for more effective strategies to manage this critical condition.

Only a few studies have examined the benefits of heated chemotherapy on cancer cell cytotoxicity and cancer eradication. Nevertheless, the putative anticancer mechanisms of HIPEC include heat-related cell damage and inhibition of DNA repair processes, as well as T-cell activation by heat-shock protein-enhanced antigen presentation, natural killer cell activation by cytokine release, enhancement of cell apoptosis and increased blood flow due to hyperthermia [8]. However, the detailed molecular events that occur following treatment with heated chemotherapy have not been thoroughly defined in the context of gastric cancer. Doing so may reveal new targets or treatment protocols that may reduce the high recurrence rate of gastric cancer with peritoneal carcinomatosis. In the present study, we established a gastric cancer hyperthermic chemotherapy treatment cell model and used it to investigate the molecular and cellular effects of heated chemotherapy in terms of cell apoptosis, cell proliferation, expression of therapeutic-resistance proteins and regulation of the immunotherapy marker, programmed death-ligand 1 (PD-L1).

2. Materials and methods

2.1. Reagents

5-Fluorouracil (5-Fu), U0126 (ERK inhibitor), glutaraldehyde, crystal violet and propidium iodide (PI) were purchased from Merck (Germany). GDC-0068 (Akt inhibitor) and Stattic (STAT3 inhibitor) were purchased from MedChemExpress (USA).

2.2. Cell culture and treatment

The AGS human gastric adenocarcinoma cell line and SNU-1 human gastric carcinoma cell line were purchased from the Bioresource Collection and Research Center (Taiwan). Cells were maintained in RPMI 1640 (Thermo Fisher Scientific, USA) supplemented with penicillin/streptomycin (Sartorius, Germany) and 10% fetal bovine serum (Peak, USA). The cultures were maintained at 37 °C and 5% CO₂. In order to mimic HIPEC, the cells were treated as described in a previous study [9]. Briefly, each well of a six-well plate was seeded with 8 × 10⁵ AGS or SNU-1 cells and incubated at 37 °C overnight. Then, cells were treated with or without 5-Fu and incubated at 37 or 43 °C for 1 h, followed by incubation at 37 °C for 23 h. Cells were harvested at 24 h for analysis.

2.3. Western blot analysis

Cells were lysed in RIPA buffer and centrifuged at 12,000 rpm for 10 min. The supernatants were collected and incubated at 100 °C for 10 min. Protein samples were stored at −20 °C before use. Protein samples were separated on gradient gels by electrophoresis and transferred onto 0.2-μm PVDF membranes (Cytiva, USA). PVDF membranes were sequentially probed with primary and secondary antibodies. Primary antibodies were purchased from Cell Signaling Technology (USA). Secondary antibodies were purchased from Jackson ImmunoResearch Inc (USA). Protein bands were visualized using ECL reagent (Merck, Germany). Protein band intensities were analyzed by ImageJ software (bundled with 64-bit Java 8).

2.4. Apoptosis analysis

Apoptosis was detected with annexin V/PI double staining, Poly(ADP-ribose) polymerase (PARP) cleavage, caspase-3 activation and cell morphology. Annexin V/PI double staining was performed as previously described [10]. Fluorescein isothiocyanate annexin V was purchased from BioLegend (USA). Cleaved PARP and casapase-3 were detected and measured by western blotting. The rounded morphology of apoptotic cells was observed and photographed with a light microscope using a 4× objective lens (Olympus, Germany) [11].

2.5. Colony formation assay

Cells were treated with 5-Fu/hyperthermia as described in Figure 1(A). After treatment, detached cells (dead cells) were removed, and adhered cells (alive cells) were collected. A total of 2000 cells was seeded into each well of a six-well plate. Cells were maintained in a cell culture incubator for 5 days. The colony formation assay was performed as previously described [12].

Figure 1. Cytotoxicity of hyperthermia and 5-Fu in AGS and SNU-1 gastric cancer cells. (A) Experimental design. AGS and SNU-1 cells were treated with or without 5-Fu at 37 or 43 °C for 1 h, followed by incubation at 37 °C for 23 h. (B) Vehicle or 5-Fu (25 μM) was treated at 37 or 43 °C. Cell morphology was inspected under a 4× objective lens. (C–H) Cells were treated with or without increasing dose of 5-Fu at 37 or 43 °C. Apoptosis was detected by annexin V/PI double staining, PARP cleavage and caspase-3 activation. *: p < 0.05 compared to 0 μM 5-Fu at 37 °C. @: p < 0.05 compared to 0 μM 5-Fu at 43 °C. #: p < 0.05; significant difference between indicated dose of 5-Fu at 37 and 43 °C.

2.6. Reactive oxygen species analysis

After treatment, cells were loaded with the fluorescent reactive oxygen species (ROS) indicator 2',7'-Dichlorofluorescin diacetate (DCF-DA) (10 μM, Merck, Germany) for 15 min. Then, cells were rinsed with phosphate buffer saline (PBS), and DCF-DA intensity was measured by flow cytometry (Beckman Coulter Life Science, USA).

2.7. Cell surface PD-L1 analysis

Cells were harvested and incubated with extracellular domain-specific Alexa Fluor 488-conjugated PD-L1 antibody (Cell Signaling Technology, USA) at 4 °C for 1 h. Excess antibody was washed off the cells with PBS. Fluorescence intensity indicating surface PD-L1 was analyzed by flow cytometry (Beckman Coulter Life Science, USA).

2.8. Statistical analysis

Data are presented as mean ± SEM. Student’s t-test or one-way ANOVA were used to assess the statistical differences. p-Values smaller than 0.05 were considered statistically significant.

3. Results

3.1. Cytotoxicity of hyperthermia and 5-Fu on gastric cancer AGS and SNU-1 cells

In this study, we wanted to understand the mechanisms by which HIPEC acts to kill gastric cancer cells. To mimic the effects of HIPEC or heated chemotherapy on gastric cancer cells in vitro, we treated AGS and SNU-1 cells with or without 5-Fu at 37 or 43 °C for 1 h. Then, we incubated the cells at 37 °C for 23 h (Figure 1(A)). The temperature of 43 °C was considered hyperthermia. After the treatment, cytotoxicity was detected by assessing cell morphology (Figure 1(B)), annexin V/PI double staining (Figure 1(C–E)), PARP cleavage (Figure 1(F,G)) and caspase-3 activation (Figure 1(F,H)). As shown in Figure 1(B), the 37 °C vehicle-treated cells mostly adhered to the well. 5-Fu or hyperthermia treatments alone only slightly disrupted the confluent layer of cells, while the combination treatment reduced cell confluence and increased numbers of rounded cells (a morphological feature of apoptosis). Annexin V/PI double staining confirmed that apoptosis was modestly induced by 5-Fu or hyperthermia alone, and that 5-Fu in combination with hyperthermia caused enhanced cytotoxicity in the gastric cancer cells. In addition, 5-Fu-induced cleavage of both PARP and caspase-3 could also be enhanced by hyperthermia. Thus, our results consistently showed that the combination of 5-Fu and hyperthermia induced indicators of apoptosis at higher levels than either treatment alone.

3.2. Cancer cell regrowth after hyperthermia and 5-Fu treatment

We next wanted to evaluate how hyperthermia and 5-Fu might affect cancer recurrence, so we treated cells as described in Figure 1(A). At the end of the treatment, non-viable detached cells were discarded, and viable cells were collected and re-seeded. Cancer cell regrowth was then determined by a colony formation assay (Figure 2(A)). We found that none of the treatments (i.e., 5-Fu and hyperthermia alone or in combination) could completely suppress cancer cell regrowth after treatment (Figure 2(B)). However, the combination of 5-Fu with hyperthermia exhibited stronger inhibitory activity toward cell regrowth than either of the treatments alone.

Figure 2. Cell regrowth after hyperthermia and 5-Fu treatment. (A) Experimental design. Cells were treated with or without 5-Fu at 37 or 43 °C for 1 h, followed by incubation at 37 °C for 23 h. Dead cells (suspended cells) were washed away. Attached cells were collected and re-seeded onto new six-well plates. (B) Cell regrowth was monitored by colony formation assay. Yellow arrows indicate colonies.

3.3. Effects of hyperthermia and 5-Fu on cell survival signals

Since we found that the combination of 5-Fu and hyperthermia could not completely prevent cancer cell regrowth after treatment, we were interested to test how hyperthermia and 5-Fu would affect cell survival signals. Treatments were performed as described in Figure 1(A), and then phosphorylation/activation of STAT3, Akt and ERK1/2 were analyzed by western blotting (Figure 3(A–F)). We found that the levels of both phospho-STAT3 and STAT3 were increased by 5-Fu treatment alone, but these effects were markedly suppressed by hyperthermia (Figure 3(A–C)). Meanwhile, 5-Fu-induced phosphorylation of Akt and ERK were enhanced by hyperthermia (Figure 3(A,D–F)). However, the effects of 5-Fu on total Akt were partially suppressed by hyperthermia (Figure 3(A,F)). In order to test whether STAT3, Akt and ERK1/2 play roles in 5-Fu-induced cytotoxicity, each of these cell survival signals was individually inhibited prior to 5-Fu treatment. Inhibitors of STAT3, Akt and ERK1/2 included Stattic (Figure 3(G,H)), GDC-0068 (Figure 3(I,J)), U0126 (Figure 3(K,L)), respectively. Cytotoxicity was analyzed according to PARP cleavage and annexin V/PI double staining. Interestingly, suppression of STAT3, Akt or ERK1/2 could enhance 5-Fu cytotoxicity.

Figure 3. Effects of hyperthermia and 5-Fu on survival signals. (A–F) Cells were treated with or without increasing doses of 5-Fu at 37 or 43 °C for 1 h, followed by incubation at 37 °C for 23 h. Levels of phospho-STAT3 (Tyr705), STAT3, phospho-Akt (Ser473), Akt, phospho-ERK1/2 (Thr202/Tyr204), ERK1/2 and β-actin were measured by western blotting. Band intensities were quantified using ImageJ. *: p < 0.05 compared to 0 μM 5-Fu at 37 °C. p < 0.05: significant difference between indicated dose of 5-Fu at 37 and 43 °C. (G,H) Cells were pretreated with Stattic (5 μM) for 1 h, followed by 5-Fu (25 μM) at 37 °C for 23 h. Apoptosis was detected by PARP cleavage and annexin V/PI double staining. Cells were pretreated with GDC-0068 (100 nM) for 1 h, followed by 5-Fu (25 μM) at 37 °C for 23 h. (I,J) Apoptosis was detected by PARP cleavage and annexin V/PI double staining. (K,L) Cells were pretreated with U0126 (10 μM) for 1 h, followed by 5-Fu (25 μM) at 37 °C for 23 h. Apoptosis was detected by PARP cleavage and annexin V/PI double staining. *: p < 0.05: significance between indicated groups. Veh: vehicle; 5 + S: 5-Fu + Stattic; 5 + G: 5-Fu + GDC-0068; 5 + U: 5-Fu + U0126; Sta: Stattic; GDC: GDC-0068; U0: U0126.

3.4. Impacts of hyperthermia and 5-Fu on chemoresistance-associated proteins and oxidative status

We also wanted to understand how hyperthermia and 5-Fu affect chemoresistance-associated proteins, so we again performed treatments as described in Figure 1(A). Following the treatment period, expression levels of multidrug resistance 1 (MDR1) and thymidylate synthase were analyzed by western blotting. Results showed that 5-Fu alone elevated the levels of MDR1 (Figure 4(A,B)) and thymidylate synthase (Figure 4(A,C)), and the 5-Fu-mediated increases in MDR1 and thymidylate synthase were attenuated by hyperthermia. We next tested the effects of 5-Fu and hyperthermia on ROS levels in the cells. Both 5-Fu and hyperthermia alone slightly increased the levels of ROS (Figure 4(D)), and the combination of 5-Fu with hyperthermia caused the highest levels of ROS. Moreover, the level of catalase was reduced by 5-Fu or hyperthermia alone (Figure 4(E,F)). Hyperthermia did not further suppress the levels of catalase in 5-Fu-treated cells. In contrast, the SOD1 level was only affected by the combination of 5-Fu with hyperthermia (Figure 4(E,G)), and SOD2 level was not affected by 5-Fu alone, hyperthermia alone or the combination of 5-Fu and hyperthermia (Figure 4(E,H)).

Figure 4. Effects of hyperthermia and 5-Fu on chemoresistance-related proteins and oxidative status. (A–C) Cells were treated with or without increasing doses of 5-Fu at 37 or 43 °C for 1 h, followed by incubation at 37 °C for 23 h. Levels of MDR1 and thymidylate synthase were measured by western blotting. *: p < 0.05 compared to 0 μM 5-Fu at 37 °C. @: p < 0.05 compared to 0 μM 5-Fu at 43 °C. #: p < 0.05; significant difference between indicated dose of 5-Fu at 37 and 43 °C. (D) Oxidative stress was measured by DCF-DA flow cytometry. Numbers show percent positive cells. (E–H) Cells were treated with or without 5-Fu (25 μM) at 37 or 43 °C for 1 h, then incubated at 37 °C for 23 h. Levels of catalase, SOD1 and SOD2 were measured by western blotting. Quantification of band intensities was performed using ImageJ. *: p < 0.05: significance between indicated groups. Veh: vehicle.

3.5. Impacts of hyperthermia and 5-Fu on PD-L1

Since advanced gastric cancer patients may sometimes receive anti-PD1/PD-L1 treatments, we wondered how HIPEC might affect the targets of this therapeutic strategy. To determine the impacts of hyperthermia and 5-Fu on PD-L1 regulation, treatments were performed as described in Figure 1(A). Levels of glycosylated and non-glycosylated PD-L1 were analyzed by western blotting. We found that 5-Fu alone elevated glycosylated PD-L1, and this stimulation was further enhanced by combination with hyperthermia (Figure 5(A,B)). In contrast, non-glycosylated PD-L1 was drastically suppressed by hyperthermia (Figure 5(A,C)). We also detected the levels of cell surface PD-L1 by flow cytometry, finding that 5-Fu increased the level of surface PD-L1, while hyperthermia inhibited this activity (Figure 5(D,E)).

Figure 5. PD-L1 expression after hyperthermia and 5-Fu treatment. (A–C) Cells were treated with or without increasing doses of 5-Fu at 37 or 43 °C for 1 h, followed by incubation at 37 °C for 23 h. Levels of glycosylated PD-L1 and non-glycosylated PD-L1 were measured by western blotting. Quantification of bands was performed using ImageJ. (D,E) Surface PD-L1 was detected by flow cytometry. *: p < 0.05 compared to 0 μM 5-Fu at 37 °C. @: p < 0.05 compared to 0 μM 5-Fu at 43 °C. #: p < 0.05; significant difference between indicated dose of 5-Fu at 37 and 43 °C.

3.6. Roles of STAT3, Akt and ERK1/2 in 5-Fu-induced upregulation of PD-L1

After discovering that HIPEC may affect PD-L1 surface expression, we asked whether cell survival signals may participate in this action. To determine the potential roles of STAT3, Akt and ERK1/2 in 5-Fu-induced PD-L1 upregulation, cells were preincubated with Stattic, GDC-0068 or U0126 and then treated with 5-Fu. Glycosylated PD-L1 was detected by western blotting, and cell surface PD-L1 was assessed by flow cytometry. We found that suppression of STAT3 and Akt potentiated 5-Fu-induced increases in glycosylated PD-L1 (Figure 6(A,B)). Yet, the suppression of ERK attenuated the 5-Fu-induced increase in glycosylated PD-L1 (Figure 6(C)). Furthermore, Stattic did not alter 5-Fu-mediated elevation of cell surface PD-L1 (Figure 6(D)), but both GDC-0068 (Figure 6(E)) and U0126 (Figure 6(F)) could suppress 5-Fu-induced upregulation of cell surface PD-L1.

Figure 6. Roles of STAT3, Akt and ERK in 5-Fu-mediated PD-L1 upregulation. (A) Cells were pretreated with Stattic (5 μM) for 1 h, followed by 5-Fu (25 μM) at 37 °C for 23 h. Levels of glycosylated PD-L1 were measured by western blotting. (D) Surface PD-L1 was detected by flow cytometry. (B) Cells were pretreated with GDC-0068 (100 nM) for 1 h, followed by 5-Fu (25 μM) at 37 °C for 23 h. Levels of glycosylated PD-L1 were measured by western blotting. (E) Surface PD-L1 was detected by flow cytometry. (C) Cells were pretreated with U0126 (10 μM) for 1 h, followed by 5-Fu (25 μM) at 37 °C for 23 h. Levels of glycosylated PD-L1 were measured by western blotting. (F) Surface PD-L1 was detected by flow cytometry. *: p < 0.05: significance between indicated groups. @: p > 0.05. Veh: vehicle; 5 + S: 5-Fu + Stattic; 5 + G: 5-Fu + GDC-0068; 5 + U: 5-Fu + U0126; Sta: Stattic; GDC: GDC-0068; U0: U0126.

4. Discussion

Patients with peritoneal metastasis of gastric cancer have very poor prognosis, with a median survival time of only 3–4 months [4]. In addition, these patients typically have poor response to conventional systemic chemotherapy, as the peritoneum has relatively few blood vessels compared to other organs [13]. Currently, the most effective treatment for the condition is a combination of CRS and HIPEC, which greatly improves the 1-, 3- and 5-year survival rates, as well as the time of disease-free survival [14,15]. For this treatment, surgeons will first remove the tumors from the peritoneal cavity as thoroughly as possible. Then, chemotherapeutic agents will be heated to a temperature between 42 and 43 °C and perfused into peritoneal cavity. In this step, unresectable or remaining tumors are often eliminated by the enhanced cytotoxic effects of heated chemotherapeutic agents. The higher temperature is expected to dilate blood vessels and increase tumor blood flow, which should result in more abundant accumulation of chemotherapeutic agents in the tumor mass, allowing concentrations to reach toxic levels. DNA damage and apoptosis are caused by the chemotherapeutics, and release of death-associated molecular pattern molecules from dead cancer cells can prime T and natural killer cell-mediated anti-tumor responses [8]. Beyond these general events, the molecular and cellular mechanisms of action of HIPEC are not fully understood. Although CRS + HIPEC is the most effective available therapeutic approach for gastric cancer with peritoneal metastasis, its long-term survival benefits are still not satisfactory. Clinical studies have revealed that the 5-year survival rate for CRS + HIPEC is no more than 30% [14,16]. Thus, a better understanding of the mechanisms underlying HIPEC therapeutic response could aid in the development of novel therapeutic strategies with improved therapeutic response and long-term clinical outcomes.

5-Fu disrupts DNA and RNA synthesis and is used to treat various types of cancer [17]. While this chemotherapeutic agent exhibits potent anticancer activity, it also causes many undesirable side effects, such as severe mucositis, cardiotoxicity, renal toxicity and febrile neutropenia, which often lead to treatment discontinuation [18,19]. Since higher doses of 5-Fu carry higher risks of side effects [20], it is important to minimize the dose as much as possible. In this study, we found that even a low dose of 25 μM 5-Fu in combination with hyperthermia showed greater anticancer activity than the highest tested dose of 5-Fu (500 μM) without hyperthermia (Figure 1). Thus, our findings indicate that hyperthermia can greatly increase the chemosensitivity of gastric cancer cells to 5-Fu. As such, hyperthermia may be an essential component of 5-Fu treatment in terms of reducing the risk of side effects [21]. Even though the combination of hyperthermia with 5-Fu induced more apoptosis than 5-Fu or hyperthermia alone (Figure 1), the combined treatment still failed to completely inhibit cancer regrowth (Figure 2). This observation might help to explain why nearly 60% of patients still experience recurrence after receiving CRS + HIPEC [22]. In addition to inducing apoptosis, both 5-Fu and hyperthermia are able to suppress tumors via modulation of multiple anticancer cytokines and anticancer immunity [23–25].

STAT3, Akt and ERK are associated with gastric cancer cell survival and prognosis [26–29]. We found that 5-Fu in combination with hyperthermia greatly reduced the levels of both STAT3 and phosphorylated STAT3 (Figure 3(A–C)). In addition, the STAT3 inhibitor Stattic potentiates 5-Fu cytotoxicity, suggesting that STAT3 activation may play a protective role against 5-Fu cytotoxicity. Meanwhile, both Akt and ERK1/2 were highly activated and phosphorylated in 5-Fu + hyperthermia-treated cells (Figure 3). Similar to the inhibition of STAT3, suppression of Akt and ERK1/2 by GDC-0068 and U0126, respectively, could also enhance 5-Fu cytotoxicity (Figure 3(I–L)). Thus, we suspect that these two survival signals might also play protective roles against 5-Fu-based HIPEC therapy and contribute to therapeutic resistance or cancer recurrence.

Overexpression of MDR1 is positively correlated with poor prognosis of patients with gastric cancer [30,31]. We found that 5-Fu treatment alone increased the levels of MDR1 (Figure 4(A,B)), while hyperthermia blocked 5-Fu-mediated upregulation of MDR1. Gastric cancer patients with high expression of thymidylate synthase typically exhibit low response to 5-Fu/cisplatin chemotherapy [32], and induction of thymidylate synthase is considered to be a common 5-Fu resistance mechanism [33]. In this study, we found that the levels of thymidylate synthase were highly upregulated by 5-Fu alone (Figure 4(A,C)). However, this 5-Fu-mediated elevation of thymidylate synthase could be suppressed by hyperthermia. Thus, our findings revealed that hyperthermia sensitizes gastric cancer cells to 5-Fu, possibly via suppression of proteins related to 5-Fu chemoresistance, such as MDR1 and thymidylate synthase.

Modulation of redox balance is thought to be a promising therapeutic approach for many cancers [34]. Excessive ROS will damage intracellular proteins, nucleic acids, lipids and organelles, which can lead to cell death [34]. Moreover, antioxidant proteins, such as catalase, SOD1 and SOD2, are associated with 5-Fu resistance and poor prognosis in various types of cancer [34–36]. We found that the combination of hyperthermia and 5-Fu effectively reduced the levels of catalase and SOD1 (Figure 4(E,G)) and greatly elevated intracellular levels of ROS (Figure 4(D)). These findings suggest that ROS may play a crucial role in 5-Fu/hyperthermia-mediated cytotoxicity.

PD1/PD-L1 blockade has produced excellent results as a treatment for various types of cancer [37]. Thus, this strategy may be a new hope for gastric cancer with peritoneal metastasis [38]. PD-L1 glycosylation is essential for its stability and its interaction with PD1 [39]. Interestingly, our results showed that although 5-Fu-induced glycosylated PD-L1 was greatly enhanced by hyperthermia (Figure 5(A–C)), the 5-Fu-induced elevation of cell surface PD-L1 was inhibited by hyperthermia (Figure 5(D,E)). These findings suggest that hyperthermia might somehow suppress PD-L1 trafficking to the cell surface. It is widely accepted that high PD-L1 expression can allow cancer cells to evade immune surveillance, and the intensity of PD-L1 expression in gastric cancer tissue is known to be a critical determinant of the therapeutic response to anti-PD1 or anti-PD-L1 therapies [38]. Furthermore, clinical studies have shown that tumors with high PD-L1 expression typically have better response to anti-PD1/PD-L1 therapy [40–42]. Our findings suggest that patients who receive CRS + 5-Fu-based HIPEC may exhibit suppression of PD-L1 at the cell surface, which could contribute to poor response to anti-PD1 or anti-PD-L1 therapies.

Our results also showed that Stattic (Figure 6(A)) and GDC-0068 (Figure 6(B)) both enhance 5-Fu-induced elevation of glycosylated PD-L1, indicating that STAT3 and ERK1/2 may play suppressive roles in PD-L1 glycosylation. In contrast, U0126 attenuated 5-Fu-induced increases in glycosylated PD-L1 (Figure 6(C)), which is suggestive of an inducing role for Akt in PD-L1 glycosylation. Even though STAT3 appears to play a suppressive role in 5-Fu-mediated PD-L1 glycosylation, inhibition of STAT3 does not alter the surface level of PD-L1. In addition, both GDC-0068 (Figure 6(E)) and U0126 (Figure 6(F)) can completely suppress 5-Fu-induced elevation of cell surface PD-L1. These findings strongly suggest that Akt and ERK are involved in enhancing PD-L1 trafficking to the cell surface. Intriguingly, both Akt and ERK are highly activated by 5-Fu/hyperthermia, yet the levels of cell surface PD-L1 are reduced after 5-Fu/hyperthermia treatment. Thus, we speculate that 5-Fu/hyperthermia suppresses PD-L1 trafficking to the cell surface via some other unknown signaling mechanism. Furthermore, this unidentified signaling mechanism appears to be a dominant regulator of PD-L1 trafficking, as compared to STAT3, Akt and ERK.

Together, our results provide new insights into how cancer cells respond to 5-Fu-based HIPEC. The insights gained from our study may be helpful in the development of new strategies to improve the therapeutic efficacy of HIPEC.

Author contributions

Bor-Chyuan Su: conceptualization; writing—original draft. Guan-Yu Chen: investigation; methodology. Chun-Ming Yang: investigation. Wei-Ting Chuang: investigation. Meng-Chieh Lin: investigation. Pei-Ling Hsu: methodology. Chu-Wan Lee: visualization. Chih-Cheng Cheng: data curation. Shih-Ying Wu: writing—review and editing. Bo-Syong Pan: writing—review and editing. Hsin-Hsien Yu: supervision (equal); writing—original draft (equal).

Acknowledgment

The authors thank Marcus Calkins for language editing.

Disclosure statement

The authors declare that they have no competing interests.

Data availability statement

All data generated during this study and included in this published article.

Additional information

Funding

This research was funded by the Ministry of Science and Technology (MOST, Taiwan), MOST 109-2320-B-038-010-MY2; 110-2320-B-038-023. This research was also funded by the Taipei Medical University—Wan Fang Hospital, 110TMU-WFH-20 and 112-wf-eva-21. This research was also funded by the TMU Research Center of Cancer Translational Medicine from the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE), Taipei, Taiwan.