HA15 inhibits binding immunoglobulin protein and enhances the efficacy of radiation therapy in esophageal squamous cell carcinoma

Abstract

Proteomic profiling is a promising approach to identify novel predictors of radiation response. The present study aimed to identify potential biomarkers of radiation response by serum proteomics in esophageal squamous cell carcinoma (ESCC) patients and find efficacious therapeutic drugs to enhance the efficacy of radiation therapy (RT). Serum binding immunoglobulin protein (BIP) was identified and validated as a treatment response predictor in ESCC patients treated with RT. Novel BIP inhibitor HA15 showed antitumor activity in ESCC cells by viability assay. Tumor cell colony formation and apoptosis assay revealed targeting BIP was associated with significant improvements of radiation sensitivity. Further analyses revealed that HA15 enhanced radiation‐induced endoplasmic reticulum (ER) stress and immunogenic cell death (ICD) in ESCC. Clinical data indicated that high expression of BIP was associated with poor survival in patients of ESCC. In conclusion, proteomics analysis suggested BIP was a promising predictor of radiation response in locally advanced ESCC. The BIP inhibitor HA15 acted as an ER stress inducer and ICD stimulator; RT combined with HA15 was effective in suppressing the growth of ESCC in vitro and in vivo. Pretreatment BIP was an essential prognostic biomarker in locally advanced ESCC patients treated with RT.

Proteomic profiling and in vitro analyses revealed binding immunoglobulin protein (BIP) regulates radiation resistance by alleviating endoplasmic reticulum (ER) stress. Targeting BIP with HA15 enhances ER stress and immunogenic cell death, thereby improving the efficacy of radiation therapy. BIP played a crucial role in predicting clinical response and survival for esophageal squamous cell carcinoma patients treated with radiation therapy.

Abbreviations

5‐FU

fluorouracil

ATF6

activating transcription factor 6

AV

annexin V

BIP

binding immunoglobulin protein

CALR

calreticulin

CHOP

C/EBP homologous protein

CR

complete response

DDP

cisplatin

ER

endoplasmic reticulum

ERAD

ER‐associated degradation

ESCC

esophageal squamous cell carcinoma

GO

Gene Ontology

HMGB1

high mobility group box protein 1

ICD

immunogenic cell death

IHC

immunohistochemical

IRE1α

inositol‐requiring enzyme 1α

MS

mass spectrometry

OS

overall survival

PD

progressive disease

p‐eIF2α

phospho‐eukaryotic initiation factor 2α

PI

propidium iodide

p‐PERK

phospho‐protein kinase RNA‐like ER kinase

PR

partial response

RIP1

receptor‐interacting protein 1

RT

radiation therapy

SD

stable disease

UPR

unfolded protein response

XBP‐1(s)

X‐box binding protein 1

1. INTRODUCTION

Esophageal cancer is one of the most frequently diagnosed gastrointestinal cancers across the world. ¹ The two most common type of esophageal cancers are adenocarcinomas and squamous cell carcinomas. ² Adenocarcinoma is the most prevalent type of esophageal cancer in the western world, and squamous cell carcinoma is the most common form of cancer of the esophagus in East Asia. ³ , ⁴ Generally, detecting and treating cancer in the early stage can lead to a cure. However, esophageal cancer often displays little or no symptoms in the early phase, and numerous newly diagnosed individuals were in locally advanced stage or advanced stage. ⁵ , ⁶ Radiation therapy is a type of cancer treatment that uses high‐energy beams to eradicate tumor cells. Over the last few decades, RT is an important therapeutic strategy in the management of locally advanced ESCC or advanced ESCC. ⁵

For esophageal cancer patients who were eligible for RT, the 5‐year OS rate ranged from 5% to 50%. ³ Due to the existence of radiation resistance, approximately 50% of recurrence was observed within the gross tumor volume for locally advanced ESCC treated with RT. ⁷ As RT becomes increasingly administered, further investigation of the underlying molecular mechanisms of radiation resistance and the precise selection of patients that could benefit from this treatment protocol are vitally important. In recent years, various cutting‐edge technologies such as genomic, transcriptomic, and epigenomic tools have been utilized to investigate the intrinsic radiation sensitivity of tumor cells. ⁸ , ⁹ Although multiple predictors of cancer radiation resistance have been proposed, these results do not always correlate well with protein expression level.

Unlike the genome, the physiological or pathological processes in the organism could be directly reflected with protein expression. Proteomics could provide significant biological information of the processes underlying healthy and diseased cellular processes at the protein level. Current proteomic tools allow large‐scale screening and identification of proteins, the study of their functions, and identification of protein–protein interactions. ¹⁰ Mass spectrometry has become one of the powerful analytical techniques for large‐scale protein study in proteomics. ¹¹ Recent findings have revealed that proteomics might be helpful to identify potential biomarkers of radiation response. ¹² A group from Japan reported several predictors of radiation sensitivity for esophageal cancer, however, most of the patients eligible for analysis were treated for adenocarcinoma, with very little emphasis placed on ESCC. ¹³

The mechanism of radiation response is incompletely elucidated in ESCC and a better understanding of this will allow investigators to develop new treatment strategies to kill cancer cells. ¹⁴ In the present study, we used serum MS to detected potential radiation response predictors in patients of ESCC. Moreover, we attempted to clarify the mechanism of radiation sensitivity, to screen possible small molecule inhibitors to improve the radiation sensitivity of ESCC both in vitro and in vivo.

2. MATERIALS AND METHODS

Proteome profiling analysis, biomarker verification, and mechanism investigation are described in Appendix S1.

3. RESULTS

3.1. Serum MS identifies ER stress involved in regulating radiation response in ESCC

To identify effective biomarkers predicting clinical response to RT, we undertook serum‐based MS from patients who were initially diagnosed with ESCC. Briefly, the patients included in the present analysis had locally advanced, inoperable, but nonmetastatic ESCC (medically inoperable primary disease, poor surgical candidates, and those who refused surgery); serum samples were collected prior to anticancer treatment. All of the patients were treated with standard intensity‐modulated RT combined with chemotherapy. The chemotherapy regimen comprised four courses of 5‐FU and DDP. The total doses of RT were 59.4–66 Gy in 33 daily fractions (1.8–2 Gy/fraction) over 6 weeks, the first week concurrent with chemotherapy. Treatment protocols are elucidated in Figures 1A and S1A–D. Tumor response was evaluated after a full course of chemoradiotherapy (Figure 1B). The general characteristics of included patients are shown in Table S1. Of the 112 patients eligible for analysis, 8.9% patients achieved CR, 58.9% PR, 25.9% had SD, and 6.3% showed PD (Figure 1C). Mass spectrometry‐based serum proteomics analyses were undertaken to identify protein signatures for the prediction of treatment response in ESCC patients (Figure 1D–F). Interestingly, comprehensive proteomic profiling revealed a number of proteins in serum that were differentially expressed between the CR group and PD group. More than 800 proteins were detected and relatively quantified, 86 proteins were upregulated and 62 proteins were downregulated (Figures 1D and S1E). Both GO term enrichment and Kyoto Encyclopedia of Genes and Genomes pathways were carried out; the two analytical approaches showed enrichment of ER stress‐related proteins and complement‐related proteins (Figure 1E,F, Tables S2 and S3). As complements had been recognized as a central mediator of RT‐promoted tumor‐specific immunity and clinical response, ¹⁵ we focused on abnormally expressed proteins that involved in ER stress.

FIGURE 1.

Serum samples were collected from esophageal squamous cell carcinoma (ESCC) patients prior to anticancer treatment; proteomic analysis was carried out to identify potential biomarkers for predicting treatment response. (A) Flowchart of the chemoradiotherapy (CRT) protocol for ESCC patients. (B, C) Treatment responses were evaluated in the included ESCC patients. (D) Volcano plot of differentially expressed proteins in the complete response (CR) group compared with the progressive disease (PD) group; depicting protein data p values vs. fold change, red points represent data points with p < 0.05 and a fold rate >1.2. (E) Gene Ontology (GO) analysis of differentially expressed proteins in the CR group compared with the PD group. GO analysis was undertaken according three terms: molecular function, biological process, and cellular component. (F) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. (G) Pretreatment levels of binding immunoglobulin protein (BIP) in serum from patients with ESCC undergoing CRT were measured by ELISA. (H) The Cancer Genome Atlas database analysis of substantial expression of heat shock protein A5 (BIP) between nontumorous tissues, esophageal adenocarcinoma tissues, and ESCC tissues. *p ˂ 0.05; **p ˂ 0.01; ***p ˂ 0.001; ****p ˂ 0.0001. CT, chemotherapy; n.s., not significant (p > 0.05); PR, partial response; RT, radiation therapy; SD, stable disease

3.2. High level of BIP in serum is correlated with poor response in ESCC patients treated with RT

Considering the best scoring and the relative intensities, BIP was selected for further validation as a key regulator of ER stress. Pretreatment serum BIP level was measured by ELISA in all the included patients. As shown in Figure 1G, upregulated BIP expression was observed in the pretreatment serum samples of poor responders (both the SD and PD groups). In contrast, the level of serum BIP was lower in the pretreatment serum samples of good responders (both the CR and PR cohorts) than the poor responders. There was insignificant difference between the CR and PR groups; similarly, no significant difference was achieved in the SD group compared to the PD group. We also investigated the gene expression profiling data in The Cancer Genome Atlas. The results indicated a higher expression of heat shock protein A5, which encodes BIP, in ESCC than in normal tissues (Figure 1H). Overall, serum proteomic profiling indicated that ER stress was involved in regulating treatment efficacy, and pretreatment BIP levels were predictive of ESCC patient response to RT.

3.3. HA15 targets BIP and attenuates tumor cell proliferation

With respect to BIP as an essential mediator of ER stress, we screened the anticancer compound library and identified numerous compounds targeting BIP. The top 10 candidate BIP inhibitors identified through this screening are shown in Table S4. We found HA15 was the strongest BIP inhibitor. HA15 was initially developed from thiazolidinedione, this novel compound exerted anticancer effects in several malignant tumors by suppressing BIP. ¹⁶ However, the effect of HA15 on human ESCC cell lines has not been evaluated. Thus, we measured BIP expression in various ESCC cell lines and a normal esophageal epithelia cell line, SHEE. Binding immunoglobulin protein was commonly overexpressed in ESCC cell lines compared with the SHEE cell line, and KYSE510 was associated with the highest expression of BIP (Figure 2A,B). Cell viability assay was then carried out. We found a preferential cytotoxic activity of HA15 against ESCC cells, and the effect was dose‐dependent (Figure 2C,I). Conversely, no deleterious effects were noticed in SHEE cells treated with HA15 (Figure 2J). Although there was a weak correlation between BIP expression and HA15 sensitivity, this compound was effective in suppressing ESCC cells. HA15 at 10 μM is safe with minimal adverse effects and was applied in the subsequent analysis. Accordingly, the unique role of HA15 in suppressing ESCC cells while sparing normal cells made it a promising inhibitor for anticancer treatment.

FIGURE 2.

Binding immunoglobulin protein (BIP) expression in multiple esophageal squamous cell carcinoma (ESCC) cell lines. Targeting BIP with HA15 inhibits the proliferation of ESCC cells, however, there is limited toxicity in normal cells. (A) The expression of BIP in multiple ESCC cells and normal esophageal epithelial cells (SHEE). (B) Quantification of BIP expression from (A). (C–I) Dose–survival curves of HA15 on multiple ESCC cell lines (KYSE30, KYSE70, KYSE140, KYSE150, KYSE410, KYSE450, and KYSE510) were estimated by MTT assay at 24 h. (J) Dose–survival curves of HA15 on normal esophageal epithelial cells were estimated by MTT assay at 24 h. Data are means ± SD (n = 3). *p ˂ 0.05; **p ˂ 0.01; ***p ˂ 0.001; ****p ˂ 0.0001. n.s., not significant (p > 0.05)

3.4. HA15 sensitizes RT in ESCC with high expression of BIP

We then estimated whether BIP inhibitor could be utilized as an RT‐boosting medication in eradicating ESCC cells by clonogenic assays. As presented in Figure 3A,B, HA15 significantly reduced the surviving fraction of KYSE70 cells and KYSE510 cells compared with RT alone. HA15 seems much more efficient in KYSE70 cells than KYSE510 cells when combined with a single dose of 2 Gy RT. In the meantime, we also conducted siRNA interference experiments to evaluate the efficacy of targeting BIP in improving radiation sensitivity. As shown in Figure 3C–F, knockdown of BIP has limited side‐effects on cell viability. Consistently, inhibiting BIP combined with RT significantly attenuated survival fractions in both KYSE510 and KYSE70 cells (Figure 3G,H).

FIGURE 3.

Targeting binding immunoglobulin protein (BIP) enhances the sensitivity of radiotherapy (RT) in esophageal squamous cell carcinoma cells. (A) Colony formation assay was performed in KYSE70 cells following RT (0, 2, 4, 6, 8, 10, and 12 Gy), or RT + HA15 treatment. (B) Colony formation assay was performed in KYSE510 cells following RT (0, 2, 4, 6, 8, 10, and 12 Gy), or RT + HA15 treatment. (C) Quantitative RT‐PCR of KYSE70 cells infected with siRNAs targeting BIP. (D) Dose–survival curves of KYSE70 cells infected with siRNAs targeting BIP were evaluated by MTT assay at 24 h. (E) Quantitative RT‐PCR of KYSE510 cells infected with siRNAs targeting BIP. (F) Dose–survival curves of KYSE510 cells infected with siRNAs targeting BIP were evaluated by MTT assay at 24 h. (G) Colony formation assay was performed in KYSE70 cells infected with siRNAs targeting BIP. (H) Colony formation assay was conducted in KYSE510 cells infected with siRNAs targeting BIP. Data are means ± SD (n = 3). *p ˂ 0.05; **p ˂ 0.01; ***p ˂ 0.001; ****p ˂ 0.0001. n.s., not significant (p > 0.05); siNC, negative control siRNA.

To further confirm the efficacy of suppressing BIP in enhancing radiation response, clonogenic assays were undertaken in ESCC cells with low expression of BIP. In comparison to RT alone, 10 μM HA15 was inefficient to decrease the survival fractions in both KYSE150 and KYSE450 cells when combined with RT (Figure S1F,G). Taken together, our findings indicated that targeting BIP was an appealing strategy to boost the strength of RT against ESCC cells.

3.5. Binding immunoglobulin protein inhibitor HA15 combined with RT induces tumor cell apoptosis

Radiation therapy was capable of activating stress adaptation responses that involve the ER. ¹⁷ Investigators have paid attention to alterations that occur in ER stress. ¹⁸ Briefly, if ER stress is prolonged and tumor cells fail to restore ER protein homeostasis, apoptotic cell death ensues. ¹⁶ Hence, we examined whether inhibition of BIP with HA15 causes cell death by triggering apoptosis. As seen in Figures 4A,B and S2, blocking BIP with small molecule inhibitor HA15 induced AV and PI double staining in ESCC cells. This indicated that targeting BIP decreased protein homeostasis and caused ER stress‐mediated apoptosis. Radiation therapy‐triggered ER stress was also detected in the current analysis (Figures 4A,B and S2). Additionally, the combination of RT and BIP inhibitor resulted in a higher apoptotic cell death rate than RT alone in tumor cells, indicating a synergistic effect of BIP inhibitor and RT. These results showed that both BIP inhibitor and RT were able to trigger apoptosis in tumor cells.

FIGURE 4.

Binding immunoglobulin protein (BIP) inhibitor HA15 enhanced the efficacy of radiotherapy (RT)‐induced apoptosis in KYSE70 cells, however, this compound is ineffective in SHEE cells. BIP inhibitor combined with RT enhanced and prolonged the activation of endoplasmic reticulum (ER) stress. (A) Annexin V/propidium iodide staining was carried out following 24 h of DMSO (negative control [NC]), HA15 (10 μM), RT (2 Gy), or RT/HA15 treatment in KYSE70 cells. (B) Quantification of cell apoptosis from (A). (C) KYSE510 cells were treated with vehicle, HA15 (10 μM), RT (2 Gy), or RT/HA15, and the ER stress markers were analyzed by western blot. (D, E) Quantification of ER stress markers from (C). Data are means ± SD. *p ˂ 0.05; **p ˂ 0.01; ***p ˂ 0.001; ****p ˂ 0.0001. ATF6, activating transcription factor 6; CHOP, C/EBP homologous protein; IRE1α, inositol‐requiring enzyme 1α; HSP90, heat shock protein 90; n.s., not significant (p > 0.05); p‐eIF2α, phospho‐eukaryotic initiation factor 2α; p‐PERK, phospho‐protein kinase RNA‐like ER kinase; RT, radiation therapy; XBP‐1(s), X‐box binding protein 1

We then evaluated the apoptotic cell death rate in SHEE cells. Although RT was efficient in triggering apoptosis, the extent of AV and PI double staining in the combined group (RT/HA15) remained similar to that in the RT group (Figure S3). The outcome confirmed there was limited toxicity of HA15 on normal cells.

Overall, our data clearly illustrated that targeting BIP improved radiation response by enhancing apoptosis in ESCC cells, and this enhancement was specific for cancer cells.

3.6. HA15 combined with RT cause enhanced and prolonged ER stress

Upon ER stress, the influx of new proteins into the ER were prevented. The two cascades of signaling pathways, including UPR and ERAD, were activated to restore protein homeostasis and alleviate ER stress. ¹⁹ Among them, the three primary modulators of UPR, PERK, IRE1α, and ATF6, contributed to increase the amount of ER chaperones (such as BIP), improve folding capacity, and avoid ER stress‐induced cell apoptosis. ¹⁸ The ERAD was activated and promoted the cells’ capacity to clear misfolded proteins. ²⁰ However, severe ER stress could result in CHOP‐mediated cell apoptosis. ¹⁸ Thus, we hypothesized that suppressing BIP, the key regulator of ER stress, might cause enhanced and prolonged ER stress, eventually resulting in CHOP‐mediated apoptotic cell death.

We studied the key regulators of ER stress by western blot analysis of ESCC cells treated with HA15. As shown in Figure 4C–E, HA15 induced amplification of UPR indicators including p‐PERK, p‐eIF2α, IRE1α, XBP1(s), and CHOP. The results indicated that HA15 was able to trigger ER stress. Meanwhile, we evaluated the effect of RT on ER stress at the same time, and a moderate increase of UPR markers was observed. Unsurprisingly, the expression of UPR markers had strongly increased after the combination of HA15 and RT. We also detected a significant amplification of CHOP, which implied enhanced tumor cell apoptosis. These data suggested that RT combined with BIP inhibitor was capable of inducing enhanced and prolonged UPR, thereby causing irresolvable ER stress and CHOP‐mediated cell death.

Collectively, BIP was able to regulate treatment response by alleviating radiation‐induced ER stress; HA15 activated UPR and acted as a novel radiosensitizer that weakened ESCC cells by triggering ER stress. The combination of RT and BIP inhibitor prolonged intracellular ER stress and upregulated CHOP‐mediated apoptosis.

3.7. Targeting BIP combined with RT triggers ICD

Immunogenic cell death is essential for antitumor immunity. ²¹ The extracellular release of ATP and HMGB1 from dying tumor cells, and the translocation of CALR from ER to cell surface, have been well‐recognized as ICD‐associated biomarkers. ²² Both HMGB1 and ATP served as danger‐associated molecular patterns, and ICD contributes to the maturation of dendritic cells and the stimulation of cytotoxic T cell‐mediated adaptive immunity. In addition to simply eradicating tumor cells, high dose radiation‐induced ICD has been depicted in several types of cancers; nevertheless, there was limited efficacy for moderate dose RT (<6 Gy) in triggering ICD. ²³ , ²⁴

The application of ICD inducer is a promising strategy to improve antitumor immune response of moderate dose RT. Chemotherapy medicines such as doxorubicin and oxaliplatin have been well recognized as ICD inducers. ²⁵ , ²⁶ The esophagus is a muscular tube, and it is difficult to deliver a single high dose RT for ESCC. ²⁷ 5‐Fluorouracil and DDP were the widely utilized cytotoxic drugs in ESCC. To date, both 5‐FU and DDP remain controversial in stimulating ICD in tumor models. ²⁸ , ²⁹ , ³⁰ The launch of novel ICD inducers could contribute to overcoming radiation resistance by improving the antitumor immune response for ESCC. Previous analysis revealed the induction of sustained ER stress was capable of eliciting ICD. ³¹ Therefore, we estimated the immunogenicity of cancer cells after treatment with BIP inhibitor and/or RT.

As shown in Figure 5A–F, RT‐induced CALR cell surface exposure, extracellular ATP, and HMGB1 release were observed in both KYSE70 and KYSE510 cells. This effect appeared to occur in a dose‐dependent manner, and all the ICD markers increased rapidly after high doses of RT. In addition, targeting BIP was accompanied by ICD in ESCC cells. After treated with 10 μM HA15, the number of relative light units was substantially higher in the combined group compared with untreated control group (Figure 5A,B). The outcomes implied that radiation‐triggered ATP release was enhanced by BIP inhibitor. In agreement with the increased secretion of ATP, the amount of CALR on the cell surface was significantly boosted by HA15 and RT (Figure 5C,D). Furthermore, there was a synergistic effect between BIP inhibitor and RT in inducing the secretion of HMGB1 from tumor cells to extracellular space, especially for an increased concentration of HA15 (Figure 5E,F).

FIGURE 5.

Suppressing binding immunoglobulin protein enhanced radiation‐mediated ICD in esophageal squamous cell carcinoma cells, the potential molecular mechanism was estimated. (A) Extracellular ATP release was measured in KYSE70 cells treated with HA15, radiation therapy (RT), or RT/HA15. (B) Extracellular ATP release was measured in KYSE510 cells treated with HA15, RT, or RT/HA15. (C) Calreticulin (CALR) cell surface exposure was measured in KYSE70 cells treated with HA15, RT, or RT/HA15. (D) CALR cell surface exposure was measured in KYSE510 cells treated with HA15, RT, or RT/HA15. (E) Extracellular high mobility group box protein 1 (HMGB1) release was measured in KYSE70 cells treated with HA15, RT, or RT/HA15. (F) Extracellular HMGB1 release was measured in KYSE70 cells treated with HA15, RT, or RT/HA15. (G) KYSE510 cells were treated with vehicle, HA15 (10 μM), RT (2 Gy), or RT/HA15, and the markers of cell apoptosis (cl‐caspase3 and cl‐caspase 8), necrosis (receptor‐interacting protein 1; RIP1), and autophagy (LC3A/B II) were analyzed by western blot. Data are means ± SD. *p ˂ 0.05; **p ˂ 0.01; ***p ˂ 0.001; ****p ˂ 0.0001. HSP90, heat shock protein 90; NC, negative control; n.s., not significant (p > 0.05); RLU, relative light unit

To validate the effect of cytotoxic drugs in inducing ICD, ESCC cells were incubated with 5‐FU or DDP. We found the production of both HMGB1 and ATP in extracellular space after treatment with 5‐FU, and there was a synergistic effect between 5‐FU and RT; however, 5‐FU has no effect on the translocation of CALR to the cell membrane (Figure S4). Consistently, RT‐induced ATP and HMGB1 liberation could be enhanced by DDP, and an unaltered exposure of CALR to the cell surface was detected after treatment (Figure S5). Consequently, HA15 could be utilized as an ICD inducer in ESCC cells. This compound was much more efficient than 5‐FU or DDP in inducing ICD.

3.8. Mechanisms of HA15 combined with RT in inducing ICD

It was reported that both ER stress and apoptosis are involved in CALR cell surface translocation, and that CALR on the cell membrane acted as an “eat me” signal for dendritic cells. ²⁴ Receptor‐interacting protein 1 is a key regulator of tumor cell necrosis. ³² After exposure to RT, HMGB1 was released from tumor cells undergoing necrosis. ³³ Additionally, ATP extracellular release was regulated by autophagic machinery. ³⁴ Next, we analyzed the correlation between the major forms of cell death and ICD.

As observed in Figure 5G, suppression of BIP was capable of triggering cancer cell apoptosis (upregulation of cl‐caspase3 and cl‐caspase 8), necrosis (overexpression of RIP1), and autophagy (high expression of LC3A/B II). Radiation‐stimulated tumor cell apoptosis, necrosis, and autophagy were also detected. As expected, HA15 combined with RT resulted in enhanced tumor cell death. In total, targeting BIP synergistic with RT was highly efficient in triggering ICD by promoting cell apoptosis, necrosis, and autophagy.

3.9. HA15 enhances the therapeutic efficacy of RT in mice

To confirm the therapeutic potential of BIP inhibitor, we analyzed the treatment efficacy of RT combined with HA15 in vivo (Figure 6A). The results of the animal study (Figure 6B–E) showed that untreated tumor‐bearing mice developed fast‐growing tumors, and HA15 alone led to a moderate inhibition of tumor development. In contrast, RT could significantly reduce tumor burden compared to the untreated control or HA15 alone group. In parallel with in vitro observations, the combination of RT and HA15 dramatically decreased tumor volume compared with single treatments. Moreover, HA15 was associated with limited toxicity in mice. These data clearly uncovered the antitumor activity of HA15 in vivo and confirmed that BIP inhibitor was a promising radiosensitizer in ESCC.

FIGURE 6.

HA15 or/and radiation therapy (RT) inhibited patient‐derived xenograft (PDX) tumor growth in vivo. Immunohistochemical (IHC) staining indicated low expression of binding immunoglobulin protein (BIP) was associated with favorable survival in esophageal squamous cell carcinoma (ESCC). (A) Flowchart of the irradiation protocol for the PDX (3 cases of ESCC) mouse model. (B) HA15 (30 mg/kg), RT (2 Gy*5f), or RT/HA15 decreased the tumor volume compared to vehicle treatment. (C) HA15 had no effect on mice body weight. (D) Tumor weight was decreased in HA15, RT, or RT/HA15 groups compared to vehicle group. (E) Representative photographs of tumors. (F) BIP expression in normal esophageal epithelial cells (I) and ESCC tissues (II–VI). (G) Statistical analysis of IHC staining intensity; low expression group (n = 50), high expression group (n = 62). (H) Low BIP expression in ESCC patients treated with RT is associated with a favorable overall survival. *p ˂ 0.05; **p ˂ 0.01; ***p ˂ 0.001; ****p ˂ 0.0001. n.s., not significant (p > 0.05)

3.10. Overexpression of BIP in tumor tissues is correlated with poor survival

Given all the above data, BIP conferred radiation resistance in tumor cells. Previous studies reported that BIP expression was associated with patient survival in malignant tumors. ³⁵ , ³⁶ To identify subsets of ESCC that were sensitive or resistant to RT, we analyzed BIP expression in tumor tissues by IHC. Generally, BIP was broadly upregulated in ESCC tumor tissues compared with normal epithelium in esophagus (Figure 6F). Among the included ESCC samples, 62 patients were associated with high expression of BIP, the other 50 patients were detected with low expression of BIP (Table S5). High BIP expression was associated with a shorter OS in patients with ESCC compared with the low BIP expression group (18.3 months vs. 29.6 months, respectively, p = 0.036; Figure 6G,H). The 1‐year and 3‐year OS were 82.1% and 40.8%, respectively, in the low BIP expression group, and 69.6% and 23.4%, respectively, in the high BIP expression group (Figure 6H). These observations were consistent with the results both in vitro and in vivo.

Accordingly, our results convincingly suggested that BIP could be utilized as a prognostic factor in patients with locally advanced ESCC treated with definitive chemoradiotherapy. For patients with high BIP expression, targeting BIP could be an encouraging strategy to improve their survival.

4. DISCUSSION

In the present analysis, clinical serum sample‐based proteomics revealed that BIP was involved in regulating treatment response through ER stress in ESCC patients treated with RT. The expression of BIP was varied in multiple ESCC cell lines and a normal esophageal epithelial cell line. Binding immunoglobulin protein conferred radiation resistance of tumor cells by activating stress adaptation responses in ER. Consistently, targeting BIP with novel small molecular HA15 caused apoptotic cell death and ICD, thereby improving the efficacy of RT. Additionally, BIP inhibitor was capable of enhancing radiation‐induced ER stress and ICD in tumor cells, confirming that the combination of RT and HA15 was a promising strategy in ESCC. Additionally, our data described BIP as a prognostic factor in patients of ESCC, with high expression of BIP associated with poor survival.

Persistent ER stress has been depicted as one of the hallmarks of cancer. ¹⁹ The current study revealed ER stress was involved in regulating radiation response: under weak or mild ER stress conditions, the signaling pathways prevented tumor cell death by upregulating BIP. Binding immunoglobulin protein as a major ER chaperone was essential in acquiring malignant characteristics of tumor, and high expression of BIP has been observed in various types of human tumors. ³⁷ Yang et al. reported that BIP was a key promotor of tumor cell proliferation during RT. ³⁸ Consistent with previous findings, our analysis showed that high levels of serum BIP were associated with poor response to RT in ESCC patients; moreover, overexpression of BIP in tumor tissues correlated with unfavorable survival. These data implied that BIP could be a novel target in the generation of appropriate strategies for improving treatment efficacy and patient prognosis. Of note, cell surface BIP acted as a receptor for viral entry and signal transduction, whereas intracellular BIP was involved in regulating ER stress. ³⁹ To date, the role of serum BIP in stress response has not been fully understood, the relationship between serum level and IHC expression of BIP was poorly defined, and further investigations are needed.

As BIP has been described as an attractive therapeutic target, investigators were dedicated to developing innovative anticancer drugs. It has been shown that BIP can be expressed on the tumor cell surface, and Ab targeting BIP showed anticancer activity and improved the efficacy of RT. ⁴⁰ Ruthenium‐based BIP inhibitors showed potent cytotoxicity against malignant tumors by inducing extensive ER stress. ⁴¹ Ruthenium complexes were effective in human colorectal cancer cells and were associated with enhanced efficacy of RT. ⁴² Importantly, several anticancer ruthenium drugs, including KP1019, BOLD‐100, NAMI‐A, and TLD‐1433, have entered phase I clinical trials, and the results were encouraging. ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ Furthermore, recent findings indicated a marked anticancer effect of HA15 by targeting BIP to increase ER stress, and this novel compound also showed the ability to overcome BRAF inhibitor resistance in melanoma cells. ¹⁶ In our analysis, the combination of HA15 and RT caused irreversible ER stress and enhanced ICD in ESCC (Figure 7). The development of radiation resistance is a major obstacle in anticancer therapy; HA15 inhibits the proliferation of ESCC, including those types of ESCC cells with acquired resistance to radiation. This novel bioactive compound provided alternatives for improve the efficacy of RT by triggering ER stress.

FIGURE 7.

Targeting binding immunoglobulin protein (BIP) (HA15) enhances radiation induced endoplasmic reticulum (ER) stress and immunogenic cell death (ICD) in tumor cells, thereby improving the response to radiation therapy (RT) and eliciting antitumor immunity. ATF6, activating transcription factor 6; CALR, calreticulin; CHOP, C/EBP homologous protein; DAMP, danger‐associated molecular pattern; ERAD, ER‐associated protein degradation; HMGB1, high mobility group box protein 1; IRE1α, inositol‐requiring enzyme 1α; p‐eIF2α, phospho‐eukaryotic initiation factor 2α; p‐PERK, phospho‐protein kinase RNA‐like ER kinase; UPR, unfolded protein response; XBP‐1(s), X‐box binding protein 1

Radiation stimulating tumor‐specific immune response exerted antineoplastic effects at the systemic level, also known as the abscopal effect. ⁴⁷ However, the abscopal effect was mainly seen in tumors treated with high dose RT. ⁴⁸ For tumors (e.g., esophageal cancer) treated with moderate or low dose RT, the abscopal‐like reactions were seldom documented. In our analysis, there was limited effect of moderate dose RT in eliciting ICD compared with high dose RT. HA15 could be utilized as an ICD inducer, as the combination of HA15 and RT contributed to the establishment of a peritumoral pro‐immunogenic milieu. In this setting, RT combined with ICD inducers might yield optimal therapeutic response.

Despite the high visibility that has been given to the impact of BIP inhibitor on ER stress‐mediated radiation efficacy, there are several interesting questions worthy of further evaluation. First, validation of complement‐related proteins and ER stress markers as predictive factors of treatment response will be required in a large independent patient cohort, and prospective validation is preferred. Second, the intrinsic and extrinsic mechanisms underlying ICD remains to be elucidated. Given the antitumor activity of HA15 in ESCC, further preclinical experiments and clinical trials are needed. Additionally, under the premise of maintaining local tumor control, the application of anti‐immunosuppressive and anti‐inflammatory medicines for managing postradiation symptoms should be investigated.

In summary, our data suggested BIP‐mediated alleviation of ER stress played an essential role in radiation response. Novel BIP inhibitor HA15 synergized with RT was able to improve treatment efficacy by inducing irreversible ER stress and enhanced ICD. Pretreatment BIP could be considered as a potential biomarker for predicting treatment response and prognosis in ESCC patients treated with RT.

AUTHOR CONTRIBUTIONS

H.L. and H.G. conceived the idea and designed the study. H.L. performed most of the experiments, analyzed data, and wrote the manuscript. L.‐X.W., D.‐J.Z., Y.‐N.S., S.‐J.W., and S.S. provided help for in vivo and in vitro experiments. H.G. supervised the entire project. The author(s) read and approved the final manuscript.

FUNDING INFORMATION

This work was supported by the National Natural Science Foundation of China (no. 81773230), the Science and Technology Research Plan of Henan (no. 212102310619), the Medical Science and Technology Research Program of Henan (no. LHGJ20210171), and the Bethune Cancer Radiotherapy Translational Medicine Research Program (no. flzh202115).

CONFLICT OF INTEREST

The authors declare that they have no competing interests.

ETHICS STATEMENT

Approval of the research protocol by an institutional review board: The study was approved by the Institutional Review Board of the Affiliated Cancer Hospital of Zhengzhou University Research (182106000062).

Informed consent: The authors declare that the collection of tissue samples and clinicopathologic data were obtained from patients at the Affiliated Cancer Hospital of Zhengzhou University with written informed consent.

Registry and registration no. of the study/trial: N/A.

Animal studies: The authors declare that the mice were cared for in strict accordance with the institutional guidelines for animal care and approved by the Committee on the Ethics of Animal Care and Use of Zhengzhou University.

Supporting information

Appendix S1.

Luo H, Wang L, Zhang D, et al. HA15 inhibits binding immunoglobulin protein and enhances the efficacy of radiation therapy in esophageal squamous cell carcinoma. Cancer Sci. 2023;114:1697‐1709. doi: 10.1111/cas.15712