Identifying adipocyte-derived exosomal miRNAs as potential novel prognostic markers for radiotherapy of esophageal squamous cell carcinoma

Abstract

BACKGROUND

Radiation resistance limits radiotherapy efficacy in esophageal squamous cell carcinoma (ESCC). The tumor microenvironment, particularly adipocytes, plays a role in promoting cancer progression. Extracellular vesicles and microRNAs (miRNAs) regulate gene expression and hold prognostic potential for esophageal carcinoma. Elucidating radioresistance mechanisms and identifying radiosensitization targets can help enhance radiotherapy efficacy for esophageal cancer.

AIM

To investigate the potential role of miRNAs derived from adipocyte exosomes as prognostic markers for radiotherapy efficacy in ESCC.

METHODS

Free adipocytes were isolated from human thoracic adipose tissue. A co-culture model of adipocytes and ESCC cells was established to observe colony formation and cell survival post-irradiation. ESCC cell apoptosis was assessed by flow cytometry. Western Blot and immunofluorescence assays were performed to evaluate DNA damage in ESCC cells post-irradiation. Adipocyte-derived exosomes were isolated by ultracentrifugation and identified by electron microscopy. A similar set of experiments was performed on ESCC cells to analyze cell survival, apoptosis, and DNA damage post-radiation exposure. Exosomes from adipose tissue and serum exosomes from ESCC patients pre- and post-radiotherapy were subjected to high-throughput miRNA-sequencing and validated using real-time quantitative polymerase chain reaction. The correlation between potential target miRNAs and the short-term prognosis of radiotherapy in ESCC was evaluated by receiver operating characteristic curve analysis.

RESULTS

Co-culturing adipocytes with ESCC cells enhanced radioresistance, as evidenced by increased colony formation. Adipocyte co-culture reduced ESCC cell apoptosis and DNA damage post-radiation. Adipocyte-derived exosomes similarly conferred radioresistance in ESCC cells, decreasing apoptosis and DNA damage post-irradiation. High-throughput miRNA-sequencing identified miR-660-5p in serum and adipose tissue exosomes. Patients with high expression of serum exosome miR-660-5p showed poor prognosis after radiotherapy.

CONCLUSION

Adipocyte-derived exosomal miR-660-5p is a potential biomarker for evaluating radiotherapy efficacy in ESCC.

Key Words: Esophageal squamous cell carcinoma; Adipocyte; Exosomes; MicroRNA; Radiotherapy

Core Tip: This study revealed for the first time the impact of adipocyte-derived extracellular vesicle microRNAs on the radiosensitivity of esophageal squamous cell carcinoma (ESCC), providing valuable resources for studying the tumor microenvironment and ESCC, as well as predicting patient responses to radiotherapy.

INTRODUCTION

Esophageal cancer is a highly prevalent malignancy worldwide. In China, it accounts for half of all newly diagnosed cancer cases[1]. Esophageal cancer is mainly categorized into esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EA), with ESCC being the predominant type in Asian populations[2,3]. Comprehensive treatment, primarily involving radiotherapy, is the standard therapeutic approach for locally advanced esophageal cancer. However, over half of patients experience treatment failure due to local and intra-field cancer recurrence[4,5]. Radioresistance is the main factor limiting radiotherapy effectiveness in esophageal cancer cells[6,7]. Notably, both cancer cells and the surrounding tumor microenvironment influence the radiosensitivity of esophageal cancer.

Obesity is recognized as a significant factor contributing to cancer morbidity and progression[8]. Adipocytes, the main constituents of adipose tissue, reside in the tumor microenvironment and promote the occurrence and development of various cancers. Research indicates that adipocytes facilitate tumor progression through soluble factors, such as leptin or interleukin-6, and extracellular matrix remodeling[9,10]. Cancer-associated adipocytes (CAA) accompanying tumors enhance the invasive capacity of cholangiocarcinoma by supplying free fatty acids and adipokines to tumor cells[11]. However, the impact of adipocytes in the tumor microenvironment on the radiosensitivity of ESCC and the underlying mechanisms are not well characterized in contemporary literature.

Adipose tissue functions as an endocrine organ, releasing adipocyte-specific bioactive elements that regulate various physiological and pathological processes[12]. Nevertheless, the capacity of adipocytes to remotely modulate tumor behavior is not clear. We hypothesize that exosomes derived from adipocytes play a key role in this process.

Exosomes, vesicle-like bodies (40-100 nm in diameter) secreted by cells, are ubiquitous in peripheral blood, urine, and other bodily fluids[13]. As extracellular messengers, they facilitate the crosstalk between tumor cells and surrounding microenvironment components. Exosomes transport various molecules such as microRNAs (miRNAs) and long non-coding RNAs (lncRNA), proteins, and lipids, influencing tumor progression, metastasis, and chemoresistance[14]. Exosomes, containing rich miRNA content, circulate through tissues and fluids, influencing tumor radiosensitivity by regulating DNA damage repair, cell cycle, and apoptosis[15]. miRNAs also shape the tumor microenvironment, affecting radiosensitivity[16]. For instance, miR-21 transferred via extracellular vesicles (EVs) mediates ionizing radiation-induced bystander effects[17]. By transferring mRNA, miRNA, and proteins, exosomes facilitate crucial tumor microenvironment interactions, modulating tumor radiosensitivity[18,19].

The objective of this study was to identify potential miRNAs that may mediate the promotion of ESCC radioresistance by adipocyte-derived exosomes. We first investigated the role of adipocytes and adipocyte-derived exosomes in the radioresistance of ESCC cells. Next, we employed high-throughput miRNA sequencing to investigate the changes in the expression profile of miRNAs in serum and adipose tissue-derived exosomes. Additionally, we analyzed serum exo-miR-660-5p levels in ESCC patients to assess their clinical value. Our findings indicate that adipocyte-derived exosomal miR-660-5p enhances ESCC radiation resistance, providing a potential therapeutic target to improve the radiosensitivity of ESCC.

MATERIALS AND METHODS

Cell culture

ESCC cells KYSE30 and TE-1 were obtained from Wuhan PuNuoSai Life Technology Co. Ltd. (Wuhan, China). Cell lines were cultured in RPMI 1640 medium (Hyclone, United States), supplemented with 10% fetal bovine serum (Gibco, United States) and a 1% penicillin-streptomycin solution (Beyotime Biotechnology, Nantong, China).

Cell irradiation

Cell irradiation was performed using a linear accelerator from RAD SOURCE (Suwanee, GA, United States), delivering a dose rate of 1.15 Gy/minute at 160 kV. Clonogenic assays were conducted with absorbed doses of 0, 2, 4, 6, and 8 Gy. Apoptosis, immunofluorescence, and Western Blot experiments utilized 4 and 6 Gy.

Sample collection

Human thoracic adipose tissue samples and serum samples from ESCC patients were collected at The Affiliated Tumor Hospital of Nantong University, with approval from the institutional ethics committee (Ethics Research Grant 2022). The study was conducted after obtaining written informed consent from all participants. Between February 2022 and July 2022, 99 blood samples were collected. Patients were eligible for inclusion if they met the following criteria: (1) Karnofsky performance status score ≥ 70; (2) Age ≤ 75 years; (3) Inoperable cases or those who refused surgery; (4) Lesion size ≤ 10 cm, with no signs of esophageal perforation on X-ray; (5) No severe heart, liver, or kidney dysfunction with normal blood routine parameters; and (6) No severe cachexia or cardiovascular diseases and anticipated survival > 3 months. Pregnant and lactating women, patients with other concomitant malignant tumors, and those lacking follow-up data were excluded. Clinical staging adhered to the 7^th edition of the American Joint Committee on Cancer Staging Manual. All patients (Table 1) received involved-field radiation therapy with a total dose of 60-66 Gy (2 Gy per fraction, 5 days per week). They also received two cycles of paclitaxel combined with platinum-based neoadjuvant chemotherapy and two cycles of adjuvant chemotherapy post-radiation. Blood samples were collected one day before treatment. Short-term efficacy evaluation included barium esophagography for primary lesions and RECIST criteria for metastatic lesions. Follow-up examinations were conducted every 3 months post-treatment until death or June 2024, consisting of physical examinations, barium swallow tests, and chest computed tomography scans.

Table 1 Clinical characteristics of patients with esophageal squamous cell carcinoma.

Characteristics	ESCC, n
Sex
Male	33
Female	67
Age
< 60 years	21
≥ 60 years	79
Length
< 5 cm	54
≥ 5 cm	46
Therapeutic modality
Radiotherapy	58
Chemoradiotherapy	42
T stage
T1–2	52
T3–4	48
Lymph node metastasis
N0–1	93
N2–3	7
TNM stage
I-II	75
III-IV	25

TNM: Tumor-node-metastasis; ESCC: Esophageal squamous cell carcinoma.

Adipocyte isolation and co-culture

Thoracic adipose tissue was minced and mixed with 0.2% collagenase (Biofrox, Germany) at a 3:1 ratio in a 50 mL centrifuge tube. The mixture was shaken at 37 °C for 30 minutes and centrifuged at 1000 rpm for 5 minutes. After centrifugation, adipose tissue layers were separated: The top layer was oil, the middle layer contained adipocytes, and the bottom layer included fibrous connective tissue and collagenase solution. The middle layer of adipocytes was used for experiments. The isolated adipocytes were added to a polycarbonate membrane chamber (8 μm pore size) placed in a culture dish containing ESCC cells, establishing a co-culture model.

Exosome extraction and identification

Adipocyte culture supernatant or serum was centrifuged (2000 × g, 4 °C for 10 minutes), and the upper layer was filtered (0.22 μm, Millipore, United States). The filtrate underwent ultracentrifugation (35000 rpm/210000 × g, 4 °C for 70 minutes), discarding the supernatant. The pellet was resuspended in PBS, centrifuged again (35000 rpm/210000 g, 4 °C for 70 minutes), and the supernatant was removed. The final pellet was resuspended in aseptic PE tubes. Extracellular vesicle protein concentration was measured by BCA, with a target of approximately 5 μg/μL. Samples were stored at 4 °C (short-term, approximately one week) and -80 °C (long-term, 3 months). The size distribution and concentration of EVs were analyzed using nanoparticle tracking analysis (NTA) with a ZetaView particle tracker from ParticleMetrix (Meerbusch, Germany). The structure of EVs was examined by transmission electron microscopy (TEM; JEM-1200EX, JEOL Ltd., Japan). TSG101 and CD9 were used as exosomal markers.

Analysis of exosome uptake by confocal microscopy

Exosome solutions were labeled with 2 μL of Mem Dye stock solution, incubated at 37 °C for 30 minutes, and centrifuged at 3000 × g for 5 minutes. 100 μL of PBS was added to the filter tube and centrifuged at 3000 × g for 5 minutes; these steps were repeated. Subsequently, labeled exosomes (1-50 μL) were added to cells and incubated at 37 °C, 5% CO₂ for 1-24 hours. Following supernatant removal, PBS washing, and adding fresh culture medium with serum, uptake was observed and imaged using a fluorescence microscope (Olympus, Tokyo, Japan).

Clonogenic assay

Cells in the logarithmic growth phase were digested into single-cell suspensions. Various cell numbers were seeded in six-well plates, and exposed to five dose points of 0, 2, 4, 6, and 8 Gy radiation, with three replicates per dose. After 24-hour co-culture with adipocytes and exosomes, cells were irradiated and subsequently cultured for 14 days. Cells were fixed with methanol and stained with Giemsa for 30 minutes. Colonies containing more than 50 cells were counted. Sensitizer enhancement ratios (SER) were calculated using the multi-target single-hit model: SER = D₀ value of radiation alone/D₀ value of radiation + treatment group. The following equation was used for the multi-target single-hit model: SF = 1 - [1 - exp (-D/D₀)]N, Dq = D₀logN, where SF is the cell survival fraction, D is the radiation dose (Gy), D₀ is the mean lethal dose, Dq is the quasi-threshold dose, and N is the extrapolation number. Cell survival curves were plotted based on radiation dose.

Flow cytometric analysis of cell apoptosis

ESCC cells were seeded into six-well plates in different groups and cultured in the complete medium for 24 hours. Cells were then treated under different experimental conditions: Co-cultured with adipocytes or exosomes for 24 hours followed by 6 Gy irradiation. Post-irradiation, cells were digested with trypsin for 48 hours, centrifuged to collect cell pellets, resuspended in PBS to make single-cell suspensions, centrifuged to discard the supernatant, resuspended in 1 × buffer, and stained with Annexin V 7ADD/PE (BD, United States). Then, the cells were left at room temperature in dark for 15 minutes before apoptosis detection using flow cytometry.

Immunofluorescence assay

Cells were seeded in glass-bottom dishes and treated under different conditions 24 hours before irradiation. Post-irradiation (0.5 hours), cells were fixed with 4% paraformaldehyde for 10 minutes, washed gently with PBS three times, permeabilized with 1% Triton × 100 for 10 minutes, and blocked with serum at room temperature for 1 hour. After washing, cells were incubated overnight at 4 °C with mouse monoclonal anti-γH2AX antibody (1:200, Abcam, United Kingdom), followed by incubation with Cy3-labeled goat anti-mouse IgG (Beyotime, China) for 1 hour. The cells were observed using an Olympus fluorescence microscope (Olympus, Tokyo, Japan), and γH2AX foci in each cell were counted using Image Pro Plus software (Media Cybernetics). The average number of γH2AX foci per nucleus was determined from three independent experiments.

MiRNA sequencing and bioinformatics analysis

MiRNA libraries were constructed using the QIAseq miRNA Library Kit (QIAGEN, Germany). Experimental steps included 3' ligation, 5' ligation, reverse transcription, QIAseq miRNA NGS bead preparation, cDNA purification, library amplification with HT plate indices, and library amplification with tube indices. After quality control, libraries were sequenced on an Illumina HiSeq 2500 using the SE50 strategy. Cutadapt software was used to trim adapters from the ends of raw reads while retaining reads longer than 17 nucleotides. The FANse3 ultra-high-precision sequence alignment algorithm was employed to align reads obtained from each sample to reference sequences (human mature miRNAs, miRBase version 22.1). Mapping parameters were set as -E5% - indel -S14. Transcripts per million were calculated to normalize RNA sequencing depth. miRNAs with read counts ≥ 10 were considered expressed, while those with counts < 10 were considered non-expressed. For gene sequencing data, differential expression analysis was performed using the edgeR package, a statistical method based on the negative binomial distribution. miRNAs were considered differentially expressed when |logFC| > 1 and false discovery rate (FDR) < 0.01. miRNAs with logFC > 1 were considered upregulated, while those with logFC < 1 were considered downregulated. Raw FASTQ data were subjected to quality checks and length selection before sequencing reliable fragments. Differential miRNA expression between different samples was analyzed for target gene prediction, and subjected to Gene Ontology and KEGG Pathway enrichment analyses.

Western blot

Cells from different treatment groups were lysed with RIPA buffer (Solarbio, China), and protein concentrations were determined using the BCA Protein Assay kit (Beyotime, China). After boiling denaturation with the loading buffer, SDS-PAGE electrophoresis was performed, followed by membrane transfer. The membrane was blocked with 5% BSA for 1.5 hours, then incubated overnight with primary antibody at 4 °C. After washing thrice with TBST, the membrane was incubated with secondary antibody at room temperature for 1 hour, followed by three additional washes. Chemiluminescence was used for detection, and Image J software was used for quantifying the target bands and the internal controls. The following antibodies were employed: Anti-γ-H2AX (Cell Signaling Technology, United States; 9718S, 1:1000); anti-TSG101 (Cell Signaling Technology, United States; 28405S, 1:1000); anti-CD9 (Cell Signaling Technology, United States; 98327S, 1:1000); anti-GAPDH (Cell Signaling Technology, United States; 5174S, 1:1000); anti-α-Tubulin (Cell Signaling Technology, United States; 2125S, 1:1000).

Real-time quantitative polymerase chain reaction

To validate the results, the selected miRNAs were subjected to real-time quantitative polymerase chain reaction (qRT-PCR) analysis. Total RNA was extracted using TRIzol reagent (Invitrogen, United States). The HyperScript' III miRNA 1^st StrandcDNA Synthesis Kit (EnzyArtisan, Shanghai) was used to perform reverse transcription. Real-time PCR was conducted using NovoStart SYBR qPCR SuperMix plus (Novoprotein, Shanghai, China). The reference gene used was U6. The primers are listed in Table 2. The results are shown as relative expression levels calculated by the 2−ΔΔCt method.

Table 2 List of primer sequences of related RNAs.

Primer	Sequences (5' to 3')
hsa-miR-660-5p-F	CGCGGCATACCCATTGCATATC
hsa-miR-660-5p-R	ATCCAGTGCAGGGTCCGAGG
hsa-miR-660-5p-RT	GTCGTATCCAGTGCAGGGTCCGAGGTAT TCGCACTGGATACGACCAACTC
hsa-miR-378a-3p-F	AATCCGGAACTGGACTTGGAGTC
hsa-miR-378a-3p-R	ATCCAGTGCAGGGTCCGAGG
hsa-miR-378a-3p-RT	GTCGTATCCAGTGCAGGGTCCGAGGTAT TCGCACTGGATACGACGCCTTC
hsa-miR-193b-3p-F	AACACGCAACTGGCCCTCAAA
hsa-miR-193b-3p-R	ATCCAGTGCAGGGTCCGAGG
hsa-miR-193b-3p-RT	GTCGTATCCAGTGCAGGGTCCGAGGTAT TCGCACTGGATACGACAGCGGG
hsa-mir-30a-F	CCGCTCGTGTAAACATCCTCGA
hsa-mir-30a-R	ATCCAGTGCAGGGTCCGAGG
hsa-mir-30a-RT	GTCGTATCCAGTGCAGGGTCCGAGGTAT TCGCACTGGATACGACCTTCCA

Validation of exosomal miRNA

To assess the predictive value of differentially expressed miRNAs in ESCC radiotherapy response, receiver operating characteristic (ROC) curves were generated using MedCalc 20.218 software. Statistical significance was defined as P < 0.05.

Statistical analysis

All experiments were performed in triplicate and none of the samples were excluded from the analysis. Owing to the normal distribution of all variables, the results are presented as mean ± SD, and data analysis was performed using the Student's t-test. Statistical analysis was performed using SPSS 22.0 (Chicago, IL, United States) and GraphPad Prism 8.0 (CA, United States) software, with the significance level set at P < 0.05.

RESULTS

Adipocytes enhance radioresistance of ESCC cells

After co-culture with adipocytes and X-ray exposure, the clonogenic survival of KYSE30 (Figure 1A) and TE-1 (Figure 1B) cells was noticeably greater compared to the control group, with SER of 0.832 and 0.825, respectively (Table 3). Moreover, 48 hours post-irradiation with co-culture, the apoptotic rates of ESCC cells KYSE30 (Figure 1C) and TE-1 (Figure 1D) were significantly decreased compared to the control. These results indicate that adipocytes can inhibit Ionizing radiation-induced apoptosis and promote radioresistance, while the addition of the exosome inhibitor GW4869 in co-culture with adipocytes effectively increased the sensitivity to radiotherapy (Supplementary Figure 1).

Figure 1 Adipocytes enhance the resistance of esophageal cancer cells to radiation therapy. A: Representative images of colony formation and corresponding survival curves for KYSE30 cells; B: Representative images of colony formation and corresponding survival curves for TE-1 cells; C: Flow cytometry analysis of apoptosis in KYSE30 cells; D: Flow cytometry analysis of apoptosis in TE-1 cells. Live cells are located in the lower left quadrant, early apoptotic cells in the lower right quadrant, and late apoptotic cells in the upper right quadrant. The total number of apoptotic cells includes both early and late apoptotic cells, and the apoptosis rates of KYSE30 and TE-1 cells were evaluated. Mean values from three independent experiments are presented and analyzed using the t-test. ^aP < 0.01; ^bP < 0.001.

Table 3 The D0, N, and SER values of cells subjected to different treatment conditions.

Cell	Group	N	Dq	D0	SER
KYSE-30	Control	1.710	1.105	2.060	0.832
	+ Adipocyte	1.615	1.186	2.475
	Control	1.337	0.705	2.426	0.883
	+ Exosome	1.513	1.137	2.746
TE-1	Control	1.395	0.745	2.239	0.825
	+ Adipocyte	1.248	0.601	2.713
	Control	1.724	1.022	1.876	0.751
	+ Exosome	1.885	1.584	2.498

Sensitizer enhancement ratios value was simulated using the multi-target single-hit model. D₀: Mean lethal dose; N: Extrapolation number; SER: Sensitizer enhancement ratios.

Co-culture with adipocytes attenuates the dynamics of DNA damage repair in cells

Within 24 hours before irradiation, co-culture with adipocytes significantly inhibited X-ray-induced double-strand breaks. Laser confocal microscopy revealed a reduction in γ-H2AX foci in the nuclei of KYSE30 (Figure 2A) and TE-1 (Figure 2B) cells exposed to ionizing radiation after co-culture with adipocytes. Western blot analysis indicated decreased expression of γ-H2AX protein in KYSE30 (Figure 2C) and TE-1 (Figure 2D) cells following co-culture with adipocytes, consistent with immunofluorescence findings, suggesting that adipocytes may alleviate ionizing radiation-induced DNA damage in ESCC cells. These results demonstrate that adipocytes decrease IR-induced DNA damage and promote radioresistance in ESCC.

Figure 2 The confocal immunofluorescence staining images of KYSE30 and TE-1 cells (γ-H2AX labeled in red and nuclear staining in blue). A: KYSE30; B: TE-1 cells. Quantitative analysis of γ-H2AX foci numbers is presented in a scatter plot (^aP < 0.01); C and D: Results of Western blot analysis showing γ-H2AX protein levels in KYSE30 (C) and TE-1 cells (D) co-cultured with adipocytes and exposed to radiation, compared to the control group that received only irradiation.

Isolation and identification of exosomes from adipocyte origin

Transmission electron microscopy revealed vesicles with a diameter of approximately 100 nm and a double-layer membrane structure, confirming the presence of exosomes secreted by adipocytes (Figure 3A). The size, polydispersity index, and Zeta potential of the exosomes were measured. As shown in Figure 3B, the Number distribution and Intensity distribution of exosomes were highly homogenous, with no excess peaks observed, indicating an average size of exosomes ranging from 80 to 120 nm. The expression of exosomal marker proteins TSG101 and CD9 is depicted in Figure 3C. Figure 3D illustrates the internalization of exosomes into the cytoplasm of cells after co-incubation for 4 hours, with an increasing number of exosomes entering the cells over time.

Figure 3 Characterization of extracted exosomes. A: Transmission electron microscopy image of exosomes (scale bar = 100 nm); B: Nanoparticle tracking analysis of exosome concentration and size; C: Western blot detection of exosomal-specific markers TSG101 and CD9; D: Uptake of exosomes by KYSE30 cells.

Adipocyte-derived exosomes enhance radioresistance of ESCC cells

The combined impact of adipocyte-derived exosomes and X-ray irradiation on the clonogenic survival of KYSE30 (Figure 4A) and TE-1 (Figure 4B) cells was assessed. Cells treated with exosomes before irradiation exhibited higher survival rates than the control, with SER of 0.883 and 0.751, respectively (Table 3). After 48 hours of co-culture following irradiation, the apoptotic rates of KYSE30 (Figure 4C) and TE-1 (Figure 4D) cells were significantly lower in the exosome-treated groups than in irradiation-only groups. These results indicate that adipocyte-derived exosomes can inhibit ionizing radiation-induced apoptosis and promote radioresistance.

Figure 4 Radiation response and apoptosis in esophageal squamous cell carcinoma cells. A and B: Radiation response and corresponding survival curves of the colony formation assay for KYSE30 (A) and TE-1 (B) cells; C and D: Flow cytometry analysis of apoptosis in KYSE30 (C) and TE-1 (D) cells, with the lower left quadrant representing live cells, the lower right quadrant representing early apoptosis, and the upper right quadrant representing late apoptosis. Accumulative cell apoptosis includes both early and late stages. Mean values from three independent experiments are presented and analyzed using the t-test. ^aP < 0.01.

Adipocyte-derived exosomes alleviate DNA damage in ESCC cells after radiation

We previously demonstrated that adipocyte-derived exosomes affect the radiosensitivity of cancer cells. Here, we further investigated whether adipocyte-derived exosomes influence the dynamics of DNA damage repair. Laser confocal microscopy revealed a decrease in γ-H2AX foci points in the nuclei of KYSE30 (Figure 5A) and TE-1 (Figure 5B) cells post-irradiation in the treated groups. Western Blot analysis indicated reduced irradiation-induced expression of γ-H2AX protein in KYSE30 (Figure 5C) and TE-1 (Figure 5D) cells with exosome treatment. These findings, consistent with immunofluorescence results, suggest that adipocyte-derived exosomes can mitigate DNA damage in ESCC cells induced by ionizing radiation.

Figure 5 DNA damage in esophageal squamous cell carcinoma cells after radiation. A: Co-localization immunofluorescence staining of KYSE30; B: TE-1 cells, (γ-H2AX in red and nuclear staining in blue). Scatter plot of quantitative analysis of γ-H2AX foci number (^aP < 0.01); C and D: Western blot analysis of γ-H2AX protein levels in irradiated KYSE30 (C) and TE-1 (D) cells, respectively.

Differential expression of miRNAs and functional enrichment analysis

High-throughput sequencing of samples from adipocyte-derived exosomes and blank controls revealed 330 differentially expressed genes, with 219 upregulated and 111 downregulated (P < 0.05 & |log2FC| > 1) (Figure 6A). Subsequently, a comparison of high-throughput sequencing results from serum exosomes of ESCC patients before and after radiotherapy revealed 20 differentially expressed miRNAs. GO enrichment analysis of differentially expressed miRNAs in serum exosomes revealed enrichment in the Biological Process category, particularly in "positive regulation of transcription". Furthermore (Figure 6B), KEGG pathway analysis identified the top 20 entries based on log10Pvalue, highlighting several key signaling pathways associated with the target genes of these miRNAs (ErbB signaling pathway, FoxO signaling pathway, Wnt signaling pathway, Ras signaling pathway, and MAPK signaling pathway) (Figure 6C). The clustering heatmap is displayed in Figure 6D. Four intersecting miRNAs were identified: MiR-30a-5p, miR-378a-3p, miR-193b-3p, and miR-660-5p (Figure 6E).

Figure 6 Next-generation sequencing analysis of miRNA expression. A: Differential expression of miRNAs in adipocyte-derived exosomes; B: Gene Ontology analysis of differentially expressed miRNAs in serum-derived exosomes; C: Kyoto Encyclopedia of Genes and Genomes analysis of differentially expressed miRNAs in serum-derived exosomes; D: Heat map of differentially expressed miRNAs in serum-derived exosomes from esophageal cancer patients before and after radiotherapy; E: Venn diagram illustrating overlapping differentially expressed miRNAs in adipocyte-derived exosomes and serum-derived exosomes from esophageal cancer patients before and after radiotherapy.

Exo-miR-660-5p as a potential biomarker for predicting radiotherapeutic efficacy in ESCC

ESCC patients after radiotherapy were divided into two groups based on RECIST 1.1 criteria: 58 cases with partial response were classified as the effective radiotherapy group, while 41 cases with stable disease or progressive disease were classified as the ineffective radiotherapy group. The expression levels of the 4 selected miRNAs were validated by qRT-qPCR. Analysis revealed elevated expression levels of serum exosome miR-193b-3p and miR-660-5p in the ineffective radiotherapy group, whereas a significant decrease was observed in the effective group (P < 0.001). These findings suggested a potential role of serum exo-miR-193b-3p and exo-miR-660-5p as biomarkers for evaluating radiotherapy efficacy.

The predictive capability of serum exosome miR-193b-3p and miR-660-5p in distinguishing between good and poor responders to chemo-radiotherapy was assessed using ROC curve analysis. Serum exo-miR-660-5p demonstrated strong predictive ability in differentiating responders (area under the curve: 0.829, 95%CI: 0.740-0.897, P < 0.0001; Figure 7). However, there were no significant differences in serum exo-miR-193b-3p levels between good and poor responders (Figure 7). The differences in serum exo-miR-660-5p levels between good and poor responders and its strong discriminatory power in distinguishing between the two groups suggest its potential role as a biomarker for predicting patient responses to radiotherapy.

Figure 7 Exo-miR-660-5p as a potential biomarker for predicting radiotherapeutic efficacy in esophageal squamous cell carcinoma. A: Representative findings of Controls and patients with esophageal squamous cell carcinoma (ESCC); B: Esophageal and computed tomography (CT) examination findings before and after radiotherapy in responsive esophageal cancer patients; C: Esophageal and CT examination findings before and after radiotherapy in non-responsive esophageal cancer patients; D: Real-time quantitative polymerase chain reaction validation of selected miRNAs. Four upregulated miRNAs (miR-30a-5p, miR-378a-3p, miR-193b-3p, miR-660-5p) were validated in the sample set. MiR-193b-3p and miR-660-5p showed significant differential expression (^aP < 0.05; ^bP < 0.01); E: Receiver operating characteristic curve analysis to assess the predictive ability of miR-193b-3p and miR-660-5p for radiochemotherapy sensitivity in ESCC. Serum exosomal miR-660-5p exhibited a statistically significant difference, suggesting its potential role as a biomarker for predicting radiotherapy sensitivity in ESCC. ESCC: Esophageal squamous cell carcinoma.

DISCUSSION

Most contemporary research on esophageal cancer radiosensitivity is focused on functional alterations of key genes related to cancer cell proliferation, apoptosis, DNA damage, and repair[20]. The present study aimed to explore the impact and regulatory mechanisms of the tissues surrounding esophageal cancer on its radiosensitivity. Studies have shown that excessive fat accumulation, leading to obesity or overweight, is a risk factor for esophageal cancer[21]. Adipose tissue, primarily composed of adipocytes, serves as the body’s largest energy reservoir. Upon breakdown, it releases glycerol and fatty acids, providing essential sources of energy for various tissues[22]. Additionally, adipose tissue serves as a vital endocrine organ, producing and secreting important signaling molecules such as adipokines, inflammatory cytokines, and enzymes[23]. In the present study, ESCC cells co-cultured with adipocytes and exposed to high-energy X-rays showed increased resilience. Specifically, KYSE30 and TE-1 cells co-cultured with adipocytes exhibited significantly enhanced clonogenic formation rates post-irradiation compared to cells cultured alone. Additionally, apoptosis was decreased and DNA double-strand breaks in the nuclei were notably reduced in TE-1 cells co-cultured with adipocytes. These results suggest a close association between adipocytes in the tumor microenvironment and the radiation resistance of ESCC.

EVs, especially exosomes, have garnered widespread attention due to their involvement in cancer progression and their potential as diagnostic and prognostic markers. Studies have demonstrated the involvement of EVs in cancer growth, metastasis, and angiogenesis. Moreover, EVs facilitate intercellular crosstalk, particularly among adipocytes, influencing tissue microenvironments. Small EVs shed by adipocytes stimulate fatty acid oxidation and migration in melanoma cells, and these effects were enhanced in obesity[24]. EVs were first discovered in serum culture medium in 1983[25]. EVs play important roles in immune surveillance, cell apoptosis, tumor progression, and several other physiological and pathological processes, including gene reprogramming in target cells[26]. Studies have reported the role and mechanism of adipocyte-derived EVs in regulating the biological effects of adipocytes[27]. However, no studies have investigated the modulatory effect of adipocyte-derived EVs on the radiosensitivity of esophageal cancer and the underlying mechanisms. This study investigated the effects of EVs isolated from clinical thoracic fat tissue specimens. EVs derived from adipocytes were found to increase the clonogenic formation rate of ESCC cells after X-ray exposure, reduce apoptosis in irradiated cells, and inhibit radiation-induced DNA double-strand breaks, indicating that adipocyte-derived EVs may promote radiation resistance of ESCC.

Exosomes possess a typical lipid bilayer and can selectively enrich various proteins, miRNAs, and lncRNAs, facilitating their transport from donor cells to recipient cells[28]. Notably, miRNAs are the most enriched nucleic acid molecules in exosomes. miRNAs, small non-coding RNAs (18-25 nt), are evolutionarily conserved and have emerged as promising diagnostic and prognostic biomarkers for various cancers[29]. The expression profiles of miRNAs correlate with specific clinical pathological parameters across different cancer subtypes, suggesting their potential as biomarkers based on tumor origin, histology, invasiveness, or chemotherapy sensitivity[30]. A recent miRNA study analyzed pre-treatment tumor biopsies and identified four differentially expressed miRNAs (miR-145-5p, miR-152, miR-193b-3p, and miR-376a-3p) that, in combination, could predict the chemoradiation response in ESCC patients[31].

However, research on the role of exosomes in cancer is hindered by significant limitations, including the lack of validation of screened miRNAs in clinical samples and the uncertainty surrounding the biomarker potential of miRNAs in exosomes. In this study, adipose-derived exosomes were found to increase the radioresistance of ESCC cells by transferring miRNAs. Furthermore, high-throughput sequencing was used to screen differentially expressed miRNAs in exosomes derived from adipocytes and serum exosomes pre- and post-ESCC radiotherapy, revealing overlapping miRNAs. To our knowledge, this is the first study to reveal the changes in adipose-derived exosome miRNA expression in ESCC. Subsequently, the four upregulated miRNAs identified were validated by qRT-PCR. The results indicated an association between high expression of exo-miR-660-5p and poorer radiotherapy efficacy. Identifying miRNAs that can predict the sensitivity of ESCC patients to radiotherapy before treatment can help inform the subsequent treatment plans. In this study, we identified exo-miR-660-5p as a potential novel prognostic marker associated with radiotherapy efficacy in ESCC. These findings need to be validated using a larger and independent patient cohort. Additionally, further studies are required to explore the potential role of exosomal miRNAs as biomarkers.

CONCLUSION

This study demonstrates a strong link between tumor microenvironment adipocytes and ESCC radioresistance. Adipocyte-derived exosomes modulate ESCC radiosensitivity, with enclosed miRNAs serving as key mediators of ESCC radiation resistance. Serum exosomal miR-660-5p upregulation is a potential prognostic marker for adverse outcomes. Further validation in larger, independent cohorts is necessary to confirm predictive accuracy. Our findings can potentially aid in the treatment decision-making process for locally advanced ESCC patients by predicting radiotherapy and chemotherapy responsiveness, ensuring that only those who are likely to respond well will receive such treatment.

ACKNOWLEDGEMENTS

We wish to thank Professor Yang Jiao and Dr. Yi-Ting Tang for the timely help with reviewing and revising this paper.