Background
NPC is an Epstein-Barr virus (EBV)-associated cancer, prevalent in southeast Asia and north Africa [
1]. According to the statistics published in 2018, cancer arising from the nasopharynx is responsible for 129,079 cases and 72,987 deaths each year [
2]. Despite the early detection and advances in radiotherapy, treatment for patients who partially responded to irradiation continue to be unsatisfactory primarily due to the acquisition of radioresistance and tumor recurrence following irradiation [
3,
4]. However, the underlying mechanisms remain elusive.
The tumor microenvironment (TME) consists of stromal cells, including fibroblasts, macrophages, endothelial cells, other immune cells, extracellular matrix, as well as the bioactive substances secreted by these cells [
5]. As one of the most predominant stromal cell subtypes in solid tumors, cancer-associated fibroblasts (CAFs) contribute to most aspects of cancer progression, including proliferation, migration, invasion, angiogenesis, induction of chemoresistance and escape from immune-mediated killing [
6‐
8]. It has been reported that CAFs can enhance cervical tumor growth after irradiation and promote the epithelial mesenchymal transition of pancreatic tumors via crosstalk between tumor cells and CAF-like vascular endothelial growth factors (VEGF) and fibroblast growth factors (FGF) [
9,
10]. Radiobiological research has shown that the relationship between NPC and CAFs remains poorly understood. Recently, immune checkpoint inhibitors have emerged at the frontier in anticancer research for patients with recurrent or metastatic head and neck cancer following irradiation treatment [
11,
12]. Although several breakthroughs have been made, the total efficacy of immunotherapy is finite. Tranilast is a CAF inhibitor [
13] that has been reported to suppress CAF proliferation and inhibit the adverse effects induced by CAFs within the immune microenvironment [
14]. For example, it has been reported that Tranilast treatment was associated with a reduced level of transforming growth factor-β1 production by CAFs. However, the precise mechanism by which Tranilast suppresses CAF activity in irradiation-related research remains unknown.
Our research has found that CAFs can activate the NF-κB pathway in irradiated NPC cells to reduce DNA damage, which could be interrupted by tranilast. These findings suggest that Tranilast may have potential value in increasing tumor sensitivity to irradiation. Thus, the latent function of Tranilast in NPC warrants further investigation in the future.
Materials and methods
Cell culture and regents
The human NPC cell lines, 5-8F, 6-10B, and HK-1 were donated by the Sun Yat-sen University Cancer Center. Cells were cultured in 1640 medium (Gibco) supplemented with 10% FBS (Gibco) and 1% penicillin/streptomycin (Thermo Scientific). Human CAFs and normal fibroblasts (NFs) were extracted from fresh NPC tissues and matched to normal nasopharynx tissues from patients diagnosed with local recurrence NPC patients treated with radical radiotherapy previously. Fibroblast isolation was performed as previously described [
15]. Briefly, fresh tissues were cut into pieces of 2 to 3 mm and cultured in high-glucose DMEM medium (Gibco) for approximately one week until fibroblasts appeared. α-SMA was used to identify CAFs and NFs as described [
6]. Specimens were obtained with written informed consent from patients with NPC enrolled at Nanfang Hospital, Southern Medical University. Characteristics of 4 CAF donors were documented in supplementary Table
1. In functional studies, NFs and CAFs were used within 10 passages. All cells were cultured in a humidified incubator containing 5% CO
2 at 37 °C.
Human recombinant IL-8, HGF, and TNF-α were purchased from Peprotech (Suzhou, China). An antagonist of IL-8 receptor (Repertaxin), inhibitor of NF-κB activation (caffeic acid phenethyl ester [CAPE]), and a lactate inhibitor (sodium dichloroacetate, DCA) were purchased from Selleck Chemicals (Houston, USA); Tranilast was purchased from Target Mol (Massachusetts, USA).
Establishment of radiation-resistant (IRR) 5-8F IRR cells
Establishment of IRR cell lines was performed as previously described [
16]. In brief, cells were treated twice with irradiation at a graded dose of 2, 4, 6, 8 and 10 Gy, respectively. A total IR dose of 60 Gy was administered with the entire selection procedure. The final surviving 5-8F cells were verified, defined as radioresistant 5-8F cells, and termed 5-8F IRR. The parental 5-8F cells were treated without irradiation. Cells within 10 passages were used for experiments once the 5-8F-IRR cells were successfully established.
Cell viability assay
CAFs were subjected to a cell viability assay using Cell Counting Kit-8 (Dojindo, Japan) according to the manufacturer’s instructions. In brief, CAFs were cultured in medium containing Tranilast at a concentration of 200 μM and 400 μM, respectively for two days and subsequently cultured at the density of 1.5 × 103 cells per well in triplicate in 96-well plates containing Tranilast-free medium. After culturing for 8 h, CAFs were applied to CCK-8 assay and a subsequent detection was performed at the indicated time points. To detect the impact of Repertaxin and CAPE on NPC cell line proliferation, cells were cultured in medium containing a wide-range of reagent concentrations (0.01, 0.1, 1, 10, and 100 μM for CAPE and 1, 10, 40 μM for Repertaxin) for 2 - 4 days and then subjected to a CCK-8 assay.
A total of 5-8F, 6-10B, and HK-1 cells were cultured in six-well plates for 12 h at various densities according to the aim of each experiment. The cell density was listed as follows: 300 (0 Gy), 400 (2 Gy), 1000 (4 Gy), 2000 (6 Gy), 4000 (8 Gy) per well for 5-8F and 6-10B and 500 (2 Gy), 10,000 (8 Gy) per well for HK-1. After a 2-week incubation, the six-well plates were obtained and colonies with more than 50 cells were counted. The surviving fraction (SF) was estimated using the following formula: SF = (number of colonies formed/number of cells seeded × plating efficiency of the control group), where plating efficiency was calculated as the ratio between colonies observed and the number of cells plated. Dose-response colony survival curves were plotted accordingly.
Cytokine assay
CAF and NF cells were cultured in serum-free medium when grown to 80% confluence. After incubating for two days, the medium was collected and concentrated before dialysis. After labeling with biotin, the samples were applied to an antibody array chip (R&D; Cat ARY022B, LabEx) overnight. The following day, the chips were reacted with streptavidin-conjugated fluorescence dye, and detected with a chemiluminescence imager (Chemi Scope 6300). Data were normalized to the total protein.
RNA sequencing and tissue microarray
Total RNA of parent 5-8F and 5-8F IRR cells were extracted using TrIzol reagent (Invitrogen Life Technologies, CA) according to the manufacturer’s instructions. The prepared RNA was then subjected to sequencing on a BGISEQ-500 platform at Beijing Genomics Institution (
www.genomics.org.cn, BGI, Shenzhen, China). The level of gene expression was quantified and the NOISeq method was performed to screen for differentially expressed genes (DEGs) as previously described [
17]. A tissue microarray was conducted by Shanghai Biotechnology Corporation using Affymetrix Genechip according to manufacturer’s protocol [
18].
Radiosensitivity index (RSI) analysis and gene set enrichment analysis
An RSI index served as an indicator for radiosensitivity, and was calculated as previously described [
19]. In brief, head and neck squamous carcinoma (HNSCC) patients in the TCGA database and NPC patients in the GSE12542 dataset were divided into two groups based on the median of CAFs score using an online tool (
https://gfellerlab.shinyapps.io/EPIC_1-1/). Following Racle’s study, 20 markers were used to identify CAFs [
20]. Gene Set Enrichment Analysis (GSEA) software version 3.0 (Broad Institute, USA) was used to analyze GSE48501 and a human microarray containing radioresistant and radiosensitive NPC samples. A threshold of
P ≤ 0.05 was applied for the analysis. Data for GSE48501 and GSE12542 were downloaded from the NCBI Gene Expression omnibus (GEO,
http://www.ncbi.nlm.nih.gov/geo/).
Lactate detection
Lactate was detected using a lactic acid measurement kit (Junji Biotechnology Co, China) according to the manufacturer’s protocol. Briefly, blank medium and conditioned medium produced by NFs and CAFs were prepared and subjected to a chromogenic reaction with the indicated reagents. The absorbance value was detected by a microplate reader machine (BIO-RAD 689). Data were normalized to the lactate standards.
Comet assay
A comet assay was performed using a DNA Damage Detection Kit (Keygen, China). The cells were irradiated as described at dose of 2 Gy and treated in accordance with the experimental design. The following day, cells were harvested and suspended in PBS containing 1% low-melting agarose and layered onto adhesive microscope slides previously covered with 0.5% normal-melting agarose. The cells were dipped in a specific lysed buffer at 4 °C for 2 h. Next, the DNA was uncoiled and unwound in an alkalescent electrophoresis buffer for 30 min. Electrophoresis was performed and the cells were stained with 4′,6-diamidino-2-phenylindole (DAPI) solution for 10 min in a dark room. The slides were examined with an Eclipse fluorescence microscope (Nikon, Japan). Analysis of comet assay was performed as described [
21].
Real-time quantitative PCR
Total RNA was extracted from CAFs using TrIzol reagent (Invitrogen Life Technologies, CA) as instructed by the manufacturer’s protocol. SuperScript III First-Strand Synthesis SuperMix (Thermo Fisher Scientific, MA) was used for reverse transcription according to the manufacturer’s protocol. Real-time quantitative PCR (qRT-PCR) was implemented to measure specific mRNA expression using an ABI7500 FAST system with TaqMan Reverse Transcription Reagents and SYBR Green PCR MasterMix (Applied Biosystems, CA, USA). GAPDH was used as a loading control. The specific primers were listed as follows: GAPDH, forward: 5′ -GGAGCGAGATCCCTCCAAAAT-3′ and reverse: 5′-GGCTGTTGTCATACTTCTCATGG-3′. IL-8, forward: 5′-GTGCAGTTTTGCCAAGGAGT-3′ and reverse: 5′-CTCTGCACCCAGTTTTCCTT-3′. HGF: forward: 5′-TGGGCCATTCTATTCCCCC-3′ and reverse: 5′-CATGGGGTCAAGCTTCCAGT-3′. The △Ct expression values for amplification were calculated by normalizing to an internal control.
Conditioned medium derived from human NFs and CAFs
CAFs and matched non-malignant NFs were cultured in 25 mL culture flask with high-glucose DMEM (Gibco) supplemented with 10% FBS at same density overnight, respectively. The cells were refreshed with serum-free DMEM and the supernatants were harvested after 48 h. Next, the cell pellet was removed after centrifugation and the conditioned medium (CM) was acquired and stored at − 80 °C for future experimentation.
Western blotting analysis
Cells were lysed with RIPA lysis buffer (Beyotime, Shanghai, China) supplemented with phosphatase and a protease inhibitor cocktail (Roche). The proteins were subjected to 10% SDS-PAGE followed by transfer to PVDF membrane and subsequent incubation with specific primary antibodies at 4 °C overnight. Membranes were then incubated with respective secondary antibodies (FuDe, China) for 1 h at room temperature. The primary antibodies included mouse monoclonal anti-GAPDH (60004–1-IG, Proteintech, USA), rabbit monoclonal anti-α-SMA (ab124964, Abcam, USA), rabbit monoclonal p65 (8242, CST, USA), rabbit monoclonal p-p65 (3039, CST, USA), mouse monoclonal IKB-α (4841, CST, USA), and rabbit monoclonal γ-H2AX (9718, CST, USA). GAPDH was used as an internal reference.
Immunofluorescence
Cells were seeded into a confocal dish and incubated overnight. The cells were washed three times with phosphate buffered saline (PBS) and fixed with 4% paraformaldehyde for 10 min. Next, 0.5% Triton X-100 was used for permeabilization and the samples were blocked with goat serum albumin. Subsequently, cells and tissues were incubated with primary antibodies targeted to α-SMA (1:400) and (or) rabbit monoclonal IL-8 (60004–1-IG, Proteintech, USA) at concentration of 1:100 or γ-H2AX (1:200) at 4 °C overnight, followed by an incubation with an anti-rabbit Alexa fluor-594-conjugated and (or) anti-rabbit Alexa fluor-488-conjugated secondary antibody, respectively. The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI), and images were observed with an Eclipse fluorescence microscope (Nikon, Japan). Expression of α-SMA and IL-8 were determined by mean fluorescence intensity. Analysis of nuclear γ-H2AX foci was performed as described [
21].
IL-8 RNA knockdown in CAFs
A knockdown of IL-8 in CAFs was performed using small interfering RNA (siRNA) (RIBOBIO, China) with Lipofectamine 2000 (Invitrogen, CA) for 10 h of transfection. The specific sequence targeting IL-8 was listed as follows: GCCAAGGAGTGCTAAAGAA.
Patients and clinical samples
All patients were treated with intensity-modulated radiation therapy with a radical dose ranging from 60 Gy to 70 Gy. Radioresistance was defined as the local relapse or recurrence of the ever irradiated area after 6 months of radiotherapy via enhanced computed tomography or enhanced magnetic resonance imaging [
11]. Human NPC samples were obtained from Nanfang Hospital, Southern Medical University, including 6 cases of radioresistant tissues and 16 cases of radiosensitive tissues. The collection of clinical samples for research was approved by the Ethics Committee of the Nanfang Hospital.
Immunohistochemistry assay
Each of the staining processes were conducted according to the manufacturer’s protocol. Primary antibodies consisted of rabbit monoclonal anti-α-SMA (ab124964, Abcam, USA) at a dilution of 1:400 and FAP (ab207178, Abcam, USA) at the dilution of 1:100. The levels of α-SMA and FAP expression were evaluated based on the staining intensity and percentage of positively-stained CAFs. Final scores were assessed as the total evaluation of staining intensity (0 for negative, 1 for weak, 2 for moderate and 3 strong) plus percent positivity (0 for 0%, 1 for 1% - 25%, 2 for 26% - 50%, 3 for > 50%) [
22]. The IHC evaluation was performed by two independent pathologists [
23].
Mouse irradiation assays
Three-week-old male BALB/c nude mice were purchased from Southern Medical University Laboratory Animal Center (Guangzhou, China). Experiments involving animals were approved by the Institutional Animal Care and Use Committee of the Southern Medical University. Mouse models were generated by a subcutaneous inoculation of 5-8F cells alone or combined with human cancer-associated fibroblasts at a 1:1 ratio. The number of cancer cells in each injection was 1 × 10
6 for one flank side. Mice were divided into six groups as required and received irradiation treatment at a 8Gy × 3 schedule 9 days after injection [
24,
25]. Before irradiation, one co-injection group was subjected to an orthotopic injection of 200 μM Tranilast in 100 μL daily for three days [
13]. The tumor volume was determined at the indicated time points with the following formula: tumor volume (mm
3) = ½ × longest diameter 2 × shortest diameter. Mice were sacrificed and the tumors were excised after 2 weeks of the first irradiation. Treatment of irradiation was given at a dose rate of 500 cGy/min with a linear accelerator (Varian 2300EX, Varian, Palo Alto, CA) that generated 6MV X-ray.
Statistical analysis
Statistical analysis of the data was performed using SPSS software version 20.0. Both paired and unpaired t-tests were performed to analyze the data between two experimental groups. Mann-Whitney tests were used to calculate the P values for IHC staining quantification. Data were presented as the means ± standard deviation (SD). Statistical significance was defined as a P-value less than 0.05. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significance.
Discussion
NPC is an EBV-related cancer, which is especially prevalent in Guangdong province, China [
34]. Clinically, the challenges following relapse after irradiation are unavoidable and one of the primary reasons for the poor prognosis of cancer treatment [
16,
35]. However, the potential mechanisms for tumor relapse are complex and remain poorly understood. NPC is a type of solid carcinoma with several elements of the TME. To date, the TME has gained increasing attention as a nutrient source required for cancer aggressiveness. Current research has mainly focused on the TME rather than only on the tumors, emphasizing the importance of the TME in tumor progression.
The TME is composed of cellular and non-cellular components and has been reported to promote tumor recurrence following irradiation [
13,
36]. Of the plethora of cell types within the TME, CAFs are predominant, present in abundance and exhibiting distinct properties in numerous cancers [
37]. In head and neck cancer, a high degree of CAF infiltration was reported to accelerate tumor progression via the regulation of metabolism activity [
6]. Moreover, CAFs resident in NPC have been reported to promote tumor migration and invasion, thereby contribute to a poor prognosis. Despite these findings, studies involving CAFs and irradiation in NPC are inadequate. According to our findings, there was substantial CAF infiltration in the radioresistant NPC tissue compared with radiosensitive NPC tissue. Generally, tumors with abundant stromal components are usually stiff. This finding was consistent with the observation that patients with poor clinical efficacy usually present with stiff tumor [
38]. Moreover, CAFs were found to both enhance survival and induce radioresistance in irradiated NPC cells, illustrating the multifunctionality of CAFs within the TME for NPC. As previously documented, CAFs modified the TME primarily through the continuous secretion of inflammatory cytokines essential for tumor proliferation, migration, invasion, metastasis and, resistance to treatment (e.g., vascular endothelial growth factor [VEGF] and HGF) [
39‐
41]. Pathways that respond to irradiation, including the Wnt /β-catenin, β1-integrin, and P38 signaling pathways, have been identified [
10,
42,
43]. In addition, IL-8 and HGF have been reported to be secreted by CAFs to promote HNSCC aggressiveness and were associated with worse prognosis [
44,
45]. Based on our research, CAFs were found to enhance cell proliferation after irradiation via the secretion of IL-8 to trigger NF-κB activation with up-regulation of p-p65 in NPC cells. Disruption of NF-κB signaling impaired the recovery of tumor cells after irradiation promoted by CAFs.
In general, radiation can destroy DNA and result in damage ranging from nucleotide lesions to single- and double-strand breaks (SSBs and DSBs) [
46], thereby killing tumor cells. The underlying mechanisms responsible for radioresistance are highly variable, among which include DNA damage repair [
47,
48]. Recently, CAFs have been reported to assist cancer cell recovery from irradiation through autophagy, which is related to the DNA damage repair pathway [
13]. Therefore, we hypothesized that CAFs could reduce the level of DNA damage to NPC cells following irradiation. Detection of the γ-H2AX protein that participated in the process of DNA damage [
49,
50] showed that CAFs actually impaired tumor cell DNA damage. Moreover, NF-κB signaling was reported to engage in irradiation resistance in glioblastoma and pathogenesis of several cancer [
51,
52]. Accordingly, we showed that a blockade of IL-8/NF-κB signaling interrupted NPC survival after irradiation, emphasizing the significance of the NF-κB pathway during the process of radioresistance.
Even though CAFs can have multiple cells-of-origin, such as stellate cells, mesenchymal stem cells, mesothelial cells and several other potential sources, resident fibroblasts are considered to be the most important source of CAFs [
53]. In our study, CAFs were proved to promote irradiated tumor proliferation when co-injection with tumor cells in vivo. Similarly, it was reported that CAFs co-injected with tumor cells could be detected and persisted in tumor tissues [
54,
55]. Nonetheless, it was also demonstrated by several studies that CAFs in xenograft tumors disappear shortly after co-injection [
56]. We speculated an early enhancement of proliferation in tumor cells when co-injection with CAFs. This effect could persist for an indicated time once the critical signaling pathways were activated in tumor cells, even though injected CAFs gradually disappeared later. Insights into the mechanisms require further investigation.
Due to the predominant abundance within TME and the versatile properties essential for tumorigenesis, CAFs were considered to be a potential target cell type for therapeutic options. Previous studies have implied that Tranilast, a CAF inhibitor, could inhibit CAF proliferation and activation, as well as suppress the release of bioactive cytokines produced by CAFs [
57]. Consistent with these reports, our data demonstrated that Tranilast inhibited CAF proliferation and reduced CAF activation as evidenced by lower α-SMA expression. Functionally, Tranilast impeded NPC cell growth after irradiation and restored the radiosensitivity of tumor cells against CAFs. Mechanically, Tranilast blocked the activation of NF-κB pathway and reversed the DNA damage induced by irradiation. All data indicate that Tranilast may be a promising inhibitor against CAFs to accelerate NPC elimination after irradiation. And the potential function of Tranilast to sensitize NPC to irradiation needs further research in the future.
Conclusions
In summary, our findings revealed the importance of CAFs in NPC progression following irradiation by enhancing the survival of tumor cells, which promotes radioresistance. Mechanistically, IL-8 secretion by CAFs activated the NF-κB pathway in NPC, thus reducing the level of DNA damage caused by irradiation. Importantly, treatment with Tranilast inhibited CAF functionality and sensitized the tumor cells to irradiation; however, additional clinical samples are warranted to further validate the relationship among CAFs, radioresistance in tumor cells.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.