Background
Nasopharyngeal carcinoma (NPC) is an aggressive form of cancer that lacks differentiation. Due to the radiosensitivity and anatomic location of NPC, radiotherapy is the standard treatment [
1]. While radiotherapy is currently regarded as the most effective treatment for NPC, the emergence of radiation resistance and tumor recurrence following irradiation are major hindrances to successful treatment [
2]. To improve the prognosis for NPC patients and increase 5-year survival rates, researchers have focused on preventing radioresistance and enhancing radiosensitivity [
3]. Furthermore, the underlying mechanisms that lead to the growth and metastasis of NPC need to be fully clarified, highlighting the need for effective therapeutic strategies for the treatment of patients with NPC.
Radiation-induced cell death occurs through the induction of deoxyribonucleic acid (DNA) damage, which may then lead to apoptosis, senescence and autophagy [
4]. Autophagy is primarily described as a mechanism of radioresistance in cancer, but its exact role is complex and unclear, which has impeded progress in the development of agents targeting autophagy for cancer treatment [
5]. Autophagy, the lysosomal breakdown of cytosolic components, is a crucial homeostatic process that is triggered when subjected to stress and appears to be related to DNA damage [
6]. This process is critical for the maintenance and structural reconstruction of cell homeostasis by eliminating aging cells or those with internal organ damage, misfolded proteins, and invading pathogens. The formation of autophagosomes is the morphological hallmark of autophagy and requires the involvement of multiple factors, including Beclin1, LC3B, and p62 [
7]. Several signaling molecules restrain this process, with the adenosine 5′-monophosphate (AMP)-activated protein kinase (AMPK)/mammalian target of rapamycin (mTOR) signaling pathway being the definitive pathway implicated in autophagic control [
8].
The DNA-binding protein RPA (RPA1, 2, and 3) is a heterotrimeric protein that functions in the ataxia telangiectasia mutated protein (ATM)/ATR-mediated DNA damage response (DDR) [
9]. Moreover, (1Z)-1-[(2-hydroxyanilino) methylidene] naphthalen-2-one (HAMNO) is a new protein interaction inhibitor of RPA that specifically blocks RPA1′s interaction with ATR, and thus, checkpoint engagement in the context of replication stress is suppressed [
10,
11]. Exome-wide association analysis in our earlier investigation revealed RPA1 to be a unique predictive biomarker for NPC. Evidence from both gain- and loss-of-function experiments indicated that RPA1 aided in the development, invasion, metastasis, and radioresistance of NPC cells. Additionally, our lab suggested that increasing radiation sensitivity in NPC by therapeutically targeting RPA1 may be possible, but the mechanism behind this effect remains unknown [
12].
Here, we demonstrated that pharmacological suppression of RPA alone in the current research inhibited proliferation via apoptotic mechanisms and increased radiosensitivity in NPC cells. Mechanistically, the improved effectiveness of concurrent RPA and radiation and/or autophagy inhibition with chloroquine (CQ) may be explained by the fact that RPA inhibition activated AMPK, which induced autophagy. Overall, our findings indicate that a treatment approach including inhibitor combinations that concurrently block various metabolic pathways, including autophagy, may be extremely successful for enhancing NPC sensitivity to radiation.
Methods
Cell culture
SYSUCC provided human NPC cell lines (5-8F, S26, and CNE2), while HEK293T cells were purchased from ATCC. All cell lines were grown in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, USA) with 10% fetal bovine serum (FBS, Gibco, USA) at 37 °C in a humidified environment containing 5% carbon dioxide. Using a mycoplasma testing kit (Vazyme, China), we found that none of the cell lines were infected with the pathogen. Cell lines were maintained in a healthy physiological state by passaging them every two days when they reached approximately 80–90% confluency and were not passaged for more than two months.
Western blotting and antibodies
With a phosphatase and protease inhibitor cocktail (Beyotime, China), cells were lysed and sonicated in RIPA (Beyotime, China) lysis solution. Proteins were separated by SDS‒PAGE and then transferred to a PVDF membrane (Merck Millipore, USA). Following 1 h at room temperature in a buffer containing 5% skim milk (BD Biosciences, USA) and 0.1% Tween 20 (Beyotime, China), the membranes were blocked. The membrane was then treated with the primary antibodies specified overnight at 4 °C, followed by incubation with goat anti-mouse or anti-rabbit secondary antibodies conjugated with horseradish peroxidase (HRP) (Cell Signaling Technology, USA) for 1 h. PierceTM ECL western blotting substrate (Thermo Fisher, USA) and Bio-Rad ChemiDoc Touch were used to detect and analyze the proteins. Antibodies against the following targets were used in this study: phospho-mTOR, mTOR, LC3B, P62, cleaved caspase 3, γ-H2A (5536, 2983, 3868, 23214, 9661, 2577, Cell Signaling Technology, USA), phospho-RPA2 (381220, Zen BioScience, China), actin (A5441, Sigma‒Aldrich, USA), RPA2 (ab76420, Abcam, USA), and Ki-67 (14-5698-82, Thermo Fisher, USA).
Immunofluorescence staining
Glass coverslips were used to cultivate cells, which were then treated with HAMNO for the durations specified before being fixed in 4% paraformaldehyde for 20 min. Following a 10-min permeabilization in 0.3% Triton X-100 (Beyotime, China) in phosphate buffer saline (PBS), a 1-h blocking in 5% goat serum (Gibco, USA) and 0.3% Triton X-100 in PBS, and an overnight incubation at 4 °C with the appropriate primary antibodies, the cells were analyzed. Three washes in PBS were followed by incubation with a 1:750 dilution of an Alexa-conjugated secondary antibody after the cells had been stained. After 10 min in antifade mounting solution containing 4′,6-diamidino-2-phenylindole (DAPI, Beyotime, China), the cells were observed under a 63 × objective in a confocal laser scanning microscope (Carl Zeiss, Germany). The software package Zen 3.4 blue edition was used to calculate the relative fluorescence intensities of LC3B and γ-H2A. In each of three separate tests, ≥ 50 cells were counted from each group.
Cell proliferation, colony formation and tumorsphere formation
Increased doses of HAMNO (NSC111847, InvivoChem, USA) were used to stimulate cell growth in CNE2, S26, and 5-8F cells plated at 1000 cells per well in 96-well plates. At the given period, cell viability was determined by counting viable cells after staining with Trypan blue (Gibco, USA). For analysis of cell confluence, plates were automatically monitored and recorded every 4 h by the IncuCyte S3 system (Essen BioScience, USA), and cell proliferation rates were evaluated by IncuCyte 2021C software. Colony formation assays were performed by seeding duplicate wells of 6-well plates with 1000 cells and then exposing them to X-ray doses of 0, 1, 2, 3, 4, and 5 Gy via a Rad Source R2000 irradiator. Cells were fixed with 4% paraformaldehyde for 10 min and then stained with 0.5% crystal violet solution after plate colony formation (12–14 days). With a light microscope, colonies with > 50 cells were tallied. The survival fraction was normalized to the number of colonies of nontreated cells. For 3D anchorage-free colony formation, 2000 S26 and 5-8F cells in DMEM enriched with 10% FBS were plated into 96-well ultralow attachment microplates (Corning, USA) and monitored by the IncuCyte S3 system. The culture medium was refreshed every three days throughout the experimental period. The volume of 3D colonies was calculated using the following formula: Volume = (4/3) πR3. For soft agar colony formation, cells were suspended in 100 μL of medium consisting of 10% FBS with 0.3% agar (Sigma‒Aldrich, USA) at 100 cells per well and seeded in 96-well plates with 0.6% bottom agar. The sizes of the colonies were determined by taking photographs of tumor spheres utilizing a phase-contrast microscope (Olympus microscope IX71, Japan). There were ≥ three separate experiments performed in triplicate.
Immunohistochemistry (IHC)
IHC was performed according to the standard protocol. Antigen retrieval was carried out by boiling paraffin-embedded sections in EDTA (pH 8.0, 0.05% Tween 20, 1 mM EDTA) (Sigma‒Aldrich, USA) for 20 min. After being treated with primary antibodies against Ki-67 and cleaved caspase 3 diluted 1:200 in 3% bovine serum albumin (BSA) in PBS at 4 °C overnight, the sections were blocked with 3% goat serum. The Dako REAL EnVision Detection System (Dako, Denmark) was employed for immunostaining via HRP conjugates. Two expert pathologists utilized the immunoreactivity score (IRS) technique to quantify cleaved caspase 3 and Ki-67 expression. The proportion of tumor cells that were positive was measured and assigned a score between 1 (25%) and 4 (> 75%). Zero, one, two, and three indicated no, light yellow, moderate, and dark brown staining, respectively. The extent was multiplied by the intensity to obtain the final IHC score.
Transmission electron microscopy
Ten-centimeter dishes were used for cultivating 5-8F or S26 cells, and then, they were treated with 10 μM HAMNO for 24 h. After being scraped gently with a cell scraper (Corning, USA), the samples were fixed in 2.5% glutaraldehyde (Sigma‒Aldrich, USA) for 5 min at room temperature after drug treatment. Following overnight fixation in 2.5% glutaraldehyde at 4 °C, the cells were centrifuged at a slower speed (3000 rpm min−1) to collect the precipitate. Electron microscopy (JEM-1400flash) was used to examine the samples at Sun Yat-sen University.
Quantitative RT–PCR
The RNeasy Mini Kit (Qiagen, Germany) was used to extract total cellular RNA, and the QuantiTect Reverse Transcription Kit (Qiagen, Germany) was used to reverse-transcribe 1 μg of total RNA. Using the QuantStudio 7 Flex (Thermo Fisher, USA) and TB Green® Premix Ex Taq™ II (TaKaRa, Japan), we carried out real-time quantitative PCR. All PCRs were run in triplicate, and the 2 − ΔΔCq technique utilized for assessing the amplification products and relative expression levels were calculated via the housekeeping gene β-actin as a standard. Additional file
1: Fig. S3 contains the amplification primers for SQSTM1, WIP1, GABARAPL1, and β-actin.
Glucose uptake assay, lactate production and intracellular ATP level measurement
After 2 h of incubation with 100 μg/mL 2-deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]-d-glucose (2-NBDG, Cayman Chemical, USA) in glucose-free medium, fluorescence was measured via the IncuCyte S3 system at excitation and emission wavelengths of 485 nm and 535 nm. Following the manufacturer’s protocol, lactate was measured in cell lysates using the CheKine™ Micro Lactate Assay Kit (Abbkine, China). A total of 1 × 106 cells were centrifuged at 4 °C after being lysed in 200 μL/well passive lysis buffer on ice for the ATP assay. The ATP content in the supernatant was determined using the Molecular Probes® ATP Determination Kit (Invitrogen, USA) in adherence to the recommended protocol. There were three sets of each experiment.
In vivo xenograft evaluation
The Sixth Affiliated Hospital of Sun Yat-sen University’s Committee on the Ethics of Animal Studies examined and authorized all animal studies (IACUC-2022052701). All authors followed all applicable ethical guidelines for the use of animals in scientific studies. We ordered four-week-old female BALB/c nude mice from Beijing Laboratory Animal Co., Ltd. Mice were kept in microisolator cages with a 12-h light/12-h dark cycle, 18–22 °C temperatures, and 45% humidity. Subcutaneous injections of S26 cells (1 × 106 cells in 100 μL of PBS) were made in the flanks of mice. Every three days after therapy, the tumor volume was determined using the following formula: volume = length × width2 × 0.5. When the xenograft tumor diameters reached ~ 5 mm, treatment was initiated. Animals were randomly grouped before receiving vehicle control, treatment, or combination therapy. HAMNO (2 mg/kg, 5% DMSO, 40% PEG300, 5% Tween 80) and DMSO as a vehicle control were administered intraperitoneally every 3 days twice. For combination therapy, mice were either nonirradiated or irradiated with 6 Gy once, followed by treatment with HAMNO (1 mg/kg) twice on day 2 and day 5 after irradiation. Injections of 60 mg/kg hydroxychloroquine (Sigma‒Aldrich, USA) were given intraperitoneally on days 1, 3, and 6. Mice were euthanized at the experimental end point, and xenografts were harvested, fixed in 10% formalin overnight, and paraffin embedded for histologic analysis.
Flow cytometry apoptosis assay
S26 and 5-8F cells were collected 48 h following HAMNO, DMSO or irradiation therapy and subsequently subjected to an apoptosis assay. Apoptosis analyses were performed via an Annexin V-APC/7-AAD Apoptosis Detection Kit (Multi Sciences, China). Briefly, we centrifuged the samples at 300×g for 5 min to separate detached cells from the supernatant and after EDTA-free trypsin recovery. Annexin V incubation reagent (1% Annexin V-APC and 1 × 7-AAD solution) was used to incubate the samples at room temperature for 30 min in the dark after they were resuspended in 500 μL of 1 binding buffer. A cytoFLEX flow cytometer was used to measure the apoptosis frequency, and the data were processed using CytExpert 2.2. APC − /7-AAD − cells were considered viable cells, APC + /7-AAD − cells were considered early apoptotic cells, and APC + /7-AAD + cells were considered late apoptotic or dead cells. The mean and standard deviation were calculated for n = 3 independent biological replicates.
Gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA)
The GSE12452 dataset was downloaded from the Gene Expression Omnibus (GEO) database. High RPA1/RPA3 expression was defined as the top 50% of individuals in the GSE12452 dataset, while low RPA1/RPA3 expression was defined as the bottom 50% of patients. We assessed autophagy-associated genes and made a customized “autophagy‒lysosome signature” via a previously described method [
13]. Then, we used GSEA using the R package clusterProfiler to find statistically significant functional variations between the two groups [
14]. Pathways with a normalized enrichment score (|NES|) > 1) and a p value < 0.05 were considered significantly enriched. The GSVA score was calculated for each sample in the TCGA COAD based on the expression of genes in the gene set using the GSVA R package [
15]. The GSVA scores of the two groups were compared using the Mann‒Whitney U test, and a p value < 0.05 was considered statistically significant. GSVA was utilized to assess the change and variation in pathway activities based on previously published gene signatures [
16].
Statistical analysis
Except when otherwise noted, all statistical testing was performed using GraphPad Prism version 8.3.0. Unless otherwise indicated, data that had a normal distribution are reported as the mean ± SD of triplicate experiments. Student’s unpaired t test was used to compare one single treatment to one control. Multiple treatment or condition studies were analyzed using one- or two-way ANOVA followed by a Dunnett or Tukey multiple comparison test to compare the means of each treatment to those of a predetermined control group. For the data that were not normally distributed, they were reported as the median with a 95% confidence interval (CI). Statistical significance was assessed using the following thresholds throughout the paper: *, p 0.05; **, p 0.01; ***, p 0.001. The primary raw data are available at
www.researchdata.org.cn, which is the Research Data Deposit. This document includes its data sources.
Discussion
Genomic instability is a well-known cancer hallmark, and developing drugs targeting the DDR is a potential therapeutic strategy for various solid cancers. Chemotherapeutic agents and therapeutic radiation initiate DNA damage and induce cell cycle arrest or apoptosis of tumor cells, which consequently leads to treatment resistance [
18]. Therefore, agents that suppress DDR signaling pathways are supposed to potentiate the cytotoxic impact of chemotherapy and radiotherapy as ‘sensitizing agents’ and thus overcome resistance. The key component of DDR, ATR-CHK1, might be pharmacologically inhibited as a therapeutic strategy for cancer [
19]. Clinical studies of certain ATR and CHK1 inhibitors demonstrated significant cytotoxicity, and numerous candidate molecules with enhanced safety profiles are now being explored (e.g., NCT03682289, NCT05071209, NCT02203513). However, the occurrence of off-target effects caused by ATR and CHK1 inhibition could involve pathways other than the DDR and cause cell death in both normal and cancer cells [
11]. Furthermore, the embryonic mortality of mice lacking ATR and CHK1 suggested that directly blocking the ATR and CHK1 pathways might have deleterious effects. Thus, ATR-CHK1 pathway downstream components could be useful for treating cancer [
11]. Following DSBs or at stalled replication forks, RPA binds to single-strand DNA (ssDNA) and therefore activates the ATR/CHK1 pathways, resulting in activation of the G2/M and intra-S phase cell cycle checkpoints and initiation of DNA repair. The RPA heterotrimers are downstream of the ATR substrate and may be a considerable pharmacological target for cancer therapy [
20].
Because replication stress is higher in tumor cells and the leading cause of genome instability, targeting replication stress has enabled the discovery of new cancer vulnerabilities [
21]. The RPA-ssDNA platform is a key sensor that triggers the DDR in reaction to genotoxic stressors, making it vital for genome integrity [
22,
23]. HAMNO is a small molecule inhibitor of RPA that blocks checkpoint engagement in reaction to replication stress by inhibiting the link between RPA1 and ATR/ATRIP [
10,
11]. This molecule works effectively with etoposide to induce replication stress and selectively increase cell death in cancer cells that exhibit constant DNA replication stress. This strategy has benefits in the clinic, as synergism of RPAis and other chemical drugs would enhance therapeutic efficacy while minimizing undesirable side effects [
11]. We previously determined that RPA1 increases tumor growth and resistance to therapeutics in NPC, which affects the prognosis for individuals with NPC, using bioinformatic investigations and biological evidence [
12]. Inhibition of RPA by HAMNO was shown to have a significant antitumoral impact on NPC cells both in vivo and in vitro. The standard therapeutic approach for NPC is radiotherapy. Nevertheless, radioresistance remains the major factor in suboptimal therapeutic outcomes and poor prognosis. Our findings showed that pharmacological inhibition of RPA increased the radiation-induced DDR and enhanced the radiosensitivity of NPC cells. These findings suggest that HAMNO may be useful in NPC treatment.
Autophagy functions as a double-edged sword in the mechanisms of radioresistance in NPC. While certain factors, such as lncRNA CASC19 and LACTB2, promote radioresistance by inducing autophagy, showing a positive correlation [
24,
25], LUC7L2 conversely enhances resistance by inhibiting autophagic flux [
26]. Our research demonstrated that inhibiting RPA increased autophagic flux, rendering NPC cells more responsive to autophagy inhibition. Moreover, radiation promotes autophagy in radioresistant NPC cells, and in turn, inhibition of autophagy reverses radioresistance [
27]. Consistent with these findings, our study delved into the role of RPA in NPC radiotherapy. Notably, our investigation revealed that the concurrent inhibition of RPA and autophagy heightened the radiosensitivity of NPC cells, suggesting a promising avenue to enhance therapeutic efficacy. This novel mechanistic insight provides a foundation for the development of combination therapeutic strategies targeting both RPA and autophagy to overcome radioresistance specifically in NPC.
Additionally, our data demonstrated a notable escalation in DNA damage following RPA inhibition, suggesting a compelling relationship between DNA damage and RPAi-induced autophagy. DNA damage is a potential stimulus for autophagic initiation—an adaptive mechanism intended to mitigate cellular harm and restore homeostasis [
28,
29]. Thus, one of the reasons for the heightened autophagic flux observed upon RPA inhibition might be a cellular strategy to counteract the stress induced by DNA damage.
Our investigations revealed that RPA inhibition induced a notable reduction in glycolytic activity. This observation raises intriguing questions about the potential links among RPA, metabolic pathways, and autophagic flux. We also found that RPAi activated AMPK and inhibited the mTOR signaling pathway. Given AMPK’s role as a key sensor in glycolysis and energy homeostasis [
30], the activation of AMPK upon RPA inhibition not only aligns with autophagic initiation but also supports the idea that RPA inhibition might trigger a metabolic shift away from glycolysis. Consequently, we hypothesize that the enhanced autophagic flux resulting from RPA inhibition is, in part, attributed to the attenuation of glycolysis. The suppression of glycolytic activity potentially curtails the cell’s energy supply and prompts AMPK-driven autophagy as an alternative energy source. This intricate relationship among RPA inhibition, metabolic shifts, and autophagy further underscores the complexity of cellular responses under RPA modulation. While our observations hint at an intriguing interdependence between DNA damage and RPAi-induced multiple metabolic pathways, further targeted experiments are necessary to comprehend their underlying interactions.
Our study has several noteworthy strengths. Foremost, we elucidated the mechanisms underlying RPA inhibition-induced autophagy in NPC cells, offering new insights into autophagic regulation specific to NPC. The induction of autophagic flux, activation of the AMPK/mTOR pathway, and upregulation of autophagy-related genes were identified as key factors in this process. Second, our multifaceted investigation unveiled not only the intricate interplay among RPA inhibition, the DDR, and autophagy but also the involvement of cellular metabolic pathways. This revelation underscores the depth of cellular responses orchestrated under RPA modulation. Moreover, the investigation of combining autophagic inhibition (using CQ or genetic inhibition of ATG5) with RPA inhibition showed that this combination therapy is more effective in enhancing NPC’s antitumor response to radiation than monotherapy. This finding underscores the potential synergistic effects of a combination approach targeting multiple pathways.
We also acknowledge certain limitations in our study. Our study primarily assessed the short-term efficacy and therapeutic outcomes of the proposed treatment approach. Future research is warranted to investigate the long-term results, such as treatment durability, potential adverse reactions, and toxicity. Additionally, future studies exploring the direct links among RPA inhibition, glycolytic modulation, and autophagic induction are required for a comprehensive understanding of the intricate crosstalk within the cellular machinery.
Conclusions
Our study has revelated the potential of RPA inhibition in the treatment of NPC. We demonstrated that pharmacological inhibition of RPA with HAMNO exhibited potent antitumor effects both in vitro and in vivo and enhanced the NPC response to radiotherapy.
We further elucidated the underlying mechanisms of RPA inhibition. We observed that inhibiting RPA activates the AMPK/mTOR signaling pathway, resulting in the induction of autophagic flux, which was supported by an increase in the expression of autophagy-related genes. Additionally, our research revealed a dual impact of RPA inhibition on cell metabolism. This molecule impairs glycolysis while simultaneously stimulating autophagy, creating a greater reliance on this cellular recycling mechanism.
Our findings lead us to propose a promising treatment approach for NPC. By utilizing RPA complex inhibitors to disrupt metabolic processes, particularly glycolysis, we induce a higher reliance on autophagy for the survival of NPC cells. This increased dependence on autophagy makes the cells more susceptible to the effects of autophagy inhibitors, such as CQ/hydroxychloroquine. Therefore, this combination therapy has the potential to enhance the effectiveness of radiotherapy through the synergistic effects of RPA inhibition and autophagy inhibition.
However, further research is required to validate the efficacy and safety of RPA inhibitors in a clinical setting. Although our study successfully revealed the mechanisms of RPA inhibition, a comprehensive understanding of the intricate molecular interactions involved remains imperative. Our future focus will be on revealing the precise mechanisms involved and identifying potential therapeutic targets.
Overall, our research highlights the therapeutic value of targeting RPA and metabolic pathways in NPC, providing a promising avenue for improving treatment outcomes and potentially overcoming the challenging treatment resistance of NPC to radiotherapy.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.