https://orcid.org/0000-0003-2091-0696
Figures
Abstract
DNA damage-induced by radiotherapy is a critical factor in promoting the death of colorectal cancer cells (CRC). Although high mobility group box 1 (HMGB1) reportedly plays a vital role in tumor radioresistance by modulating DNA damage repair, the precise mechanisms remain unclear. In this study, HMGB1 knockdown markedly enhanced cell apoptosis after radiation. HMGB1 downregulation significantly inhibited DNA damage repair and reactive oxygen species (ROS)-mediated redox homeostasis after irradiation in CRC cells. Mechanistically, HMGB1 interacts with KU70 via its region spanning residues 95–163. This interaction subsequently activates the non-homologous end joining (NHEJ) pathway to facilitate DNA damage repair, ultimately leading to reduced radiation-induced cell apoptosis. KU70 silencing showed the same effect as HMGB1 depletion mediated cell apoptosis and DNA damage response both in vitro and in vivo. Additionally, HMGB1 and KU70 were overexpressed in CRC tissues. Analysis of the GEPIA database indicated that elevated levels of both genes showed a trend toward association with poor patient prognosis, although this did not reach statistical significance. The current study revealed that HMGB1 may promote DNA damage repair through KU70 and its mediated NHEJ pathway to affect apoptosis in CRC cells after irradiation. Thus, targeting the HMGB1/KU70/NHEJ axis may be a potential therapeutic target to promote the response of CRC to radiotherapy and in-depth study of the specific mechanism of this axis in CRC radioresistance will help to the develop more effective treatment strategies.
Citation: Liu X, Han Y, Kuang R, Sheng W, Zhang Y, Jia X, et al. (2026) HMGB1 reduce DNA damage by binding KU70 to activate NHEJ pathway in colorectal cancer cells after radiation. PLoS One 21(3): e0345635. https://doi.org/10.1371/journal.pone.0345635
Editor: Zu Ye, Zhejiang Cancer Hospital, CHINA
Received: December 29, 2025; Accepted: March 7, 2026; Published: March 24, 2026
Copyright: © 2026 Liu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data generated or analyzed during this study are included in this published article and its Supporting information files.
Funding: This work was supported by the National Natural Science Foundation of China (No. 82203543), the Natural Science Foundation of Shandong Province (No. ZR2022QH267 and ZR2023LZL008), the Startup Project of Binzhou Medical University (No. BY2021KYQD08), and the Medical and Health Science and Technology Development Project of Shandong Province (No. 202309030619). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: No authors have competing interests.
Abbreviations: Co-IP, co-immunoprecipitation; CRC, colorectal cancer; DSBs, Double-strand breaks; FBS, fetal bovine serum; HMGB1, high mobility group protein 1; HR, homologous recombination; IHC, immunohistochemistry; IR, ionizing radiation; NC, negative control; NHEJ, non-homologous end joining; P/S, penicillin–streptomycin; PBS, phosphate buffered saline; PI, propidium iodide; PLA, proximity ligation assay; ROS, reactive oxygen species; si-RNA, small interfering RNA
Introduction
Colorectal cancer (CRC) stands as the most frequently diagnosed malignancy and a leading cause of cancer-related mortality globally for both men and women [1]. Postoperative chemoradiotherapy has proven effective in reducing local recurrence and improving overall survival in CRC patients [2]. However, the emergency of radioresistance in CRC cells significantly compromises the efficacy of radiotherapy, often resulting in incomplete tumor eradication and an elevated risk of recurrence. Among the diverse types of DNA damage induced by ionizing radiation (IR), double-strand breaks (DSBs) are recognized as the most cytotoxic lesions. The fate of irradiated cells is critically influenced by unrepaired or misrepaired DSBs, while contributes to both cell death and carcinogenesis [3]. Consequently, targeting DSB repair mechanisms, particularly the error-prone non-homologous end joining (NHEJ) pathway, represents a fundamental strategy for radiosensitization.
High mobility group box 1 (HMGB1) protein is a highly conserved, non‑histone nuclear protein that contributes to genomic stability by binding to DNA and histones to stabilize nucleosome structure, thereby differing protection against genotoxic insults from mutagens and radiation. It also modulates histone–DNA interactions to influence chromatin compaction [4]. HMGB1 is aberrantly expressed in various malignancies, including CRC, and its expression is associated with radiotherapy resistance, chemotherapy resistance, and DNA damage in tumors [5]. Previous work has demonstrated that HMGB1 inhibition sensitizes esophageal squamous cell carcinoma to radiation by impairing DNA damage repair [6]. In the context of CRC, we have recently shown that glycyrrhizin-mediated suppression of HMGB1 attenuates tumor progression through the modulation of NHEJ [7]. Nevertheless, the molecular mechanism by which HMGB1 regulates NHEJ activity in irradiated CRC cells remain to be fully elucidated.
The core NHEJ machinery is initiated by the KU70/KU80 heterodimer, which binds to DSB ends and recruits downstream effectors, including XRCC4 and DNA ligase IV [8]. It is noteworthy that HMGB1 interacts with KU70 in contexts such as nasopharyngeal carcinoma, influencing the resistance to irradiation and cisplatin [9]. Our prior studies indicate that radiation induces both the expression of HMGB1 and its translocation to the cytoplasm [10]. However, a fraction of HMGB1 remains within the cell nucleus post-irradiation, raising the question of whether it participates in DNA damage repair and thereby mediates radioresistance in CRC. In this study, we investigated the functional and molecular crosstalk between HMGB1 and KU70 in the DNA damage response of CRC. We tested the hypothesis that HMGB1 promotes radioresistance by scaffolding the KU70-dependent assembly of the NHEJ repair complex and evaluated its prognostic relevance in clinical specimens based on the GPEIA database.
Materials and methods
Cell culture
HT29 (RRID: CVCL_0320) and HCT116 (CVCL_0291) human CRC cell lines, obtained from Procell (Wuhan, China), were cultured in DMEM (CellMax, Beijing, China) supplemented with 10% fetal bovine serum (FBS; Vistech, Nanjing, China) and 1% penicillin–streptomycin (Solarbio, Beijing, China) at 37 °C in a humidified atmosphere of 5% CO₂ atmosphere. Short tandem repeat analysis (for cell identification) and mycoplasma detection were performed before the experiment (2024.06).
Cell transfections
HMGB1-stable knockdown cell lines was established as previously described [10]. si‑KU70 (RIBOBIO, Guangzhou, China) and its negative control were transfected into cells using Lipo2000 (Vazyme, Nanjing, China) according to the manufacturer’s instructions.
Irradiation
In vitro experiments employed a single radiation dose of 4 Gy administered using an X-RAD 320 cabinet irradiator (Precision X-Ray, USA) at a dose rate of 2 Gy/min. For in vivo experiments, mice received a single localized tumor irradiation dose of 10 Gy.
Quantitative RT-PCR (qPCR)
Total RNA was extracted using RNA-Easy Reagent (Vazyme). Subsequently, 1 μg of RNA was reverse transcribed with the HiScript II One-Step RT-PCR Kit (Vazyme) and subjected to PCR analysis in triplicate on an ABI 7500 system (Applied Biosystems, MA, USA) using ChamQ SYBR qPCR Master Mix (Vazyme). Primer sequences are listed in S1 Table.
Western blot
Cells were lysed using RIPA lysis buffer (Solarbio, Beijing, China) supplemented with a protease inhibitor cocktail. Proteins (30 μg) were separated by 12% SDS-PAGE, transferred onto a PVDF membrane (Merck Millipore, Darmstadt, Germany) and blocked with 5% non-fat milk. The membranes were subsequently incubated overnight at 4°C with primary antibodies. HRP-conjugated secondary antibodies were used for detection with an ECL substrate (Tacan, Beijing, China), and signal intensities were quantified using ImageJ software. Detailed information for all primary antibodies employed in this study is provided in S2 Table.
Comet assay
DNA damage was assessed using the KeyGen comet detection kit (Jiangsu, China). In brief, cells were adjusted to a destiny of 1 × 106 cells/ml and embedded in 0.7% low-melting-point agarose. The embedded cells were then lysed for 2 h at 4°C, followed by alkaline electrophoresis buffer (0.186 g Na2EDTA, 6 g NaOH dissolved in 500 ml water) for 1 h. Electrophoresis was performed at 25 V/cm for 30 min. Subsequently, the samples were neutralized with a 0.4 M Tris-HCl (pH 7.5) and stained with propidium iodide (PI). For each experimental group, 100 nuclei were randomly selected and analyzed for tail moment using CASP software.
Co-immunoprecipitation (Co-IP)
1 mg of whole-cell lysate was pre-cleared and then subjected to overnight incubation at 4°C with 4 μg of anti-HMGB1, anti-KU70 or control IgG, together with protein A/G magnetic beads (Absin, Shanghai, China). Finally, the beads were washed, eluted and analyzed by western blot.
Immunofluorescence (IF) and Proximity ligation assay (PLA)
Cells grown on glass coverslips were fixed with 4% paraformaldehyde (PFA) for 15 min, permeabilized with 0.2% Triton X‑100 for 10 min, and blocked for 1 h (using 1% BSA for immunofluorescence or Duolink® blocking buffer for proximity ligation assay). Primary antibodies (listed in S2 Table) were applied and incubated overnight at 4 °C. Subsequently, samples were incubated with either Alexa Fluor‑conjugated secondary antibodies (1:500, 1 h) for immunofluorescence or with the Duolink® PLA reagents (#DUO92101, Sigma‑Aldrich, NJ, USA) according to the manufacturer’s protocol. Fluorescence signals were captured using a Leica confocal microscope, and mean fluorescence intensity was quantified with ImageJ. For γH2A.x foci quantification, images were analyzed in ImageJ using automated thresholding and particle analysis. Cells containing more than five distinct nuclear foci were scored as positive, whereas those showing diffuse pan-nuclear staining or lacking clear foci were excluded. At least 300 randomly selected cells were assessed per experimental condition, and results are expressed as the percentage of positive cells.
Molecular docking
The 3D structures of HMGB1 and KU70 were obtained from the UniProt database (https://www.uniprot.org/). Molecular docking was carried out using the HDOCK server under default parameters. The resulting complex with the lowest binding energy was selected and visualized using PyMOL version 2.5.5.
Detection of caspase-3 activity
Cells were harvested 24 h after irradiation (0 or 4 Gy) and caspase-3 activity was assayed using a colorimetric kit (Beyotime, Shanghai, China) according to the manufacturer’s instructions.
Reactive oxygen species (ROS)
Cells were cultured for 24 h post-irradiated, then incubated with 10 µM DHE (#CA1420, Solarbio) for 30 min at 37 °C. DHE fluorescence intensity in CRC cells was subsequently measured using a 96-well plate reader with excitation at 488 nm and emission at 525 nm.
Xenotransplantation assays
Female BALB/c nude mice (n = 30, aged 6–8 weeks, purchase from PengYue, Jinan) were housed under SPF conditions with a 12-h light/12-h dark cycle and acclimatized for 1 week. A total of 5 × 10⁶ HT29 cells suspended in 100 µL PBS was subcutaneously inoculated into the right flank of each mouse. When tumor volume reached approximately 20 mm3, the mice were randomly allocated into groups (n = 5 per group) and subjected to intratumorally injection of si-KU70 (0.8 nmol per 20 g body weight) or localized tumor irradiation at a dose of 10 Gy. Mice were anesthetized via inhalation of 3% isoflurane immediately before irradiation. General health was monitored daily, and tumor dimensions were measured every other day to calculate volume using the formula: (length × width²)/ 2. At the experimental endpoint (24 d), xenograft tumors were harvested, photographed, and weighed. Humane endpoints were strictly observed, with any mouse developing a tumor exceeding 1000 mm³ promptly euthanized. At the conclusion of the study, all mice were euthanized by CO₂ asphyxiation. All animal procedures were conducted in accordance with the ARRIVE guidelines and were approved by the Animal Experimental Ethics Committee of Binzhou Medical University (Approval No. 2022−121).
Immunohistochemistry (IHC) and TUNEL assay
Tumor samples were fixed in 10% formalin for 24 h, embedded in paraffin and sectioned at a thickness of 5-μm. Following antigen retrieval in citrate buffer (pH 6.0) and blocking with 3% H₂O₂ and 10% goat serum, sections were incubated overnight at 4°C with primary antibodies against γH2A.x or Ki67. After incubation with an HRP‑polymer secondary antibody (Vazyme), signals were developed using DAB. Apoptosis was assessed by TUNEL staining performed with a one‑step commercial kit (Elabscience, Hubei, China) according to the manufacturer’s instructions, and nuclei were counterstained with DAPI. Positive cells were counted blindly in five randomly selected high‑power fields per section.
Prognosis and clinical phenotype analysis
The expression of KU70 and corresponding survival data were obtained from the GEPIA2 database. The correlation of HMGB1 and KU70 in CRC was analyzed using TIMER2.0. The association with pathological stage was examined via the “Stage Plot” module, and survival curves were generated by Kaplan-Meier analysis, with statistical significance assessed by the log-rank test.
Statistical analysis
All statistical analyses were conducted using GraphPad Prism version 9.0 (RRID: SCR_002798). Data are presented as the mean ± SD from at least three independent experiments, each performed in triplicate. Differences between two groups were assessed using a two‑tailed Student’s t-test. For comparisons involving more than two groups, one‑way analysis of variance (ANOVA) was employed. Statistical significance was defined as follows: *P < 0.05, **P < 0.01; n.s. indicates not significant.
Results
HMGB1 alleviates cell apoptosis by promoting the DNA damage repair in CRC cells
Our previous findings demonstrated that HMGB1 is frequently upregulated in CRC cell lines, particularly in HCT116 and HT29 cells [10]. In the present study, lentivirus-mediated knockdown of HMGB1 significantly decreased its expression at both mRNA and protein levels in HCT116 and HT29 cell lines (Fig 1a). As shown in Fig 1b, HMGB1 knockdown led to a significant increase in caspase‑3 activity, indicating enhanced apoptosis, consist with validation by Annexin V staining reported previously [10]. To investigate whether HMGB1‑associated radiation‑induced apoptosis is linked to the DNA damage response, we examined the formation of γH2A.x foci, a marker of DSBs. IF staining revealed that γH2A.x foci rapidly accumulated at 0.25 h after X‑ray irradiation, and HMGB1 knockdown further elevated the number of γH2A.x foci in CRC cells (Fig 1c). Consistent with these observations, the protein level of γH2A.x increased following 4 Gy irradiation, and this increase was more pronounced in HMGB1‑depleted cells compared with controls (Fig 1d). Furthermore, comet assay results showed that HMGB1 knockdown resulted in a significantly longer tail length after irradiation, reflecting elevated DNA damage (Fig 1e). Collectively, these data demonstrated that HMGB1 attenuates radiation‑induced apoptosis in CRC cells by promoting DNA damage repair.
a. mRNA and protein expression levels of HMGB1 in HCT116 and HT29 cells were assessed by qPCR and western blot following transfection with shRNA targeting HMGB1 (sh-HMGB1) or a negative control shRNA (sh-NC). β-actin was used as a loading control. b. Caspase-3 activity in HMGB1‑knockdown CRC cells after exposure to X‑ray irradiation. c. IF staining of γH2A.x in HMGB1‑knockdown CRC cells at indicated time points following 4 Gy X‑ray irradiation (scale bar: 50 µm). d. Western blot analysis of γH2A.x protein levels in HMGB1‑knockdown CRC cells after 4 Gy irradiation, with β-actin as the loading control. e. DNA damage was evaluated by comet assay in HMGB1‑knockdown CRC cells at 24 h after 4 Gy X‑ray irradiation (scale bar: 20 µm). Tail length per cell was quantified in 100 cells per group. Data are presented as the mean ± SEM from three experiments. **P < 0.01, *P < 0.05.
HMGB1 maintains redox homeostasis and mitigates oxidative DNA damage in CRC cells after irradiation
IR induces DNA damage not only directly through radiation-induced DNA strand breaks, but also indirectly via the generation of free radicals, including reactive oxygen species (ROS), reactive nitrogen species, and H2O2 [11]. Therefore, we further investigated whether HMGB1 influences ROS levels in irradiated CRC cells. As shown in Fig 2a, irradiation significantly promoted ROS production, and knockdown of HMGB1 further elevated basal ROS levels in CRC cells. Moreover, 8-OHdG, a primary marker of oxidative DNA damage, was present at significantly higher levels in the HMGB1 knockdown group compared with the negative control (NC) group after X-ray irradiation (Fig 2b). These findings indicate that elevated basal ROS in HMGB1-knockdown cells sensitizes them to radiation-induced oxidative injury.
a. ROS production in HMGB1‑knockdown CRC cells after 4 Gy irradiation. b. Expression of 8‑OHdG in HMGB1‑knockdown CRC cells after 4 Gy irradiation, with corresponding mean fluorescence intensity shown (scale bar: 10 µm). **P < 0.01, *P < 0.05.
HMGB1 promotes the NHEJ pathway by binding to KU70 in CRC cells
To further investigate the mechanism by which HMGB1 influences radiation-induced DNA damage repair, we quantified the expression levels of key genes associated with the core repair pathways, homologous recombination (HR) and NHEJ in CRC cells following irradiation using qPCR. The results revealed that HMGB1 depletion significantly reduced the expression of NHEJ-related genes KU70, KU80, XRCC4, after irradiation, whereas the expression of the HR-associated gene RAD51 remains unaltered (Fig 3a). Subsequently, we found that irradiation upregulated the protein levels of KU70, KU80, XRCC4, and LIG4, while HMGB1 knockdown suppressed their expression (Fig 3b and Supplementary Fig s1a in S1 Fig). Prior studies have indicated that HMGB1 facilitates resistance to radiation by binding to KU70 in nasopharyngeal carcinoma [9]. Therefore, we hypothesized that HMGB1 activates the NHEJ pathway through its interaction between HMGB1 and KU70, the core factor of the NHEJ pathway, and demonstrated that this interaction was enhanced following radiation exposure (Fig 3c). Moreover, PLA assay results showed a stronger nuclear co-localization of HMGB1 and KU70 in CRC cells after 4 Gy X‑ray irradiation (Fig 3d). To evaluate the binding affinity between KU70 and HMGB1, we performed molecular docking analysis, which yielded a binding energy of −221.94 kcal/mol and indicated that the primary interaction interface resides within amino acid residues 95–163 of HMGB1 (Fig 3e). Collectively, these findings indicate that HMGB1 directly binds to KU70 in irradiated CRC cells, thereby activating the NHEJ pathway to facilitate DNA damage repair.
a. qPCR analysis of the key HR pathway gene (RAD51) and NHEJ genes (KU70, KU80, XRCC4) in HMGB1‑knockdown CRC cells after 4 Gy X‑ray irradiation. b. Western blot analysis of NHEJ‑related proteins (KU70, KU80, XRCC4, LIG4) in HMGB1‑knockdown CRC cells irradiated at 4 Gy; β‑actin served as the loading control. c. Co‑IP assay detecting the formation of the HMGB1-KU70 immune complex in CRC cells after 4 Gy irradiation. d. PLA showing the formation and subcellular localization of HMGB1–KU70 complex in CRC cells post-irradiation. e. Molecular docking model of HMGB1 and KU70; the inset highlights the interaction between HMGB1 and amino acid residues near the binding site. **P < 0.01, *P < 0.05.
HMGB1/KU70/NHEJ pathway alleviates cell DNA damage in CRC cells after irradiation
To further investigate whether HMGB1 regulates the DNA damage repair through KU70, we utilized three distinct siRNA sequences targeting KU70 in a transient transfection assay to assess whether HMGB1 influences the radiosensitivity of CRC cells in a KU70-dependent manner. qPCR and western blot analyses showed that both KU70 mRNA and protein levels were significantly reduced, with siRNA2 exhibiting the most potent interference efficiency (Fig 1b–1c). To further validate the functional interaction between HMGB1 and KU70 in CRC cells, rescue experiments were performed by transfection with si-KU70. As shown in Fig 4a, apoptosis was significantly increased in both HMGB1-knockdown and KU70-interfered CRC cells after irradiation. Western blot analysis further revealed decreased accumulation of KU70, KU80, XRCC4, and LIG4 proteins in knockdown HMGB1 or KU70 CRC cells exposed to 4 Gy X-ray irradiation (Fig 4b). These results suggest that HMGB1 contributes to decreased radiosensitivity by enhancing the NHEJ pathway in CRC cells. Additionally, the assessment of the formation of γH2A.x foci showed that KU70 interference effectively increased the number of DNA double-strand breaks, which is consistent with the situation observed in CRC cells lacking HMGB1 after irradiation (Fig 4c). Western blot analysis showed a markable increase in γH2A.x in both HMGB1-knockdown and KU70-interfered CRC cells after exposure to 4 Gy X-ray irradiation (Fig 4d). Similarly, comet assay results indicated that HMGB1 knockdown combined with KU70 interference promoted an increase in tail length in irradiated CRC cells (Fig 4e). Collectively, these results indicate that the HMGB1/KU70/NHEJ pathway plays a critical role in the radioresistance-mediated by DNA damage repair of CRC.
a. Caspase-3 activity was measured in HMGB1-knockdown or KU70-interfered CRC cells after 4 Gy irradiation. b. Following 4 Gy irradiation, the expression of NHEJ-related proteins in HMGB1-knockdown or KU70-interfered CRC cells were analyzed by western blot using β-actin as a loading control. The relative fold-change in protein expression was quantified. c. γH2A.x focus formation was assessed at indicated time points in KU70-knockdown CRC cells after irradiation (scale bar: 20 µm). d. γH2A.x protein level in HMGB1-knockdown or KU70-interfered CRC cells were detected by western blot after 4 Gy irradiation, with β-actin as the loading control. The relative fold change of γH2A.x was quantified. e. DNA damage was evaluated by comet assay in HMGB1-knockdown or KU70-interfered CRC cells at 24 h post 4 Gy X-ray irradiation (scale bar: 20 µm). Tail DNA per cell was calculated from 100 cells. Data are presented as the mean ± SEM from the three experiments. **P < 0.01, *P < 0.05.
HMGB1/KU70 axis maintains redox homeostasis in CRC cells after irradiation
To further investigate whether HMGB1 facilitates KU70-mediated maintains redox homeostasis, cellular ROS levels and 8-OHdG formation were assessed using a ROS detection assays and IF, respectively. Si-KU70 alone phenocopied HMGB1 knockdown, elevating basal ROS accumulation after irradiation (Fig 5a). IF analysis of 8-OHdG revealed that both HMGB1 downregulation and KU70 interference significantly increased 8-OHdG levels in CRC cells following irradiation (Fig 5b). These data further support the conclusion that HMGB1 binds to KU70 and sequesters it within the nucleus to maintains redox homeostasis and protects against oxidative DNA damage.
a. Intracellular ROS levels in CRC cells after irradiation, following transfection with KU70 siRNA or HMGB1 knockdown vectors. b. Expression of 8-OHdG in CRC cells at 4 Gy irradiation after transfection with KU70 siRNA or HMGB1 knockdown vectors (scale bars: 20 µm). **P < 0.01, *P < 0.05.
HMGB1/KU70 axis mitigates CRC radiosensitivity in vivo
Given the inhibitory effect of the HMGB1/KU70 axis on the radiosensitivity of CRC cells in vitro, this study further validated its role in vivo using a subcutaneous xenograft mouse model. HT29 cells transfected with an HMGB1 shRNA plasmid or a scrambled shRNA plasmid (control group) were subcutaneously inoculated into nude mice, followed by treatment with or without 10 Gy local irradiation or intratumorally injection of si-KU70. Results showed that both tumor growth and weight in mice inoculated with sh-HMGB1 cells were significantly lower than those in the control group. After 10 Gy irradiation, tumor size and weight were also markedly reduced in the HMGB1‑knockdown group compared to the control. Consistent with the effects of HMGB1 knockdown, intratumorally injection of si-KU70 similarly suppressed tumor growth in ectopic xenografts, supporting that suppression of HMGB1/KU70 inhibits the tumorigenicity of CRC cells in vivo (Fig 6a–c). Furthermore, IHC analysis revealed a significant decrease in Ki67-postive cells following irradiation in both HMGB1 knockdown or KU70-interfered tumors (Fig 6d and Fig s1d in S1 Fig). Conversely, HMGB1 knockdown or si-KU70 treatment increased the proportion of γH2A.x‑positive cells in irradiated xenograft tissues (Fig 6e and Fig s1e in S1 Fig). In addition, compared with the scrambled shRNA control, both HMGB1 knockdown and si-KU70 treatment led to elevated levels of TUNEL‑positive cells (Fig 6f and Fig s1f in S1 Fig). Collectively, these in vivo findings indicate that HMGB1 enhances radioresistance in CRC through the KU70‑mediated NHEJ pathway.
a. Schematic of the experimental design and representative tumor images from female nude mice (n = 5 per group). implanted with sh-NC or sh-HMGB1-HT29 cells, followed by treatment with or without si-KU70 and 10 Gy X‑ray irradiation. b. Terminal tumor weights from xenograft in each treatment group. Data are presented as mean ± SEM (n = 5). c. Tumor growth curves of xenografts derived from indicated CRC cell lines under described treatment. Data are shown as mean ± SEM (n = 5). d. IHC staining of Ki67 in tumor sections from the xenograft model across treatment conditions (scale bar: 50 μm). e. IHC staining of γH2A.x in tumor tissues, indicating DNA double-strand breaks (scale bar: 50 μm). f. TUNEL staining of tumor tissues to assess apoptosis (scale bars: 20 μm). g. Proposed mechanistic model. Ionizing radiation induces DNA DSBs and ROS. HMGB1 recruits KU70 to facilitate assembly of the NHEJ repair complex (KU70/KU80/XRCC4/LIG4), thereby promoting DSB repair and suppressing apoptosis. Disruption of HMGB1 or KU70 impairs this axis, resulting in sustained DNA damage, activation of caspase‑3, and inhibition of tumor growth. **P < 0.01, *P < 0.05.
Our previous study revealed that HMGB1 is abnormally expressed and associated with poor prognosis in CRC patients [7]. To further examine the relationship between HMGB1 and KU70 and their clinicopathological relevance in CRC, we analyzed KU70 expression in CRC and normal tissues using the GEPIA database. As shown in Fig s2a in S2 Fig, KU70 was highly expressed in CRC tumors. Moreover, aberrant KU70 expression correlated with advanced TNM stage (Fig s2b in S2 Fig). Pearson correlation analysis based on the TIMER database indicated a positive correlation between HMGB1 and KU70 expression in CRC tissues (Fig s2c in S2 Fig). Although not statistically significant (P > 0.05), a trend toward poorer disease‑free survival was observed in patients with high co-expression of HMGB1 and KU70 (Fig s2d in S2 Fig). These data suggest that HMGB1 and KU70 expression may serve as a candidate indicator for patient stratification, pending validation in prospective cohorts.
Discussion
Radiotherapy is a well-established treatment that eliminates tumor cells by inducing DNA damage, generating ROS, and disrupting cellular structural integrity, leading to tumor shrinkage and cell death [12]. Different tumors exhibit markedly variable radiosensitivity. However, resistance to IR remains a major obstacle in the treatment of CRC. As an evolutionarily conserved protein, multiple studies have reported that HMGB1 is involved in radioresistance in esophageal squamous cell carcinoma [13], non-small cell lung cancer [14], and glioblastoma [15]. In CRC, most studies have focused on the effects of HMGB1 on the progression [16], however, whether HMGB1 affects the radiosensitivity of CRC is rarely studied. In this study, we found that knockdown of HMGB1 increased baseline levels of DNA damage and apoptosis, consistent with its role in maintaining chromatin architecture. However, after radiation exposure, the damage response was significantly enhanced, as evidenced by increased fold induction of γH2A.x and persistent DSBs. This indicates that HMGB1 is specifically required for the efficient repair of radiation-induced DNA damage. This baseline effect does not undermine our main conclusion; on the contrary, it further highlights the importance of HMGB1 in both maintaining baseline genomic stability and in stress-induced DNA repair processes.
A widely studied strategy to improve tumor radiosensitivity is to increase primary radiation-induced DNA damage while inhibiting the cell repair of sublethal and lethal damage. The most critical lethal DNA lesion is the DSB, which activates components such as DNA-PKcs, KU70, and KU80 to initiate the DNA damage repair process [17]. HMGB1 typically functions as a DNA chaperone, facilitating the activity of DNA repair enzymes and transcription factors [18]. Therefore, HMGB1 can bind damaged DNA and influence tumor resistance to radiation or chemotherapy [19]. Consistently, we demonstrate that HMGB1 contributes to the radioresistance in CRC by promoting DNA damage repair. Since KU-DNA binding is an initiating step in NHEJ, HMGB1 likely plays a role in the early stages of NHEJ or the broader DNA damage response. Further elucidation of the underlying molecular mechanisms revealed that HMGB1 directly interacts with KU70 through its amino acid region 95–163 to stabilize the NHEJ complex and promote repair. Notably, HMGB1’s effect on KU70 is context-dependent. In Alzheimer’s disease neurons, extracellular HMGB1 activates TLR4 signaling, leading to KU70 phosphorylation and impaired DNA binding, thereby disrupting NHEJ [20]. This dichotomy highlights that HMGB1’s functional output depends on its subcellular localization and the specific signaling environment. Additionally, NHEJ has been linked with chemoresistance or radioresistance in a variety of tumors, including CRC [21], glioma, breast cancer [22], and nasopharyngeal carcinoma [23]. Our findings suggest that HMGB1 contributes to radiotherapy resistance, potentially through its involvement in NHEJ. Consequently, targeting the HMGB1/KU70/NHEJ axis represents a promising strategy for improving outcomes in CRC, particularly for patients exhibiting radiotherapy resistance, potentially through the use of HMGB1/KU70 antagonists.
Studies have shown that IR can elevate intracellular ROS levels in mouse stromal tumor cells, thereby causing DNA damage, promoting tumor cell destruction, and ultimately inhibits tumor growth, which ROS also modulate signaling pathways linked to survival or death responses following chemotherapy or radiation therapy [24,25]. Meng et al. reported that HMGB1 inhibition attenuates toluene diisocyanate-induced occupational asthma by regulating the ROS/AMPK/autophagy pathway [26]. Beyond its direct role in ROS regulating, endogenous HMGB1 has been shown to regulate pyroptosis via the ROS/ERK1/2/caspase-3/GSDME pathway in neuroblastoma [27]. Our results align with these previous reports, indicating that HMGB1 primarily regulates baseline oxidative stress after irradiation, thereby further protecting against oxidative DNA damage. Although we did not employ NAC to scavenge ROS, the studies have demonstrated that NAC can reverse the formation of γH2A.x foci and modulate the levels of DNA damage repair proteins [28,29]. However, Zheng et al. showed that 5-FU-induced ROS accumulation leads to HIF-1α activation, which directly binds to HMGB1promotor in CRC cells [30]. These findings suggest that HMGB1 and irradiation-induced ROS may form a positive feedback loop, modulating tumor cell behavior and ultimately contributing to radioresistance or chemoresistance.
In conclusion, this study reveals that HMGB1 promotes DNA damage repair in CRC both in vitro and in vivo. We further demonstrated that nuclear HMGB1 interacts with KU70 and subsequently activates the NHEJ pathway, resulting in the inhibition of cell apoptosis in CRC (Fig 6g). Therefore, HMGB1 may serve as a predictive biomarker for radiotherapy response. Combining radiotherapy with DNA damage repair inhibitors could represent an effective therapeutic strategy to overcome radioresistance in CRC patients with high HMGB1 expression.
Limitations of the study
This study has several limitations. First, although the 95–163 amino acid region of HMGB1 was identified as the core binding domain for KU70, detailed domain mapping and experimental validation of their interaction remain to be completed, which limits the mechanistic understanding of binding specificity. Second, the functional relationship between HMGB1 and KU70 was primarily assessed through loss-of-function approaches; the lack of gain-of-function and direct rescue experiments precludes a definitive establishment of their causal role in radiation-induced DNA damage repair. Third, our findings are derived solely from cellular and animal models and have not yet been validated in clinical cohorts or human tissues, thereby limiting their translational relevance to colorectal cancer radiotherapy. Future studies should address these gaps through targeted domain mapping, complementary functional assays, and validation using clinical samples.
Supporting information
S1 Fig. HMGB1/KU70 axis contributes to radioresistance in CRC cells.
a. Analysis by ImageJ of protein expression levels of KU70, KU80, XRCC4, and LIG4 in HMGB1-knockdown CRC cells following exposure to 4 Gy X-ray irradiation. b. Expression of KU70in CRC cells after transfection with KU70 siRNA, as determined by qPCR. c. Western blot detection of KU70 expression after transient transfection with si-KU70; β-actin served as the loading control. d. Number of Ki67-positive cells in tumor tissues from a nude mouse xenograft model treated with 10 Gy irradiation combined with si-KU70. e. Number of γH2A.x-positive cells in tumor tissues from the same model following 10 Gy irradiation plus si-KU70 treatment. f. Number of TUNEL-positive cells in tumor tissues under identical treatment conditions. Data are presented as mean ± SEM (n = 5 mice per group). **P < 0.01, *P < 0.05.
https://doi.org/10.1371/journal.pone.0345635.s001
(PDF)
S2 Fig. Figure s2.
Correlation analysis of HMGB1 and KU70 expression in CRC tissue samples. a. Highly expression of KU70 in CRC tumors. b. Expression of KU70 in CRC samples at different clinical stages, as analyzed using the GEPIA database. c. Correlation analysis of HMGB1 and KU70 expression levels in CRC samples, based on data from the TIMER database. d. Kaplan-Meier survival curves illustrating diseasefree survival in CRC patients stratified by HMGB1 and KU70 expression levels, derived from the GEPIA database. **P < 0.01, *P < 0.05.
https://doi.org/10.1371/journal.pone.0345635.s002
(PDF)
S1 Table. Primers for RT-qPCR assay of genes.
https://doi.org/10.1371/journal.pone.0345635.s003
(DOCX)
References
- 1. Siegel RL, Kratzer TB, Giaquinto AN, Sung H, Jemal A. Cancer statistics, 2025. CA Cancer J Clin. 2025;75(1):10–45.
- 2. Wang L, Lynch C, Pitroda SP, Piffkó A, Yang K, Huser AK, et al. Radiotherapy and immunology. J Exp Med. 2024;221(7):e20232101. pmid:38771260
- 3. Amodio V, Vitiello PP, Bardelli A, Germano G. DNA repair-dependent immunogenic liabilities in colorectal cancer: opportunities from errors. Br J Cancer. 2024;131(10):1576–90. pmid:39271762
- 4. Chen R, Zou J, Zhong X, Li J, Kang R, Tang D. HMGB1 in the interplay between autophagy and apoptosis in cancer. Cancer Lett. 2024;581:216494. pmid:38007142
- 5. Tang D, Kang R, Zeh HJ, Lotze MT. The multifunctional protein HMGB1: 50 years of discovery. Nat Rev Immunol. 2023;23(12):824–41. pmid:37322174
- 6. Han G, Ling R, Sun C, Wang X, Zhou Y, Yu L, et al. HMGB1 knockdown increases the radiosensitivity of esophageal squamous cell carcinoma by regulating the expression of molecules involved in DNA repair. Oncol Lett. 2021;22(1):503. pmid:33986864
- 7. Han Y, Sheng W, Liu X, Liu H, Jia X, Li H, et al. Glycyrrhizin ameliorates colorectal cancer progression by regulating NHEJ pathway through inhibiting HMGB1-induced DNA damage response. Sci Rep. 2024;14(1):24948. pmid:39438689
- 8. Zhu H, Tian Y, Chen H, Qian Y, Li J, Niu D, et al. Targeting DNA damage response pathways in tumor drug resistance: Mechanisms, clinical implications, and future directions. Drug Resist Updat. 2025;83:101287. pmid:40795793
- 9. Zhu X, Cong J, Lin Z, Sun J, Yang B, Li A. Inhibition of HMGB1 Overcomes Resistance to Radiation and Chemotherapy in Nasopharyngeal Carcinoma. Onco Targets Ther. 2020;13:4189–99. pmid:32523355
- 10. Han Y, Liu X, Sheng W, Kuang R, Zhang Y, Jia X, et al. Cytoplasmic HMGB1 promotes and interacts with BECN1 through ZNF460 to induce autophagy and accelerate radioresistance in colorectal cancer cells. Front Immunol. 2025;16:1642915. pmid:41164196
- 11. Saini S, Gurung P. A comprehensive review of sensors of radiation-induced damage, radiation-induced proximal events, and cell death. Immunol Rev. 2025;329(1):e13409. pmid:39425547
- 12. Deng S, Vlatkovic T, Li M, Zhan T, Veldwijk MR, Herskind C. Targeting the DNA Damage Response and DNA Repair Pathways to Enhance Radiosensitivity in Colorectal Cancer. Cancers (Basel). 2022;14(19):4874. pmid:36230796
- 13. Dong J, Zhang X, Du X, Zou N, Shen W, Ma M, et al. HMGB1 overexpression promotes a malignant phenotype and radioresistance in ESCC. J Cancer. 2022;13(9):2717–26. pmid:35812184
- 14. Bai L, Zhang J, Gao D, Liu C, Li W, Li Q. Downregulation of high mobility group box 1 enhances the radiosensitivity of non-small cell lung cancer by acting as a crucial target of microRNA-107. Exp Ther Med. 2021;22(1):679. pmid:33986844
- 15. Qiu H, Shao Z, Wen X, Qu D, Liu Z, Chen Z, et al. HMGB1/TREM2 positive feedback loop drives the development of radioresistance and immune escape of glioblastoma by regulating TLR4/Akt signaling. J Transl Med. 2024;22(1):688. pmid:39075517
- 16. Zhao W, Chen A, Yuan N, Hao X, Wang C, Lu X, et al. The Role of High Mobility Group Box B-1 in the Prognosis of Colorectal Cancer Based on the Changes in the Intestinal Mucosal Barrier. Technol Cancer Res Treat. 2024;23:15330338231198972. pmid:38200714
- 17. Goff NJ, Mikhova M, Schmidt JC, Meek K. DNA-PK: A synopsis beyond synapsis. DNA Repair (Amst). 2024;141:103716. pmid:38996771
- 18. Balana AT, Mukherjee A, Nagpal H, Moon SP, Fierz B, Vasquez KM. O-GlcNAcylation of High Mobility Group Box 1 (HMGB1) Alters Its DNA Binding and DNA Damage Processing Activities. J Am Chem Soc. 2021;143(39):16030–40.
- 19. Farhana A, Koh AE-H, Tong JB, Alsrhani A, Kumar Subbiah S, Mok PL. Nanoparticle-Encapsulated Camptothecin: Epigenetic Modulation in DNA Repair Mechanisms in Colon Cancer Cells. Molecules. 2021;26(17):5414. pmid:34500845
- 20. Tanaka H, Kondo K, Fujita K, Homma H, Tagawa K, Jin X, et al. HMGB1 signaling phosphorylates Ku70 and impairs DNA damage repair in Alzheimer’s disease pathology. Commun Biol. 2021;4(1):1175. pmid:34635772
- 21. Luo J, Chen J-W, Zhou J, Han K, Li S, Duan J-L, et al. TBX20 inhibits colorectal cancer tumorigenesis by impairing NHEJ-mediated DNA repair. Cancer Sci. 2022;113(6):2008–21. pmid:35348274
- 22. Shaheer K, Prabhu BS, Ali HS, Lakshmanan-M D. Breast cancer cells are sensitized by piperine to radiotherapy through estrogen receptor-α mediated modulation of a key NHEJ repair protein- DNA-PK. Phytomedicine. 2024;122:155126.
- 23. Feng P, Wang Y, Liu N, Chen Y, Hu Y, Huang Z, et al. High expression of PPP1CC promotes NHEJ-mediated DNA repair leading to radioresistance and poor prognosis in nasopharyngeal carcinoma. Cell Death Differ. 2024;31(5):683–96. pmid:38589496
- 24. Lee Y-P, Lin C-R, Chen S-S, Chen R-J, Wu Y-H, Chen Y-H, et al. Combination treatment of cordycepin and radiation induces MA-10 mouse Leydig tumor cell death via ROS accumulation and DNA damage. Am J Cancer Res. 2023;13(4):1329–46. pmid:37168338
- 25. Kumar H, Gupta NV, Jain R, Madhunapantula SV, Babu CS, Kesharwani SS, et al. A review of biological targets and therapeutic approaches in the management of triple-negative breast cancer. J Adv Res. 2023;54:271–92. pmid:36791960
- 26. Meng X, Guo S, Zhang X, Jiao B, Yang X, Li M, et al. HMGB1 inhibition reduces TDI-induced occupational asthma through ROS/AMPK/autophagy pathway. Ecotoxicol Environ Saf. 2023;266:115575. pmid:37839183
- 27. Fan C-Y, Ye F-H, Peng M, Dong J-J, Chai W-W, Deng W-J, et al. Endogenous HMGB1 regulates GSDME-mediated pyroptosis via ROS/ERK1/2/caspase-3/GSDME signaling in neuroblastoma. Am J Cancer Res. 2023;13(2):436–51. pmid:36895972
- 28. Chen J, Ying K, Sun J, Wang Y, Ji M, Sun Y. NEDD4L affects KLF5 stability through ubiquitination to control ferroptosis and radiotherapy resistance in oesophageal squamous cell carcinoma. J Cell Mol Med. 2024;28(18):e70062. pmid:39317954
- 29. Chentunarayan Singh N, Yadav N, Sharma RK, Gupta P, Sarkar J, Mitra K. Oxidative stress-mediated DNA damage promotes selective degradation of nuclear components via noncanonical autophagy in triple-negative breast cancer cells. Free Radic Biol Med. 2026;242:37–53. pmid:41109361
- 30. Zheng H, Yu S, Zhu C, Guo T, Liu F, Xu Y. HIF1α promotes tumor chemoresistance via recruiting GDF15-producing TAMs in colorectal cancer. Exp Cell Res. 2021;398(2):112394. pmid:33242463