Polymorphism in the Hsa-miR-4274 seed region influences the expression of PEX5 and enhances radiotherapy resistance in colorectal cancer

Abstract

Identifying biomarkers for predicting radiotherapy efficacy is crucial for optimizing personalized treatments. We previously reported that rs1553867776 in the miR-4274 seed region can predict survival in patients with rectal cancer receiving postoperative chemoradiation therapy. Hence, to investigate the molecular mechanism of the genetic variation and its impact on the radiosensitivity of colorectal cancer (CRC), in this study, bioinformatics analysis is combined with functional experiments to confirm peroxisomal biogenesis factor 5 (PEX5) as a direct target of miR-4274. The miR-4274 rs1553867776 variant influences the binding of miR-4274 and PEX5 mRNA, which subsequently regulates PEX5 protein expression. The interaction between PEX5 and Ku70 was verified by co-immunoprecipitation and immunofluorescence. A xenograft tumor model was established to validate the effects of miR-4274 and PEX5 on CRC progression and radiosensitivity in vivo. The overexpression of PEX5 enhances radiosensitivity by preventing Ku70 from entering the nucleus and reducing the repair of ionizing radiation (IR)-induced DNA damage by the Ku70/Ku80 complex in the nucleus. In addition, the enhanced expression of PEX5 is associated with increased IR-induced ferroptosis. Thus, targeting this mechanism might effectively increase the radiosensitivity of CRC. These findings offer novel insights into the mechanism of cancer radioresistance and have important implications for clinical radiotherapy.

Keywords

colorectal cancer / polymorphism / miR-4274 / PEX5 / radiotherapy resistance

1 Introduction

Colorectal cancer (CRC) is one of the most common cancers and remains the leading cause of cancer-related deaths worldwide. Concurrently, there is a growing trend of diagnosing CRC in younger individuals at advanced stages [1]. Existing evidence suggests that radiotherapy for CRC significantly improves clinical outcomes [2]. However, the efficacy of radiotherapy in clinical applications is limited by inter-individual endogenous radioresistance. Consequently, it is imperative to investigate the fundamental mechanisms and identify potential targets to combat radioresistance in CRC.

Resistance to ionizing radiation (IR) involves various biological mechanisms. Radiotherapy results in DNA breaks in rapidly dividing cancer cells, encompassing single-strand breaks (SSBs) and double-strand breaks (DSBs). In human cells, breaks in the DNA strands are primarily repaired via homologous recombination and non-homologous end-joining (NHEJ), with the latter being the most common method. SSBs and DSBs activate DNA NHEJ repair, suggesting that some cancer cells may exhibit increased NHEJ repair system activity, leading to radiation therapy resistance and improved survival and proliferation. The Ku70/Ku80 heterodimer, Ku, is a subunit of the DNA-dependent protein kinase complex along with DNA-PKcs, a catalytic kinase subunit. This complex responds to DSBs and facilitates damaged DNA repair through the NHEJ pathway [3,4].

MicroRNAs (miRNAs) are compact, ~20-nucleotide, non-coding RNA molecules that regulate post-transcriptional events by pairing with the 3′ untranslated regions (3′UTR) of target mRNA [5]. Considering that more than half of the genes are regulated by miRNAs, they hold crucial biological significance. The function of miRNAs depends on the pairing of target mRNAs. The seed region, which comprises nucleotides 2–7 of the miRNA, plays a crucial role in this pairing process. Genetic variations in miRNAs disrupt the interactions between miRNAs and their targeted mRNAs, further altering the associated biological functions [6–8]. Notably, single-nucleotide polymorphisms (SNPs) within the crucial seed regions of mature miRNAs have the potential to influence the pairing with target mRNAs, thereby affecting their functions. Our previous study reported that rs1553867776 in the miR-4274 seed region can predict survival in patients with rectal cancer receiving adjuvant postoperative chemoradiation therapy [9,10]. However, the specific functions and molecular mechanisms of miR-4274 and its SNP in CRC remain unknown. Indeed, miR-4274 plays a key role in gastric and basal-like breast cancers [11,12]. However, the pathogenic and therapeutic implications of miR-4274 and its target genes in CRC remain obscure.

Recent studies have demonstrated that IR can trigger ferroptosis, which modulates radiosensitivity during cancer therapy [13]. Ferroptosis is a form of programmed cell death that promotes lipid peroxidation, however, the specific molecular mechanisms remain unclear [14,15]. Nevertheless, previous studies have demonstrated that peroxisomes regulate ferroptosis by synthesizing polyunsaturated ether phospholipids [16–18]. The peroxisomal biogenesis factor 5 (PEX5) plays an important role in peroxisomal functioning, primarily serving as a receptor that identifies the peroxidase target signal, PTS1, and transports it to peroxisomes [19]. Furthermore, ferroptosis is an essential factor in regulating the sensitivity of CRC to radiotherapy [20]. However, few studies have focused on the role of PEX5 in ferroptosis.

The primary aims of the current study are to (1) investigate the molecular mechanism underlying rs1553867776 in the miR-4274 seed region and (2) evaluate the impact of this genetic variant on the sensitivity of CRC cells to radiotherapy. Our results implicate rs1553867776 in miR-4274 and PEX5 in radiotherapy resistance in CRC. The rs1553867776 in miR-4274 regulates PEX5 at the protein level via post-transcriptional regulation. Meanwhile, PEX5 regulates the transfer of Ku70 into the nucleus and affects the formation of the Ku70/Ku80 complex. Additionally, PEX5 regulates radiotherapy sensitivity through the ferroptosis pathway. Finally, we demonstrate that rs1553867776 in miR-4274 seed region and PEX5 are potential biomarkers for predicting radiosensitivity.

2 Materials and Methods

2.1 Cell lines and reagents

HCT116, HCT8, RKO, LoVo, SW480, SW620, SW1116, MGC-803 and HepG2 cells were purchased from the Cell Bank of the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, and the School of Basic Medicine, Peking Union Medical College. Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum (FBS) was used to culture HCT116, RKO, LoVo, SW1116 and HepG2 cells, while Roswell Park Memorial Institute (RPMI)-1640 medium containing 10% FBS was used to culture HCT8, SW480, SW620, and MGC-803 cells under a 5% CO₂ humidified atmosphere at 37 °C.

2.2 Western blot

CRC cells were lysed using RIPA lysis buffer (NCM Biotech, WB3100) containing ProtLytic Protease Inhibitor Cocktail (NCM Biotech, P001) and ProtLytic Phosphatase Inhibitor Cocktail (NCM Biotech, P003). A BCA kit (Thermo Fisher Scientific, A53225) was used to measure the protein concentration of each sample. Subsequently, 10 µg of protein was separated by SDS-PAGE (Meilunbio, MA0286) and transferred to PVDF membranes (Millipore). Antibodies against Ku70 (ab92450), ACSL4 (ab155282), SLC7A11 (ab175186), GPX4 (ab125066), γ-H2AX (phospho S139) (ab81299) were from Abcam. And the antibodies against PEX5 (12545-1-AP), Ku80 (16389-1-AP), GAPDH (60004-1-Ig), WDR33 (22614-1-AP), Tubulin (11224-1-AP) and Lamin B1 (12987-1-AP) were from Proteintech. The signal was identified using the super-sensitive ECL luminescence reagent (Melunbio, MA0186) and detected using the Amersham Imager 600. Thermo Scientific PageRuler (Thermo Scientific, 26617) was used to differentiate between protein sizes. Quantification of the protein bands was performed by grayscale scanning using ImageJ software.

2.3 RNA extraction and qRT-PCR analysis

Total RNA was extracted using the RNA-Quick Purification Kit (ES Science, RN001). Subsequently, total RNA was reverse transcribed by employing PrimeScript RT Master Mix (TaKaRa, RR0036A). Quantitative reverse transcription PCR (qRT-PCR) was performed in triplicate using SYBR green reagent (TaKaRa, RR820A). Primer sequences used for qRT-PCR are listed in Table S1. MiR-4274 RNA levels were quantified relative to the U6 RNA levels, and mRNA levels were quantified relative to the GAPDH mRNA levels.

2.4 Establishment of CRC cell lines with PEX5 overexpression or knockout

Lentiviral particles for stable overexpression (PEX5-OE) or knockout (PEX5-KO) of PEX5 were procured from GeneChem. HCT116 and HCT8 cells were infected with lentivirus, maintained in complete medium for 24 h, and subsequently selected using puromycin. The CRISPR/Cas9 system was used to delete a portion of the genomic material for PEX5 in CRC cell lines. This was accomplished via CRISPR to create single-guide RNA (sgRNAs) sequences that targeted the genetic sequence of PEX5. These sequences were then inserted into the pUC19-U6-sgRNA plasmid. The pCAG-Cas9-EGFP and pUC19-U6-sgRNA plasmids were added to HCT116 and HCT8 cells, and the fluorescent cells were sorted into a 96-well plate for growth using flow cytometry. The efficacy of PEX5-OE or PEX5-KO was assessed using Western blot and qPCR assays, respectively.

2.5 Plasmid construction and site-directed mutagenesis

The wild-type 3′UTR of PEX5 was obtained using the restriction enzymes XhoI (NEB, R0146V) and NotI (NEB, R0189V), and was inserted into the psiCHECK-2 vectors (Promega, C8021). The vectors containing the mutation-type 3′UTR of PEX5 were generated using In-Fusion HD Cloning Plus (TaKaRa, 638909).

2.6 Luciferase reporter assays

Luciferase reporter assays were performed using the Nano-Glo Dual-Luciferase Reporter (NanoDLR™) Assay System, according to the manufacturer’s instructions (Promega, N1620). In brief, CRC cells were seeded in 48-well plates (6 × 10⁴ cells/well). After cell adhesion, 10 nmol of mimics or inhibitors (Beijing Tsingke Biotech Co., Ltd.), along with reporter plasmids, were co-transfected into the cells using Lipofectamine 2000. Cells were harvested 24 h after transfection. Firefly and Renilla luciferase activities were detected using GENE5 software.

2.7 Cell viability and colony formation assays

CRC cells transfected with miR-4274 mimics or inhibitors were seeded into 96-well plates (2000 cells/well). The cells were assessed daily for 96 h using a Cell Counting Kit-8 (CCK-8; Dojindo). Ten microliters of the CCK-8 reagent were added to each well and incubated at 37 °C for 1 h. The absorbance at 450 nm was subsequently measured. For the colony formation experiments, cells were seeded in six-well plates (2000 cells/well) containing complete growth medium with or without IR (2 Gy) treatment. After seven days, the colonies were fixed with methanol and stained with 0.5% crystal violet. The number of colonies was quantified using ImageJ software.

2.8 Cell migration assays

For the cell migration assays, HCT8 cells (1.5 × 10⁵ cells/well) were seeded into the upper chambers of Transwell plates equipped with 8-μm pore polycarbonate membrane inserts using medium without FBS. The lower chambers were filled with 150 μL medium containing FBS. Following a 16-h incubation period, cells that migrated to the underside of the membrane were washed with PBS, fixed, and stained with 1% toluidine blue. Three randomly selected fields were enumerated, and the counts were averaged. Each sample underwent triplicate assays, and each experiment was repeated at least twice. Quantification of migrated cells was performed using ImageJ software.

2.9 Extreme limiting dilution assays

CRC cells were distributed into 96-well plates at different cell densities (6.25, 12.5, 25, 50, 100, and 200 cells/well), with or without IR (2 Gy). Sphere numbers were counted and analyzed using ELDA after 7–14 days.

2.10 Comet assay

The Comet Assay Kit (Abbkine) was used to investigate DNA DSBs, and the tail moment was evaluated using Open Comet software.

2.11 RNA-sequencing analysis

Total RNA was extracted from the NC and PEX5-KO cells with or without IR (24 h after 2 Gy) and subjected to sequencing. The results were mapped to the GRCh38 human genome using Hisat2 (v. 2.0.5). Differential expression analyses of the PEX5-KO and control groups were performed using the DESeq2 R package (v. 1.20.0). Significant differential expression was determined by considering a corrected P value < 0.05 and an absolute log₂(fold change) threshold > 0.

2.12 Immunoprecipitation assays

Cells were lysed in RIPA lysis buffer (Beyotime, P0013D) while maintaining protein–protein interactions for 1 h at 4 °C. The Dynabeads Protein G kit (Invitrogen, 10007D) was used to capture the antibody–protein complexes, followed by elution with the elution buffer. The resulting samples were subjected to either Western blot or mass spectrometer analysis (PTM BIO).

2.13 Nuclear and cytoplasmic protein analysis

The cells were washed with PBS, and then their nucleus and cytoplasm were separated using the Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime, P0027) according to the manufacturer’s instructions. After isolation and extraction of cytoplasmic and nuclear proteins, they will be stored at −80 °C.

2.14 Immunofluorescent assays

CRC cells adhered to coverslips in 24-well plates (5 × 10⁴ cells/well). Subsequently, the cells were subjected to either IR or left untreated, fixed in a 4% paraformaldehyde solution for 10 min, and treated with cold methanol for 10 min. Coverslips were blocked with 3% bovine serum albumin (BSA) at 37 °C for 1 h. Subsequently, primary antibodies targeting PEX5 (Proteintech, 12545-1-AP) and Ku70 (Proteintech, 66607-1-Ig) were applied to the coverslips and incubated overnight at 4 °C. The slides were then incubated with a cross-adsorbed secondary antibody (Invitrogen) for 1 h at 37 °C. Finally, a fluorescent mounting medium with DAPI (ZSGB-BIO) was applied to mount the coverslips, and confocal microscopy was performed to acquire image fluorescence of PEX5, Ku70 and DAPI.

2.15 Lipid malondialdehyde (MDA) assays

MDA levels were measured using an MDA Assay Kit (Dojindo, M496) according to the manufacturer’s guidelines. The formation of the MDA–thiobarbituric acid (TBA) adduct, resulting from the reaction between MDA in the sample and TBA, was quantified via fluorescence intensity measurements using a microplate reader (excitation: 540 nm, emission: 590 nm). The MDA concentration was determined by referencing the standard curve. Evaluation of reaction outcomes was performed using GENE5 software, with MDA levels reported in µM.

2.16 Animal experiments

Male NOD-Prkdc-Il2rg null (NPI) mice aged 4–6 weeks (Beijing IDMO Co.) were subcutaneously injected with wild-type, NC, or PEX5-KO CRC cells (2 × 10⁶) and housed for 30 days. The miR-4274 agomiR-mimics and NC agomiR-mimics (Beijing Tsingke Biotech Co.) were injected into wild-type tumor tissues at 1 nmol (in 50 μL of PBS) per mouse every 5 days. Half of the tumor tissues were exposed to 2 Gy/day radiation for 5 days.

2.17 Radiotherapy

The cells and animals underwent radiotherapy at the radiotherapy department of the Cancer Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College. Irradiation was performed at room temperature at a dose rate of 2 Gy/min. Colorectal cancer (CRC) cells were harvested at 2, 6, 12 and 24 h after exposure to ionizing radiation (2 Gy). Additionally, the animals received daily radiation exposure of 2 Gy for a duration of 5 days.

2.18 Bioinformatics analysis

Six databases were used to predict target genes, including miRWalk [21], Target Miner [22], PolymiRTS (compbio.uthsc.edu/miRSNP) [23], DIANA TOOLS [24], miRDB [25] and Target Scan [26]. The Gene Expression Omnibus (GEO) database was used to predict hsa-miR-4274 expression. The expression of PEX5 was analyzed using The University of ALabama at Birmingham CANcer data analysis portal (UALCAN) [27]. Additionally, TIMER (v. 2.0) [28] was used to analyze survival data, while Metascape [29] was used for enrichment analysis.

2.19 Statistical analysis

Student’s t-test was used to determine the differences between groups. The study conducted survival analysis using the Kaplan–Meier method. All statistical analyses were performed using R Studio and Prism 9. A P < 0.05 was considered statistically significant.

3 Results

3.1 miR-4274 functions as a radioresistant oncogene in CRC

In our previous study, we identified an SNP, rs202195689, in the miR-4274 seed region associated with overall and disease-free survival in postoperative synchronous chemoradiotherapy patients with rectal cancer. This 5-nucleotide insertion-deletion (indel) polymorphism has been associated with rs1553867776 and is located within the miR-4274 seed region of its gene (Fig. S1A). Overall and disease-free survival times were significantly reduced in CRC patients with at least one deletion genotype compared to those with an insertion genotype.

Seven common CRC cell lines (HCT8, HCT116, RKO, SW480, SW620, SW1116, and LoVo), one gastric cancer cell line (MGC-803) and one liver cell line (HepG2) harbored only the rs1553867776 insertion allele, as confirmed by Sanger sequencing (Fig. S1B). This is consistent with our previous study showing that most patients carried insertion-type mutations at this SNP position [9,10]. To further explore the role of miR-4274 in CRC, the expression of miR-4274 in the blood of CRC patients and healthy controls was compared using data from GEO GSE106817 and GSE59856 (Fig.1). We observed significant overexpression of miR-4274 in CRC patients compared to healthy donors. Furthermore, a significant number of patients with tumors in their digestive tracts have elevated levels of serum miR-4274, indicating that the influence of miR-4274 is widespread across different tumor types.

Fig.1 miR-4274 functions as a radioresistant oncogene in CRC. (A) miR-4274 expression in blood was upregulated in patients with digestive system tumor than healthy control group. Colorectal, colorectal cancer; colon, colon cancer; stomach, stomach cancer; esophageal, esophageal cancer; liver, liver cancer. (B) Transfection miR-4274 mimics in HCT116 (left) and HCT8 (right) cells significantly increased the cell proliferation and induced radioresistance. (C) Transfection miR-4274 inhibitor in HCT116 (left) and HCT8 (right) cells significantly inhibited cell proliferation and increase the radiosensitivity. Growth curves data are mean ± SEM from 3 independent experiments and most error bars are with the symbols. (D) Western blot analysis demonstrated that the upregulation of miR-4274 resulted in a decrease in γ-H2AX protein levels, while the downregulation of miR-4274 led to an increase in γ-H2AX protein levels treated with radiation. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant of Student’s t-test. NC, NC mimics; miR [del], deletion-type miR-4274 mimics; miR [ins], insertion-type miR-4274 mimics; NC inh, NC inhibitor; miR inh [del], deletion-type miR-4274 inhibitor; miR inh [ins], insertion-type miR-4274 inhibitor. IR, irradiation; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Full size|PPT slide

To further understand the role of miR-4274 rs1553867776 in CRC, two types of miR-4274 mimics were respectively transfected into HCT116 and HCT8 cells. The insertion-type miR-4274 mimics were designed using Sanger sequencing results of CRC cell lines, and the deletion-type mimics were designed using the miRNA databases. In both HCT116 and HCT8 cell lines, the miR-4274 mimics (particularly the deletion-type) promoted cell proliferation compared to the NC group, as confirmed by proliferation and colony formation assays (Fig.1, S2A, S2B, S3A and S3B). After transfection with the miR-4274 inhibitor, the proliferation and colony formation of HCT116 and HCT8 cells was significantly inhibited (Fig.1, S2C, S2D, S3C and S3D). Similar results were seen in the gastric cancer cell line MGC-803 and liver cancer cell line HepG2 (Fig. S4). The results of cell proliferation experiments and colony formation assays suggest that miR-4274 may play a role in promoting cancer in patients with CRC and that the deletion-type elicits a more considerable effect, which is consistent with our previous survival-time results [9,10]. However, the results of cell migration assays didn’t show the significant differences. The role of miR-4274 in tumor metastasis in CRC needs further investigation (Fig. S5A and S5B).

To further investigate the correlation between the function and clinical expression of miR-4274, we examined the response of CRC cells with miR-4274 mimics and inhibitor to radiotherapy. While the miR-4274 mimics partially desensitized CRC cells to radiation (Fig.1), the miR-4274 inhibitor induced radiosensitivity (Fig.1). Since IR can induce a series of cellular DNA damage responses, we transfected miR-4274 and NC mimics into CRC cells and treated them with IR (2 Gy). We then measured the γ-H2AX protein levels—a well-known marker of DNA DSBs—to monitor cellular DNA damage. First we measured the protein level of γ-H2AX in wild-type HCT116 and HCT8 cells at 2, 6, 12 and 24 h after IR (Fig. S6A). Then we found the γ-H2AX level was lower in cells transfected with insertion-type miR-4274 mimics in 24 h after IR. Contrary results were observed in cells transfected with the miR-4274 inhibitor (Fig.1). These results suggest that rs1553867776 in miR-4274 might contribute to CRC radioresistance.

Considering that our formal analysis was based on the survival data of patients with rectal cancer who underwent postoperative chemoradiation therapy, we also explored the impact of miR-4274 on the sensitivity of CRC cells to chemotherapy. No significant difference was observed in the IC₅₀ after treatment of cells transfected with miR-4274 mimics with different concentrations of 5-fluorouracil (5-FU) (Fig. S1C and S1D). Therefore, our subsequent analyses focused on investigating miR-4274 factors affecting the sensitivity of tumor radiotherapy.

3.2 PEX5 is a target of miR-4274 in CRC

Since miRNAs function by targeting the mRNAs of specific genes, we searched six databases (miRWalk, Target Miner, PolymiRTS, DIANA TOOLS, miRDB and Target Scan) for possible targets of miR-4274. And we identified 53 candidates were showed in the all six databases (Table S2). Given that the rs1553867776 variant influences the function of miR-4274, we explored the pairing sites of miR-4274 and the predicted 53 candidates, focusing on the region that includes the bases of rs1553867776 in miR-4274. We found that PEX5 was the only target whose mRNA binding sites contained this region (Fig.2). Therefore, we selected PEX5 as the target gene for further studies.

Fig.2 PEX5 is a target of miR-4274 in CRC. (A) Venn diagram of the predicted results in 6 miRNA data sets and cancer databases followed by flowchart of miR-4274 target selection. (B) Schematic diagrams showing the putative binding sequence of miR-4274 in the wild-type PEX5 and the mutant PEX5 3′UTR. The miR-4274 was divided into miR-4274 mimics [del] and [ins] by simulating transcription products of different genotypes in rs1553867776. (C, D) The relative luciferase activity of psiCHECK-2 vector bearing PEX5 3′UTR was reduced by the miR-4274 mimics, especially the miR-4274 mimics [del], and increased by the miR-4274 inhibitor; however, mutation of the PEX5 3′UTR reversed these trends. Data are presented as mean ± SEM; *, P < 0.05; **, P < 0.01; ****, P < 0.0001 and ns, not significant of Student’s t-test. (E) Protein level of PEX5 was downregulated by the miR-4274 mimics, especially the deletion type, and upregulated by the miR-4274 inhibitor. NC, NC mimics; miR [del], deletion-type miR-4274 mimics; miR [ins], insertion-type miR-4274 mimics; NC inh, NC inhibitor; miR inh [del], deletion-type miR-4274 inhibitor; miR inh [ins], insertion-type miR-4274 inhibitor; WT, wild-type PEX5 3′UTR; MUT, mutant PEX5 3′UTR.

Full size|PPT slide

Luciferase assays showed that miR-4274 mimics decreased and miR-4274 inhibitor increased wild-type PEX5 3′UTR luciferase expression in HCT8 and HCT116. According to the predicted results, the wild-type PEX5 3′UTR was paired with the deletion type miR-4274 mimics, indicating that PEX5 is a direct target of miR-4274 (Fig.2 and S7). The results also showed that the role of the deletion type miR-4274 mimics was more pronounced than expected, especially in HCT8 cells (Fig.2). The results of the luciferase assays for the lysates from cells transfected with the miR-4274 inhibitor further verified these results (Fig.2). The Western blot results suggested that miR-4274 regulates the protein level of PEX5 via post-transcriptional regulation. The protein levels of PEX5 were lower in cells transfected with miR-4274 deletion-type mimics. On the contrary, in cells transfected with miR-4274 inhibitor, there was an increase in the level of PEX5 protein (Fig.2). Western blot further established that PEX5 is the target gene of miR-4274. These results suggested that SNPs in the miRNA seed regions could potentially influence the translation efficiency of target mRNA by affecting the binding of miRNA to the target mRNA 3′UTR, and thus play a regulatory role in cancer development.

3.3 PEX5 mediates CRC radiosensitivity

To explore the function of PEX5 in CRC, we investigated the expression of PEX5 protein by analyzing clinical tumor proteomics data from the National Cancer Institute’s Clinical Proteomic Tumor Analysis Consortium (CPTAC) using the UALCAN portal [27]. Results revealed that PEX5 protein expression was notably higher in normal tissues than in tumor tissues, which is a common phenomenon across various tumor types (Fig.3). The abundance of PEX5 protein noticeably declined with tumor progression in CRC (Fig.3). Survival analysis based on data from the Cancer Genome Atlas Rectum Adenocarcinoma Program (TCGA-READ, n = 166), TCGA Kidney Renal Clear Cell Carcinoma Program (TCGA-KIRC, n = 533), and TCGA Brain Lower Grade Glioma Program (TCGA-LGG, n = 516) indicated that patients with lower PEX5 protein expression experienced shorter survival outcomes (Fig.3, S8A and S8B). It is common in patients with cancer for the PEX5 gene to function as a tumor suppressor. Having identified the abnormal expression pattern of PEX5 in cancer tissues, we investigated how it modulates tumor cell radiosensitivity. Notably, the proliferation of HCT116 and HCT8 cell lines was significantly increased when PEX5 was downregulated, especially with IR, resulting in many cells exhibiting radioresistance when PEX5 was silenced (Fig.3–3I, S9A, S9B, S10A and S10B). In contrast, the CRC cells were more sensitive to IR when PEX5 was overexpressed (Fig.3, 3G, S9C, S9D, S10C, S10D and S11). These results support our hypothesis that PEX5 participates in DNA repair and that its reduction is associated with CRC radioresistance. Thus, the crucial role of PEX5 as a mediator of radiosensitivity in CRC cells has been demonstrated. However, the results of the cell migration assays did not reveal significant differences (Fig. S5C and S5D). Further investigation is required to elucidate the role played by PEX5 in tumor metastasis in CRC.

Fig.3 PEX5 mediates radiosensitivity of CRC. (A) Expression level of PEX5 in the CPTAC across cancers. Breast, breast cancer; colon, colon cancer; liver, liver cancer; ccRCC, clear cell renal cell carcinoma; UCEC, uterine corpus endometrial carcinoma; lung, lung cancer; PAAD, pancreatic adenocarcinoma, N, normal tissues ; T, tumor tissues. The numbers in brackets represent the sample size. (B) Expression level of PEX5 in the different stages in colon cancer from CPTAC. (C) Kaplan–Meier estimates of survival time by expression of PEX5 from TCGA-READ, n = 166. (D, E, H, I) Silencing PEX5 expression by CRISPR/Cas9 in HCT116 and HCT8 cells significantly decreased sensitivity of cells to IR treatment. (D, E) shows proliferation curves of cells. (H, I) shows fractions of cell survival by limiting dilution assays of HCT116 (H) and HCT8 (I) cells. (F, G) PEX5 overexpression significantly increased sensitivity of cells to IR treatment. Data are mean ± SEM from at least three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 and ****, P < 0.0001 of Student’s t-test. KO#1, PEX5-KO#1; KO#2, PEX5-KO#2; OE, PEX5-OE.

Full size|PPT slide

3.4 PEX5 is associated with the radiation response of CRC cells

Next, we investigated the radioresistance mechanism associated with PEX5 silencing. RNA sequencing revealed several upregulated and downregulated genes in PEX5-KO CRC cells (Fig.4 and Table S3). Gene set enrichment analysis identified downregulated genes enriched in the response to radiation, including DDB2 [30], FANCD2 [31], CASP9 [32], TP53INP1 [33], among others. This provides evidence that PEX5-KO leads to radiotherapy resistance (Fig.4 and 4C). In addition, certain upregulated genes were enriched in positive regulation of cell migration pathways (e.g., MYADM [34], SOX9 [35], LAMC2 [36], INSR [37]) and wound healing (e.g., PPARD [38], EVPL [39], SCNN1A [40], EREG [41]). These genes may contribute to the effect of PEX5 on CRC tumor growth and, ultimately, prognosis (Fig.4, 4D and 4E). These findings were further verified by real-time PCR using the PEX5-KO and PEX5-OE cells treated with IR (Fig.4).

Fig.4 PEX5 is associated with response to radiation in CRC cells. (A) Volcano plot displays the 429 upregulated genes (fold change > 1; P < 0.05) and 298 downregulated genes (0 < fold change < 1; P < 0.05) in PEX5-KO cells. (B) Metascape gene enrichment analysis of the 429 upregulated genes (up) and 298 downregulated genes (down). (C–E) The bubble plots shows the specific genes response to radiation pathways (C), positive regulation of cell migration (D), and in wound healing (E). (F) The heatmap shows the q-RT PCR results of HCT116 cells treated with IR. (G) The images of comet assays showing knockdown of PEX5 significantly diminished DNA damage in HCT116 cells, which could be reversed by overexpression of PEX5. Scale bar, 100 µm. (H) Bar charts show the statistics of tail moments in comet assays. Data are mean ± SEM from 3 replicate experiments and 10 fields were randomly selected for each experiment. ****, P < 0.0001 of Student’s t-test. KO#1, PEX5-KO#1; KO#2, PEX5-KO#2; OE, PEX5-OE.

Full size|PPT slide

To further investigate the role of PEX5 in radiosensitivity and DNA damage repair, RNA was extracted from both the NC and PEX5-KO HCT116 cells 24 h after ionizing radiation (IR). RNA sequencing analysis unveiled 130 upregulated and 152 downregulated genes in PEX5-KO HCT116 cells compared to the control group (Figure S12A). Gene set enrichment analysis identified differential genes prominently enriched in the DNA repair pathway (Figure S12B, Table S4).

DNA comet assays showed that PEX5 overexpression triggered more IR-induced DNA damage than in control cells while silencing PEX5 reduced the DNA damage in IR-treated CRC cells (Fig.4 and 4H). These results further indicate that PEX5 is important in DNA damage and repair.

3.5 PEX5 participates in Ku-mediated DNA NHEJ repair

To investigate how PEX5 might function in DNA repair, we performed immunoprecipitation assays on the HCT116 cell lysates, followed by mass spectrometry analysis (Table S5).

To investigate the role of PEX5 in radiotherapy sensitivity, we analyzed the mass spectrometry results and found that PEX5 interacted with X-ray repair cross-complementing protein 6 (XRCC6), also known as Ku70 [42,43]. Immunofluorescence staining of HCT116 and HCT8 cells showed that PEX5 was distributed in the cytoplasm, Ku70 was distributed in the nucleus and cytoplasm, and PEX5 and Ku70 co-localized in the cytoplasm (Fig.5 and S13). The interaction between PEX5 and Ku70 was confirmed by a co-immunoprecipitation assay (Co-IP) using an anti-PEX5 antibody and HCT116 cell lysates, followed by Western blot (Fig.5).

Fig.5 PEX5 participates in Ku-mediated DNA NHEJ repair. (A) Immunofluorescence analysis of PEX5 and Ku70 co-staining in HCT116 and HCT8 cells. Scale bar, 100 µm. (B) Western blot analysis of PEX5 and Ku70 in HCT116. Cell lysates were immunoprecipitated with antibody against PEX5 or IgG. (C) Immunofluorescence analysis of Ku70 co-staining in HCT116 cells with PEX5-OE or PEX5-KO treated with or without IR. Scale bar, 100 µm. (D) Western blot analysis of Ku70, Ku80 and γ-H2AX levels in HCT116 and HCT8 cells with PEX5 OE or PEX5-KO. (E) Western blot analysis of PEX5, Ku70 and Ku80 in HCT116 cells with PEX5 OE or PEX5-KO. Cell lysates were separated from cytoplasm and nucleus with PEX5-KO and PEX5-OE. The blue and red bands are protein markers. KO#1, PEX5-KO#1; KO#2, PEX5-KO#2; OE, PEX5-OE.

Full size|PPT slide

Ku70 reportedly plays an important role in DNA NHEJ repair in the nucleus [42]. Therefore, we hypothesized that PEX5 interacts with Ku70 in the cytoplasm and prevents Ku70 from entering the nucleus, leading to CRC cell radiosensitivity. Western blotting and immunofluorescence results verified that when PEX5 was knocked-out in HCT116 and HCT8 cells, the Ku70 protein was upregulated, while in PEX5-OE CRC cells, Ku70 was downregulated, particularly in cells exposed to IR (Fig.5, 5D and S6B). To further confirm our hypothesis, we performed nuclear and cytoplasmic separation of the CRC cells. Isolated proteins were further verified via Western blot. Ku70 protein was increased in the nucleus of PEX5-KO cells and in the cytoplasm of PEX5-OE cells (Fig.5). This was further confirmed by immunofluorescence (Fig.5). Therefore, PEX5, as a target gene of miR-4274, prevents Ku70 from entering the nucleus and participating in DNA damage repair, leading to CRC cell death after IR.

3.6 PEX5 participates in IR-induced ferroptosis

Ferroptosis is an iron-dependent cell death program that plays an important role in cancer development and radiosensitivity. Peroxisomes are the organelles necessary for ferroptosis [14,15]. However, the function of PEX5 in ferroptosis is unclear. Therefore, we investigated whether the role of PEX5 in radiotherapy sensitivity in CRC is related to ferroptosis. In the RNA-sequencing results, ferroptosis drivers were downregulated, while inhibitors were upregulated in CRC (Fig.6). Among the three cell death program inhibitors, Ferrostatin-1 (Fer-1) inhibited the proliferation of PEX5-OE HCT116 cells. The pro-ferroptotic effect of PEX5 partly counteracted the ferroptosis inhibitory effect of Fer-1, resulting in a lower cell survival rate relative to the NC group in PEX5-OE cells (Fig.6). Western blot analysis revealed that the abundance of ACSL4 [44–46]—an important ferroptosis driver—was changed to PEX5, particularly in IR-exposed cells. Two ferroptosis inhibitors, SLC7A11 [47] and GPX4 [13], were in contrast to PEX5 (Fig.6). As the final product of the reaction between free radicals and polyunsaturated fatty acids, MDA is a marker of lipid peroxidation that indirectly reflects cellular ferroptosis levels in cells [9]. We found that PEX5 overexpression increased IR-induced ferroptosis while PEX5-KO decreased IR-induced ferroptosis in CRC cells (Fig.6 and 6E). Hence, PEX5 might act as a ferroptosis driver that influences the radiosensitivity of CRC cells. Peroxisomes can synthesize polyunsaturated ether phospholipids (PUFA-ePLs) to induce ferroptosis [16]. Meanwhile, PEX5 protein is an essential part of peroxisomes, which could explain why PEX5 promotes the radiosensitivity of CRC cells through ferroptosis. However, the specific mechanism requires further investigation.

Fig.6 PEX5 participates in the IR-reduced ferroptosis. (A) The bubble plot shows the different regulation genes related to ferroptosis. (B) The viability ratio of 3 different inhibitors cell death program inhibitors, Ferrostatin-1 (Fer-1, 4 µM), Necrostatin-1 (Nec-1, 0.5 µM) and Z-VAD(OMe)-FMK (Z-vad, 10 µM) in HCT116 PEX5-OE cells. (C) Western blotting analysis of PEX5, ACSL4, SLC7A11 and GPX4 in HCT116 and HCT8 cells with PEX5 KO or OE. (D, E) HCT116 (D) and HCT8 (E) cells were treated with or without IR, followed by detecting the concentration of MDA. MDA levels were expressed as µM. Data are mean ± SEM from at least three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 and ****, P < 0.0001 of Student’s t-test. KO#1, PEX5-KO#1; KO#2, PEX5-KO#2; OE, PEX5-OE.

Full size|PPT slide

3.7 The effects of miR-4274 and PEX5 on CRC progression and radiosensitivity in vivo

We subcutaneously transplanted HCT116 and HCT8 cells into NPI mice and injected the insertion and deletion-type miR-4274 agomiR-mimics, or NC agomiR-mimics separately. Transfection of the insertion-type miR-4274 agomiR-mimics enhanced the progression of xenograft tumors compared to the deletion-type and control group, especially after IR (Fig.7, 7B and 7E). We transplanted PEX5-KO HCT8 cells (PEX5-KO#1) and control cells into NPI mice, and found that PEX5 downregulation significantly improved tumor growth in vivo compared to the control group (Fig.7, 7D and 7F). Similar results were obtained for HCT116 cells (Figure S14). To further verify our research, we also transplanted PEX5-OE CRC cells and control cells into NPI mice. Upregulation of PEX5 significantly reduced tumor growth and increased CRC radiosensitivity (Figure S15).

Fig.7 The effects of miR-4274 and PEX5 on CRC progression and radiosensitivity in vivo. (A, B) The effects of miR-4274 on the volumes of CRC tumors, n = 3. NC, NC ago-mimics; miR [del], deletion-type miR-4274 ago-mimics; miR [ins], insertion-type miR-4274 ago-mimics. (C, D) The effects of PEX5-KO on the volumes of CRC tumors, n = 3. *, P < 0.05; **, P < 0.01 of Student’s t-test at the last measurement. (E) The effects of miR-4274 on the weights of CRC tumors, n = 3. NC, NC ago-mimics; miR [del], deletion-type miR-4274 ago-mimics; miR [ins], insertion-type miR-4274 ago-mimics. (F) The effects of PEX5-KO on the weights of CRC tumors, n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001 and ****, P < 0.0001 of Student’s t-test. (G) The schematic illustration for the possible mechanisms of miR-4274 rs1553867776 regulates PEX5 expression and PEX5-mediated radioresistance in CRC.

Full size|PPT slide

Taken together, these results indicate that PEX5 is closely associated with high radiotherapy efficacy and improved survival outcomes in patients with CRC. Hence, PEX5 might represent an effective cancer radiotherapy response marker.

4 Discussion

In this study, using CRC as a model, we identified a significant association between rs1553867776 in the miR-4274 seed region and cancer radiosensitivity. For the first time, we identified PEX5 as the target gene of miR-4274. Moreover, compared with the insertion genotype, the deletion genotype of rs1553867776 in miR-4274 downregulates the abundance of PEX5 protein through post-transcriptional regulation. High expression of PEX5 significantly induces tumor response to radiotherapy. Mechanistically, PEX5 enhances DNA DSBs, as reflected by the cellular γ-H2AX level, likely by combining with Ku70 in the cytoplasm to impede IR-induced DNA damage repair by the Ku70/Ku80 complex in the nucleus. In addition, enhanced PEX5 expression is associated with the promotion of IR-induced ferroptosis, and targeting this mechanism can increase the radiosensitivity of cancer cells. Therefore, the miR-4274–PEX5 axis is an important contributor to tumor radioresistance and may serve as a novel clinical treatment target for improving radiotherapy efficacy.

While radiation therapy is a cornerstone of cancer treatment for many cancers, including CRC, primary or secondary radioresistance often occurs, contributing to cancer relapse and poor prognosis [2]. Currently, the mechanisms underlying radioresistance are not well understood, limiting the effectiveness of radiotherapy and personalized treatment decisions. Our previous studies reported that rs1553867776 in miR-4274 is a biomarker of clinical outcomes in patients with CRC [9,10]. Meanwhile, in the current study, we found that miR-4274 may promote CRC tumor proliferation and radiation resistance in vivo and in vitro, particularly with the deletion-type rs1553867776. Our results are consistent with our previous findings, showing that deletion-type rs1553867776 significantly correlates with tumor progression and poor survival in patients with CRC [9,10]. The seed region, nucleotides 2–7 from the 5′-end of the miRNA, plays a pivotal role in identifying target mRNAs. In the present study, we demonstrated a specific mechanism for rs1553867776 in the miR-4274 seed region. Different miR-4274 genotypes regulate the translation of PEX5 mRNA through base pairing, as evidenced by our luciferase reporter assays and Western blot results. Therefore, rs1553867776 may be regarded as a variant involved in regulating PEX5 protein levels.

PEX5 is a receptor that binds to the peroxisome membrane. While most previous studies have focused on PEX5-mediated protein import [19], the genetic effects and underlying mechanisms of action of PEX5 in CRC remain unknown. In our study, PEX5 functionally reduced CRC cell proliferation and increased radiosensitivity in vivo and in vitro. Indeed, PEX5 mediates radiosensitivity and participates in DNA NHEJ repair. Mechanistically, PEX5 interacts with Ku70 in the cytoplasm to prevent the formation of the Ku70/Ku80 complex to initiate NHEJ repair. These results implicate PEX5 as a novel regulator of the NHEJ pathway. Based on our results, we propose that PEX5 acts as a receptor protein for the binding of Ku70 in peroxisomes, which requires verification in future studies.

Another noteworthy finding is that PEX5 plays a role in ferroptosis. It is also associated with radiotherapy-induced cell death. IR increases lipid peroxidation and triggers ferroptosis, which can enhance radiosensitivity. Meanwhile, peroxisomes provide substrates that contribute to ferroptosis [15–17]. However, for the first time, we revealed that PEX5 mediates cancer radiosensitivity through ferroptosis; the specific underlying molecular mechanism requires further exploration. Nevertheless, targeting PEX5 and inducing ferroptosis in CRC to enhance radiotherapy is a promising novel approach.

Notably, this study is the first to investigate the molecular mechanisms of genetic variations in miRNA seed regions associated with CRC radiotherapy sensitivity. This provides definitive proof for SNPs’ prediction of treatment efficacy for patients with CRC. Furthermore, our study has significant therapeutic value regarding transferring a functional anti-miRNA oligonucleotide to exosomes, which may overcome IR resistance in CRC and increase the effectiveness of cancer treatment. With advancements in technology and progress in new drug development, there is promising potential for drugs targeting miR-4274 to enhance the effectiveness of radiotherapy in patients. The combination of elevating PEX5 and ferroptosis inducers applied in clinical settings holds the promise of increasing radiotherapy sensitivity in cancer patients, thereby addressing issues of radiotherapy resistance. This area has broad prospects for the future.

This study has several limitations. First, the mechanism underlying the effect of PEX5 on CRC cell proliferation and migration requires further investigation. Second, the role of PEX5 in IR-induced ferroptosis warrants further evaluation. Third, other studies have reported an oncogenic role for PEX5 in hepatocellular carcinoma, possibly due to the heterogeneity of different organs [48]. However, a study has revealed that peroxisome proteins, including PEX5, were lower in some liver tumor tissues compared to non-tumor liver tissues. The study proved the significant heterogeneity of liver cancer itself. The PEX5 protein may also play a role in inhibiting cancer in certain liver cancer cell [49]. Finally, our research is limited to the in vitro and in vivo levels. Hence, further verification in clinical trials is necessary to translate these findings into practical applications.

In summary, our multiomics analysis and functional assays identified how miR-4274 targets PEX5 and plays a critical role in radioresistance (Fig.7). We provide new insights into the molecular mechanism underlying the role of PEX5 in radioresistance; that is, PEX5 prevents Ku70 from entering the nucleus, participates in IR-induced ferroptosis, and enhances the radiosensitivity of cancer cells. Our research suggests that miR-4274 and PEX5 may have significant implications for developing precise and effective cancer radiotherapy. These molecules may serve as valuable CRC prognostic markers and therapeutic targets.

参考文献

[1] Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin 2023; 73(1): 17–48

[2] Dekker E, Tanis PJ, Vleugels JLA, Kasi PM, Wallace MB. Colorectal cancer. Lancet 2019; 394(10207): 1467–1480

[3] Pucci S, Polidoro C, Joubert A, Mastrangeli F, Tolu B, Benassi M, Fiaschetti V, Greco L, Miceli R, Floris R, Novelli G, Orlandi A, Santoni R. Ku70, Ku80, and sClusterin: acluster of predicting factors for response to neoadjuvant chemoradiation therapy in patients with locally advanced rectal cancer. Int J Radiat Oncol Biol Phys 2017; 97(2): 381–388

[4] Liu W, Miao C, Zhang S, Liu Y, Niu X, Xi Y, Guo W, Chu J, Lin A, Liu H, Yang X, Chen X, Zhong C, Ma Y, Wang Y, Zhu S, Liu S, Tan W, Lin D, Wu C. VAV2 is required for DNA repair and implicated in cancer radiotherapy resistance. Signal Transduct Target Ther 2021; 6(1): 322

[5] Bartel DP. Metazoan MicroRNAs. Cell 2018; 173(1): 20–51

[6] Salzman DW, Weidhaas JB. SNPing cancer in the bud: microRNA and microRNA-target site polymorphisms as diagnostic and prognostic biomarkers in cancer. Pharmacol Ther 2013; 137(1): 55–63

[7] Shen C, Yan T, Wang Z, Su HC, Zhu X, Tian X, Fang JY, Chen H, Hong J. Variant of SNP rs1317082 at CCSlnc362 (RP11-362K14.5) creates a binding site for miR-4658 and diminishes the susceptibility to CRC. Cell Death Dis 2018; 9(12): 1177

[8] Wang W, Yang C, Nie H, Qiu X, Zhang L, Xiao Y, Zhou W, Zeng Q, Zhang X, Wu Y, Liu J, Ying M. LIMK2 acts as an oncogene in bladder cancer and its functional SNP in the microRNA-135a binding site affects bladder cancer risk. Int J Cancer 2019; 144(6): 1345–1355

[9] Chen H, Yin L, Yang J, Ren N, Chen J, Lu Q, Huang Y, Feng Y, Wang W, Wang S, Liu Y, Song Y, Li Y, Jin J, Tan W, Lin D. Genetic polymorphisms in genes regulating cell death and prognosis of patients with rectal cancer receiving postoperative chemoradiotherapy. Cancer Biol Med 2023; 20(4): 297–316

[10] Huang Y, Feng Y, Ren H, Zhang M, Li H, Qiao Y, Feng T, Yang J, Wang W, Wang S, Liu Y, Song Y, Li Y, Jin J, Tan W, Lin D. Associations of genetic variations in microRNA seed regions with acute adverse events and survival in patients with rectal cancer receiving postoperative chemoradiation therapy. Int J Radiat Oncol Biol Phys 2018; 100(4): 1026–1033

[11] Landeros N, Corvalan AH, Musleh M, Quiñones LA, Varela NM, Gonzalez-Hormazabal P. Novel risk associations between microRNA polymorphisms and gastric cancer in a Chilean population. Int J Mol Sci 2021; 23(1): 467

[12] Shkurnikov M, Nikulin S, Nersisyan S, Poloznikov A, Zaidi S, Baranova A, Schumacher U, Wicklein D, Tonevitsky A. LAMA4-regulating miR-4274 and its host gene SORCS2 play a role in IGFBP6-dependent effects on phenotype of basal-like breast cancer. Front Mol Biosci 2019; 6: 122

[13] Liu S, Zhang HL, Li J, Ye ZP, Du T, Li LC, Guo YQ, Yang D, Li ZL, Cao JH, Hu BX, Chen YH, Feng GK, Li ZM, Deng R, Huang JJ, Zhu XF. Tubastatin A potently inhibits GPX4 activity to potentiate cancer radiotherapy through boosting ferroptosis. Redox Biol 2023; 62: 102677

[14] Lei G, Mao C, Yan Y, Zhuang L, Gan B. Ferroptosis, radiotherapy, and combination therapeutic strategies. Protein Cell 2021; 12(11): 836–857

[15] Wang X, Zhou Y, Min J, Wang F. Zooming in and out of ferroptosis in human disease. Front Med 2023; 17(2): 173–206

[16] Zou Y, Henry WS, Ricq EL, Graham ET, Phadnis VV, Maretich P, Paradkar S, Boehnke N, Deik AA, Reinhardt F, Eaton JK, Ferguson B, Wang W, Fairman J, Keys HR, Dančík V, Clish CB, Clemons PA, Hammond PT, Boyer LA, Weinberg RA, Schreiber SL. Plasticity of ether lipids promotes ferroptosis susceptibility and evasion. Nature 2020; 585(7826): 603–608

[17] Ravindran R, Bacellar IOL, Castellanos-Girouard X, Wahba HM, Zhang Z, Omichinski JG, Kisley L, Michnick SW. Peroxisome biogenesis initiated by protein phase separation. Nature 2023; 617(7961): 608–615

[18] Chen X, Kang R, Kroemer G, Tang D. Organelle-specific regulation of ferroptosis. Cell Death Differ 2021; 28(10): 2843–2856

[19] Fujiki Y, Okumoto K, Honsho M, Abe Y. Molecular insights into peroxisome homeostasis and peroxisome biogenesis disorders. Biochim Biophys Acta Mol Cell Res 2022; 1869(11): 119330

[20] Yan H, Talty R, Aladelokun O, Bosenberg M, Johnson CH. Ferroptosis in colorectal cancer: a future target. Br J Cancer 2023; 128(8): 1439–1451

[21] Sticht C, De La Torre C, Parveen A, Gretz N. miRWalk: An online resource for prediction of microRNA binding sites. PLoS One 2018; 13(10): e0206239

[22] Bandyopadhyay S, Mitra R. TargetMiner: microRNA target prediction with systematic identification of tissue-specific negative examples. Bioinformatics 2009; 25(20): 2625–2631

[23] Bhattacharya A, Ziebarth JD, Cui Y. PolymiRTS Database 3.0: linking polymorphisms in microRNAs and their target sites with human diseases and biological pathways. Nucleic Acids Res 2014; 42(Database issue): D86–D91

[24] Paraskevopoulou MD, Georgakilas G, Kostoulas N, Vlachos IS, Vergoulis T, Reczko M, Filippidis C, Dalamagas T, Hatzigeorgiou AG. DIANA-microT web server v5.0: service integration into miRNA functional analysis workflows. Nucleic Acids Res 2013; 41(W1): W169–W173

[25] Chen Y, Wang X. miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Res 2020; 48(D1): D127–D131

[26] Agarwal V, Bell GW, Nam JW, Bartel DP. Predicting effective microRNA target sites in mammalian mRNAs. eLife 2015; 4: e05005

[27] Chandrashekar DS, Karthikeyan SK, Korla PK, Patel H, Shovon AR, Athar M, Netto GJ, Qin ZS, Kumar S, Manne U, Creighton CJ, Varambally S. UALCAN: An update to the integrated cancer data analysis platform. Neoplasia 2022; 25: 18–27

[28] Li T, Fu J, Zeng Z, Cohen D, Li J, Chen Q, Li B, Liu XS. TIMER2.0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Res 2020; 48(W1): W509–W514

[29] Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, Benner C, Chanda SK. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 2019; 10(1): 1523

[30] Huang S, Fantini D, Merrill BJ, Bagchi S, Guzman G, Raychaudhuri P. DDB2 is a novel regulator of Wnt signaling in colon cancer. Cancer Res 2017; 77(23): 6562–6575

[31] Burdak-Rothkamm S, Rothkamm K, McClelland K, Al Rashid ST, Prise KM. BRCA1, FANCD2 and Chk1 are potential molecular targets for the modulation of a radiation-induced DNA damage response in bystander cells. Cancer Lett 2015; 356(2 2 Pt B): 454–461

[32] Han C, Liu Z, Zhang Y, Shen A, Dong C, Zhang A, Moore C, Ren Z, Lu C, Cao X, Zhang CL, Qiao J, Fu YX. Tumor cells suppress radiation-induced immunity by hijacking caspase 9 signaling. Nat Immunol 2020; 21(5): 546–554

[33] Sándor N, Schilling-Tóth B, Kis E, Fodor L, Mucsányi F, Sáfrány G, Hegyesi H. TP53inp1 gene is implicated in early radiation response in human fibroblast cells. Int J Mol Sci 2015; 16(10): 25450–25465

[34] Bracker TU, Sommer A, Fichtner I, Faus H, Haendler B, Hess-Stumpp H. Efficacy of MS-275, a selective inhibitor of class I histone deacetylases, in human colon cancer models. Int J Oncol 2009; 35(4): 909–920

[35] Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature 2012; 487(7407): 330–337

[36] Huang D, Du C, Ji D, Xi J, Gu J. Overexpression of LAMC2 predicts poor prognosis in colorectal cancer patients and promotes cancer cell proliferation, migration, and invasion. Tumour Biol 2017; 39(6): 1010428317705849

[37] Vasilogianni AM, Al-Majdoub ZM, Achour B, Peters SA, Rostami-Hodjegan A, Barber J. Proteomic quantification of receptor tyrosine kinases involved in the development and progression of colorectal cancer liver metastasis. Front Oncol 2023; 13: 1010563

[38] Liu Y, Deguchi Y, Tian R, Wei D, Wu L, Chen W, Xu W, Xu M, Liu F, Gao S, Jaoude JC, Chrieki SP, Moussalli MJ, Gagea M, Morris J, Broaddus RR, Zuo X, Shureiqi I. Pleiotropic effects of PPARD accelerate colorectal tumorigenesis, progression, and invasion. Cancer Res 2019; 79(5): 954–969

[39] Köse B, Laar RV, Beekhuizen HV, Kemenade FV, Baykal AT, Luider T, Güzel C. Quantitative proteomic analysis of MCM3 in ThinPrep samples of patients with cervical preinvasive cancer. Int J Mol Sci 2023; 24(13): 10473

[40] Lou J, Wei L, Wang H. SCNN1A overexpression correlates with poor prognosis and immune infiltrates in ovarian cancer. Int J Gen Med 2022; 15: 1743–1763

[41] Cheng WL, Feng PH, Lee KY, Chen KY, Sun WL, Van Hiep N, Luo CS, Wu SM. The role of EREG/EGFR pathway in tumor progression. Int J Mol Sci 2021; 22(23): 12828

[42] Sui H, Hao M, Chang W, Imamichi T. The role of Ku70 as a cytosolic DNA sensor in innate immunity and beyond. Front Cell Infect Microbiol 2021; 11: 761983

[43] Yang MD, Tsai CW, Chang WS, Tsou YA, Wu CN, Bau DT. Predictive role of XRCC5/XRCC6 genotypes in digestive system cancers. World J Gastrointest Oncol 2011; 3(12): 175–181

[44] Liao P, Wang W, Wang W, Kryczek I, Li X, Bian Y, Sell A, Wei S, Grove S, Johnson JK, Kennedy PD, Gijón M, Shah YM, Zou W. CD8⁺ T cells and fatty acids orchestrate tumor ferroptosis and immunity via ACSL4. Cancer Cell 2022; 40(4): 365–378.e6

[45] Quan J, Bode AM, Luo X. ACSL family: The regulatory mechanisms and therapeutic implications in cancer. Eur J Pharmacol 2021; 909: 174397

[46] Lei G, Zhang Y, Koppula P, Liu X, Zhang J, Lin SH, Ajani JA, Xiao Q, Liao Z, Wang H, Gan B. The role of ferroptosis in ionizing radiation-induced cell death and tumor suppression. Cell Res 2020; 30(2): 146–162

[47] Lang X, Green MD, Wang W, Yu J, Choi JE, Jiang L, Liao P, Zhou J, Zhang Q, Dow A, Saripalli AL, Kryczek I, Wei S, Szeliga W, Vatan L, Stone EM, Georgiou G, Cieslik M, Wahl DR, Morgan MA, Chinnaiyan AM, Lawrence TS, Zou W. Radiotherapy and immunotherapy promote tumoral lipid oxidation and ferroptosis via synergistic repression of SLC7A11. Cancer Discov 2019; 9(12): 1673–1685

[48] Cai M, Sun X, Wang W, Lian Z, Wu P, Han S, Chen H, Zhang P. Disruption of peroxisome function leads to metabolic stress, mTOR inhibition, and lethality in liver cancer cells. Cancer Lett 2018; 421: 82–93

[49] Zhu H, Lin Y, Lu D, Wang S, Liu Y, Dong L, Meng Q, Gao J, Wang Y, Song N, Suo Y, Ding L, Wang P, Zhang B, Gao D, Fan J, Gao Q, Zhou H. Proteomics of adjacent-to-tumor samples uncovers clinically relevant biological events in hepatocellular carcinoma. Natl Sci Rev 2023; 10(8): nwad167

Acknowledgments

This work is supported by grants from the National Natural Science Foundation (Grant No. 81972859 to W.T.), CAMS Innovation Fund for Medical Sciences (CIFMS) (Grant No. 2021-I2M-1-013 to D.L. and W.T.), and State Key Laboratory of Molecular Oncology Grants (Grant No. SKLMO-2021-03 to W.T. and SKLMO-KF-2023-03 to D.L.)

Electronic Supplementary Material

Supplementary material is available in the online version of this article at https://doi.org/10.1007/s11684-024-1082-6 and is accessible for authorized users.

Compliance with ethics guidelines

Conflicts of interest Qixuan Lu, Ningxin Ren, Hongxia Chen, Shaosen Zhang, Ruoqing Yan, Mengjie Li, Linlin Zheng, Wen Tan, and Dongxin Lin declare that they have no conflict of interest.

The study was approved by the appropriate institutional and/or national research ethics committee and the study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Informed consent was obtained from all patients for being included in the study. All institutional and national guidelines for the care and use of laboratory animals were followed.

版权

2024 Higher Education Press