YEATS2 promotes DNA repair and induces anoikis resistance by enhancing chromatin accessibility to drive prostate cancer metastasis
Haoran Li, Yarong Song, Yukun Cong, Chuxiong Wang, Kang Chen, Chunyu Liu, Menghao Zhou, Yunjie Ju, Jinyu Chen, Liang Chen, Yifei Xing
TL;DR
This study shows that YEATS2 helps prostate cancer spread by making DNA repair easier and preventing cell death, offering new treatment ideas.
Contribution
The novel role of YEATS2 in promoting prostate cancer metastasis through chromatin accessibility and DNA repair is revealed.
Findings
YEATS2 expression is elevated in metastatic prostate cancer and linked to poor outcomes.
YEATS2 promotes metastasis by enhancing RAD50 expression via chromatin accessibility.
Mirin inhibits lymph node metastasis in prostate cancer cells in vivo.
Abstract
Despite advancements in therapeutic strategies, metastatic prostate cancer (mPCa) remains challenging to treat, with limited clinical efficacy and poor prognosis. Anoikis resistance in tumor cells is crucial for their survival in the vascular system and plays a key role in metastasis. Therefore, investigating the molecular mechanisms of metastasis and anoikis resistance is essential for identifying novel therapeutic targets and strategies. In this study, we found that YEATS domain-containing 2 (YEATS2) plays a critical role in promoting PCa metastasis by suppressing anoikis. We observed that YEATS2 expression was elevated in mPCa and associated with poor clinical outcomes. Knockdown of YEATS2 reduced the metastatic potential of PCa cells both in vivo and in vitro, whereas its overexpression inhibited anoikis and promoted metastasis by upregulating the expression of the DNA damage repair…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8- —https://doi.org/10.13039/501100001809National Natural Science Foundation of China (National Science Foundation of China)
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsProstate Cancer Treatment and Research · Cancer-related gene regulation · Epigenetics and DNA Methylation
Introduction
Prostate cancer (PCa), the most common tumor of the male urinary system, accounts for 7.3% of newly diagnosed cancers worldwide, with over 1.4 million new cases and more than 370,000 related deaths annually, making it one of the leading causes of cancer-related mortality in men [1]. Although multiple treatment options are available for PCa, therapeutic outcomes for metastatic disease remain limited. Patients with locally advanced prostate cancer typically undergo radical prostatectomy (RP) combined with radiation therapy (RT), but overall oncological outcomes remain poor [2]. Most tumor cells undergo programmed cell death (anoikis) after detaching from the extracellular matrix, while a minority of cells develop anoikis resistance and continue to survive in the vascular system (blood vessels and lymphatics), with resistance to anoikis being a prerequisite and a critical step for the metastasis of tumor cells [3–5]. Therefore, investigating the molecular mechanisms underlying PCa metastasis and anoikis resistance is essential for identifying novel anti-tumor targets and developing effective treatment strategies.
Obtaining samples from PCa patients with bone metastases remains challenging, whereas both primary tumors and paired lymph node metastatic lesions from the same patient are more readily accessible in cases of lymph node metastasis. Additionally, lymph node metastasis animal models are relatively well-established. Therefore, our study aims to investigate the potential mechanisms through which PCa cells acquire anoikis resistance, with a focus on lymph node metastasis. By integrating sequencing data from public databases [6, 7], transcriptomic data from anoikis-resistant PCa cells and proteomic data from clinical metastatic PCa tissues, we identified consistent upregulation of SRPRB, YEATS2, and FAM126A, and downregulation of OSR2 in metastatic tissues and anoikis-resistant PCa cells. Notably, YEATS2 upregulation was significantly associated with poor prognosis in PCa, correlating with a marked reduction in overall survival and progression-free survival. YEATS2 is a component of the histone acetyltransferase complex known as ATAC (Ada-Two-A-Containing), and it participates in the regulation of histone post-translational modifications [8, 9]. YEATS2 promotes tumor progression or metastasis in non-small cell lung cancer and pancreatic cancer [10, 11]. However, the impact of YEATS2 on prostate cancer development and progression, as well as its role in promoting anoikis resistance in tumor cells, remains unexplored.
In this study, we found that YEATS2 enhances histone acetylation, increases chromatin accessibility, and upregulates the expression of the DNA damage repair-related gene RAD50 by recognizing the histone modification H3K27ac, thereby inducing anoikis resistance and promoting prostate cancer metastasis.
Materials and methods
Clinical samples
Specimens of PCa tissues and their lymph metastatic tissues were obtained from patients who underwent radical cystectomy for prostate carcinoma at the Department of Urology, Union Hospital, affiliated with Tongji Medical College, between 2023 and 2024. All specimen collections for the organization were conducted with patient-informed consent and approved by the Research Ethics Committees of Union Hospital, Tongji Medical College, and Huazhong University of Science and Technology.
Differential gene expression and protein analysis
Using R Studio, we conducted an analysis of sequencing data related to metastatic and primary prostate cancer lesions from the public database platform Gene Expression Omnibus (GEO) (GSE6752, GSE6919). This was combined with differentially expressed genes from our previously established anoikis-resistant PC-3 cells (GSE100629) to identify common differentially expressed genes. The proteomics results were assisted by Genechem (Shanghai, China). The thresholds were set as Fold change >1.5 or Fold change < 1.5^–1^, and P < 0.05. (Details are provided in Supplementary files 1–4).
Cell lines and cell culture
The 22Rv1 (RRID: CVCL_1045), PC-3 (RRID: CVCL_0035), and RM-1 (RRID: CVCL_B459) cell lines were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The parental cells (PC-3) were cultured in ultra-low attachment six-well plates (3471, Corning, USA) for 14 days, then transferred to standard six-well plates. Cells that adhered within 24 h were considered anoikis-resistant cells and were named PC-3-AnoR. The methods for establishing prostate cancer cell lines from primary and metastatic lesions are described below. All cells were cultured in RPMI 1640 medium (HY1640, TBDscience, China) supplemented with 10% fetal bovine serum (FBS500-H, HYCEZMBIO, China) and 1% penicillin/streptomycin (HY-C-XP976, HYCEZMBIO), and incubated at 37 °C in a humidified atmosphere of 5% CO_2_ and 95% air.
Mouse popliteal lymph node metastasis model
All animal experiments were approved by the Animal Care Committee of Tongji Medical College. Experimental animals used were BALB/c Nude (male, 4 weeks) and C57BL/6 (male, 4 weeks) mice, purchased from the Animal Center of the Chinese Academy of Medical Sciences and housed at the SPF-level animal experimental center. 22Rv1 and PC-3 cells were injected into BALB/c Nude mice, while RM-1 cells were injected into C57BL/6 mice. A suspension of 50 μL PBS containing 2 × 10^6^ prostate cancer cells was injected into one footpad of the mouse, with an equal volume of PBS injected into the contralateral footpad as a control. Mice were collected when the tumor volume in the footpad reached approximately 200 mm³.
The acquisition of primary and metastatic cell lines
The footpad primary tumor tissue and the ipsilateral popliteal lymph nodes were isolated and soaked in 2% penicillin/streptomycin PBS for 1 h. They were then thoroughly minced in an EP tube and placed in serum-free RPMI-1640 culture medium containing 2 mg/ml Type I collagenase (SCR103, Sigma, USA) at 37 °C in a 5% CO2 incubator. Once no obvious tissue clumps were observed, dissociation was terminated by adding 2% penicillin/streptomycin PBS, and the cells were then centrifuged, resuspended, and plated. Primary tumor cells were cultured and established cell lines capable of stable passaging. To confirm that the cells obtained from the popliteal lymph nodes were metastatic tumor cells, 22Rv1 and PC-3 cells were labeled with GFP and RFP, respectively, before being injected into the footpad. The primary tumor cells and lymph node metastatic cells derived from the 22Rv1, PC-3, and RM-1 cell lines were named as follows: 22Rv1-Pt, 22Rv1-Mt, PC-3-Pt, PC-3-Mt, RM-1-Pt, and RM-1-Mt, respectively.
Western blot analysis
Total protein was extracted from cells using RIPA lysis buffer (G2002-100ML, Servicebio, China). Protein concentration was determined with a BCA Protein Assay Kit (HBCA-500, HYCEZMBIO). Equal amounts of protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred to a polyvinylidene difluoride (PVDF) membrane (IPFL00010, Millipore, USA). To block the PVDF membrane, 5% skimmed milk or 5% BSA was applied for 1 h. The membrane was washed three times with 1X Tris-buffered saline containing 0.05% Tween 20 (TBST), with each wash lasting 10 min. Following this, the membrane was incubated with primary antibodies overnight at 4 °C. The next day, it was treated with the corresponding secondary antibodies (HRP-conjugated Affinipure Goat anti-mouse antibody, SA00001-1, or HRP-conjugated Affinipure Goat anti-rabbit antibody, SA00001-2, ProteinTech, China) at room temperature for 1 h. Bands were visualized via chemiluminescence using an electrochemiluminescence (ECL) system. The primary antibodies used were: SRPRB (A10591, Abclonal, China)), YEATS2 (24717-1-AP, Proteintech), FAM126A (26243-1-AP, Proteintech), OSR2 (A18178, Abclonal), γH2AX (AP0687, Abclonal), RAD50 (29390-1-AP, Proteintech), VAV3 (A15385, Abclonal), BCL2 (A0208, Abclonal), BRCA1 (A0212, Abclonal), PARP1 (A27791, Abclonal), NR2C2 (ab109301, Abcam, UK), Rabbit control IgG (AC005, Abclonal), Histone H3 (68345-1-Ig, Proteintech), H3K9ac (A7255, Abclonal), H3K14ac (A7254, Abclonal), H3K27ac (A22264, Abclonal), H3K27cr (PTM-545RM, PTM BIO, China), anti-pan Kbz (PTM-762, PTM BIO), anti-Flag (66008-4-Ig, Proteintech), anti-GST (AE077, Abclonal), MRE11(A2559, Abclonal), β-actin (66009-1-Ig, Proteintech), GAPDH (60004-1-Ig, Proteintech).
Quantitative real-time PCR (qRT-PCR)
Total RNA was extracted using Ultrapure RNA Kit (CW0581, CWBIO, China). The synthesis of cDNA from the extracted RNA was carried out using the HiScript cDNA synthesis kit (R323-01, Vazyme, China). Following the manufacturer’s protocol, quantitative RT-PCR (qRT-PCR) was conducted with ChamQ SYBR qPCR Master Mix (Q311-02, Vazyme) on a StepOne Plus real-time PCR system (Life Technologies, USA). ACTB served as the loading control for normalization. The sequences of the primers used in this study are detailed in Supplementary file 5.
Immunohistochemistry (IHC)
As previously described [12], IHC staining was conducted using primary antibodies targeting YEATS2 (PA5-36939, Invitrogen, USA). To evaluate immunoreactivity, an H-score was calculated based on both the percentage and staining intensity of positively stained PCa cells. The staining intensity was categorized into four levels: (0) no staining, (1) weak staining, (2) moderate staining, and (3) strong staining. The H-score was determined using the formula: (percentage of weak staining ×1) + (percentage of moderate staining ×2) + (percentage of strong staining ×3), with possible scores ranging from 0 to 300.
Plasmid transfection
The following plasmids were constructed using PAIVIBIO (Wuhan, China): plasmids for YEATS2 overexpression, knockdown, and truncation; NR2C2 overexpression and knockdown; RAD50 overexpression and knockdown; MRE11 knockdown. Plasmids and vectors were introduced into PCa cells using a DNA transfection reagent (TF20121201, NEOFECT) following the guidelines provided by the manufacturer. Plasmid knockdown sequences are listed in Supplementary file 6.
Cell viability assay
Cell counting kit-8 (CCK-8) assay (HYCCK8, HYCEZMBIO) was used to determine cell viability. Five thousand cells were plated per well in ultra-low attachment six-well plates. After the corresponding time in suspension culture, all cells were collected, resuspended in 200 μL of complete culture medium, and then transferred to a 96-well plate. The cells were incubated at 37 °C in a 5% CO_2_ incubator for 24 h. Following the manufacturer’s instructions, the CCK-8 detection reagent was added, and the cells were incubated at 37 °C for 1 h. The absorbance at 450 nm for each group was measured using a microplate reader (Tecan, Switzerland). The different time points in the suspension plates served as observation endpoints, and the above steps were repeated.
Cell migration and invasion assays
To evaluate the migratory and invasive potential of these cells, they were pre-starved in serum-free medium for 12 h. Transwell assays were performed using 24-well plates fitted with 8.0 μm pore-size polycarbonate membrane inserts (353097, Corning). For the migration assay, the lower chambers were filled with 600 μL of complete medium. Subsequently, cell numbers were quantified. Approximately 4 × 10^4^ cells were suspended in 200 μL of serum-free medium and added to the upper chamber; cells were incubated at 37 °C with 5% CO₂ (48 h for 22Rv1 cells and 24 h for PC-3 cells). After incubation, remove any cells remaining in the upper chambers. The cells that migrated through the membrane were fixed with 4% paraformaldehyde for 30 min and then stained with 0.1% crystal violet for 30 min at room temperature. After staining, the cells were washed with phosphate-buffered saline (PBS) three times and counted in three randomly chosen fields under a 200× inverted phase-contrast microscope (Olympus, Japan). For the invasion assay, the procedure was similar to the migration assay, but with the addition of 60 μL of Matrigel matrix (C0372, Beyotime, China) in each upper chamber.
Wound healing assay
Cells were plated in 6-well dishes and scratched vertically with a 1250 μL pipette when they reached approximately 90% confluence. Following three washes with PBS, the cells were cultured in serum-free medium. Microscopic images were captured at 0 h and 24 h post-scratch to assess cell migration.
Xenograft generation and analysis
For the animal experiments, four-week-old male athymic BALB/c nude mice were obtained from Beijing Vital River Laboratory Animal Technology. All procedures were conducted following the ARRIVE guidelines. The mice were randomly assigned to three groups, with five mice in each group and no blinding was applied during the experiments. 22Rv1 cells (YEATS2-Sh-vector, YEATS2-Sh #1, YEATS2-Sh #2) were used to establish a popliteal lymph node metastasis model by injecting 2 × 10^6^ of these cells into the footpad of each mouse. Five to six weeks later, when the footpad tumor volume in the footpad reached approximately 200 mm³, the mice were sacrificed to harvest the popliteal lymph nodes. Similarly, 22Rv1 and RFP-labeled PC-3 cells (RAD50-Sh-vector, RAD50-Sh #1, RAD50-Sh #2) were used to establish a popliteal lymph node metastasis model by injecting 2 × 10^6^ of these cells into the footpad of each mouse. The lung metastasis model was established by tail vein injection of 1 × 10^6^ RFP-labeled PC-3-Mt cells. One week after tumor cell injection, each mouse received tail vein injections of either DMSO or Mirin (HY-117693, MCE, China) every three days for a total of twelve injections. Mirin was administered at doses of 50 or 100 mg/kg. Two weeks after the final injection, lung tissues were collected from all mice according to their respective groups. Fluorescence images of the xenografts in nude mice were captured using the In-Vivo FX PRO Imaging System (Bruker Corporation, USA).
Immunofluorescence
Cells were plated on coverslips and incubated for 12 h before being collected and fixed with 4% paraformaldehyde for 15 min. To permeabilize the cells, 0.3% Triton X-100 was applied for 5 min at room temperature. Following this, the cells were blocked with 3% bovine serum albumin for 60 min at room temperature. The primary antibodies γH2AX (AP0687, Abclonal) were then added and incubated overnight at 4 °C. The following day, the cells were treated with fluorescently labeled secondary antibodies (Cy3-conjugated Affinipure Goat Anti-Rabbit IgG(H + L), SA00009-2, ProteinTech) for 1 h in the dark. After washing the cells with PBS three times for 10 min each, the nuclei were stained with DAPI for 10 min. Fluorescence images were captured using a Nikon A1Si laser scanning confocal microscope (Nikon, Japan).
RNA- sequencing and ATAC- sequencing
Based on the criteria of Fold change ≥1.5 and P < 0.05, upregulated genes in 22Rv1-YEATS2 compared to 22Rv1-Vector were obtained through RNA sequencing, with assistance from NovelBio (Shanghai, China). Based on the criteria of log_2_(FC) ≥ 1 and P < 0.05, upregulated ATAC peaks in 22Rv1-YEATS2 compared to 22Rv1-Vector were obtained through ATAC-seq, with assistance from Novogene (Beijing, China). (Details are provided in Supplementary files 7 and 8).
Protein acquisition and peptide pulldown assay
The cDNAs encoding the YEATS domains of human YEATS2 (amino acids 231–310) were tagged with GST and then inserted into the pGEX-6P-1 vector (PAIVIBIO, China). The pGEX-6P-1 vector was induced with 1 mM Isopropyl β-D-thiogalactoside (IPTG) (HY15921, MCE) for 12 h, and the GST-YEATS domain was purified following the protocol of the GST-tag Protein Purification Kit (P2262, Beyotime). Subsequently, Ultrafiltration Spin Columns (FUF510, Beyotime) were used to purify GST-tagged proteins and remove GSH to minimize interference. Histone H3 peptide and its H3K27ac-modified form were synthesized at Sangon Biotech (Shanghai, China). Two micrograms of biotinylated histone peptides were incubated overnight at 4 °C with 2 μg of GST-fused proteins in binding buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.05% NP-40, 1 mM protease inhibitor cocktail). The peptide-YEATS domain complex was then isolated using Streptavidin Magnetic Beads (P2151, Byeotime), and subsequently performed a Western blot.
Co-immunoprecipitation
Cells in the logarithmic growth phase were collected from 10 cm dishes, resuspended, and washed with PBS. They were then lysed on ice for 1 h using lysis buffer (NP-40: cocktail: PMSF = 100: 2: 1). The mixture was then centrifuged at 12,000 rpm for 5 min at 4 °C, and the supernatant was collected. To remove non-specific binding, IgG was used, and the samples were divided into three groups: InPut, IgG, and IP. Corresponding antibodies (4 μg) were added and incubated overnight at 4 °C. The bound proteins were then captured using A + G magnetic beads, washed, and heated with sample buffer before proceeding with western blot analysis.
Luciferase reporter assay
To investigate whether the transcriptional regulation of RAD50 is influenced by NR2C2, the following plasmids were co-transfected into PC3 cells: the NR2C2 overexpression plasmid, the pGL3-basic luciferase reporter plasmid containing the RAD50 promoter, and the pRL-TK Renilla luciferase reporter vector. To validate the specific binding sites of NR2C2 on the RAD50 promoter, experiments were conducted using wild-type and mutant plasmids containing the RAD50 promoter, along with the NR2C2 overexpression plasmid in PC3 cells. After 48 h of culture, the activities of firefly and Renilla luciferases were measured using the Dual-Luciferase® Reporter Assay System (RG027, Beyotime).
Chromatin immunoprecipitation (ChIP)
ChIP assays were performed using BeyoChIP™ ChIP Assay Kit (P2080S, Beyotime) according to manufacturer’s instructions. Immunoprecipitation was carried out using specific antibodies against the target protein or histone modification. Multiple primer pairs were designed to cover the regions of interest. ChIP-enriched DNA was analyzed by qPCR using these primers, and representative results are shown. Primers used in this study were synthesized by Sangon Biotech (Shanghai, China) (Supplementary file 5).
Statistical analysis
Statistical analyses were performed using Prism 10.0 (GraphPad) and SPSS 23.0 (IBM Corporation). Results are presented as mean ± SD. Survival data were analyzed through the Kaplan–Meier method, with comparisons made using the log-rank test. A two-tailed unpaired Student’s t test was utilized to determine p-values, with significance set at p < 0.05. Normality tests were conducted to determine whether parametric or nonparametric methods should be used, and an F-test was performed to compare variances between groups, with Welch’s correction applied where necessary. Each experiment included three biological replicates.
Results
YEATS2 is highly expressed in mPCa and associated with poor prognosis
Dysregulations in pathways such as apoptosis, ferroptosis, and autophagy contribute to anoikis resistance [13, 14]. However, the underlying mechanisms of anoikis resistance remain unclear. By mining sequencing data related to metastatic PCa from public databases [6, 7], analyzing our group’s anoikis-resistant PCa cell chip data, we identified 69 upregulated and 112 downregulated differentially expressed genes, which may be involved in the regulation of anoikis resistance and metastasis in PCa cells (Fig. 1A–C). Proteomic analysis of clinical samples from mPCa identified 27 upregulated and 31 downregulated differentially expressed proteins (Fig. 1D). Based on this, we identified four candidate genes that may regulate PCa metastasis: SRPRB, YEATS2, FAM126A, and OSR2 (Fig. 1E). We found that the mRNA and protein levels of SRPRB, YEATS2, and FAM126A were upregulated in metastatic tumor tissues and anoikis-resistant PCa cells, whereas OSR2 was downregulated (Figs. 1F, G and S1A). Survival analysis from the TCGA PRAD revealed that only increased YEATS2 expression was associated with poor prognosis, as indicated by reduced overall survival and progression-free survival (Fig. 1H, I). Based on these results, YEATS2 was selected as the target gene for further investigation in this study.Fig. 1YEATS2 is highly expressed in metastatic lesions of PCa and associated with poor prognosis.A The differentially expressed genes in metastatic prostate cancer (mPCa) datasets GSE6919 and GSE6752 are identified using thresholds of FC ≥ 1.5 (upregulated, red) and FC ≤ 1.5^–1^ (downregulated, blue), with P < 0.05. B The differentially expressed genes in anoikis-resistant (AnoR) PCa cells are identified using thresholds of FC ≥ 1.5 (upregulated, red) and FC ≤ 1.5^–1^ (downregulated, blue), with P < 0.05. C Venn diagram showing the overlap of differentially expressed genes in mPCa and anoikis-resistant PCa cells. D The proteomic analysis of differentially expressed proteins in mPCa specimens from our center was performed using thresholds of FC ≥ 1.5 and P < 0.05 for upregulation, and FC ≤ 1.5^–1^ and P < 0.05 for downregulation. PCa-PL: primary PCa lesions; PCa-LM: lymph node metastases of PCa. E The intersection of differentially expressed genes and proteins. F Relative mRNA levels of SRPRB, YEATS2, FAM126A, and OSR2 in mPCa tissues and anoikis-resistant PCa cells. β-actin was used as an internal control. G Protein levels of SRPRB, YEATS2, FAM126A, and OSR2 in mPCa tissues and anoikis-resistant PCa cells. β-actin was used as an internal control. H Overall survival and progression-free survival in patients from TCGA PRAD with YEATS2 low versus high expression. I Overall survival and progression-free survival in patients from TCGA PRAD with SRPRB, FAM126A, and OSR2 low versus high expression. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
Overexpression of YEATS2 promotes anoikis resistance and metastasis in PCa cells
Due to its simplicity in preparation and relatively low time cost, the popliteal lymph node metastasis model established by injecting tumor cells into the footpad of mice is widely used for studies of tumor metastasis in vivo [15, 16]. To further verify the expression of YEATS2 in metastatic tumor cells, we established a popliteal lymph node metastasis model using mice and isolated primary PCa cell lines from footpad tumors and metastatic PCa cell lines from popliteal lymph nodes (Fig. 2A–D). The protein level of YEATS2 in metastatic tumor cells (Mt) was higher than that in primary tumor cells (Pt) (Figs. 2E and S1B). Consistently, IHC staining for YEATS2 showed a similar trend, with higher expression observed in the corresponding lymph node metastases (Fig. S1C). We also found that metastatic tumor cells derived from metastases exhibited stronger anoikis resistance and metastatic potential compared to those derived from the primary tumor (Fig. 2F–H). We conducted overexpression and knockdown experiments of YEATS2 in PCa cells derived from primary and metastatic tumors (Fig. 3A). We found that in PCa cells, overexpression of YEATS2 significantly enhanced anoikis resistance and metastatic potential in vitro, while knockdown of YEATS2 had the opposite effect (Fig. 3B, C). In the popliteal lymph node metastasis mice model, we observed that knockdown of YEATS2 significantly inhibited the metastasis of PCa cells in vivo (Fig. 3D, E). Collectively, these findings suggest that YEATS2 promotes anoikis resistance and metastasis in PCa.Fig. 2. Metastatic PCa cells have stronger anoikis resistance and metastatic potential.A Schematic diagram of establishing a PCa lymph node metastasis mouse model and isolating primary/metastatic PCa cells. B Mouse popliteal lymph nodes after footpad injection of PBS solution or prostate cancer cell suspension. C Representative H&E images of popliteal lymph node metastases and footpad primary lesions. Scale bars: 500 μm and 50 μm. D Light and fluorescence microscopy images of primary and metastatic PCa cells. 200×. E Protein levels of YEATS2 in primary and metastatic PCa cells. β-actin was used as an internal control. F Cell viability of primary and metastatic PCa cells under suspension culture. G Migration and invasion abilities of primary and metastatic PCa cells. Scale bars: 100 μm. H Wound healing assay of primary and metastatic PCa cells. Scale bars: 500 μm. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.Fig. 3YEATS2 promotes anoikis resistance and metastasis in PCa cells.A Protein levels of YEATS2 in PCa cells with YEATS2 overexpression or knockdown. β-actin was used as an internal control. B Cell viability of PCa cells in the indicated groups under suspension culture. C Migration and invasion abilities of PCa cells in the indicated groups. Scale bars: 100 μm. D Images of excised popliteal lymph nodes acquired using a digital single-lens reflex camera. E Volume of metastatic foci in the popliteal lymph nodes of mice in the indicated groups. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
YEATS2 enhances anoikis resistance by promoting DNA damage repair in PCa cells
YEATS2 can form a histone acetyltransferase ATAC (Ada-Two-A-Containing) complex with other proteins, participating in the regulation of histone post-translational modifications [9]. A study indicates that the ATAC complex can act as a driver of tumor progression [8]. Lysine acetylation, a common post-translational modification of histones, regulates chromatin dynamics and accessibility in eukaryotic cells, influencing numerous cellular functions and playing a critical role in tumorigenesis and cancer progression [17, 18]. Histone acetylation increases the affinity of histone tails, leading to nucleosome relaxation and facilitating transcription, and is typically associated with transcriptional activation [19]. Given the relatively high expression of YEATS2 in anoikis-resistant and metastatic cells, we hypothesized that YEATS2 might play a role in promoting the transcription of certain genes, contributing to the development of anoikis resistance and subsequent metastasis in PCa. So, we analyzed gene expression profiles of mPCa samples from both the TCGA PRAD database and the metastatic prostate cancer cohort (SU2C/PCF Dream Team) [20], and identified genes associated with YEATS2 expression. We found that 341 and 421 genes were positively correlated with YEATS2 expression in the two datasets, while 15 and 5 genes were negatively correlated (Figs. 4A and S1D). Gene Ontology (GO) pathway enrichment analysis of the genes positively correlated with YEATS2 in both datasets revealed that these genes were enriched in pathways related to chromatin organization and remodeling, cell cycle, and DNA damage response and repair (DDR) (Figs. 4B and S1D). After detachment from the extracellular matrix, tumor cells sustain survival through metabolic reprogramming, which increases the production of reactive oxygen species (ROS), widely recognized as mediators of DNA damage, thereby inducing various forms of DNA damage [21, 22]. Additionally, through SWATH proteomics analysis of primary and bone mPCa tissues, Diego et al. identify a subset of bone mPCa cells that upregulated genes associated with DNA damage repair and cell proliferation [23]. We also found that GSEA analysis based on cell chip data revealed an enrichment of differentially expressed genes in anoikis-resistant PCa cells across multiple DNA damage repair (DDR) gene sets (Fig. 4C). γH2AX is a widely used sensitive molecular marker for DNA damage [24]. By detecting the DNA damage marker γH2AX, we found that suspension culture induced DNA damage in PCa cells, with primary tumor cells exhibiting more pronounced damage than metastatic tumor cells, while overexpression of YEATS2 significantly alleviated suspension culture-induced DNA damage (Fig. 4D). Additionally, low-dose cisplatin-induced DNA damage inhibited anoikis resistance in PCa cells, while overexpression of YEATS2 could reverse this effect (Figs. 4E, F and S1E). In summary, YEATS2 enhances anoikis resistance in PCa cells by promoting DNA damage repair.Fig. 4YEATS2 promotes DNA damage repair and induces anoikis resistance in PCa cells.A Genes correlated with YEATS2 expression in mPCa samples from the TCGA PRAD, with red indicating positively correlated genes (n = 341) and blue indicating negatively correlated genes (n = 15). B Enrichment analysis of YEATS2 positively correlated genes. C Enrichment analysis of DDR-related gene sets in differentially expressed genes from anoikis-resistant PCa cells. D Levels of γH2AX in PCa cells as indicated. γH2AX is shown in red, and nuclei were stained with DAPI. 600×. E Western blot analysis of DNA damage in PCa cells after treatment with different concentrations of cisplatin. GAPDH was used as an internal control. F Cell viability of PCa cells in the indicated groups under suspension culture. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
YEATS2 induces anoikis resistance by upregulating RAD50 expression through enhanced chromatin accessibility
YEATS2 enhances chromatin accessibility in the promoter regions of target genes, thereby increasing their transcriptional activity, which may contribute to anoikis resistance and metastasis in PCa cells. ATAC-seq (Assay for Transposase-Accessible Chromatin with high throughput sequencing) is an emerging high-throughput sequencing technique that utilizes transposase to investigate chromatin accessibility. It is currently the method of choice for studying chromatin accessibility [25]. To investigate the potential target genes regulated by YEATS2, we performed ATAC-seq on PCa cells overexpressing YEATS2. We found that overexpression of YEATS2 enhanced chromatin accessibility in 8,020 chromatin regions, encompassing 4,370 genes (Fig. 5A). We also performed RNA-seq on PCa cells overexpressing YEATS2 and identified 145 upregulated genes (Fig. 5B). By integrating these results with the DDR gene set (DNA damage response, GO:0006974), we identified the DDR-related genes BCL2, RAD50, and VAV3 (Fig. 5C). These genes may have acted as potential downstream targets of YEATS2, mediating the regulation of DNA repair and promoting anoikis resistance in PCa cells detached from the extracellular matrix (ECM). A positive correlation between RAD50 and YEATS2 mRNA expression levels was observed in mPCa samples from both the TCGA PRAD dataset (Fig. 5D) and the SU2C/PCF metastatic prostate cancer cohort (Fig. S2A). We also found that overexpression of YEATS2 in PCa cells led to upregulation of RAD50 at both the mRNA and protein levels, while no significant changes were observed in other genes (Figs. 5E, F and S2B). Moreover, data from the public database (Fig. 5G, H), our team’s chip data (Fig. 5I), and western blot analysis (Fig. S2C–E) demonstrated the relatively high expression of RAD50 in metastatic PCa cells and anoikis-resistant PCa cells. We validated the overexpression efficiency of the RAD50 plasmid in PCa cells (Fig. S2F). Rescue experiments showed that, under suspension culture conditions, overexpression of RAD50 partially reversed the DNA damage induced by YEATS2 knockdown in PCa cells (Figs. 5J and S2G). In vitro, overexpression of RAD50 reversed the inhibitory effects of YEATS2 knockdown on anoikis resistance and metastatic capability in PCa cells (Fig. S3A, B). In vivo experiments showed a similar trend, indicating that in PCa cells, RAD50 overexpression could reverse the reduction in metastatic capacity caused by YEATS2 knockdown (Fig. S3C). Our findings suggest that YEATS2 and RAD50 may regulate anoikis resistance and metastasis by enhancing DNA damage repair capacity. In addition, PARP1 and BRCA1 upregulation within DNA damage repair pathways has been reported to promote anoikis resistance and tumor metastasis [26, 27]. Considering that YEATS2, RAD50, PARP1, and BRCA1 are all involved in DNA damage repair, we further investigated the association between YEATS2/RAD50 and PARP1/BRCA1 expression. The results showed that RAD50 overexpression in PCa cells could partially reverse the downregulation of PARP1 and BRCA1 caused by YEATS2 knockdown (Fig. S3D). These findings suggest that YEATS2 promotes anoikis resistance and metastasis in PCa cells by upregulating RAD50 expression.Fig. 5YEATS2 upregulates RAD50 expression by enhancing chromatin accessibility.A ATAC-seq analysis of chromatin accessibility in PCa cells overexpressing YEATS2. Red indicates regions with increased accessibility (n = 8020), with |log_2_(FC)| ≥ 1 and P < 0.05. B RNA-seq analysis of differentially expressed genes in PCa cells overexpressing YEATS2. Red indicates upregulated genes (n = 145), with FoldChange ≥ 1.5 and P < 0.05. C Venn diagram showing overlap between genes with increased chromatin accessibility upon YEATS2 overexpression, RNA-seq upregulated genes, and DDR-related genes (GO:0006974). D Correlation between YEATS2 expression and the expression levels of BCL2, RAD50, and VAV3 in mPCa samples from the TCGA PRAD. E Relative RNA levels of BCL2, RAD50, and VAV3 in PCa cells overexpressing YEATS2. β-actin was used as an internal control. F Protein levels of BCL2, RAD50, and VAV3 in PCa cells overexpressing YEATS2. β-actin was used as an internal control. G Relative expression of RAD50 in primary and metastatic PCa samples from the GSE6752 dataset. H Relative expression of RAD50 in the indicated groups of the mPCa cohort. I Relative expression of RAD50 in parental PCa cells and anoikis-resistant (AnoR) PCa cells. J Protein levels of γH2AX in PCa cells from the indicated groups. GAPDH was used as an internal control. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
YEATS2 increases the recruitment of the transcription factor NR2C2 to the RAD50 gene promoter region
To explore the specific mechanism by which YEATS2 regulates RAD50 expression, we used the Integrative Genomics Viewer (IGV) to visualize accessible regions of the RAD50 gene in the ATAC-seq data. This analysis revealed that these accessible regions are primarily located in the RAD50 gene promoter (Fig. 6A). In addition, ChIP-qPCR analysis revealed significant enrichment of YEATS2 at the RAD50 promoter region, with markedly higher enrichment observed in metastatic cells compared with primary cells (Fig. S4A). The dual-luciferase reporter assay indicated that YEATS2 promotes RAD50 gene transcription (Fig. 6B). The gene promoter contains numerous transcription factor binding sites, and chromatin reader domains can recruit transcription factors to regulate gene transcription activity [28]. As a subunit of the ATAC complex and one of the chromatin readers, YEATS2 is known to target histones associated with gene promoters. Therefore, we utilized the STRING and BioGRID databases to predict YEATS2-interacting proteins and identify potential transcription factors of RAD50. This analysis suggested that the transcription factor NR2C2 might mediate the regulation of RAD50 expression by YEATS2 (Fig. S4B). Co-immunoprecipitation (Co-IP) confirmed that YEATS2 could interact directly or indirectly with NR2C2 (Fig. 6C). To investigate the role of NR2C2 in YEATS2-mediated transcription of RAD50, we overexpressed and knocked down NR2C2 in PCa cells (Fig. S4C, D). The dual-luciferase reporter assay confirmed the promotive effect of NR2C2 on RAD50 transcription (Fig. 6D). We found that YEATS2 and NR2C2 collaboratively promote transcription and translation of RAD50 (Figs. 6E–G and S4E). Moreover, NR2C2 expression is positively correlated with YEATS2 expression (Fig. S4F). Western blot analysis demonstrated that NR2C2 expression was elevated in PCa cells with high or ectopic YEATS2 expression (Fig. S4G, H). We identified NR2C2 binding sites on the RAD50 promoter by predicting potential sites using the JASPAR database and the Position Weight Matrix (PWM) for the NR2C2 motif (Fig. S4I, J). We validated these transcription factor binding sites (TFBS) by constructing dual-luciferase reporter vectors containing mutated sequences of the predicted binding sites (Fig. 6H). ChIP-qPCR assays and dual-luciferase reporter indicated that the NR2C2 binding site on the RAD50 promoter is located at TFBS #1 (Fig. 6I–K). To further verify the effect of YEATS2 on the recruitment of NR2C2 to the RAD50 promoter, we overexpressed YEATS2 in 22Rv1-Pt cells and knocked down YEATS2 in 22Rv1-Mt cells, respectively. ChIP-qPCR analysis showed that the enrichment of NR2C2 at the RAD50 promoter region was positively correlated with the level of YEATS2 (Fig. 6L). Our Co-IP results have shown that YEATS2 and NR2C2 are associated either directly or indirectly. To further investigate the nature of their interaction, we used AlphaFold3 to predict potential direct binding sites between YEATS2 and NR2C2. However, both the predicted pTM and ipTM scores were below 0.6, suggesting that their interaction is unlikely to be direct and may instead be mediated by proteins or nucleic acids (Fig. S4K). We performed Co-IP assays after nuclease treatment, and the interaction between YEATS2 and NR2C2 was abolished following nuclease treatment (Fig. S4L). Taken together with the ChIP-qPCR results, these findings indicate that the association between YEATS2 and NR2C2 is likely bridged by the RAD50 promoter region. Although YEATS2 and NR2C2 do not appear to directly interact, our results indicate that YEATS2 plays a crucial role in facilitating NR2C2 recruitment to the RAD50 promoter (Fig. 6L), thereby promoting the transcriptional activation of RAD50 in collaboration with NR2C2.Fig. 6YEATS2 increases the enrichment of transcription factor NR2C2 in the RAD50 gene promoter region.A Integrative Genomics Viewer (IGV) visualization of the upregulated open fragments in the RAD50 gene (P = 0.013). B Relative transcriptional activity of RAD50 in the indicated groups. C Co-immunoprecipitation (Co-IP) assay showing the interaction between YEATS2 and NR2C2. D Effect of NR2C2 overexpression on the relative transcriptional activity of RAD50. E Relative mRNA levels of RAD50 in the indicated groups. β-actin was used as an internal control. F Protein levels of RAD50 in the indicated groups. β-actin was used as an internal control. G Relative transcriptional activity of RAD50 in the indicated groups. H Construction of dual-luciferase reporter plasmids with mutations at various TFBS sites. I Relative NR2C2 enrichment in different transcription factor binding sites on RAD50 gene promoter. IgG was used as a negative control. J Agarose gel electrophoresis showing NR2C2 enrichment in different transcription factor binding sites on RAD50 gene promoter. IgG was used as a negative control. K Dual-luciferase assay to validate the binding site of NR2C2 on the RAD50 promoter. L ChIP-qPCR analysis of NR2C2 enrichment at the RAD50 promoter region upon YEATS2 overexpression in 22Rv1-Pt cells and YEATS2 knockdown in 22Rv1-Mt cells. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
YEATS2 promotes RAD50 expression by upregulating the levels of histone H3K9ac and H3K14ac
YEATS2, an ATAC-specific subunit, recognizes histone modification sites such as H3K27ac and recruits the ATAC complex to target genes rich in these modifications, where the complex’s HAT module primarily acetylates histones at H3K9ac and H3K14ac [9, 29]. Some studies have indicated that YEATS2 can recognize the modification sites H3K27cr and H3K27bz [30, 31]. The acetylation of histones H3K9 and H3K14, which are important epigenetic marks for gene transcription activation, promotes gene expression by altering chromatin structure and recruiting transcription factors [32]. The Cistrome Data Browser (Cistrome DB) database revealed enrichment of histone H3K9 and H3K14 acetylation sites in the RAD50 promoter region in PCa cells (Fig. 7A). This suggests that the transcription activation of RAD50 may be influenced by H3K9ac and H3K14ac. We treated PCa cells with the histone deacetylase inhibitor Trichostatin A to evaluate the effects of H3K9ac and H3K14ac on RAD50 transcription and observed RAD50 mRNA levels gradually increased with rising concentrations of Trichostatin A (Fig. 7B). Western blot analysis showed that Trichostatin A increased H3K9ac and H3K14ac levels in a dose-dependent manner, while also promoting RAD50 protein expression (Figs. 7C and S5A). Our previous results showed that YEATS2 is relatively upregulated in anoikis-resistant and metastatic PCa cells. To investigate the relationship between YEATS2 and the expression of H3K9ac and H3K14ac, we performed western blot analysis and found that PCa cells with higher YEATS2 expression exhibited increased levels of both H3K9ac and H3K14ac (Figs. 7D and S5B). In addition, the overexpression of YEATS2 led to an upregulation of H3K9ac and H3K14ac, whereas knockdown of YEATS2 resulted in a downregulation of these modifications (Figs. 7E, F and S5C, D). Since YEATS2 can recognize histone modification sites such as H3K27ac, H3K27cr, and H3K27bz, we speculate that YEATS2 may regulate RAD50 expression by recognizing these specific sites and consequently influencing H3K9ac and H3K14ac levels. Western blot analysis demonstrated that in PCa cells, sodium benzoate (NaBz) increased the level of histone H3K27bz modification (Fig. 7G). However, NaBz treatment at various concentrations did not significantly elevate RAD50 mRNA or protein levels in either the vector control (Fig. 7H, I) or YEATS2 overexpression group (Fig. S5E, F). Similarly, crotonic acid treatment increased the level of H3K27cr modification in PCa cells (Fig. 7J). However, ChIP-qPCR analysis indicated that H3K27cr was not markedly enriched at the RAD50 promoter, irrespective of crotonic acid treatment (Fig. S5G), and there were no significant changes in RAD50 mRNA (Figs. 7K and S5H) or protein levels (Figs. 7L and S5I). Therefore, we speculated that the regulation of RAD50 expression by YEATS2 may not be mediated through H3K27bz or H3K27cr. We hypothesized that H3K27ac may be involved in the regulation of RAD50 expression by YEATS2. We found an enrichment of H3K27ac in the promoter region of RAD50 (Fig. 7M), suggesting that RAD50 expression may be regulated by H3K27ac. Notably, although relatively high levels of H3K9ac and H3K14ac were observed in anoikis-resistant and metastatic PCa cells, H3K27ac levels were not significantly upregulated in these cells (Fig. 7N). Further ChIP-qPCR analysis showed that while H3K27ac was significantly enriched at the RAD50 promoter, its enrichment was not affected by YEATS2 expression (Fig. S5J). These results indicate that the observed differences in H3K9ac and H3K14ac levels are more likely attributable to changes in YEATS2 levels rather than to variations in H3K27ac abundance at the RAD50 promoter. We speculate that upregulated YEATS2, by recognizing the existing H3K27ac mark, may promote increased H3K9ac and H3K14ac levels, thereby influencing RAD50 transcription. A485 is a specific inhibitor of H3K27ac that can reduce its expression levels [33]. Although overexpression or knockdown of YEATS2 did not significantly affect H3K27ac enrichment at the RAD50 promoter region, A485 treatment markedly reduced H3K27ac enrichment at this site (Fig. S6A). Moreover, ChIP-qPCR analysis revealed that H3K9ac and H3K14ac were significantly enriched at the RAD50 promoter, and their enrichment was co-regulated by YEATS2 and H3K27ac (Fig. S6B, C). In summary, YEATS2 promotes RAD50 expression by enhancing H3K9ac and H3K14ac levels, a process in which H3K27ac plays a critical role. These findings motivated us to further elucidate how H3K27ac mediates the YEATS2-dependent regulation of RAD50.Fig. 7YEATS2 promotes the expression of RAD50 by upregulating the levels of histone H3K9ac and H3K14ac.A The Cistrome Data Browser (Cistrome DB) database identifies enrichment of H3K9ac and H3K14ac in the RAD50 promoter region. B Relative mRNA levels of RAD50 in PCa cells after treatment with different concentrations of Trichostatin A. C Protein levels of H3K9ac, H3K14ac, and RAD50 in PCa cells treated with different concentrations of Trichostatin A. β-actin and Histone H3 were used as internal controls. D Protein levels of H3K9ac and H3K14ac in anoikis-resistant and metastatic PCa cells. Histone H3 was used as an internal control. E Protein levels of H3K9ac and H3K14ac in PCa cells with YEATS2 overexpression. Histone H3 was used as an internal control. F Protein levels of H3K9ac and H3K14ac in PCa cells with YEATS2 knockdown. Histone H3 was used as an internal control. G Protein levels of histone benzoylation in the indicated groups. Histone H3 was used as an internal control. H Relative mRNA levels of RAD50 in PCa cells after treatment with different concentrations of NaBz. I Protein levels of RAD50 in PCa cells treated with different concentrations of NaBz. β-actin was used as internal controls. J Protein levels of histone crotonylation in the indicated groups. Histone H3 was used as an internal control. K Relative mRNA levels of RAD50 in PCa cells after treatment with different concentrations of crotonic acid. L Protein levels of RAD50 in PCa cells treated with different concentrations of crotonic acid. β-actin was used as an internal control. M The enrichment of H3K27ac in the RAD50 promoter region. N Protein levels of H3K27ac in anoikis-resistant and metastatic PCa cells. Histone H3 was used as an internal control. Statistical analysis is shown on the bar graphs. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
YEATS2 upregulates RAD50 expression by recognizing H3K27ac via its YEATS domain
The YEATS domain of YEATS2 is located between amino acids 231 and 310 (Fig. 8A). We constructed a GST-tagged YEATS domain plasmid, biotinylated histone H3 peptides, and peptides modified with H3K27ac. The peptide pulldown assay showed that the YEATS domain can bind to H3K27ac extracellularly (Fig. 8B). To explore the interaction between YEATS2 and H3K27ac in PCa cells, we constructed a plasmid overexpressing flag-tagged YEATS2 in primary PCa cell lines. We also constructed a truncated YEATS2 plasmid in which the YEATS domain was deleted (Fig. S6D). Co-immunoprecipitation (Co-IP) results showed that YEATS2 can recognize H3K27ac in PCa cells (Figs. 8C and S6E). In addition, Co-IP performed after nuclease treatment showed similar results (Fig. S6F). However, the interaction was abolished in Co-IP assays when YEATS domains were deleted (Fig. S6G). Re-ChIP-qPCR analysis using Flag and H3K27ac antibodies demonstrated that YEATS2 and H3K27ac co-occupy the RAD50 promoter region (Fig. S6H). Taken together, these results suggest that YEATS2 may bind H3K27ac via its YEATS domain and co-occupy the RAD50 promoter to regulate its transcription. Western blot showed that overexpression of YEATS2 increased RAD50 expression without significantly altering H3K27ac levels, while A485 reduced H3K27ac levels and partially reversed the YEATS2-induced increase in RAD50 expression (Figs. 8D and S7A). This suggests that H3K27ac plays a crucial role in YEATS2-mediated regulation of RAD50 expression. In PCa cells with YEATS2 knockdown, we transfected plasmids overexpressing wild-type YEATS2 or the truncated YEATS2 variant. Western blot showed that overexpression of the wild-type YEATS2 plasmid reversed the decrease in RAD50 expression caused by YEATS2 knockdown, whereas overexpression of the truncated YEATS2 plasmid failed to reverse the decrease in RAD50 expression (Figs. 8E and S7B). In summary, YEATS2 promotes RAD50 expression by recognizing H3K27ac through its YEATS domain, with both factors playing key roles in this process.Fig. 8YEATS2 regulates RAD50 expression via H3K27ac recognition, and inhibition of the DNA damage repair process suppresses PCa metastasis in vivo.A Schematic diagram of the YEATS domain structure in YEATS2. B Pull-down assay showing that the YEATS domain binds to the H3K27ac peptide. C Co-immunoprecipitation assay demonstrating the interaction between YEATS2 and H3K27ac in PCa cells. D Western blot analysis illustrating the effect of the H3K27ac-specific inhibitor A485 on YEATS2-mediated regulation of RAD50 expression. β-actin and Histone H3 were used as internal controls. E Protein levels of RAD50 in the indicated groups. β-actin was used as an internal control. F Fluorescent images of lung tissues in nude mice acquired using the In-Vivo FX PRO Imaging System (Bruker Corporation, USA). Mirin concentrations: #1, 50 mg/kg; #2, 100 mg/kg. G Schematic diagram showing YEATS2-mediated DNA damage repair and promotion of anoikis resistance after ECM detachment in PCa cells. YEATS2 is upregulated after ECM detachment in PCa cells. 1. YEATS2 recognizes H3K27ac via its YEATS domain, upregulates H3K9ac and H3K14ac, and increases chromatin accessibility at the RAD50 promoter. It also recruits NR2C2 to promote RAD50 transcription. 2. Upregulation of RAD50 enhances DNA damage repair capacity, promotes PCa cell survival under detachment conditions, thereby inducing anoikis resistance, and this process can be blocked by the MRN complex inhibitor Mirin. Image created with BioRender.com, with permission. Data represent mean ± SD from n = 3 biological replicates. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
Knockdown of RAD50 or administration of MRN complex inhibitors both reduce the metastasis of PCa cells in vivo
RAD50 forms the MRN complex with MRE11 and NBS1, playing a critical role in the detection, processing, and repair of DNA damage [34]. Mirin is a specific inhibitor of the MRN complex that suppresses its role in DNA damage repair [35]. Our study demonstrated that overexpression of RAD50 significantly enhanced the survival capacity of PCa cells in suspension culture, an effect that could be reversed by Mirin (Fig. S7C, D). Our previous result demonstrated that YEATS2 knockdown can suppress the ability of PCa cells to metastasize to the popliteal lymph nodes in mice (Fig. 3D).
Western blot analysis demonstrated that Mirin treatment induced DNA damage in PCa cells (Fig. S7E, F). To further investigate the impact of Mirin-induced DNA damage on tumor cell metastasis in vivo, we employed a lung metastasis model and performed orthogonal genetic perturbation via MRE11 knockdown as a control. The results showed that both MRE11 knockdown and treatment with the MRN complex inhibitor Mirin inhibited the metastatic ability of PCa cells in mice (Figs. 8F and S7G, H). In addition, RAD50 knockdown in PCa cells resulted in a decreased metastatic capacity of the tumor cells to lymph nodes (Fig. S8A–C). These results demonstrate that the DNA damage repair capacity of PCa cells is critical for anoikis resistance and in vivo metastasis.
Discussion
To metastasize successfully, tumor cells must overcome anoikis, a form of cell death triggered by detachment from the extracellular matrix. The acquisition of anoikis resistance allows these cells to survive in the absence of attachment to the extracellular matrix, facilitating migration to distant sites and promoting metastasis. The mechanisms underlying the development of anoikis resistance in tumor cells are complex and diverse. Studies have demonstrated that dysregulation of pathways such as apoptosis, ferroptosis, and autophagy contributes to anoikis resistance [13, 14]. Despite extensive research on the mechanisms of anoikis resistance, the specific mechanisms by which PCa cells evade anoikis remain poorly understood.
By comparing the shared molecular characteristics of metastatic prostate cancer cells and anoikis-resistant cells, our study identified the relatively high expression of YEATS2, suggesting its potential role in anoikis resistance. Numerous studies have shown that YEATS2 promotes growth, metastasis, and progression in various cancers and is frequently associated with poor prognosis [36–39]. However, the role of YEATS2 in anoikis resistance and prostate cancer metastasis has not yet been reported. YEATS2 functions as a subunit of the ATAC complex, which regulates chromatin accessibility by modifying histone acetylation through its HAT module, thereby modulating transcription [9, 29]. Dynamic changes in chromatin accessibility are intimately associated with tumorigenesis, progression, and metastasis [40–43]. Therefore, we hypothesized that YEATS2 regulates the transcriptional activity of specific genes by modulating chromatin accessibility in target gene regions, thereby promoting anoikis resistance. Using ATAC-seq, RNA-seq, and enrichment analysis of potential YEATS2-regulated target genes, combined with the characteristics of oxidative stress and metabolic reprogramming in anoikis-resistant cells, we found that the DNA damage repair pathway may mediate YEATS2-induced anoikis resistance in PCa cells.
The impact of DNA damage and repair on tumor cells is complex and multifaceted. On one hand, impaired DNA repair capacity, under certain conditions, can drive tumorigenesis, progression, and metastasis [44]. On the other hand, aberrant upregulation of DNA repair capacity is associated with resistance to cancer therapy [45–47]. For instance, the upregulation of DNA damage repair-related protein RAD23B is associated with colorectal cancer progression and metastasis [48]. Similarly, certain tumors, such as metastatic melanoma, exhibit increased DNA damage repair capacity [49]. In PCa bone metastases, the upregulation of DNA damage repair-related genes has also been observed [23]. Numerous studies have shown that inhibiting DDR-related pathways can effectively reverse treatment resistance and suppress tumor progression [50–54]. In summary, the proper functioning of various DNA damage repair mechanisms is essential for tumor cell survival and metastasis. In this study, we found that suspension culture induced DNA damage in PCa cells, whereas YEATS2 overexpression reduced DNA damage and enhanced anoikis resistance. Conversely, YEATS2 knockdown exerted the opposite effect. These findings suggest that the ability of PCa cells to resist anoikis following detachment from the extracellular matrix is associated with DNA damage status and is regulated by YEATS2.
RAD50 binds to MRE11 and NBS1 to form the MRN complex, which plays a critical role in the repair of DNA double-strand breaks (DSBs) [55, 56]. The expression of MRN complex genes has been associated with cancer prognosis, with elevated RAD50 levels strongly linked to poor survival and adverse outcomes in non-small cell lung cancer and melanoma [57]. Elevated RAD50 expression is significantly associated with aggressive progression and poor survival in patients with PCa [58]. Notably, SLC26A4-AS1 suppresses thyroid cancer metastasis both in vitro and in vivo by downregulating RAD50 and MRE11 expression. In contrast, its silencing enhances the interaction between DDX5 and E2F1, promoting their binding to MRN gene promoters, thereby activating the MRN/ATM-dependent DSB signaling pathway and driving metastasis [59]. Bioinformatics analysis revealed that RAD50 is highly expressed in anoikis-resistant prostate cancer cells and metastatic lesions. Our study demonstrated that RAD50 overexpression partially mitigates DNA damage caused by YEATS2 knockdown, while also enhancing anoikis resistance, as well as the migration and invasion potential of prostate cancer cells in vitro. Conversely, inhibition of the MRN complex by Mirin or knockdown of RAD50 suppressed metastasis of PCa cells in vivo.
Nuclear receptor subfamily 2 group C member 2 (NR2C2), also known as testicular receptor 4 (TR4), is involved in various pathophysiological processes, including glucose and lipid metabolism, DNA damage repair, and oxidative stress [60–63]. The role of NR2C2 in tumors varies depending on tumor type and stage. For instance, in prostate cancer, NR2C2 suppresses tumor initiation [64]. However, in tumor progression and metastasis, NR2C2 promotes the development of prostate cancer, seminoma, and clear cell renal cell carcinoma [65–69]. In hepatocellular carcinoma, NR2C2 inhibits tumor metastasis and invasion by downregulating EphA2 expression [70]. Our study demonstrated that YEATS2 upregulates RAD50 expression by recruiting transcription factor NR2C2 and enhancing its expression, while NR2C2 knockdown partially abrogates this effect. We also identified the transcription factor binding site (TFBS) #1 (GGGTTCAT) in the RAD50 promoter region bound by NR2C2. These findings suggest that NR2C2 plays a crucial role in YEATS2-mediated promotion of PCa cell metastasis and anoikis resistance.
Histones, as scaffolding proteins that package DNA into chromatin, undergo various post-translational modifications (PTMs) that play a crucial role in epigenetic mechanism [71]. These modifications recruit recognition factors, known as “readers,” to mediate downstream regulatory processes [72]. Recently, YEATS domains have been identified as novel readers of histone lysine acetylation and various non-acetyl acylation marks. Among the known YEATS domain-containing proteins (YCPs) in humans, the most extensively studied are ENL, AF9, YEATS2, and GAS41 [73]. YEATS family members possess subtle yet distinct features that enable flexible interactions with different acylation marks, with YEATS2 showing the strongest preference for lysine benzoylation over lysine acetylation and crotonylation, due to its wider ‘tip-sensor’ pocket [30]. Moreover, the YEATS domain in YEATS2 is a selective reader of histone crotonylation [31]. As a protein that also contains a YEATS domain, YEATS4 has been associated with poor prognosis in breast, gastric, and liver cancers [74–76]. In our study, we found that YEATS2 promotes RAD50 expression by increasing its chromatin accessibility, a process closely related to the recognition of histone post-translational modifications (PTMs) by reader proteins. We hypothesized that this regulatory mechanism is mediated by YEATS2 recognizing one or more histone modifications, specifically H3K27ac, H3K27cr, or H3K27bz. Our results showed that H3K27ac is enriched at the RAD50 promoter region, and Trichostatin A treatment enhanced RAD50 expression in PCa cells. However, treatment with NaBz or crotonic acid to upregulate H3K27bz or H3K27cr levels had no effect on RAD50 mRNA or protein expression in PCa cells, regardless of YEATS2 overexpression. ChIP-qPCR assays using an H3K27cr antibody also yielded negative results. A limitation of this aspect of the study is that the absence of commercially available specific antibodies against H3K27bz necessitates reliance on this method for indirect corroboration of H3K27bz occupancy at the RAD50 promoter. These findings suggest that YEATS2 likely promotes RAD50 expression by specifically recognizing the H3K27ac. This hypothesis was further supported by peptide pull-down assays, co-immunoprecipitation (Co-IP), and use of the H3K27ac-specific inhibitor A485. In summary, YEATS2 may act as a molecular bridge linking H3K27 acetylation to the subsequent propagation of histone acetylation marks such as H3K9ac and H3K14ac, thereby promoting the transcriptional upregulation of RAD50 and inducing anoikis resistance in PCa cells. Given that YEATS2 is a core subunit of the ATAC complex, its recognition of H3K27ac likely facilitates the recruitment of the complex’s HAT module, which contains GCN5 (KAT2A) and is responsible for catalyzing H3K9/14 acetylation [29]. This model provides a mechanistic explanation for the observed increase in H3K9/14ac levels upon YEATS2 upregulation. Future studies, such as GCN5 (KAT2A) ChIP or co-occupancy analyses at the RAD50 promoter, will be essential to validate this proposed link and to further elucidate how the functional interaction between YEATS2 and the HAT module contributes to RAD50 transcriptional activation.
In conclusion, our study identified elevated YEATS2 expression in metastatic and anoikis-resistant PCa cells. YEATS2 enhances DNA damage repair and promotes anoikis resistance by recognizing the H3K27ac modification in the RAD50 promoter region through its YEATS domain. This process increases chromatin accessibility and promotes the recruitment of transcription factor NR2C2 to the RAD50 promoter, thereby enhancing RAD50 transcription and expression (Fig. 8G). This enhanced capacity for DNA damage repair following extracellular matrix detachment ultimately drives anoikis resistance and tumor metastasis. These findings provide new insights into potential therapeutic strategies for metastatic PCa.
Supplementary information
Supplementary Figures Supplementary file 1 Supplementary file 2 Supplementary file 3 Supplementary file 4 Supplementary file 5 Supplementary file 6 Supplementary file 7 Supplementary file 8 Supplementary file 9 Uncropped Western blot gel images
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Merritt N, Garcia K, Rajendran D, Lin ZY, Zhang X, Mitchell KA, et al. TAZ-CAMTA 1 and YAP-TFE 3 alter the TAZ/YAP transcriptome by recruiting the ATAC histone acetyltransferase complex. Elife. 2021;10:e 62857.10.7554/e Life.62857 PMC 814379733913810 · doi ↗ · pubmed ↗
- 2Xie Y, Sahin M, Wakamatsu T, Inoue-Yamauchi A, Zhao W, Han S, et al. SETD 2 regulates chromatin accessibility and transcription to suppress lung tumorigenesis. JCI Insight. 2023;8:e 154120.10.1172/jci.insight.154120 PMC 997750836810256 · doi ↗ · pubmed ↗