Identification of coilin in bone marrow as a potential neuroblastoma tumor progression marker transcriptionally regulated by MYCN

Abstract

Background

Current risk stratification for neuroblastoma (NB) relies on limited markers like MYCN amplification. Coilin, a key Cajal body component, regulates cellular processes. This study investigates whether coilin expression in bone marrow (BM) serves as a predictive biomarker for NB progression and elucidate its function in this disease.

Methods

The functions and molecular mechanisms of coilin were investigated by employing cell lines and animal models. Coilin mRNA levels in patient samples were measured by RT–PCR, and their relationships with clinicobiological characteristics and outcomes were analyzed.

Results

Cisplatin induced dramatic changes of coilin distribution and expression. Databases showed that high expression of coilin exerted predictive values for poor outcome in NB. Coilin promoted proliferation in vitro and in vivo. Knockdown of coilin expression inhibited cell migration and invasion, promoted apoptosis and increased the Cisplatin drug sensitivity. Moreover, coilin activates p53/p21 signaling pathway and was a direct target of MYCN. Analysis of BM samples demonstrated that high expression of coilin was obviously associated with adverse clinical biological features. Importantly, the levels of coilin at diagnosis were markedly higher than those at the time before maintenance treatment in the exact paired patients. Survival analysis presented that high coilin expression in BM is associated with poor prognosis.

Conclusions

A novel and accessible coilin-targeted liquid biopsy method was developed, capable of detecting minimal residual disease (MRD) in early-stage NB and predicting disease progression and recurrence. Coilin was transcriptionally regulated by MYCN, offering potential avenues for the development of novel drugs or intervention strategies.

Introduction

Neuroblastoma (NB) is a pediatric solid tumor which derived from the neural crest cells of sympathetic nervous system. As the most frequent cancer in infants and young children, NB accounts for approximately 15% of childhood cancer-related mortality.¹ Due to the insidious clinical manifestations and rapid progression, most patients with NB are diagnosed as high-risk patients. Even with multimodality treatment, the 5-y survival rate for high-risk patients remains below 50%, posing a significant obstacle to improve the treatment outcome of NB.²

The amplification of MYCN oncogene serves as a definitive biomarker for high-risk NB, and is a primary carcinogenic driver contributing to poor clinical prognosis.³ However, only 20%–30% of patients harbor this aberration. The majority of patients do not exhibit MYCN amplification. Additionally, NB exhibits extreme heterogeneity. Unlike leukemia, there are no recurrent fusion genes linked to NB initiation and progression. Therefore, searching for new prognostic markers related to disease progression, exploring their biological functions and regulatory mechanisms in NB is currently the focus of research.

Liquid biopsy is a minimally invasive method for obtaining tumor-derived biomarkers to monitor MRD.⁴ There are various types of body fluids that can be used for biopsy, including blood, saliva, BM and cerebrospinal fluid. Due to the fact that BM is the most common site of metastasis and recurrence in NB, it has become the preferred source of liquid biopsy for NB patients.⁵ In addition to morphological analysis, there are currently many more sensitive methods for detecting infiltrating cells in BM. For example, NB-related mRNAs can be detected by quantitative RT–PCR, with a sensitivity of up to 0.0001%.⁶

Coilin is a marker protein of Cajal bodies (CBs), which are enriched in metabolically active cells such as neurons and tumor cells,⁷ and play structural and functional roles in nuclear metabolism. CB dynamics, reflected by coilin redistribution, are highly responsive to DNA damage.⁸ Elevated coilin expression has been associated with poor prognosis in leukemia,⁹ and its downregulation causes abnormal RNA splicing – a process increasingly implicated in tumor development.¹⁰ Moreover, coilin levels influence drug sensitivity to cisplatin and daunorubicin in HeLa and Reh cells.^9,11 However, its role in NB progression remains unclear.

In this study, we demonstrated that elevated expression of coilin in the BM is associated with high malignancy and poor prognosis in NB. Functional experiments revealed that coilin is crucial for NB cell proliferation and metastasis. In vitro model confirmed the role of coilin in promoting tumor progression. Besides that, we also found coilin was the target of MYCN, and maybe a new target for NB therapy.

Materials and methods

Patients

This study included 100 newly diagnosed NB patients (median age 33.5 months, range 1–203) who received treatment at Beijing Children's Hospital from 2021 to 2022. Patients were staged using the International Neuroblastoma Risk Group Staging System (INRGSS) and stratified into risk groups according to the INRG system. Treatment followed the BCH-NB-2007 protocol.¹² Clinical data were collected retrospectively. Five BM samples from patients in long-term remission served as controls. Data collection ended on 31 May 2024.

Cell culture and drug treatments

The human NB cell line SH-SY5Y and SK-N-BE2 were obtained from the American type culture collection (ATCC, USA). All cell lines have been proven by DNA profiling within the last 3 y. All experiments were performed with mycoplasma-free cells. Cells were cultured at 37 °C with 5% CO2 in DMEM medium supplemented with 10% fetal bovine serum (Corning, USA). Research grade Cisplatin and Vincristine were purchased from MedChemExpress (MCE).

Lentivirus infection

Coilin knockdown was attained by transfection of lentiviral vectors containing coilin-targeting short hairpin RNA(shRNA) oligonucleotide. GV493 shRNA expression vector were synthesized by Shanghai GeneChem Co., Ltd, China. The targeting shRNA sequences were as follows: shcoilin-1: GCATTGGAAGAGTCGAGAGAA; shcoilin-2: CCTGGGAAATTTGATTTAGTT; shcoilin-3: GCGGGTTTCTTAAAGACAGTA; and control scramble shRNA: TTCTCCGAACGTGTCACGT. For coilin overexpression, human full-length coilin rescue cDNA was cloned into lentiviral vector CV572 with cherry tag in C-terminal (GeneChem Co.Ltd, China).

RNA isolation and quantitative real-time polymerase chain reaction (qRT–PCR)

BM samples of NB patients were obtained at diagnosis. Mononuclear cells were separated using previous method⁹ and immediately stored at −80 °C until use. The total RNAs from cell lines and specimens were extracted with Trizol reagent (Invitrogen) according to the protocols. cDNAs were synthesized using random hexamers and Moloney murine leukemia virus (MMLV) reverse transcriptase (Promega, Madison, USA) following the manufacturer's instructions. qRT‒PCR analysis for coilin expression was measured using Gene Expression Master Mix and TaqMan probe for human coilin mRNA (assay ID Hs00982300_m1). Applied Biosystems ViiA7 system (Life Technologies, USA) was used. The internal control was β-glucuronidase (GUS) gene, and the primers were Forward: 5′-GAAAATATGT GGTTGGAGAG CTCATT-3′, Reverse: 5′-CCGAGTGAAG ATCCCCTTTT TA-3′.

The PCR reactions and program were performed as previously described.² The average Ct value of bone marrow samples from five patients who had been in completely remission for more than 5 y were used as a calibrator. Relative expression of coilin gene was determined by 2−ΔΔCt method.

Western blotting analysis

Protein lysates were extracted using RIPA buffer (Thermo Fisher Scientific, USA) containing protease and phosphatase inhibitor (Thermo). BCA protein assay kit (APPLYGEN, Beijing, China) was used to measure the protein concentration. A total of 30–50 μg protein/lane was separated in 4%–20% precast gel (APPLYGEN, Beijing, China) and transferred onto 0.45 μm Polyvinylidene Fluoride (PVDF) membrane (Millipore, Germany). After blocking with 5% skimmed milk for 1 h at room temperature, the membranes were incubated with primary antibodies against coilin (sc-55594, 1:1000, Santa Cruz), P53 (cat.#2527S, 1:1000, Cell Signaling), MYCN (ab227822, 1:1000, Abcam), bcl2 (AF6285, 1:1000; Beyotime), Caspase 3 (active) (AF1150, 1:1000; Beyotime), N-Cadherin (AF5237, 1:1000; Beyotime), E-Cadherin (AF0138, 1:1000; Beyotime) and GAPDH (KC-5G5, 1:1000, Kangchen-BioTech) at 4 °C overnight. After washing three times with TBST, the blots were incubated with appropriate horseradish peroxidase conjugated secondary antibodies for 1 h at room temperature. Protein blots were visualized using a super enhanced chemiluminescence (ECL) kit (P0018FS; Beyotime).

Immunohistochemistry (IHC)

For immunohistochemical detection, paraffin-embedded sections of BM biopsy samples were hydrated and heated with citrate buffer (pH 6) for antigen retrieval. Then, the sections were incubated with diluted primary antibody of anti-coilin (sc-55594, 1:50, Santa Cruz) and anti-Ki67 (ab15580, 1:1000, Abcam) over night at 4 °C. HRP-labeled Goat anti-rabbit and anti-mouse secondary antibodies were used (GeneTech, Co., Ltd, Shanghai, China). An experienced pathologist evaluated the staining results according to the proportion and intensity of positive cells.

Cell proliferation and apoptosis assay

An MTS assay kit (G3580, Promega, USA) was used to evaluate the cell viability. NB cells were seeded in 96-well plates (3000 cells/well) and then incubated for different periods of time as needed. After that, 20 μl of MTS reagent was added into each well and incubated for 2 h at 37 °C in 5% CO2. Multiskan GO Microplate Reader (Thermo Fisher Scientific) was used to measure the optical density (OD) at 490 nm. Each experiment repeated three times.

Assessment of cell apoptosis was performed using an Annexin V-PE/7-AAD Apoptosis Detection Kit (88-8102-72, eBioscience, USA) according to the operating instruction and detected using flow cytometry (Beckman Coulter).

Colony formation assay

About 800 cells were seeded into a six-well plate with three wells per group. The cells were continuously incubated in DMEM complete culture medium and with the medium changed every 4 d. After 14 d, the colonies were fixed with 4% paraformaldehyde for 20 min and washed with PBS for three times. Then colonies were stained with 0.1% crystal violet for 10 min. After washing with PBS, the colonies in each well were photographed and analyzed.

Transwell cell migration and invasion assay

For transwell migration assays, 5 × 10⁴ NB cells in 200 μl of FBS-free medium were added to the top chambers and 600 μl of medium supplemented with 10% FBS was added to the lower chamber. After incubation for 24 h, removed the chamber and fixed them in 4% paraformaldehyde, sequentially stained with crystal violet for 20 min. Then used a cotton swab dipped in PBS to wipe off the cells on the upper layer of the chamber, without damaging the cells on the lower layer of the chamber. The migrated cells on the lower surface of the chamber were observed and photographed under a microscope (LEICA). For invasion assays, the same treatment were used in addition to applying Matrigel to the transwell inserts.

In vitro drug sensitivity assay

In vitro drug sensitivity of transfected SK-N-BE2 and SY5Y cells was determined using the cell counting MTS Assay. Cisplatin (HY-17394, MCE) was purchased from MedChemExpress(MCE). Transfected cells were plated into a 96-well plate with 8000 cells per well. Gradient concentrations of cisplatin (0.1–500 µM) was added into the well and incubated at 37 °C, 5% CO2 for 24 h. After that, MTS reagent (20 μl) was added into per well and continue to incubate for 2 h. Then measured the OD value at 490 nm as mentioned earlier. The formula used to calculate the inhibition rate was as follows: [1-(OD of cisplatin exposed well/OD of negative control well)] ×100%. Curve fitting method was used to calculate the 50% inhibitory concentration (IC50) so as to determine the cellular sensitivity of cisplatin. Each experiment should be repeated at least three times, with each treatment was triplicated.

Luciferase reporter assay

For luciferase reporter assays, we used the Dual-Luciferase Reporter Assay System (E1910, Promega). The experiment is divided into six groups with each group repeated three times. Luciferase reporter plasmids (GV238- COIL promoter, GV238- COIL promoter-mut and GV712-MYCN) were transfected into HEK-293 cells. The Renilla plasmid pRL-TK was co-transfected with the target plasmid as an internal reference. The ratio of firefly luciferase value to Renilla luciferase value in the same sample represented the relative expression level of luciferase.

Chromatin immunoprecipitation (ChIP)

The binding of MYCN to the promoter region of coilin was analyzed using ChIP-IT^® Express Kit (cat.#53008, Active Motif). ChIP assay was conducted based on the manufacturer's manual. Prepared chromatin samples were immunoprecipitated with anti-MYCN antibody (Abcam) or normal IgG (Santa Cruz) as negative control.

PCR was used to amplify the precipitated DNA and the primers used for coilin promoter were as follows: 5′-GCCTGTAGTAGCCAGGTTTCA-3′ and 5′-CGTAGCCTAACCGTCTCGGA-3′. The original non-immunoprecipitated DNA was used as an input control.

Immunofluorescence

Cells were grown on glass coverslips placed into the 12-well plates and treated with. Cisplatin and Vincristine (MCE) for 24 h, respectively. Then the cells were fixed in 4% paraformaldehyde for 20 min at room temperature and washed three times with PBS. After that, cells were permeabilized with 0.5% Triton X-100 for 10 min and blocked with 2.0% BSA for 1 h. Then anti-coilin antibody (sc-55594, 1:50, Santa Cruz) was used for incubation overnight at 4 °C, and Alexa Fluor 488-labeled Goat Anti-Mouse IgG was used as secondary antibody (A0428, 1:500, Beyotime). Next, 4′,6-diamidino-2-phenylindole (DAPI) was used to stain the nuclei. Fluorescence images were collected with ZEISS Axio ImageM2 microscope.

Neuroblastoma xenograft experiments

Female BALB/c nude mice aged six to eight weeks were acquired from Beijing Vital River Laboratory Animal Technology Co., Ltd. All mice were housed under pathogen-free environment and allocated into corresponding groups randomly and evenly. In vivo experiments were approved by the Animal Experiments and Experimental Animal Welfare Committee of Capital Medical University (2020-k-244).

To determine the impact of coilin on tumor formation ability, human SK-N-BE2 cells transfected with lentiviruses encoding control shRNA and shcoilin (1 × 10⁷, 100 μL in serum-free DMEM with 100 μL matrigel) were implanted subcutaneously. A total of 10 mice were used, with five mice in each group. The body weight of mice were monitored every other day. The formula of tumor volume was as follows: volume = 0.5 × (length × width.² In this study, the maximum tumor diameter did not exceed 1.5 cm. The mice were put to death using cervical dislocation. Tumor samples were obtained, weighed and treated with 10% formalin for subsequent immunohistochemical analysis.

Statistics

Statistical analysis and figures were generated with SPSS 16.0 software and GraphPad Prism 5 software. Each experiment was performed at least three times. Patients were divided into low and high coilin expression groups based on the ROC curve. The Fisher's exact test was performed to compare the differences between the two groups. The difference of event-free survival (EFS) between groups was determined using Kaplan–Meier method and a two-sided log-rank test. EFS was described as the date of initial diagnosis to the date meeting induction failure, relapse or death, whichever came first. Patients with continuous remission were counted at the date of last follow-up time. For all tests, results were considered significant with a P-value <  0.05.

Results

Cisplatin induced DNA damage changes the nuclear localization and expression of coilin in NB cells

Coilin has been implicated in the DNA damage response, accompanied by morphological changes in cell phenotype. However, its role in NB cells remains unclear. We treated SK-N-BE2 and SY5Y cells with different concentrations of Cisplatin and detected coilin localization through immunofluorescence analysis. The results displayed dramatic changes of coilin distribution, including signal accumulation and dispersion. Moreover, the morphological changes of coilin exhibited dose-dependent to Cisplatin concentrations (Figure 1A‐B). To verify that changes of coilin distribution were indeed related to DNA damage, we compared the response of NB cells to vincristine (VCR), which targets microtubules. In contrast, we did not observe any morphological changes of coilin in either type of cells (Figure 1C). Additionally, Cisplatin significantly reduced the level of coilin protein (Figure 1D‐E). These findings prompted us to explore the clinical relevance of coilin expression in NB patients.

Figure 1.

Nuclear localization and expression of coilin in NB cells after Cisplatin treatment. (A,B) Immunofluorescence analysis of coilin in SK-N-BE2 and SY5Y cells treated with various doses of Cisplatin for 24 h. (C) Immunofluorescence analysis of coilin in SK-N-BE2 and SY5Y cells treated with vincristine (VCR). (D) Coilin expression at protein levels in SK-N-BE2 cells treated with different doses of Cisplatin for 24 h. (E) Quantification from the western blot of coilin normalized to GAPDH. White arrows mark CBs, and red arrows denote nucleolar accumulation of coilin. The data presented are representative of at least three biologically independent experiments. ** P < 0.01, *** P < 0.001. Bar = 10 μm.

High expression of coilin exerts predictive values for poor outcome in NB through analyzing of databases

To evaluate the clinical relevance of coilin, we analyzed 498 NB patients from the GEO dataset (GSE62564). Results displayed that patients at stage IV had higher coilin expression compared to those at other stages (P < 0.001, Figure 2A). Moreover, a significant difference in coilin levels was also observed between patients in the high-risk group and those in the non-high-risk group (P < 0.001, Figure 2B). Survival analysis further revealed that, compared to low expression group, patients with high coilin expression had a shorter EFS and OS rate (15-y EFS: 65.6% ± 3.0% vs. 57.0% ± 3.4%, P = 0.047; 15-y OS: 80.9% ± 2.8% vs. 67.4% ± 4.3%, P = 0.001, Figure 2C‐D).

Figure 2.

High expression of coilin exerts predictive values for poor outcome in NB through analyzing of GEO database. (A) Coilin is highly expressed in stage IV than in other stages. (B) The coilin levels is higher in high-risk group than in non-high risk group. (C,D) EFS and OS in patients with high or low coilin expressions (P = 0.047, P = 0.001). ** P < 0.01, *** P < 0.001.

Coilin knockdown inhibits NB cell proliferation, migration and invasion

Considering the abnormal expression of coilin in NB, we evaluated its functional role in vitro. SK-N-BE2 and SY5Y cells were transfected with lentivirus shcoilin or shcon to reduce the expression of coilin (Figure 3A‐D). MTS assays showed that the proliferation ability of NB cells significantly decreased after coilin knockdown (Figure 3E‐F). Then the SK-N-BE2 cells were infected with lentivirus shcoilin and coilin-cherry lentivirus simultaneously. We found coilin overexpression markedly enhanced proliferation (Figure 3G). Colony formation assays further confirmed that low coilin expression reduced colony numbers, while overexpression rescued this effect (Figure 3H‐J). The above results indicate that coilin promote the growth of NB cells in vitro.

Figure 3.

Coilin knockdown inhibits NB cell proliferation. (A-D) SK-N-BE2 and SY5Y cells were transfected with lentivirus shcon or shcoilin, coilin expression was measured by western blot. Quantification analysis were performed based on the western blot results. (E,F) Cell proliferation was monitored by MTS assay in SK-N-BE2 and SY5Y cells. (G) SK-N-BE2 cells were infected with lentivirus shcoilin or shcoilin and coilin-Cherry lentivirus simultaneously, and cell proliferation was monitored by MTS assay. (H-J) Colony formation assays in SK-N-BE2 and SY5Y cells following coilin depletion or coilin overexpression. The number of colonies were compared. The data presented are representative of at least three biologically independent experiments. ** P < 0.01, *** P < 0.001.

Given the association of high coilin expression with multi-organ metastasis in NB, we explored whether coilin knockdown affects migration and invasion. Transwell assays showed that coilin knockdown significantly decreased the migration rate (Figure 4A–B). Moreover, using Matrigel-coated transwell chambers, we observed a significant reduction in the number of invasion cells following coilin knockdown (Figure 4E–F). Mechanistically, coilin knockdown increased the expression of E-cadherin and decreased the expression of N-cadherin, indicating its inhibition of epithelial mesenchymal transition (EMT) (Figure 4C–D,4G–H). Rescue experiments confirmed that overexpression of coilin reversed these effects, significantly promoting cell migration and invasion. (Figure 4I–L).

Figure 4.

Coilin knockdown inhibits NB cell migration and invasion. (A, B) The migration and invasion of SK-N-BE2 cells following coilin knockdown. (C, D) E-cadherin and N-cadherin expression were measured by western blot and quantification analysis were performed. (E, F) The migration and invasion of SY5Y cells following coilin knockdown. (G, H) E-cadherin and N-cadherin expression were measured by western blot and quantification analysis were performed. (I-L) The migration and invasion of SK-N-BE2 cells following coilin knockdown and coilin overexpression. E-cadherin and N-cadherin expression were measured by western blot and quantification analysis were performed. The data presented are representative of at least three biologically independent experiments. ** P < 0.01, *** P < 0.001. Bar = 20 µm.

Coilin knockdown promotes apoptosis and increased the Cisplatin drug sensitivity in NB cells

As coilin is involved in the disease progression of NB, we investigated whether abnormally expressed coilin proteins affected cell survival and response to chemotherapy drugs. Flow cytometry analysis showed that compared to the control group, coilin knockdown significantly increased cell apoptosis. This was accompanied by decreased expression of Bcl2 and increased levels of cleaved caspase 3. (Figure 5A–H). In rescue experiments, overexpression of coilin reversed the effect of knockdown, markedly reducing the number of apoptotic cells, increasing Bcl2 expression, and decreasing cleaved caspase 3 levels (Figure 5I–L).

Figure 5.

Coilin knockdown promotes apoptosis and increased the Cisplatin drug sensitivity in NB cells. (A, B) Flow cytometry analyzes for apoptosis in SK-N-BE2 cells infected with lentivirus shcon or shcoilin. Quantification of apoptotic cells were analyzed. (C, D) Expression of Bcl2 and cleaved caspase-3 were measured by western blot and quantification analysis were performed. (E, F) Flow cytometry analyzes for apoptosis in SY5Y cells infected with lentivirus shcon or shcoilin. Quantification of apoptotic cells were analyzed. (G, H) Expression of Bcl2 and cleaved caspase-3 were measured by western blot and quantification analysis were performed. (I, J) Flow cytometry analyzes for apoptosis in SK-N-BE2 cells following coilin knockdown and coilin overexpression. Quantification of apoptotic SK-N-BE 2 cells were analyzed. (K, L) Expression of Bcl2 and cleaved caspase-3 were measured by western blot and quantification analysis were performed. (M-O) Sensitivity of SK-N-BE2 and SY5Y cells to Cisplatin increased following downregulation of coilin and decreased following overexpression of coilin. The data presented are representative of at least three biologically independent experiments. ** P < 0.01, *** P < 0.001.

We further observed whether coilin levels affect sensitivity to Cisplatin. Results showed that compared to shcon group, Cisplatin IC50 values were significantly lower in shcoilin infected SK-N-BE2 (23.58 µM vs. 20.72 µM vs. 17.15 µM vs. 10.32 µM, P < 0.001) and SY5Y (42.34 µM vs. 39.70 µM vs. 31.87 µM vs.7.806 µM, P < 0.001) cells (Figure 5M–N). While the IC50 value was obviously increased when coilin was overexpressed in rescue experiment (23.99 µM vs.14.64 µM vs.24.50 µM, P < 0.001) (Figure 5O).

Coilin activates p53/p21 signaling pathway in NB cells

The p53/p21 signaling pathway is a core regulatory pathway in controlling cell cycle progression, cell differentiation, DNA repair, and apoptosis. Our results showed that the expression of p53 and p21 were significantly stimulated by coilin knockdown. (Figure 6A–D). Coilin overexpression rescued the effect of coilin knockdown on p53 and p21 (Figure 6E–F). Taken together, these findings indicate that coilin functions as an oncogene in NB, promoting cell proliferation and metastasis by activating p53/p21 signaling pathway.

Figure 6.

Coilin activates p53/p21 signaling pathway in NB cells. (A, C) Representative protein levels of coilin, p53, p21 and GAPDH in SK-N-BE2 and SY5Y cells transfected with lentivirus shcon or shcoilin. (B, D) Quantification from the western blot of target protein normalized to GAPDH. (E) Representative protein levels of coilin, p53, p21 and GAPDH in SK-N-BE2 cells transfected with lentivirus shcon, shcoilin and coilin-Cherry lentivirus. (F) Quantification from the western blot of target protein normalized to GAPDH. The data presented are representative of at least three biologically independent experiments. ** P < 0.01, *** P < 0.001.

Coilin knockdown effectively inhibits the growth of NB tumor

To further investigate the role of coilin in tumor progression in vivo, 1 × 10⁷ SK-N-BE2 cells with or without coilin knockdown were subcutaneously injected into BALB/c nude mice (Figure 7A–B). All five mice in the shcon group developed tumors, whereas four out of five mice in the shcoilin group formed tumors, with one mouse showing no detectable tumor formation. Tumor volumes and tumor weights of shcoilin group were notably lower than those in the shcon group (P = 0.012, P = 0.038, Figure 7C–E). There was no obvious difference in body weight between the two groups (Figure 7F). Immunohistochemical staining of mice tumor tissues showed a significantly lower number of Ki-67 positive cells in the shcoilin group compared to the shcon group (Figure 7G). These observations indicate that coilin promote NB progression in vivo.

Figure 7.

Coilin knockdown inhibits the growth of NB tumor. (A, B) SK-N-BE2 cells stably express shcon or shcoilin were injected into BALB/c nude mice and coilin expression in cells was assessed by western blot and PCR. (C) Representative tumor images of shcon and shcoilin group. (D, E) The tumor volumes and tumor weights are shown. (F) The body weight of mice in shcon and shcoilin group. (G) Representative images of immunohistochemical staining of Ki67 in mice tumor tissues. Bar = 100 µm.

Coilin is a direct target of MYCN

Based on the above results, coilin functioned as an oncogene in NB, promoting cell proliferation and metastasis, inhibiting apoptosis, and reducing Cisplatin sensitivity. Given that MYCN is the most prominent oncogene in NB and drives the activation of many genes related to tumor aggression. We speculated a regulatory relationship between MYCN and coilin. We first detected the protein levels of MYCN and coilin in SY5Y (single copy MYCN) and SK-N-BE2 (harbors amplified MYCN) cells. Both protein levels were obviously higher in SK-N-BE2 cells (Figure 8A–B). Analysis of patient data (GSE62564) revealed that coilin mRNA levels were much higher in MYCN amplified patients than in nonamplified patients (P = 0.001) (Figure 8C). Besides, we detected coilin protein in BM samples collected at our hospital, with metastatic tumor cells accounted for over 50%. Patients with MYCN amplification (case 2) exhibited high coilin expression, suggesting a protein-level correlation between MYCN and coilin (Figure 8D). This result was further supported by immunohistochemistry which showed strong positive expression of coilin in the BM sample of patient with MYCN amplification (Figure 8E).

Figure 8.

Coilin is a direct target of MYCN. (A, B) The protein levels of MYCN and coilin in SK-N-BE2 and SY5Y cells. (C) Coilin mRNA expression in patients with and without MYCN amplification from GEO database. (D) The protein levels of MYCN and coilin in bone marrow samples. (E) Representative images of H&E and immunohistochemical staining of Ki67 and coilin on bone marrow biopsy samples. (F-H) The protein and mRNA levels of coilin and MYCN following MYCN knockdown. (I) Schematic representation of MYCN binding sites in the human coilin promoter and the construction of MUT. (J) Luciferase reporter assay in HEK-293 cells grouped by COIL promoter-luc-NC + MYCN-NC, COIL promoter-luc-NC + MYCN, COIL promoter-luc + MYCN-NC, COIL promoter-luc + MYCN, COIL promoter-mut-luc- + MYCN-NC, COIL promoter-mut-luc- + MYCN. (K) ChIP assay in SK-N-BE2 cells. The data presented are representative of at least three biologically independent experiments. ** P < 0.01, *** P < 0.001. Bar = 20μm.

Further experiments found that knockdown of MYCN reduced both coilin mRNA and protein levels (Figure 8F–H), suggesting transcriptional regulation. We searched the promoter region of coilin and identified one potential MYCN binding site (Figure 8I). To further verify the transcriptional regulation, we constructed luciferase reporter assay. Mutation of MYCN binding site within the coilin promoter fragment significantly reduced coilin promoter activity in HEK-293 cells (Figure 8J). Additionally, ChIP assays confirmed the direct binding of MYCN to the −11/ + 1 region of the coilin promoter (Figure 8K). The above results definitely confirmed that coilin was a direct transcriptional target of MYCN in NB cells.

High coilin expression in BM is associated with high malignancy and poor prognosis in NB

To further clarify the clinical relevance of coilin expression in NB, we performed qRT‒PCR to measure coilin mRNA levels in BM samples from 100 newly diagnosed NB patients at Beijing Children's Hospital. The associations of coilin expression with clinical characteristics were detailed in Table 1. Elevated coilin expression was obviously associated with older age at diagnosis (≥18 months), adrenal primary tumor site (P = 0.007, P = 0.011), high-risk group (P = 0.002, Figure 9A), advanced stage (stage M), BM metastasis and disease events (all P < 0.001, Figure 9B–D). Notably, patients who had more than three metastatic organs exhibited markedly higher coilin levels (P < 0.001). Increased coilin expression also correlated with positive PHOX2B expression, high LDH and NSE levels (P < 0.001, P = 0.003, P = 0.001). We also quantified coilin levels in the exact paired 18 patients who were in high-risk group. Samples were obtained at the time of diagnosis and the time before maintenance treatment, respectively. Coilin expression reduced significantly after treatment compared to diagnosis in 17 cases (P = 0.007, Figure 9E).

Table 1.

Relationship of coilin expression to the clinicobiological characteristics of NB patients (n = 100).

Variables	Total n (%)	Low-coilin (%)	High-coilin (%)	P-values*
Gender	100	63	37
Male	55	35 (55.6)	20 (54.1)	1.000
Female	45	28 (44.4)	17 (45.9)
Age (months)
<18	23	20 (31.7)	3 (8.1)	0.007
≥18	77	43 (68.3)	34 (91.9)
Primary site
Abdomen	64	35 (55.6)	29 (78.4)	0.011
Thorax	33	27 (42.9)	6 (16.2)
Others	3	1 (1.6)	2 (5.4)
INRGSS stage
L1	9	8 (12.7)	1 (2.7)	<0.001
L2	37	31 (49.2)	6 (16.2)
M	50	21 (33.3)	29 (78.4)
MS	4	3 (4.8)	1 (2.7)
Risk group
Very low	9	8 (12.7)	1 (2.7)	0.002
Low	14	12 (19.0)	2 (5.4)
Intermediate	26	20 (31.8)	6 (16.2)
High	51	23 (36.5)	28 (75.7)
BM metastasis
No	53	45 (71.4)	8 (21.6)	<0.001
Yes	47	18 (28.6)	29 (78.4)
Number of organs with metastasis
<3	55	45 (71.4)	10 (27.0)	<0.001
≥3	45	18 (28.6)	27 (73.0)
Tumor size (cm)
<10	74	50 (79.4)	24 (64.9)	0.156
≥10	26	13 (20.6)	13 (35.1)
LDH (IU/L)
<500	63	47 (74.6)	16 (43.2)	0.003
≥500	37	16 (25.4)	21 (56.8)
NSE (ng/ml)
<100	44	36 (57.1)	8 (21.6)	0.001
≥100	56	27 (42.9)	29 (78.4)
MYCN gene
Nonamplified	88	55 (87.3)	33 (89.2)	1.000
Amplified	12	8 (12.7)	4 (10.8)
11q23 LOH
No loss	65	49 (77.8)	16 (43.2)	0.001
LOH	35	14 (22.2)	21 (56.8)
1p36 LOH
No loss	87	55 (87.3)	32 (86.5)	1.000
LOH	13	8 (12.7)	5 (13.5)
PHOX2B gene
Negative	43	37 (58.7)	6( 16.2)	<0.001
Positive	57	26 (41.3)	31 (83.8)
Event
No event	75	57 (90.5)	18 (48.6)	<0.001
Event	25	6 (9.5)	19 (51.4)

NB: neuroblastoma; INRGSS: International Neuroblastoma Risk Group Staging System; BM: bone marrow; LDH: lactate dehydrogenase; NSE: neuron-specific enolase; PHOX2B: paired-like homeobox 2B; LOH: loss of heterozygosity.

Figure 9.

High coilin expression in bone marrow (BM) is associated with high malignancy and poor prognosis in NB. (A) The expression of coilin in BM was higher in high-risk group than in non-high risk group. (B) Coilin was highly expressed in stage M than in other stages. (C) The expression of coilin in patients with BM metastasis was higher than that without BM metastasis. (D) The expression of coilin in patients with event occurred was obviously higher than those without event occurred. (E) Coilin expression at the time of diagnosis and the time before maintenance treatment. (F) The EFS of patients with high coilin levels was significantly lower than those with low coilin levels (P < 0.001). *** P < 0.001.

ROC analysis revealed that coilin expression served as a reliable predictor for disease progression or recurrence, with an area under the curve (AUC) of 0.843 (95% CI: 0.762–0.924, P < 0.001). According to the optimal cutoff value of coilin level (1.0697), patients were divided into low and high coilin expression groups (63 and 37 cases, respectively). Survival analysis showed that during the follow-up time, the estimated 3-y EFS in high expression group was 34.7% ± 12.5%, which was obviously lower than those in low expression group (3-y EFS: 89.3% ± 4.3%, P < 0.001, Figure 9F). In multivariate analysis for EFS, after adjusted for factors such as age, risk group, BM metastasis, LDH, and coilin levels, high coilin expression was an independent predictor of poor EFS in all patients (HR = 3.941, 95% CI = 1.553 to 10.002, P = 0.004, Table 2).

Table 2.

Univariate and multivariate Cox proportional hazards analyzes of event-free survival (EFS) in patients with NB.

	Total patients (n = 100)
	Univariate*	Multivariate#
Variable	P	HR	95%CI	P
Age (≥18 vs. <18 months)	0.003	NI	NI	NI
Risk group (very low vs. low vs. intermediate vs. high)	<0.001	13.649	1.837–91.649	0.01
BM metastasis (yes vs. no)	<0.001	NI	NI	NI
LDH (≥500 vs. <500 IU/L)	0.002	NI	NI	NI
Coilin (≥1.0697 vs. <1.0697)	<0.001	3.941	1.705–10.998	0.002

BM: bone marrow; LDH: lactate dehydrogenase.

Discussion

NB is a highly heterogeneous tumor, where recurrent genetic alterations are rare aside from MYCN amplification. A significant proportion of patients still face poor survival without clear prognostic indicators. BM detection has emerged as a promising approach to identify abnormally expressed genes at NB diagnosis. As a marker protein of CBs, coilin reflects tumor cell responses to cytotoxic stress.¹³ Our study reveals that coilin in BM serves as a prognostic marker with oncogenic functions in NB and is transcriptionally regulated by MYCN, suggesting potential targets for novel therapeutic strategies.

Beyond its role in snRNP biogenesis and mRNA processing, coilin participates in DNA damage repair. It was rapidly recruited to DNA lesions, undergoes morphological changes,¹⁴ and inhibited non-homologous end joining (NHEJ) by interacting with Ku protein.¹⁵ Additionally, coilin suppressed RNA polymerase I activity during DNA damage.¹⁶ Dysregulated DNA damage response (DDR) is a key driver of chemotherapy resistance. Although coilin's precise role in DDR remains unclear, recent evidence suggests it influences daunorubicin sensitivity in leukemia cells.⁹ In our study, Cisplatin treatment induced dramatic coilin relocalization, nucleolar accumulation, alongside significant protein levels reduction. These findings support coilin's involvement in cellular stress responses in NB and suggest its level may correlate with disease progression.

Database analysis revealed that coilin is dysregulated in NB and correlates with clinical outcomes. Functional experiments further demonstrated that coilin acts as a oncogenic driver. Coilin knockdown markedly suppressed cell proliferation, accompanied by upregulation of p53/p21, induction of apoptosis. More surprisingly, we found coilin may be linked to multi-organ metastasis in NB tumors, as its knockdown significantly impaired tumor cell migration and invasion. Complete coilin depletion is detrimental to cellular and developmental processes.¹⁷ Zebrafish embryos lacking coilin exhibited absent CBs, embryonic lethality, and defects in RNA splicing and snRNP maturation.¹⁸ In contrast, coilin knockdown showed marked tumor growth inhibition and reduced Ki-67 expression in vivo, reflecting decreased proliferative activity. We propose that coilin deficiency disrupts the structure and function of CBs, impairing snRNP maturation and trafficking, which alters splicing of key cell cycle regulators such as p27and CDK2, ultimately leading to cell cycle arrest. Furthermore, coilin reduction enhanced Cisplatin sensitivity in NB cells, suggesting its role in drug-induced nuclear stress responses. These findings highlight coilin as a critical driver of tumor progression and underscore its therapeutic potential for NB treatment.

Our study reveals that coilin expression in BM is closely linked to NB malignancy and correlates with poor clinical outcomes, particularly in patients with MYCN amplification, where coilin levels are significantly elevated. The coilin gene is located at 17q22, which is the most common breakpoints for unbalanced translocations involving 17q (17q22→17qter). 17q gain plays a crucial role in NB development. Both SK-N-BE2 and SH-SY5Y cell lines used in our study harbor 17q gain, suggesting coilin may have increased copy number. Previous studies have shown that 17q gain is highly associated with MYCN amplification, and some evidence suggests a potential functional synergy between these two alterations. IGF2BP1 (located at 17q21.32) has been demonstrated to form a positive feedback loop with MYCN, promoting 17q/2p chromosomal gains and enhancing MYCN stability. Although there is currently no direct evidence identifying coilin as a core driver gene of 17q gain, its genomic location suggests that its expression may be influenced by the overall gain of 17q. More importantly, the regulatory effect of MYCN on coilin expression that we observed may stem not only from direct transcriptional control but also, in part, from the gene dosage effect conferred by 17q gain. We propose that coilin may represent a potential node linking 17q genomic aberrations with the MYCN-driven oncogenic network. Given the challenges in directly targeting MYCN due to its non-enzymatic nature and broad regulatory functions,¹⁹ coilin emerges as a promising therapeutic target, particularly for MYCN-driven tumors.

Furthermore, BM-based coilin detection offers a minimally invasive and sensitive approach for prognostic stratification and MRD monitoring via liquid biopsy. Regular BM testing enables early relapse detection and dynamic assessment of treatment response, particularly beneficial for high risk patients. Recognized by the International Neuroblastoma Response Criteria Bone Marrow Working Group (INRGBMWG)²⁰ and supported by our preliminary data,²¹ BM analysis is a valuable clinical tool, with coilin showing strong potential as a novel biomarker. Future large-scale studies will further evaluate its dynamic role during treatment to advance diagnostic, prognostic, and therapeutic strategies in NB.

In summary, this study demonstrates that elevated coilin expression in BM is associated with high malignancy and poor prognosis in NB. Coilin serves as an oncogene which promotes cell proliferation and metastasis by activating p53/p21 signaling pathway and decreases the sensitivity to Cisplatin. Moreover, coilin is a direct transcriptional target of MYCN. However, coilin serves as a core component of CBs, and its association with tumorigenesis, along with the structural analysis of its interacting proteins, remains poorly understood. To date, no inhibitors or targeted therapies specifically targeting coilin have been developed. Nevertheless, with the expanding application of sequencing technologies and proteomics in clinical tumor samples, the functional roles of coilin and its contributions to tumor initiation and progression are poised to receive significant attention. Notably, in NB characterized by MYCN amplification- a high risk subtype with limited therapeutic options – coilin may emerge as a promising therapeutic target, offering potential avenues for the development of novel drugs or intervention strategies.

Funding Statement

This work was supported by the “Beijing Research Ward Excellence Program, BRWEP” (BRWEP2024W102090108), National Key R&D Program of China (2025YFC2708600/2025YFC2708602), and the National Natural Science Foundation of China (grant no. 82293660/82293665). (Beijing Research Ward Excellence Program,BRWEP)

Author contributions

The work reported in the paper has been performed by the authors. Zhixia Yue: Conceptualization, project administration, resources, supervision, and writing-review and editing. Lan Li: Methodology. Shuguang Liu: Methodology, software. Chao Gao: Investigation, software, validation. Sidou He: Methodology and resources. Tianlin Xue: Methodology. Wen Zhao: Resources, Methodology. Chunying Cui: Methodology and software. Chao Duan: Resources, Methodology. Yan Su: Funding acquisition, project administration, resources, supervision.

Disclosure statement

All authors agree with the presented findings, have contributed to the work, and declare no conflict of interest.

Data availability statement

The data generated in the present study are available from the corresponding author upon reasonable request.

Ethics approval statement

This study was approved by the Beijing Children's Hospital Institutional Ethics Committee (2020-k-244). All participants signed the informed consents according to the Declaration of Helsinki. Animal experiments were approved by the Animal Experiments and Experimental Animal Welfare Committee of Capital Medical University.

Consent for publication

The informed consent was obtained from study participants.

Abbreviations

NB

neuroblastoma

BM

bone marrow

MRD

minimal residual disease

CB

Cajal body

EFS

event-free survival

INRGSS

International Neuroblastoma Risk Group Staging System

LDH

lactate dehydrogenase

NSE

neuron-specific enolase

PHOX2B

paired-like homeobox 2B

LOH

loss of heterozygosity

ChIP

chromatin immunoprecipitation

PCR

polymerase chain reaction