Introduction
Female breast cancer (BRCA) is ranked as the most common type and the fifth leading cause of cancer-related deaths in 2020 [
1]. The mortality rate of BRCA has witnessed a stable decline during the past years, and a timely diagnosis is critical for a better prognosis of patients [
2]. The therapeutic strategy for BRCA largely depend on the molecular subtypes, which are generally classified into four classes: luminal A, luminal B, HER-2-positive, and triple-negative [
3]. Current standard treatments for BRCA include surgical resection, chemotherapy, radiation therapy, endocrine therapy, and targeted therapy [
4]. Radiotherapy is an important option for BRCA, especially for the highly malignant and advanced cancers [
5‐
7]. This sort of therapy uses ionizing radiation (such as X-rays and γ-rays) and has substantially improved the prognosis and survival of BRCA patients after mastectomy [
8]. However, the adaptive radioresistance that promotes metastatic and recurrent disease remains a major problem [
9].
By analyzing GEO datasets (
https://www.ncbi.nlm.nih.gov/gds/) GSE31863 and GSE101920, we obtained ubiquitin protein ligase E3C (UBE3C) as a candidate upregulated in BRCA tissues after radiotherapy. UBE3C (also known as HECTH2) is an E3 ligase, and mutations in the HECT domain of UBE3C can lead to pathophysiological states, such as neurological impairments and cancer in humans [
10‐
12]. In BRCA, UBE3C has been found to activate proliferation, migration, and invasion of cancer cells in vitro by triggering the nuclear translocation of β-catenin, a master factor associated with tumor development [
13]. However, there have been no reports on its role in radioresistance to date. E3 ubiquitin ligases primarily induce covalent binding of ubiquitin to target proteins [
14,
15]. Proteins ubiquitination generally modulates protein degradation and relocation, which represents a critical post-translational modification that exerts multifaceted functions in cancer-related pathways [
16]. In this study, our subsequent bioinformatics analyses predicted tumor protein p73 (TP73) as a potential downstream target of UBE3C, while FosB proto-oncogene (FOSB) was predicted as a potential upstream transcription factor of UBE3C. The TP73 gene encodes protein 73 (p73), a member of the p53 tumor suppressor protein family [
17]. TP73 translates into intricate number of isoforms with opposite functions: TAp73 and ΔEx2p73, ΔEx2/3p73, ΔNp73 and ΔN′p73 ([
18]. Similar to TP53, TAp73 is a crucial regulator in the cell response to a variety of stress, including DNA damage [
19]. Mutation of TP53, as a key tumor-suppressor, has been linked to increased radioresistance of tumor cells [
20,
21]. However, the exact role of TP73 in radioresponse in human cancers is not well understood, even though its overexpression has been linked to reduced radioresistance in colorectal cancer [
22]. As for FOSB, it has been documented as a fundamental factor supporting migration and invasion of tumor cells [
23]. However, its role in radioresistance remains untouched either. In our previous report, we identified that long non-coding RNA (lncRNA) LINC00963 plays significant roles in the radioresistance in BRCA [
24]. Intriguingly, we obtained from CatRapidomics (
http://service.tartaglialab.com/page/catrapid_omics_group) that LINC00963 has a predicted binding relationship with FOSB. Therefore, we conjectured that LINC00963 possibly affects UBE3C expression by interacting FOSB and therefore affect the protein stability of TP73 to influence radioresistance in BRCA. Collectively, this study aims to clarify the biological functions of UBE3C and its potential interacting molecules including FOSB and TP73 in radioresistance of BRCA cells both in vitro and in vivo.
Materials and methods
Two datasets GSE31863 and GSE101920 concerning molecular signature of genes related to radiosensitivity in BRCA were downloaded from the GEO database (
https://www.ncbi.nlm.nih.gov/gds). The Limma R Package was used to normalize and adjust the the datasets. Differentially expressed genes (DEGs) were screened by LogFC > 2.0 and adj.
p value < 0.05. Expression of the top 10 DEGs in BRCA, and the correlation between UB3E3C expression with BRCA patient’s clinical stages and prognosis were further queried in the Gene Expression Profiling Interactive Analysis (GEPIA;
http://gepia2.cancer-pku.cn/) system, a web-based tool that allows users to analyze RNA sequencing data of tumors and normal samples from The Cancer Genome Atlas (TCGA;
https://www.cancer.gov/research/areas/genomic) project. Correlation of UBE3C with the radiosensitivity in TCGA-BRCA was analyzed in the UCSCXena database (
https://xenabrowser.net/). Possible proteins that can bind with UBE3C were predicted in Ubibrowser (
http://ubibrowser.bio-it.cn/ubibrowser_v3/). The promoter sequence of UBE3C was predicted from UCSCbrowser (
http://genome.ucsc.edu/), and then the candidate transcription factors that can bind with UBE3C promoter were predicted from JASPAR (
https://jaspar.genereg.net/). The ChIP-seq data of the candidate transcription factors in the breast cancer cell line MCF-7 were downloaded from ENCODE (
https://www.encodeproject.org/), and the binding peaks of FOSB (ENCSR569XNP), ZNF302 (ENCFF037UPH), SP1 (ENCFF072FPF), ELK1 (ENCFF123KER) and FOXF2 (ENCFF008ABX) with the UBE3C promoter were analyzed. The genomic data were visualized and analyzed by the UCSC browser (
https://genome.ucsc.edu/). Additionally, RNA molecules that can bind with UBE3C and proteins that can bind with LINC00963 were predicted in CatRapidomics.
Cells
MDA-MB-23 L (CRM-HTB-26), SK-BR-3 (HTB-30) and HEK-293T (ACS-4500) cell lines were procured from ATCC. The MDA-MB-23 L and SK-BR-3 cell lines were cultured in RPMI-1640 containing 10% fetal bovine saline (FBS) and 1% penicillin/streptomycin. All cells were cultured in six-well plates containing 5 µL short hairpin (sh) RNA and 5 µL Lipofectamine 2000 (Thermo Fisher Scientific, Rockford, IL, USA). The shRNA carrier Ribo™ h-UBE3C Smart Silencer was provided by RiboBio Co., Ltd., (Guangzhou, Guangdong, China). ShRNA targeting UBE3C, TP73 and FOSB were provided by Origene, and the sequence information is given in Table
1.
UBE3C-#1 | 5'-CACCGCATTTGATCGCTGTGCTACCCGAAGGTAGCACAGCGATCAAATGC-3' |
UBE3C-#2 | 5'-CACCGGATGGATCTGAGAGACTTACCGAAGTAAGTCTCTCAGATCCATCC-3' |
TP73 | 5'-CACCAGCCAGTTGACAGAACTAAGGCGAACCTTAGTTCTGTCAACTGGC-3' |
Cells in the six-well plates were also incubated with 1 µg gene overexpression vector of UBE3C, TP73, or FOSB, or with 3 µL Polyjet (SignaGen Laboratories) for 24 h. The gene overexpression vectors were provided by GeneChem Co., Ltd. (Shanghai, China). In the present paper, gene silencing was designated as knockdown (kd) while gene overexpression as knockin (ki). Scramble shRNA was set as the control (Con) for shRNAs while Empty vector (Vec) was set for the control for gene overexpression vectors.
Reverse transcription quantitative polymerase chain reaction (RT-qPCR)
Total RNA from the cultured cells or tissues was isolated using the TRIzol reagent (Thermo Fisher Scientific). Reverse transcription of RNA to cDNA was conducted using a commercially acquired cDNA synthesis kit (Thermo Fisher Scientific). Thereafter, qPCR analysis was performed using the SYBR Green Master Mix (Thermo Fisher Scientific). Relative gene expression was evaluated using the 2
−ΔΔCt method with actin beta (ACTB) mRNA as the endogenous control. The primer information is provided in Table
2.
UBE3C | ATGAACCTGCTGAAGCTCCC | CTGACGAAGGAAGCACTGGT |
FOSB | TTTTCTCCTCCGCCTGTGTC | TCACACTCTCACACTCGCAC |
TP73 | GGGAGGGACTTCAACGAAGG | ATGGTGGTGAATTCCGTCCC |
LINC00963 | CTGTGTTACCCTGGCTGGAG | AAATGACTCAGGCTGGGCTC |
γH2AX | ACGACGAGGAGCTCAACAAG | CGGGCCCTCTTAGTACTCCT |
GAPDH | GATTTGGTCGTATTGGGCGC | TTCCCGTTCTCAGCCTTGAC |
Western blot (WB) analysis
Total protein was obtained by the RIPA lysis buffer (Beyotime Biotechnology Co., Ltd., Shanghai, China). According to the bicinchoninic acid analysis, the protein concentration was 1.5–2 µg/µL. After that, equal amounts of protein samples (50 µg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes. The membranes were blocked with 5% non-fat milk for 15 min and covered with the diluted primary antibodies including ACTB (1:2,000, ab8226, Abcam Inc., Cambridge, MA, USA), FOSB (1:1,000, MA5-15056, Thermo Fisher Scientific), TP73 (1:1000, ab202474, Abcam), UBE3C (1:1,000, PA5-110540, Thermo Fisher Scientific), and γH2AX (1:2,000, ab81299, Abcam) overnight at 4 ℃. Later, the membranes were further incubated with secondary antibody (1:5,000, GTX213110-01, GeneTex Inc.) at room temperature for 1 h. The protein bands were developed using the enhanced chemiluminescence kit (Thermo Fisher Scientific), and the protein level relative to ACTB was analyzed by Image J.
Exponentially growing parental or radioresistant SK-BR-3 and MDA-MB-231 cells were cultured in six-well plates at a density of 300 cells per well at 37 °C with 5% CO2. After two weeks, when visible cell colonies were observed, the cells were rinsed with PBS, and the cell colonies were fixed with formaldehyde for 10 min and stained with Giemsa working solution for 15 min. The cell colonies were counted under the microscope.
CellTiterGlo (CTG) assay
Cell viability after 8 Gy irradiation was assessed using the CellTiter-Glo®2.0 kit (Promega; G9242) according to the manufacturer’s protocol. Briefly, cells were seeded at 5000 cells/well in 96-well plates (Corning Glass Works, Corning, NY, USA) and incubated overnight. Afterward, the cells were exposed to 8 Gy irradiation for 2 h, and then the CTG reagent was added to the wells. The luminescence signal was read by a plate reader (BioTek Instruments Inc., Winooski, VT, USA). Relative survival was normalized to the DMSO treatment group.
Flow cytometry
Exponentially growing cells were seeded into culture flasks at 2.5 × 106 cells/mL. After 24 h, the cells were cultured in RPMI-1640 containing 2% FBS for 24 h. Thereafter, the cells were collected, stained following the instructions of AnnexinV FITC/PI kit (Beyotime), and then the apoptosis of cells in each group was detected by flow cytometry. The Annexin-V+/PI− cells were defined as early apoptotic cells, and the Annexin-V+/PI+ cells were defined as late apoptotic cells.
Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL)
TUNEL assay was performed using an in-situ cell death detection kit (Sigma-Aldrich Chemical Company, Merck KGaA, Darmstadt, German). The tissues were prepared as 4-µm thick sections. The sections were deparaffinized and rehydrated using standard protocols, followed by treatment with 50 µL Protein K working solution (1×) at 37 °C for 25 min. Later, the sections were warm-incubated with the TUNEL reaction mixture (1:9) containing TdT and dUTP at 37 °C for 2 h in a humidified environment and then counter-stained with DAPI and analyzed under a fluorescence microscope. The apoptosis index (AI) was analyzed by Image-ProPlus6.0 software, and the calculation formula was as follows: \({\text{ AI = (the number of TUNEL - positive nuclei in one area)/(the number of total nuclei in the same area) }}\)
Immunofluorescence staining and in-situ hybridization
To visualize the sub-cellular location of LINC00963 and proteins, BRCA cells were fixed, permeabilized, and prehybridized. After that, we performed the hybridization with Cy-3-conjugated LINC00963 probes in the dark at 37 °C overnight. Thereafter, the cells were rinsed at 42 °C in SSC buffer and blocked with 5% bovine serum albumin (BSA). Subsequently, the cells with reacted with the primary antibody at room temperature for 1 h and then with secondary antibody, followed by and DAPI staining and microscopy observation.
Xenograft mouse models
The animal study protocol was approved by the hospital institutional review board of Shanghai Tenth People’s Hospital, Tongji University. Male nude mice (BALB/c, 8 weeks old) were procured from SJA Laboratory Animal Co., Ltd. (Hunan, China). The parental or radioresistant BRCA cells with corresponding stable transfections (using lentivirus vectors) were injected into the nude mice subcutaneously. From day 15, the mice were exposed to 16 Gy irradiation once every three days. The length (L) and width (W) of the tumors were measured to calculate the volume (V) as follows: V (mm3) = L × W2 × 0.5. At the end point of experiments (day 35), the mice were euthanized by overdosed pentobarbital Na and the tumors were collected and weighed.
Additionally, metastasis of BRCA cells was analyzed in vivo. The luciferase-containing parental or radioresistant BRCA cells (2 × 106 cells) were injected into the nude mice via intracardiac injection. The fluorescence images were obtained by the imaging system (Ex/Em = 710/790 nm) to analyze the fluorescence intensity. After 35 d, the animals were euthanized, and the lung tissues were collected for hematoxylin and eosin staining.
Immunohistochemistry (IHC)
The tumor tissue sections were embedded and prepared as 5-µm sections. The sections were deparaffined, rehydrated, treated with H2O2 and Tris/EDTA buffer (pH 9.0), and then blocked with 5% BSA for 20 min. After that, the sections were incubated with the antibodies including KI67 (ab16667, Abcam), PCNA (ab18197, Abcam) and γH2AX (ab81299, Abcam) at 4 ℃ overnight, and then with the secondary antibody at room temperature for 2 h. After color development by DAB and nuclei counter-staining with hematoxylin, the sections were sealed for microscopy observation to evaluate the positive staining (brownish staining).
Co-immunoprecipitation (Co-IP)
For Co-IP assay, magnetic beads were pre-incubated with the antibody of P73 (1:100, ab202474, Abcam). After that, the cells were lysed in with Co-IP buffer and centrifuged. The supernatant was collected and incubated with the beads overnight at 4 ℃. Finally, the beads were collected and the protein levels in the immunoprecipitates were examined by WB analysis using anti-UB antibody (1:1000, ab140601, Abcam).
Ubiquitination examination in vitro
In brief, 1 µg GST-TP73, 500 ng GST-UBE3C, 10 µg FLAG-tagged ubiquitin, 200 ng E1 and E2 ubiquitination ligases were added to the reaction system, followed by the addition of ubiquitination buffer (25 mM HEPES, pH 7.4; 3 mM MgCl2; 10 mM NaCl; 0.05% Triton-X-100; 0.5 mM DTT, 3 mM Mg-ATP; 1% protease inhibitor cocktail) till a final concentration of 50 µL. After reaction at 37 °C for 1 h, the product was run on SDS-PAGE for WB analysis.
Chromatin immunoprecipitation (ChIP)-qPCR
The ChIP assay was performed according to the instructions of the ChIP kit (Cell Signaling Technology (CST), Beverly, MA, USA). In brief, cells were crosslinked in formaldehyde, and the reaction was terminated by glycine. Later, the cell lysate was collected and ultrasonicated for DNA truncation. The supernatant was collected and reacted with anti-FOSB (MA5-15056, Thermo Fisher Scientific) or normal rabbit IgG at 4 °C for 1 h, and then incubated with protein A/G magnetic beads overnight. Thereafter, the protein A beads was collected and washed, and the DNA was eluted and purified, and the UBE3C promoter fragment was quantified by qPCR analysis.
Luciferase reporter gene assay
The UBE3C promoter sequence was inserted into the pGL3-Basic vector (Promega Corporation, Madison, WI, USA) to construct luciferase reporter vector. The empty vector pGL3-enhancer plasmid was used as the negative control, and the pGL3-control expressing the firefly luciferase was selected as the positive control. The recombinant vectors pGL3-UBE3C-promoter and pGL3-UBE3C-promoter-Nluc-E were used as the experimental groups. The pRL-SV40 plasmid containing renal luciferase was used as an internal control. Overexpression vectors of FOSB or LINC00963 was transfected into cells to examine the alteration in luciferase activity.
Biotin-labeled RNA pull-down assay
Biotinylated LINC00963 was prepared using the MEGAscript™ T7 Transcription Kit (Invitrogen) and Pierce RNA 3′ End Desthiobiotinylation Kit (Thermo Fisher Scientific) following the manufacturer’s protocol. The RNA-protein pull-down was then performed using the Pierce Magnetic RNA-Protein Pulldown Kit (Thermo Fisher Scientific). Briefly, biotinylated RNA was captured with streptavidin-coated magnetic beads and incubated with whole-cell lysates of cells at 4 °C for 6 h. The recovered eluates were separated by SDS-PAGE for subsequent analysis.
RNA immunoprecipitation (RIP)
Cells were lysed in RIPA to collect protein sample. The protein sample was reacted with the primary antibodies at 4 ℃ overnight. The antibody-conjugated sample was reacted with protein A Sepharose (Sigma-Aldrich) at 4 ℃ for 2 h. After that, the sample was washed and incubated with proteinase K for 1.5 h. The RNA in the sample was extracted by TRIzol reagent and examined by RT-qPCR.
Statistical analysis
SPSS 21.0 (IBM Corp. Armonk, NY, USA) was used for data processing. Normal distribution of measurement data was examined by Kolmogorov-Smirnov test, and the data are presented as the mean ± standard deviation. Differences between groups were compared by the t test, or by the 1 or 2-way ANOVA when multiple groups are included. p < 0.05 was deemed to represent significant difference.
Discussion
Clinically, acquired radioresistance following irradiation therapy remains a major causative factor for tumor recurrence and poor prognosis and patients [
28], highlighting the need for more effective strategies to enhance radioresponse. In this study, through comprehensive bioinformatics analyses and functional experiments, we identified that the aberrant upregulation of UBE3C in BRCA cells following radiotherapy is a key contributor to radioresistance, mediated by its ubiquitination regulation on TP73. Furthermore, we found that the UBE3C upregulation is partly due to the interaction between LINC00963 and FOSB.
GEO datasets have been increasingly used as advanced and convenient tools for the quick screening of candidate genes related to specific biological processes including radioresistance in cancer [
29‐
31]. In this study, by analyzing GSE31863 and GSE101920 datasets and querying TCGA-BRCA database, we obtained UBE3C as an upregulated gene in BRCA tissue samples following radioresistance and its elevation was linked to poor radioresponse. Indeed, we identified increased mRNA and protein levels of UBE3C in induced radioresistant BRCA cell lines compared to the parental cell lines. Previous studies have indicated the promoting roles of UBE3C in the malignant phenotype of tumor cells such as growth, proliferation, and dissemination in BRCA and gastric cancer [
13,
32], which were reportedly attributive to the activation of the oncogenic β-catenin. In a recent work by Xu et al. UBE3C has been reported to accelerate proliferation, migration, invasion, angiogenesis, and resistance to death of clear-cell renal-cell carcinoma cells through by inducing ubiquitination of phosphatidylethanolamine binding protein 1 [
33]. The oncogenic property of UBE3C is thus nothing new, but its correlation with radioresistance remains elusive and intriguing. By inducing UBE3C knockdown in two radioresistant BRCA cell lines while UBE3C knock-in in two parental BRCA cell lines, we observed that the UBE3C was linked to proliferation, resistance to death and DNA damage, and tumorigenic activity of cells under irradiation exposure. Therefore, we confirmed that the UBE3C functions as a decisive factor leading to increased radioresistance.
Thereafter, by performing Co-IP, we identified TP73 as a highly reliable target of UBE3C among the candidates predicted from the Ubibrowser system. Unlike TP53, TP73 has no mutation reported [
18], but the full-length TP73, mainly TAp73 reportedly mimics p53 function including the enhancement of sensitivity to radiotherapy in experimental systems [
34,
35]. Indeed, it has been reported as a credible biomarker predictive of cancer regression and favorable prognosis of patients [
36]. Under normal physiological circumstances, the TP73 protein levels are in general quite low, but TAp73 accumulates and stabilizes in response to irritation [
19]. However, we found that the UBE3C upregulation following radiotherapy induced ubiquitination and degradation of the TP73 protein. The p53-like anti-proliferative and pro-apoptotic properties of TP73 make it as an important determinant of resistance to not only chemotherapy but also radiotherapy [
35,
37]. However, the direct influence of TP73 on chemoresistance in human cancers remains largely unknown. Here in this work, the fact that TP73 knockdown suppressed whereas its upregulation increased the sensitivity of BRCA cells to irradiation evidenced its supportive role in radiosensitivity.
By querying the CatRapidomics system, we obtained that LINC00963 could bind to FOSB protein while FOSB could bind to UBE3C promoter. We previously found that LINC00963 promoted radioresistance in BRCA by sequestering microRNA-324-3p and inducing ACK1 overexpression [
24]. The interaction with other RNAs represents one of the classic functional mechanisms of lncRNAs. In addition, they can bind to DNA or transcription factors directly to modulate gene expression in the transcription level [
38]. Upregulation of FOSB has been detected in highly metastatic triple-negative BRCA [
39]. Likewise, FOSB has been identified as one of the key transcription factors aberrantly expressed in BRCA patients. Although there has been no exact evidence concerning the function of FOSB in radioresistance, a previous work by Bandey et al. demonstrated that high expression of FOSB in tumor tissues was linked to elevated granulin precursor that was correlated with the expression of genes related to DNA repair [
40]. Here, we validated that the binding between LINC00963 and FOSB promoted the nuclear translocation of FOSB, which led to increased transcription of UBE3C. Indeed, artificial downregulation of FOSB was found to reduce radioresistance in two BRCA cell lines.
Despite the inspiring findings, there remains several limitations of the study. First, we did not introduce mutated segments of LINC00963, making the exact region of sequence correlated with its specific biological activity still elusive. Meanwhile, in terms of the ubiquitination assays, performing UB pulldown and mass spectrometry analysis would better present the ubiquitination enrichment. These experiments were not included in this work primarily due to the funding and time limitations. Furthermore, the LINC00963 to TP73 axis represents one of the possible mechanisms; however, according to the bioinformatics analyses, there might be more regulatory axis involved in the regulation of radioresistance in BRCA. We would focus on these issues in our future studies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.