UBE2T/STAT3 Signaling Promotes the Proliferation and Tumorigenesis in Retinoblastoma

Abstract

Purpose: The purpose of this paper was to investigate the expression and function of Ubiquitin-conjugating enzyme 2T (UBE2T), a human E2 ubiquitin-conjugating enzyme, in human retinoblastoma.

Methods: The expression of UBE2T in normal retina and retinoblastoma was analyzed using the Gene Expression Omnibus (GEO) databases, and its expression was immunohistochemically evaluated in 29 retinoblastoma sections and 5 normal retinas. Then CCK-8, flow cytometry, RNA-sequencing analysis, and in vivo assays were performed to explore the exact role of UBE2T in retinoblastoma.

Results: We found that retinoblastoma showed higher UBE2T expression than normal retina in GEO datasets and tissues. The immunoreactive score of UBE2T ≥4 was associated with group E in IIRC, T2-T4b in pTNM staging, poorly differentiated retinoblastoma, and high-risk histopathological factors. Knockdown of UBE2T reduced the cell viability, increased the apoptosis cells and G0/G1 cells, and inhibited subcutaneous tumor growth in vivo. Mechanistic studies showed that UBE2T knockdown induced down-regulation of phosphorylation of STAT3 and its downstream genes in vitro and in vivo. Rescue assays confirmed that STAT3 signaling pathway was involved in the effect of reduced cell viability, elevated apoptosis cells, and G0/G1 cells mediated by UBE2T knockdown.

Conclusions: Our data indicate that UBE2T significantly participates in the proliferation of retinoblastoma via the STAT3 signaling pathway, suggesting the potential of UBE2T as a therapeutic target for retinoblastoma treatment.

Retinoblastoma is a malignant tumor of the intraocular lens and originates from the developing retina in early childhood. Although the current treatment strategies for retinoblastoma, including local chemotherapy, intravenous chemotherapy, radiotherapy, and enucleation increase the overall survival rate to more than 95%,¹ the mortality rate is still high in developing countries due to late diagnosis.² Analysis of genetic origin of retinoblastoma led to the development of Knudson’s two-hit hypothesis and the roles of some oncogene like Rb1 and MYCN has been identified,^3–5 but current treatment strategies do not target these genes specifically. Hence, it is necessary for scientists to search for novel potential genes for early diagnosis and development of new therapeutic targets for patients with retinoblastoma. 

The Ubiquitin-conjugating enzyme 2T (UBE2T) belongs to the members of the ubiquitin E2 family and a ubiquitin conjugation (UBC) domain is contained. UBE2T significantly participates in the process of ubiquitin activation from the E1 activating enzyme to the substrate.⁶ The function of UBE2T is closely associated with the regulation of the normal physiological and pathogenetic processes. UBE2T could interact with FANCL and catalyze the FANCD2 mono-ubiquitination. Moreover, UBE2T is also significantly involved in the regulation of DNA damage.^7–9 Recently, UBE2T has been shown to play an oncogenic role in various human malignancies, including breast, gastric, hepatocellular, nasopharyngeal, prostate, and lung cancer, and high expression of this gene was associated with poor survival in these cancers.^10–18 However, up to now, no study was found to investigate the relationship between UBE2T and retinoblastoma. The role of UBE2T in retinoblastoma still remains largely unknown. 

In the present study, we uncover the diagnostic role of UBE2T in retinoblastoma and confirmed its clinical value in identifying high risk retinoblastoma. Moreover, we reported that knockdown of UBE2T inhibited cell proliferation, promoted cell apoptosis, and suppressed cell-cycle transition in retinoblastoma. Additional studies have demonstrated that the effect of UBE2T on retinoblastoma was associated with the activation of STAT3 signaling pathway. This study provides new insights into the mechanism of UBE2T in retinoblastoma, which may become potential therapeutic targets for this malignancy. 

Methods

Patients and Tissues

A total of 29 paraffin-embedded retinoblastoma and 5 normal retinal sections were collected and enrolled in this study. No pre-operative radiation or chemotherapy was given to the patients before the enucleation. The clinical features of patients with retinoblastoma included gender, age, grouping in International intraocular retinoblastoma classification (IIRC), and pTNM stage. The eye staging was determined by the IIRC system, which was a commonly used classification system proposed by Linn Murphree.¹⁹ Behavior of retinoblastoma invasion was assessed on the basis of high-risk histopathologic features (HRPFs), including invasion of choroid, anterior chamber, sclera, iris and ciliary body, and optic nerve. The Fujian Provincial Hospital and Fujian Medical University Union Hospital Ethics Committee approved this study. 

Immunohistochemical Staining

Formalin-fixed paraffin-embedded retinoblastoma samples were cut into 3-µm-thick sections. Blocks were rehydrated in advance; antigen was recovered by boiling citrate buffer or Tris-EDTA buffer. After being blocked with horse serum and goat serum for 20 minutes, slides were immunostained with primary antibodies overnight at 4°C. After 3-time washing using PBS, 30-minute incubation of the sections with the secondary peroxidase-labeled antibody was performed. The 3,3′-diaminobenzidine (DAB) was used to stain the slides, and then Gill's Hematoxylin No. 3 (Sigma, St. Louis, MO, USA) was used to counterstain. The following antibodies were selected for detection: UBE2T (1:200, ab154022, Abcam), Ki67 (1:200, sc-23900, Santa Cruz Biotechnology), PCNA (1:200, sc-56, Santa Cruz Biotechnology). Immunohistochemical images were taken with digital microscope camera (Nikon, Japan) and fluorescence microscopy (Olympus, Japan). 

Immunohistochemical Evaluation

The expression of UBE2T was determined by calculating immunoreactive score (IRS), which was a simple, reproducible scoring method based on the positive cell percentage and staining intense in five arbitrary fields.²⁰ Briefly, the staining of negative, weak, moderate, and strong were defined as the score 0, 1, 2, and 3, respectively. In addition, the staining extent of 0%, 1% to 25%, 26% to 50%, 51% to 75%, and 76% to 100% were defined as score 0, 1, 2, 3, and 4, respectively. The sum score of staining intensity and extent was defined as IRS. Thus, a score from 0 to 7 points is obtained. For statistical analysis, samples having IRS ≤3 were considered as low UBE2T expression, and IRS ≥4 points were considered as high UBE2T expression. The IRS of UBE2T was evaluated independently by two experienced pathologists under masked conditions. 

Cell Culture

Human retinoblastoma cell lines (Y79, Weri-Rb1, and HXO-Rb44) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The fetal bovine serum (FBS; 10%)- and penicillin/streptomycin sulfate (1%)-contained RPMI 1640 medium was applied for cell culture. 

Cell culture incubator was set to 37°C with 5% CO2. To activate STAT3 signaling, Colivelin (0.5 nM, Santa Cruz Biotechnology, Dallas, TX, USA), a STAT3 activator,²¹ was added in cell culture medium for 48 hours. 

Lentiviral Transduction

Endogenously downregulate UBE2T using three human siRNA sequences (siUBE2T-1: GCAACTGTGTTGACCTCTA; siUBE2T-2: CCAGTCAGCTAGTAGGCAT; and siUBE2T-3: CCTGGTTCATCTTAGTTAA), and the siRNA for negative control (si-NC) were sourced from Invitrogen. The integrity of all expression constructs was confirmed by DNA sequencing. Lentiviral particles were produced by transient transfection of pLKO.1 vector, the packaging construct (psPAX2), and the envelope plasmid (pMD2G) in HEK-293T cells. Lentiviral supernatants were collected at 48 hours post-transfection, and filtered with 0.45-µm size filter (Millipore, Billerica, MA, USA). Cells were infected with diluted lentiviral at multiplicity of infection (MOI) of 50. After 72 hours of infection, the UBE2T expression level was determined by quantitative real-time PCR (qRT-PCR) and Western blotting. 

Cell Proliferation Assay

After siUBE2T transfection, cell viability was measured. Briefly, after a 1 to 6-day culture of the cell (1 × 10³/well), the addition of 10 µL/well CCK-8 solution was performed. After 4 hours of incubation at 37°C, the cell proliferation detection was conducted using an Infinite pro 200 Tecan plate reader (TECAN, Mönnedorf, Switzerland) at 450 nm. 

Cell-Cycle Analysis

After digestion using trypsin solution, washing using PBS, and fixation using 70% ethanol, the propidium iodide (PI; Sigma-Aldrich, Allentown, PA, USA) was added to the RNase (QIAGEN, Hilden, Germany) treated cells. Cell cycle distribution was analyzed using an FACS Caliber (Becton-Dickinson, San Jose, CA, USA). 

Apoptosis Detection

After the cells were transfected as described above, the cells were harvested by trypsin in exponential phase and resuspended at a concentration of 1 × 10⁵/mL. Then, the mixture was incubated with Annexin V-FITC/PI for 20 minutes in a dark place. Flow cytometry was performed immediately to record the apoptotic status of the cells. 

Quantitative Real-Time-PCR

After RNA extraction and cDNA synthesis, the qRT-PCR assays were conducted in triplicate using 3 independent sets of cDNA denaturation at 95°C for 15 seconds, annealing at 60°C for 45 seconds, and elongation at 72°C for 1 minute on an ABI prism 7500 System. The 2^−ΔΔCT method was used to calculate relative expression values, which were normalized to β-actin as a control. Primers used are shown in Supplementary Table S1. 

Western Blot Analysis

Total protein was extracted and quantified from tissue or cells as previously described. Briefly, after extraction, the protein concentration was also determined by Protein quantification kit (Applygen Technologies, Beijing, China) via spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) at 562 nm. Protein samples was mixed with loading buffer (Transgene, Beijing, China) and heated to denature. After separation using 10% SDS-PAGE and transferring, the membrane was washed. Next, after 2-hour blocking using 5% skim milk and incubation overnight with primary antibody (Supplementary Table S2), the membrane was washed with PBS containing 0.1% Tween 20 (PBST). The membrane was then incubated with HRP-conjugated secondary antibody (Boster, Wuhan, China) for 2 hours. The blots were imaged by chemiluminescent (Proteintech, Wuhan, China) and views were taken by a gel documentation system (BioRad, Hercules, CA, USA), and band signal intensity was determined by ImageJ software (version 1.44), which was normalized to β-actin. 

RNA-Sequencing Analysis

RNA-sequencing (RNA-seq) was applied in siUBE2T-1 cell line and siNC Y79 cell line. Total RNA was isolated from these cells, as previously described. RNA-seq libraries were prepared with the Illumina TruseqTM RNA Sample Prep Kit and subjected to single-end 300-bp sequencing on Illumina Novaseq 6000 platform. Differentially expressed genes between the control group and the siUBE2T group were identified based on the analysis criteria of ≥2-fold change and adjust P value <0.05 with the limma package. Enrichment analyses, such as Gene Ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG), were performed to explore the downstream molecules and pathways of UBE2T in retinoblastoma. 

Xenograft Model

Six-week-old male nude mice, which weighed about 18 to 20 g, were used in this study. All mice were housed under strict pathogen-free conditions. All animal procedures were conducted in strict accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and all animal protocols were approved by Fujian Medical University Animal Experiments Ethics Committee. A total of 8 × 10⁵ Y79 cells stably transfected with si-UBE2T or si-NC were injected subcutaneously into the left flank of nude mice. Mice were randomized to the siUBE2T or the si-NC group (n = 6 for each group). Subcutaneous tumor length and width were examined daily. Subcutaneous tumor volume (TV) was calculated as: TV (mm³) = width × width × length/2. All the tumor nodules were extracted, photographed, and weighed at the end point of the procedures. 

Bioinformation and Statistical Analysis

Expression of UBE2T in retinoblastoma was analyzed in GSE97508, GSE110811,²² and GSE24673²³ in the GEO database. The gene set enrichment analysis (GSEA)²⁴ was conducted to compare the signaling pathways between patients with high and low UBE2T expression in retinoblastoma merged datasets (GSE29683²⁵ and GSE59983²⁶). The associations between UBE2T expression and the clinicopathological characteristics of the patients were analyzed using the Fisher’s exact test. SPSS20.0 (IBM, SPSS, Chicago, IL, USA) was used for the statistical analyses. Data were shown as means ± S.D. evaluated by Student’s t-test. Each experiment was conducted three times independently. The significant difference was presented as P value less than 0.05. 

Results

UBE2T is Upregulated in Retinoblastoma and is Correlated With High-Risk Histopathologic Features

First, the mRNA levels of UBE2T in retinoblastoma were analyzed in three different datasets in the GEO database (Figs. 1A-C). The UBE2T expression dramatically increased in retinoblastoma compared to that in the normal retinal of GSE97508 (Fig. 1D), GSE110811 (Fig. 1E), and GSE24673 (Fig. 1F). Next, the protein levels of UBE2T were detected in retinoblastoma samples by immunohistochemical (IHC). As shown in Figures 1G-J, UBE2T expression was found predominantly in the nucleus and was detected in all 29 patients with retinoblastoma, with a high UBE2T expression in 16 (55.1%) patients and a low UBE2T expression in 13 cases (44.8%), whereas no UBE2T expression was found in the normal retinal. Furthermore, the association between UBE2T and the clinicopathological features of retinoblastoma was conducted. The Table summarized the correlation of UBE2T expression with clinicopathological parameters. The results showed that the IRS of UBE2T ≥4 was positively associated with group E in IIRC (P = 0.001), T2-T4b in pTNM staging (P = 0.008), poorly differentiated retinoblastoma (P = 0.003), and high-risk histopathological factors (P = 0.001). To further validate the immunohistochemical results, the transcriptional level and protein level of UBE2T in normal ARPE-19 and retinoblastoma cells, including Weri-Rb1, Y79, and HXO-Rb44, were determined. The results revealed that UBE2T was highly expressed in all three retinoblastoma cell lines whereas lowly expressed in ARPE-19 cell line (Figs. 1K-M). Collectively, these results demonstrate that UBE2T was expressed highly in retinoblastoma, and its expression was positively correlated with high-risk histopathologic features. 

Figure 1.

 

UBE2T is transcriptionally upregulated in retinoblastoma. (A-C) Volcano plot showing the expression of UBE2T based on GSE97508, GSE110811, and GSE24673, respectively. (D-F) Expression of UBE2T in retinoblastoma and normal retina based on GSE97508, GSE110811, and GSE24673, respectively. (G) IHC of UBE2T in the normal retinal tissue. (H) Negative control stained with nonspecific immunoglobulin. (I, J) IHC of UBE2T in well-differentiated and poor-differentiated retinoblastoma respectively. (K) The mRNA expression of UBE2T in retinoblastoma cells lines and normal RPE cell lines. (L, M) protein expression of UBE2T in retinoblastoma and normal RPE cells. ** and *** represented the P values less than 0.01 and 0.001 compared to the ARPE19 group.

Table.

 

Relationship Between Clinicopathological Features and UBE2T Expression in Patients With Enucleated Retinoblastoma

 

UBE2T Knockdown Restrains the Proliferation, Induced Apoptosis, and G0/G1 Phase Arrests of Retinoblastoma Cells In Vitro

Due to the high-level expression of UBE2T in retinoblastoma, Y79 and Weri-Rb1 cell lines with stable UBE2T knockdown using three siRNAs were established. As shown in Figure 2A, the siUBE2T-1 and siUBE2T-2 can more effectively silence UBE2T expression than siUBE2T-3. Hence, siUBE2T-1 and siUBE2T-2 were selected to explore the biological functions of UBE2T in retinoblastoma. The results for proliferation detection showed that the Y79 and Weri-Rb1 cell proliferation in the si-NC group were significantly faster than those in the siUBE2T-1 and siUBE2T-2 groups, which demonstrated that silencing UBE2T significantly reduced the proliferation of retinoblastoma cells. Next, the apoptosis rates and cell cycle distribution were evaluated. The cells in the siUBE2T-1 and siUBE2T-2 groups exhibited more apoptosis than the cells in the si-NC group, which indicated that silencing UBE2T significantly increased the apoptosis of retinoblastoma cells. The cell cycle distribution showed that, compared to the cells in the si-NC group, the cells in the siUBE2T-1 and siUBE2T-2 groups exhibited significantly shorten S phase and prolonged G0/G1 phase (Fig. 2D), indicating the arrest of cell cycle in the cells after UBE2T silencing. Furthermore, as shown in Figure 2E, BAX and CYCS expressions in the cells of the siUBE2T-1 and siUBE2T-2 groups were significantly enhanced in comparison to those of the cells in the si-NC group. Collectively, these results demonstrate that silencing UBE2T restrained the proliferation, induced apoptosis, and G0/G1 phase arrests of retinoblastoma cells in vitro. 

Figure 2.

 

UBE2T knockdown inhibits retinoblastoma cells tumorigenesis in vitro. (A) Confirming the knockdown efficiency of UBE2T by Western blotting and qRT-PCR. (B) Measurement of proliferation by CCK-8 in the UBE2T silenced cells. (C) Measurement of apoptosis by flow cytometry in the UBE2T silenced cells. (D) Measurement of cell cycle distribution by flow cytometry in the UBE2T silenced cells. (E) The expression of BAX and CYCS were measured by WB in UBE2T knockdown Y79 and WERI-RB-1 cells, respectively. ** and *** represented the P values less than 0.01 and 0.001 compared to the si-NC group.

STAT3 Signaling is the Downstream of UBE2T in Retinoblastoma Cells

In order to determine the downstream effects of UBE2T-mediated retinoblastoma progression, RNA sequencing was performed to compare the si-NC group with the siUBE2T-1 group in the Y79 cell line. As shown in Supplementary Figures S1A, S1B, the volcano plot and principal component analysis (PCA) showed that the analysis could separate the distribution patterns between the si-NC group and the siUBE2T-1 group. The GO term analysis showed the biological process of differentially expressed genes (DEGs) were mainly enriched in multicellular organismal process, cellular component organization or biogenesis, etc. (see Supplementary Figure S1C). The enrichments of DEGs were found in complement and coagulation cascades, cytokine-cytokine receptor interaction, and Jak-STAT signaling pathway (Fig. 3A). Furthermore, GSEA was performed in the combined retinoblastoma datasets GSE29683 and GSE59983 in GEO database. As shown in Figure 3B, the UBE2T expression was positively correlated with the gene signatures in Jak-STAT signaling pathway. We noticed both KEGG and GSEA enriched in STAT signaling pathway. To confirm these results, Western blotting assays were performed and showed that the phosphorylation of STAT3 and the Bcl-2, Bcl-xl, c-Myc decreased, and p21 increased in the siUBE2T-1 and siUBE2T-2 groups compared with the si-NC group in Y79 and Weri-Rb1 cell lines (Figs. 3C-D). Collectively, these results demonstrate that STAT3 signaling is the downstream of UBE2T in retinoblastoma cells. 

Figure 3.

 

STAT3 signaling is the downstream of UBE2T in retinoblastoma cells. (A) JAK-STAT pathway was enriched (red arrow) in the si-NC group compared with siUBE2T-1 group in Y79 cell line by KEGG pathway analysis. (B) GSEA plot showing that UBE2T expression positively correlated with JAK-STAT gene signatures. (C, D) The expression of STAT3 and p-STAT3 were measured by Western blotting in the UBE2T silenced cells. (E, F) The Bcl-2, Bcl-xl, c-Myc, and p21 production in UBE2T knockdown Y79 and Weri-Rb-1 cells, respectively.

STAT3 Signaling Pathway is Crucial for UBE2T Mediated Biological Changes in Retinoblastoma Cells

To further delineate the crucial role of STAT3 in the UBE2T mediated changes in retinoblastoma cells, Colivelin, an STAT3 activator, was used in the si-UBE2T Y79 cell line. CCK-8 assay showed that Colivelin restored the effect of reduced cell viability in the siUBE2T group (Fig. 4C). Consistently, the results of flow cytometry showed that the apoptosis cells were decreased in the Colivelin + siUBE2T group compared with the siUBE2T group (Figs. 4A, 4D). Meanwhile, the G0/G1 cells were decreased, and the S phase cells were increased in the Colivelin + siUBE2T group compared with the siUBE2T group (Figs. 4B, 4E). Taken together, these findings suggest that STAT3 signaling pathway is crucial for UBE2T mediated biological changes in retinoblastoma cells. 

Figure 4.

 

Regulation of STAT3 signaling by UBE2T. (A, D) Flow cytometric assays were performed when cells were transfected with vehicle + si-NC, vehicle + siUBE2T, Colivelin + si-NC, or Colivelin + siUBE2T. The cell apoptosis rate of Colivelin + siUBE2T was significantly lower than that of vehicle + siUBE2T. (B, E) Flow cytometric assays of cell cycle distribution were performed when cells were transfected with vehicle + si-NC, vehicle + siUBE2T, Colivelin + si-NC, or Colivelin + siUBE2T. G0/G1 cell phase populations in Colivelin + siUBE2T was significantly lower than that in vehicle + siUBE2T. (C) CCK-8 assays were performed when cells were transfected with vehicle + si-NC, vehicle + siUBE2T, Colivelin + si-NC, or Colivelin + siUBE2T. The cell viability was showed. *** and **** represented the P values less than 0.001 and 0.0001 compared to the vehicle + siNC group, ### and #### represented the P values less than 0.001 and 0.0001 compared to the vehicle + siUBE2T group.

UBE2T Knockdown Suppressed the Tumorigenesis of Retinoblastoma Cells In Vivo

To evaluate the effect of silencing UBE2T in vivo, we used a xenograft murine model using Y79 cells. As shown in Figure 5A, obvious reduction of tumor volume was observed in the mice from the siUBE2T group compared to those from the si-NC group. The tumors in the si-NC group significantly grew faster and heavier when compared with the siUBE2T group (Fig. 5B). IHC showed that the expression of UBE2T was still lower in the siUBE2T groups than that in the si-NC groups at 33 days (Fig. 5C). Furthermore, the expression of Ki67 and PCNA, two proliferation markers, were performed by immunohistochemistry. As shown in Figure 5D, the expression of Ki67 and PCNA in the siUBE2T groups were lower than those in the si-NC group. Last, to further verify the STAT3 signaling pathway involved in the tumorigenesis of retinoblastoma cells in vivo, the expression of p-STAT3 and its downstream genes were examined by Western blotting in each transplanted tumor. As shown in Figure 5E, silencing UBE2T decreased the phosphorylation of STAT3 and its downstream genes, including Bcl-2 and c-Myc. These findings illustrate that silencing UBE2T suppressed the tumorigenesis of retinoblastoma cells via the STAT3 signaling pathway in vivo. 

Figure 5.

 

UBE2T knockdown suppressed the tumorigenesis of retinoblastoma cells in vivo. (A) Representative images of excised tumors from 6 BALB/C nude mice at 33 days after injection with si-NC and siUBE2T Y79 cells. (B) Average tumor volumes of excised tumors were measured. The tumors in the siUBE2T group were significantly grew faster and heavier when compared with the si-NC group. (C) Representative image of tumors stained with anti-UBE2T at 33 days. The expression of UBE2T was lower in the siUBE2T groups than that in the si-NC groups. (E) Representative images of tumors stained with anti-Ki76 and PCNA. The expression of Ki67 and PCNA were lower in siUBE2T groups than that in si-NC groups. (F) The expression of p-STAT3, Bcl-2, and c-myc in subcutaneous tumors in the siUBE2T groups were lower than that in the si-NC groups. ** represented the P values less than 0.01 compared to the si-NC group.

Discussion

Numerous studies have suggested that the ubiquitin proteasome (UPS) system plays critical roles in degrading most intracellular proteins and regulating cellular functions including proliferation, apoptosis, and signal transduction. Cumulative evidence highlights the involvement of UPS in the tumorigenesis.²⁷ Agents targeting components of the UPS have been developed, and some E2 ubiquitin–conjugating enzymes have been proved to be associated with oncogenesis,^28,29 which implies that targeting the E2 family may be a promising way for cancer therapy. Existing evidence demonstrated that UBE2T is an oncogene in multiple types of tumors, but the expression and biological molecular mechanisms involved in retinoblastoma remain unclear. 

In this paper, we focused on the role of UBE2T in retinoblastoma and found that tumor tissue had higher mRNA and protein levels than the normal retina. In addition, silencing UBE2T inhibits proliferation and growth, and promotes apoptosis of retinoblastoma in vitro and in vivo. Furthermore, we found that UBE2T is involved in the cell cycle transition of G1-to-S, indicating UBE2T plays an important tumorigenic role in retinoblastoma. These results were consistent with previous studies showing that the elevated expression of UBE2T contributes to increased proliferation, invasion and migration, and decreased apoptosis in various solid malignancies,¹⁸ indicating that UBE2T may become a potential therapeutic target in these cancers. 

For patients with retinoblastoma, clinical and histopathologic high-risk factors are important indicators for adjuvant chemotherapy after enucleation. A classical poor prognostic factor includes poorly differentiated tumors, massive or extrachoroidal extension, extending beyond the cut end of the optic nerve, and anterior chamber invasion.³⁰ However, some patients that progress to extraocular dissemination and metastasis are not predicted by these clinical and histopathologic features.³¹ Thus, it is necessary to search for additional markers that more accurately predict high risk retinoblastoma. Herein, we observed significantly high UBE2T expression in retinoblastoma and the closely positive association of high UBE2T expression with HRPFs, PDRB, higher grade of IIRC, and pTNM staging. These results indicated that UBE2T may become a promising marker to predict the need for adjuvant therapy for retinoblastoma. 

In further experiments, we studied the molecular mechanism of UBE2T on retinoblastoma by analyzing RNA-seq profile and GEO databases, and focus on the STAT pathway. To further delineate the crucial role of STAT3 in the UBE2T mediated changes in retinoblastoma cells, we activated the STAT3 pathway using Colivelin and found that it can restore the effect of reduced cell viability, elevated apoptosis cells, and G0/G1 cells in UBE2T knockdown Y79 cell line, indicating that UBE2T can enhance proliferation of retinoblastoma through activating STAT3 pathway. The persistent activation of p-STAT3 has been identified to contribute to the tumorigenesis, proliferation, and invasion of various types of cancer.³² STAT3 expression has been proved to be upregulated in retinoblastoma tissues, and inhibition of STAT3 suppressed proliferation of retinoblastoma in vitro and in vivo^33–36 A recent in vitro study using microarray followed with bioinformatics analyses revealed that JAK-STAT pathway related genes, including IL-6, contribute to the tumor stroma interactions of retinoblastoma.³⁷ Current evidence indicates that some signaling molecules, like Toll-like receptors, long non-coding RNAs, and microRNAs, can affect biological functions of STAT3 in cancer,³⁸ but the connection between UBE2T and the STAT3 signal pathway has not been reported. Our results suggest that UBE2T may be another initiator of this pathway. However, the direct mechanism of the UBE2T-STAT3 connection was not clearly elucidated in this study. Given that the function of UBE2T was to catalyze the ubiquitin to protein substrates, the specific protein that directly binds to UBE2T and activates the STAT3 signal pathway was worth further identifying. 

There are several limitations to note in our study. Choosing ARPE-19 cells as a control group has limitations. It is currently believed that retinoblastoma is thought to originate from human cone precursors,³⁹ but there are no human-derived cone cell lines available for study. In addition, no pharmaceutic approaches targeting UBE2T was applied in this study. The clinical significance could probably be strengthened using therapeutic agents directly targeting UBE2T, such as M435-1279, which was reported to inhibit the growth of gastric cancer with no carcinogenicity and low environmental toxicity.⁴⁰ 

In summary, we have revealed that UBE2T was highly expressed in retinoblastoma and played a crucial role in promoting proliferation and growth of retinoblastoma via activating STAT3 pathway. This study contributes important data regarding the molecular mechanism of UBE2T on the pathogenesis of retinoblastoma and provides a potential biomarker for identifying high risk retinoblastoma. 

Acknowledgments

Supported by the Fujian Provincial Natural Science Foundation (Grant Nos. 2021J01360, 2022J01242), the Fujian Provincial Health Technology Project (Grant No. 2020GGB003), and the Open Project of Tianjin Key Laboratory of Retinal Functions and Diseases (Grant No. 2021tjswmm003). 

Disclosure: N. Xu, None; Y. Cui, None; H. Shi, None; G. Guo, None; F. Sun, None; T. Jian, None; H. Rao, None