UBE2T is upregulated, predicts poor prognosis, and promotes cell proliferation and invasion via inhibiting autophagy in ovarian cancer

Background: Aberrant upregulation and oncogenic roles of UBE2T have been revealed in several cancers. However, the expression, clinical signicance, and functions of UBE2T has not been explored in ovarian cancer (OC). Methods: In this study, the mRNA and protein expression of UBE2T in OC were detected via analyzing the online databases and immunohistochemical staining. Moreover, relations of UBE2T expression with clincopathological features and prognosis OC patients were further analyzed. Besides, the effects of UBE2T knockdown on growth, proliferation, and invasion of OC cells were investigated by CCK-8, plate clone formation, and Transwell assays. Finally, the underlying mechanism of UBE2T associated functions in OC was analyzed. Results: The results indicated that UBE2T was signicantly upregulated in OC tissues. UBE2T expression was notably correlated with clinical features such as primary T stage, TNM stage in OC patients. UBE2T, acting as an independent prognostic indicator, was inversely associated with prognosis of OC patients. UBE2T knockdown remarkably suppressed the growth, proliferation, and invasion of OC cells indicating by impaired cell viability, less cell clones and invasive cells. Mechanistically, the oncogenic roles of UBE2T was exerted by inhibiting autophagy via maintaining AKT/mTOR activity in OC. Conclusion: Collectively, our ndings conrm UBE2T upregulation predicts poor prognosis and promotes malignant progression via suppressing autophagy in OC, suggesting the promising values of UBE2T both in diagnosis and treatment of OC.


Introduction
Ovarian cancer (OC) is one of aggressive gynecologic cancers with the poor prognosis, whose ve-year overall survival rate is less than 50%. Consistent with the circumstances for most carcinomas, the favorable therapeutic and prognostic outcomes are just achievable for OC patients at early stage.
However, the actual rate of OC patients diagnosed at early stage is as low as 20%, indicating the most OC patients are diagnosed at advanced stages, which largely accounts for the unsatisfactory prognosis of OC patients [1,2]. Obviously, the unfavorable outcomes suggest the defects of the current biomarkers and treatments of OC and it is vital to explore novel diagnostic and prognostic biomarkers, and therapeutic targets to improve the e ciency of diagnosis and treatment in OC.
Dysregulations of catalytic enzymes, involving regulation of protein stability and degradation, exert critical roles in malignant transformation and progression of cancers [3,4]. Severing as the main pathway of protein degradation, ubiquitin-proteasome system (UPS) is closely related with carcinogenesis, re ected by the aberrant expression of proteins accounting for catalyzing the ubiquitination reaction [3,5]. The substrates ubiquitination are sequentially catalyzed by ubiquitin-activating enzymes (E1), ubiquitin-conjugating enzymes (E2), and ubiquitin ligases (E3), with ubiquitin. Given their capacities of determining substrate speci city in ubiquitination, the expressions and roles of E3 ligases have been widely explored in cancers and their roles in malignances are comprehensively revealed [5,6].
Interestingly, apart from E3 ligases, the aberrant expressions and critical functions of ubiquitinconjugating enzymes, being considered as a family of constitutive enzymes before, in tumorigenesis have been indicated by the recent work [7].
Ubiquitin-conjugating enzyme E2T (UBE2T) is rstly identi ed in the study of Fanconi anaemia syndrome (FA) and plays vital roles in the development of FA via catalyzing the mono-ubiquitination of FANCD2 and FANCI. Accordingly, high mutation rate of UBE2T is observed in patients with FA [8]. Recently, the roles of UBE2T have been revealed in cancers. Upregulation of UBE2T, severing as an unfavorable prognostic indicator, has been con rmed in both solid tumors, such as gastric cancer [9,10], hepatocellular carcinoma [11,12], breast cancer [13], lung cancer [13,14], nasopharyngeal carcinoma [15], osteosarcoma [16], prostate cancer [17] and renal cell carcinoma [18], and non-solid tumor like multiple myeloma (MM) [19]. Consistently, UBE2T depletion notably suppresses malignant progression via modulation the activity AKT [18] and p53 [12]. Nevertheless, the expression, clinical signi cance, functions and the corresponding mechanism of UBE2T has not been revealed in OC.
Therefore, in this study, we examined the mRNA and protein expression, prognostic value of UBE2T in OC via mining the online data and immunohistochemistry assay in local OC cohort. Moreover, we also explored the primary functions of UBE2T in OC by detecting the effects of UBE2T knockdown on cell phenotypes, including growth, proliferation, and invasion, and analyzed the role of autophagy in UBE2T related functions. The outcomes reveal that UBE2T mRNA and protein is upregulated and severs as an indicator for poor prognosis of OC patients. Furthermore, UBE2T knockdown signi cantly activated autophagy via suppressing AKT/mTOR, and subsequently inhibits growth, proliferation, and invasion of OC cells. side) <0.05 was considered statistically signi cant.

Results
UBE2T is upregulated in OC.
Firstly, we analyzed the expression of UBE2T via several web portals. As shown in Fig. 1A, compared with the normal counterparts, the mRNA expression of UBE2T was signi cantly upregulated in OC tissues re ected by two data sets in Oncomine [23]. Accordingly, through analyzing the data in GEPIA [24], which integrated the data from TCGA and GTEx databases, upregulation of UBE2T mRNA in OC was further validated (Fig. 1B). Furthermore, Mass-spectrometry-based proteomic data from UALCAN [25] indicated upregulation of UBE2T protein in OC (Fig. 1E). Finally, we analyzed the protein level of UBE2T in a cohort of OC via IHC staining. The results showed that the staining intensity of UBE2T was signi cantly stronger in OC specimens than in normal samples (Fig. 1F). Thus, these results revealed UBE2T was upregulated in OC both at mRNA and protein level.
UBE2T correlates to the clinical features of patients with OC.
Subsequently, we explored relation between UBE2T expression and clinopathological variables. According to Oncomine data, the UBE2T mRNA level was comparative among different histological types including mucinous, serous and endometrioid type of OC (Fig. 1C), which was con rmed by the results of IHC (Fig.   1F, Table 1). Interestingly, based on the data from GEPIA, UBE2T mRNA level was inversely associated with the stage of OC patients (Fig. 1D). However, the IHC data revealed that although no signi cant difference of UBE2T protein among stage I, II, III and IV, notable upregulation of UBE2T was observed in advance stage(III+IV) OC tissues than that in early stage (I+II) (Fig. 1F, Table 1). Moreover, UBE2T level positively correlated with primary T stages in OC (Table 1). However, possibly due to the limited sample volumes, UBE2T showed no signi cant relation with distant metastasis and lymph node metastasis ( Table 1). These results revealed that UBE2T was associated with stages of OC patients indicating UBE2T may involve in the progression of OC.
UBE2T inversely associated with the prognostic outcomes and served as an independent indicator for OC patients.
Next, we investigated the relation between UBE2T and prognosis in OC via online and experimental data.
The data from Kaplan-Meier Plotter portal [26] indicated UBE2T upregulation was negatively correlated with overall survival (OS), progression-free survival (PFS), and post-progression survival (PPS) of OC patients ( Fig. 2A). Besides, GEPIA data con rmed that higher UBE2T mRNA was signi cantly correlated poor OS of OC patients (Fig. 2B). Consistently, upregulation of UBE2T protein served as a poor indicator for OC patients as well (Fig. 2C). The univariate Cox proportional hazards regression analysis revealed that primary T stage, TNM stage, distant metastasis, and the level of UBE2T were signi cantly related to the OS of OC patients ( Table 2). The multivariate Cox proportional hazards regression analysis indicated that UBE2T upregulation severed as an independent predictor for poor OS for OC patients (Table 2). Thus, these outcomes con rmed UBE2T was inversely related to the prognosis and served as an independent indicator for poor prognosis for OC patients.
UBE2T depletion inhibits growth, proliferation, and invasion of OC cells.
Next, we further explored the functions of UBE2T in OC. Therefore, we checked the expression of UBE2T, depleted UBE2T expressions with si-UBE2T transfection, and subsequently detected the in uences on OC cells. As the qPCR and western blot results shown (Fig. 3A), both mRNA and protein expression of UBE2T were remarkably higher in OC cells, OVCAR3 and SKOV3 than that in IOSE80 cell, an immortalized normal ovarian epithelial cell line. As indicated by Fig. 3B, UBE2T expression was successfully knocked down in OVCAR3 and SKOV3 cells. The CCK-8, plate clone formation and Transwell invasion assays manifested UBE2T depletion signi cantly suppressed the growth, proliferation, and invasion of OC cells, demonstrated by impaired viability (Fig. 3C), less cell clones (Fig. 3D) and invasive cells (Fig. 3E). Herein, these results unveiled UBE2T depletion could inhibit growth, proliferation and invasion of OC cells.
Thus, we revealed that UBE2T knockdown suppressed malignant progression of OC cells via activating autophagy by inhibiting AKT/mTOR.

Discussion
Here, we demonstrated that the mRNA and protein level of UBE2T was notably upregulated in OC tissues and cells. Upregulation of UBE2T was inversely correlated with stages and prognosis of OC patients. UBE2T inhibition remarkably suppressed the growth, proliferation and invasion of OC cells via activating autophagy by suppressing AKT/mTOR. Collectively, our ndings suggest the promising values of UBE2T as a prognostic biomarker and therapeutic target in OC.
The roles of UBE2T have been primarily revealed in FA related ubiquitin signaling [8]. UBE2T is critical for maintenance of genome integrity in FA involving its regulatory roles in mono-ubiquitination of FANCD2 and FANCI. Notably, FA patients bearing related gene mutations are particularly prone to malignant outcomes, suggesting the vital roles of FA related genes in tumorigenesis [8]. Indeed, aberrant upregulation of UBE2T has been observed in a host of cancers. For example, UBE2T is remarkably upregulated in MM cells, especially in the early stage. UBE2T level is signi cantly associated with IgG serotype of MM and UBE2T upregulation predicts poor prognosis, including OS and event-free survival time, of patients with MM [29]. Furthermore, UBE2T is upregulated in gastric cancer tissues and cells, positively associated with poor differentiation, advanced T stage, and short OS time in patients with gastric cancer [9,10]. Accordingly, UBE2T upregulation, severing as a poor prognostic biomarker, has been con rmed in nasopharyngeal carcinoma [20], osteosarcoma [16], lung cancer [13,14], breast cancer [16], prostate and hepatocellular cancer [11,12,17,26]. Consistently, we demonstrated that UBE2T mRNA and protein was upregulated in OC tissues and cells and UBE2T upregulation was associated with stages and poor prognosis OC patients.
The functions of UBE2T have been explored in cancers. UBE2T knockdown exerts signi cantly inhibitory effects on cell characteristics via different mechanisms. By inhibiting AKT and related pathways, UBE2T depletion remarkably suppresses cell proliferation, migration, and invasion of osteosarcoma [30], nasopharyngeal carcinoma [15], liver cancer [12], and renal cell carcinoma [18]. Besides, the suppressive roles of UBE2T knockdown on bladder cancer and hepatocellular carcinoma are mainly caused by apoptosis [31] and cell cycle arrest [11]. Furthermore, the oncogenic roles of UBE2T closely depend on their enzymatic activity. UBE2T can inactivate p53 and BRAC1, promote proliferation of hepatocellular carcinoma and breast cancer cells via directly catalyzing their mono-ubiquitination [11,32]. Consistently, UBE2T knockdown inhibited AKT/mTOR and exerted notably inhibitory effects on growth, proliferation, and invasion of OC cells.
Ubiquitin conjugating enzymes can exert essential regulatory roles in physiopathologic processes, including cancers, via modulating autophagy [8,33,34]. UBE2L6 depletion can activate autophagy and regulate the chemosensitivity of esophageal cancer cells [33]. Serving as either oncogenic or anti-tumor regulator, autophagy is widely implicated in regulation of stem maintenance, proliferation, invasion, metastasis, and therapy resistance of OC [27,28,35]. Here, we validated that UBE2T inhibition could promote autophagy, subsequently suppress the malignant characteristics of OC cells via constraining AKT/mTOR.

Conclusion
In conclusion, we initially and comprehensively explored the expression, clinical signi cance, fundamental functions, and underlying mechanisms of UBE2T in OC. The outcomes reveal that UBE2T is notably increased in OC, which severs as an indicator for poor prognosis in OC patients. UBE2T knockdown suppresses AKT/mTOR activity, subsequently activates autophagy, and eventually inhibits growth, proliferation, and invasion of OC cells. These ndings contribute a better understanding of development and progression of OC and present UBE2T as a promising biomarker and therapeutic target in diagnosis and treatment of OC, respectively.

Tissue samples, Ethical Statement and Immunohistochemistry (IHC)
An OC tissue chip, containing 70 OC tissues and 10 normal tissues (#OVC1504) was purchased from the shanghai Superbiotek Pharmaceutical Technology Inc. (Shanghai, China). The study was approved by the Academic Ethics Committee of the Second Xiangya Hospital of Central South University and performed under the instructions of Declaration of Helsinki. IHC and scoring were applied according to our previous description [20] with primary antibody UBE2T (dilution: 1:50, BBI, Shanghai, China).

Cell Culture
Normal ovarian epithelial cell line IOSE80, and OC cell lines, SKOV3 and OVCAR3, were purchased from the American Type Culture Collection (ATCC, VA, USA). SKOV3 was maintained in RPMI-1640 medium plus 10% fetal bovine serum (FBS, Thermo, MA, USA). OVCAR3 and IOSE80 were cultured in DMEM medium plus 10% FBS (Thermo, MA, USA). The cells were culture in a humidi ed incubator at 37℃ and 5% CO 2 . The chloroquine (CQ, Sigma-Aldrich, MO, USA) and MK2206 (Selleck, TX, USA) were added into the culture medium for functional experiments as indicated in gure legends.

Cell growth assay
The cell growth assay was performed as previously described [21]. The effects of UBE2T rescue and MK2206 addition on cell growth were re ected by inhibitory growth rate with si-NC group as control. The experiments were independently performed in triplicate.

Plate clone formation assay
Plate clone formation assay was carried out as previously described [22]. The experiments were independently performed for three times.

Transwell invasion assay
Transwell invasion assay was performed as previous description [22]. The experiments were independently performed for three times.

Western blot
The western blot was performed as previous description [22].

Statistical analysis
Statistical analyses and statistical charts were analyzed and produced using SPSS20.0 software and GraphPad Prism version 8. For comparisons between two groups, a Student t-test or chi-square test was carried out. Survival curves were obtained via Kaplan-Meier method, and the statistical analysis was evaluated by Log-rank test. The univariate and multivariate Cox regression was performed to analyze the relationship of among UBE2T expression, clinicopathological parameters, and survival in OC patients. For all analyses, P (two side) <0.05 was considered statistically signi cant.

Declarations
Ethics approval and consent to participate The study was approved by the Academic Ethics Committee of the Second Xiangya Hospital of Central South University (OV_XXM_20191128) and performed under the instructions of Declaration of Helsinki.
The informed consent were obtained from the all enrolled patients.

Consent for publication
Not applicable.

Availability of data and materials
The online databases, including Oncomine(www.oncomine.org), GEPIA(gepia.cancer-pku.cn), and UALCAN(ualcan.path.uab.edu), were used. Other data and materials are available from the corresponding author on reasonable request.

Funding
The present study was granted by the National Natural Science Foundation of China (nos. 81702924, 81671437, 81801425, and 81771558) and the Natural Science Foundation of Hunan Province of China (nos. 2018JJ3811).
Author's contributions X-M. X and WH designed this study, wrote and revised the manuscript. X-M. X and X-L.