Ubiquitination-directed protein degradation is important in many cancers for tumor initiation and maintenance, and E3 ligases containing HECT domains are emerging as new therapeutic targets. In contrast to many other E3 ligases, the role of HUWE1 in ovarian cancer where HUWE1 is dysregulated has been unclear. Here we report that genetic deletion of Huwe1 in the mouse inhibits transformation of ovary surface epithelium cells without significantly affecting cell survival and apoptosis, and that Huwe1 deletion after tumors have been initiated inhibits tumor growth. In Huwe1-deficient cells, expression of histone H1.3 increased, inhibiting the expression of noncoding RNA H19. H19 silencing phenocopied the effects of Huwe1 deficiency, whereas H1.3 silencing partially rescued the expression of H19 and the Huwe1-null phenotype. Inducible silencing of HUWE1 in human ovarian cancer cells produced a similar phenotype. Mechanistically, HUWE1 bound and ubiquitinated H1.3, which was consequently marked for destruction by proteasomes. Our results establish that HUWE1 plays an essential role in promoting ovarian cancer. Cancer Res; 77(18); 4773–84. ©2017 AACR.
Ubiquitination and its control of subsequent protein degradation are essential biological processes in many cells, including cancer. The specificity of these processes is regulated by the E3 ligase. The HECT family (homologous to E6-AP carboxyl terminus-type E3s) is an important class of E3 ligases, and many of its members are involved in cancer-associated processes, including apoptosis and growth arrest (1). HUWE1 is a large HECT E3 ligase with a molecular size of 480 kDa, and its role in cancer has been the subject of debate since it was first cloned. Different research groups have shown that HUWE1 targets P53 and MCL-1 for degradation and that this process promotes cell apoptosis or survival, respectively, in U2OS cells (2, 3). Furthermore, many of its substrates have been shown to play positive roles in tumorigenesis. These include a variety of tumor-promoting and tumor-suppressing substrates, such as MYC (4), histones (5), CDC6 (6), N-MYC (7), POLB (8), TopBP1 (9), PA2G4 (isoform 2) (10), Miz-1 (11), HDAC2 (12), Pol λ (13), MyoD (14), BRCA1 (15), TIAM1 (16), and H2AX (17).
In humans, it has been reported that elevated HUWE1 expression is associated with a variety of cancers, such as breast, lung, colon, liver, and pancreatic cancers (4, 18, 19), and high HUWE1 expression is associated with a worse prognosis (19). However, HUWE1 mutations and deletions have also been reported in many cancers. These include a deletion in glioblastoma (20) and mutations in various types of cancer, such as those cataloged in The Cancer Genome Atlas (TCGA) data, which include skin cutaneous melanoma, squamous non–small cell lung cancer, and uterine cancer. HUWE1 expression has also been shown to be lower in teratoma cells than in embryonic stem cells that possess very similar biological characteristics (21).
HUWE1 inactivation results in the inhibition of cell proliferation in some human cancer cell lines, but it has no such effects in other cell lines, including the colon adenocarcinoma cell line SW480 (4). HUWE1 inactivation also arrested proliferation in another colon carcinoma cell line (Ls174T) and inhibited tumor growth in a xenograft model (22). However, Huwe1 knockout in mice accelerated colon tumorigenesis initiated by loss of tumor suppressor gene Apc (23). In a 7,12-dimethylbenz[a]anthracene/phorbol myristate acetate (DMBA/PMA)-induced mouse model of skin cancer, deleting the Huwe1 gene accelerated tumor development by inducing the accumulation of the Myc/Miz complex (24). These contradictory data indicate that HUWE1 may play different roles in different cell types and in response to different genetic alterations.
Ovarian cancer is the fifth leading cause of cancer-related death in women and the first leading cause of cancer-related death in the female genital organs (25, 26). The mechanisms underlying ovarian cancer remained to be determined (27, 28). In one study, HUWE1 expression was increased in approximately 70% of human ovarian cancer samples (29). However, the roles played by HUWE1 in ovarian cancer remain to be clarified. Here, we report that Huwe1 deletion impaired transformation in mouse ovary surface epithelium (MOSE) cells and inhibited tumor formation and growth. These effects were partially dependent on the Huwe1-H1.3-H19 cascade.
Materials and Methods
Conditional Huwe1L/L knockout mice were obtained from A. Lasorella (The Institute for Cancer Genetics, Columbia University Medical Center, New York, NY; ref. 20). Six-week-old BALB/c nude mice were purchased from Vital River Laboratory Animal Technology Co., Ltd. The animal protocol was approved by the animal ethics committee of the Kunming Institute of Zoology, Chinese Academy of Sciences.
MOSE cells were isolated from Huwe1L/L mice following a previously described procedure (30). Ovaries were dissected from 6- to 8-week-old virgin mice and immediately transferred into digestion buffer [DMEM/F12 (Life Technologies) containing 4 mg/mL collagenase-Dispase (Roche), 30 mg/mL bovine albumin (Millipore), and 1 μL/mL DNase I (Sigma)]. The mixture was then incubated at 37°C in 5% CO2 for 1 hour. The cells were collected and cultured in MOSE complete medium on gelatinized plates for 24 hours to eliminate residual non-MOSE cells. SKOV-3 and HEK293T cells were cultured in DMEM containing 10% FBS (Millipore), 100 U/mL penicillin, and 100 mg/mL streptomycin (Life Technologies). All of the cells were cultured in a humidified incubator with 5% CO2 at 37°C and confirmed to be mycoplasma-free by PCR (31). The SKOV-3 cell line was purchased from China Infrastructure of Cell Line Resources in September 2016 and validated by the provider using short tandem repeat (STR) profiling. All experiments using this cell line were completed within 3 months after purchase.
The lentiviral vectors pTomo-pCMV-KRASG12D-2A-MYC-IRES-CreER-shp53 and pTomo-pCMV-KRASG12D-2A-MYC-IRES-Cre-shp53, which were used for MOSE malignant transformation, were constructed in a pTomo-Flag-HRAS-H1siMp53 vector backbone that was obtained from Dr. Inder M. Verma (The Salk Institute for Biological Studies, La Jolla, CA; ref. 32). The genetic elements KRASG12D-2A-MYC-IRES-CreER and KRASG12D-2A-MYC-IRES-Cre were inserted into the plasmids between the XbaI and SalI restriction sites to replace Flag-Hras. To construct the H1.3-expressing lentiviral vector pTomo-pCMV-H1.3-HA, the coding sequences of the mouse H1.3 gene (NM_145713.4) or human H1.3 gene (NM_005320.2) were cloned into the pTomo vector between the XbaI and BamHI restriction sites using an HA-Tag sequence (5′-TACCCATACGATGTTCCAGATTACGC T-3′) at the C-terminus. For the Huwe1, H19, and H1.3 shRNA plasmids, the target sequences were cloned into a Tet-pLKO-puro vector (Addgene) according to the manufacturer's protocol (http://www.addgene.org/21915/). The following target sequences were used: H1.3, 5′-CACTTTGTTATGGGTACATTT-3′; H19, 5′-GTCCTGGGTCCATCAATAAAT-3′; and HUWE1, 5′-CGACGAGAACTAGCACAGAAT-3′.
Lentiviral preparation and infection
Lentiviral DNA vectors were cotransfected into HEK293T cells with the packaging plasmids pCMVΔ8.9 and pMD2.G at a ratio of 10:5:2. The culture medium was collected and centrifuged at 2,500 rpm for 10 minutes, and the supernatant was filtered. The supernatant containing the lentiviral particle was centrifuged at 25,000 rpm at 4°C for 2.5 hours. The lentivirus pellet was resuspended in PBS containing 0.1% BSA and stored at –80°C in aliquots. The copy number of lentiviral particles was confirmed using quantitative RT-PCR with U5 primers (U5-F, 5′-AGCTTGCCTTGAGTGCTTCA-3′ and U5-R, 5′-TGACTAAAAGGGTCTGAGGG-3′; ref. 33). MOSE cells were seeded at 5 × 105 in 6-cm plates and infected at a multiplicity of infection (MOI) of 10.
RNA isolation and real-time PCR
Total RNA was isolated from cell lines or tumor tissues using a PureLink RNA Mini Kit (Life Technologies) according to the manufacturer's instructions. Undesired DNA was removed using a TURBO DNA-free Kit (Invitrogen). Total RNA (2 μg) was reverse transcribed using random primers with a RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific). Quantitative PCR was performed in triplicate using the SYBR Green method (Life Technologies). Relative expression levels were normalized to the level of 18S rRNA. All of the primers used in this study are shown in Supplementary Table S1.
FastQC (v0.11.2) was used to confirm high-quality reads, which were those with Q scores higher than 20 (99.99% accuracy). Clean reads were mapped to the mouse genome (version GRCm38.p4, which includes 21,936 protein-coding and 3,495 lincRNA genes) using Tophat-2.1.0. Gene expression levels were calculated as the expected fragments per kilobase of transcript per million fragments (FPKM) using express Cufflinks2.2.1. Genes were considered significantly up- or downregulated at a false discovery rate (FDR)-adjusted P ≤ 0.05 and an absolute value of fold change ≥ 2 using Cuffdiff (34). Hierarchical clustering of samples was performed in the R Package (v3.2.2) gplots (v2.17.0; https://www.R-project.org/; ref. 35). The raw RNA-seq data has been deposited into the SRA database on the NCBI website (accession no. SRP081212).
Soft agar colony formation assay
Complete medium containing 0.6% agarose (Sigma) was mixed using a pipette and then aliquoted at 1 mL per well in 6-well plates. The plates were then cooled for approximately 5 minutes at 4°C to solidify the agarose (base agar). A total of 2 × 104 cells were mixed with 1 mL complete medium containing 0.3% soft agar, and the mixture was then added to the top of the base agar. The plates were cooled again for approximately 5 minutes at 4°C, and 1 mL of complete medium was then added to each well. After 4 weeks, 1 mL of PBS containing 4% formaldehyde and 0.005% Crystal violet was added to fix and stain the colonies.
Xenograft mouse models of ovarian cancer
A total of 2 × 106 cells were resuspended in PBS containing 30% Matrigel (BD Biosciences) and subcutaneously injected into 6-week-old BALB/c nude mice. The tamoxifen (Sigma; 50 mg/kg) treatment was administered via intraperitoneal injection every other day after the subcutaneous tumors had reached a diameter of approximately 5–8 mm. Doxycycline (Sigma) was delivered in water containing 5% sucrose at a concentration of 2 mg/mL. Tumor volume was calculated using the formula 0.5 × length × width × width (36). The data are shown as the means ± SD, and the statistical analyses were performed using GraphPad Prism statistical software.
Immunoprecipitation and Western blotting
The use of human cancer tissues was approved by the Institutional Review Board at Kunming Institute of Zoology, Chinese Academy of Sciences (approval ID: SWYX-2012027). Tissues and cells were lysed in RIPA buffer [50 mmol/L Tris-HCl, 150 mmol/L sodium chloride, 1% IGEPAL CA-630 (Sigma), 0.5% sodium deoxycholate, 0.1% SDS, and 0.1 mg/ml phenylmethylsulfonyl fluoride (PMSF, Sigma)] for 30 minutes on ice. For immunoprecipitation, 500 μg of total protein was incubated with 10 μL of anti-HA antibodies and protein G agarose (Santa Cruz Biotechnology, sc-2003), and the procedure was performed according to the manufacturer's instructions. The protein samples were subjected to standard SDS-PAGE electrophoresis and immunoblotted. The antibodies used in the study are shown in Supplementary Table S2.
BrdUrd incorporation assay
Bromodeoxyuridine (BrdUrd; Thermo Scientific) was added to the cell medium at a final concentration of 10 μmol/L, and the mixture was incubated for 1 hour. The cells were fixed with 4% paraformaldehyde in PBS and then permeabilized with 0.3% Triton X-100 in PBS. The cells were then treated with 2 mol/L HCl for 30 minutes and blocked in 10% goat serum in a PBS solution for 1 hour at room temperature. The anti-BrdUrd primary antibody (Abcam ab6326, 1:250) was then added, and the mixture was incubated first for 1 hour at room temperature and then with CY3-labeled secondary antibodies.
Sections cut from paraffin-embedded samples (5 μm thick) were deparaffinized, rehydrated, and processed for antigen retrieval. The sections were blocked with 10% serum and incubated with primary antibodies at 4°C overnight. A horseradish peroxidase (HRP)-labeled secondary antibody and tyramide amplification system (Perkin Elmer) were used to fluorescently detect Huwe1 expression. Cy3- or DyLight 488–labeled secondary antibodies (Jackson Immunoresearch Laboratories) were used to label the tissue for other markers. The information for all primary antibodies used in this study are provided in Supplementary Table S2.
Huwe1 deletion impairs transformation in MOSE cells
To determine the role of Huwe1 in ovarian cancer, we established an ovarian cancer model by transforming MOSE cells in vitro via the lentiviral expression of KRASG12D, MYC, and tp53 shRNA, which are the genes that are most frequently mutated in ovarian cancer according to the TCGA database (Supplementary Fig. S1; refs. 37, 38). Overexpression of oncogenic KRASG12D and MYC in combination with inactivating p53 induced malignant transformation in vitro and serous carcinoma in nude mice (39, 40). MOSE cells were isolated from a Huwe1L/L mouse and cultured in complete medium, as described previously (30). Approximately 93% of the cells were positively labeled for CK8, which is a marker of ovarian epithelial cells (Supplementary Fig. S2). A lentivirus expressing KRASG12D, MYC, and a tp53 shRNA was used to transform MOSE cells. Henceforth, we use MOSE cells to refer to infected MOSE cells unless otherwise indicated. To delete the Huwe1 gene, a Cre or CreER element was subcloned into the lentiviral vector (Fig. 1A). The results of Western blot analysis showed that KRASG12D and MYC were expressed in infected cells and that p53 expression was efficiently silenced (Fig. 1B). Huwe1 expression was similar between MOSE cells infected with the control lentivirus and those infected with the lentivirus expressing CreER that were not treated with 4-hydroxytamoxifen (4-OHT). Huwe1 expression was not detectable at the protein level in cells expressing Cre. Upon treatment with 4-OHT, Huwe1 was deleted in MOSE cells that expressed CreER (Supplementary Fig. S3A and S3B). Consistent with previous reports, the MYC protein accumulated as a result of Huwe1 knockout (4). Cell proliferation was also inhibited by Huwe1 knockout, as shown in the growth curve and bromodeoxyuridine (BrdUrd) incorporation data (Fig. 1C and D; Supplementary Fig. S4). No obvious cell death or clear staining for the apoptosis marker cleaved caspase-3 was observed in these experiments (Supplementary Fig. S5). Cre-mediated or 4-OHT–induced Huwe1 knockout significantly blocked colony-forming capacity in MOSE cells (Fig. 1E).
To determine whether MOSE cells were transformed and form tumors in vivo, 2 × 106 cells were subcutaneously injected into immunodeficient mice. MOSE cells with intact Huwe1 (CreER) rapidly and efficiently formed tumors in all mice, whereas none of the mice that were injected with Huwe1 knockout cells formed tumors (Fig. 1F). The tumors expressed CK8 and CA-125, which are markers of ovarian epithelial cancer (Fig. 1G and H), and exhibited characteristics of adenocarcinoma when stained with hematoxylin and eosin (H and E; Fig. 1I).
Huwe1 is indispensable for tumor maintenance
To determine whether Huwe1 plays an essential role in cancer maintenance, we subcutaneously injected Huwe1L/L CreER cells into mice. Huwe1 deletion was induced by treatment with tamoxifen, which was transformed into the active metabolite 4-OHT in vivo on day 16, when tumors were not yet clearly detectable, and on day 32, when the tumor size reached 5–8 mm in diameter (Fig. 2A). Treatment with tamoxifen inhibited tumor growth and decreased tumor size (Fig. 2B). When the mice were sacrificed on day 50, the mice that received tamoxifen treatment before any tumor mass had clearly formed had only small tumors or no tumors at all, whereas the mice that received tamoxifen treatment after a tumor mass had clearly formed had significantly smaller tumor masses than those observed in the control group (Fig. 2C). Huwe1 was efficiently deleted, as shown by a decrease in Huwe1 protein expression in tumors (Fig. 2D). The loss of Huwe1 decreased the number of proliferating cells and induced cellular apoptosis in vivo (Fig. 2E and F). Tumors in all of the groups expressed the epithelial marker CK8 (Supplementary Fig. S6). Interestingly, the small tumors in the group that was treated with tamoxifen early expressed the Huwe1 protein. This finding indicates that these tumors were derived mainly from MOSE cells that escaped tamoxifen-induced Huwe1 deletion.
The effect of tamoxifen treatment in Huwe1CreER xenografts was caused by Huwe1 deletion and not tamoxifen
Tamoxifen can inhibit proliferation in ovarian cancer cells by blocking the estrogen receptor and induce apoptosis by increasing oxidative stress, and its therapeutic effects have been evaluated in human patients (41). To exclude the possibility that the antitumor effect observed in tamoxifen-treated mice was mainly the consequence of tamoxifen itself instead of Huwe1 deletion, we transformed Huwe1L/L MOSE cells using a lentivirus that lacked Cre or CreER elements. Cell proliferation and colony formation capacity were affected by when MOSE cells expressing CreER, were treated with 4-OHT, whereas 4-OHT induced no significant effect in MOSE cells that lacked CreER expression (Fig. 3A–C). When nude mice with Huwe1L/L CreER xenograft tumors were treated with tamoxifen, the sizes of the tumors were significantly decreased. However, only a small effect on tumor size was observed when mice with Huwe1L/L xenograft tumors (without CreER expression) were treated with tamoxifen (Fig. 3D and E). These results demonstrate that the anticancer effect observed in these mice was mainly caused by the deletion of Huwe1 that was induced by tamoxifen and not by tamoxifen itself.
The H1.3-H19 cascade mediates the tumor-promoting function of Huwe1
To understand the molecular mechanisms involving Huwe1 in ovarian cancer, we performed RNA-seq to analyze the differences between MOSE Huwe1L/LCreER and Huwe1L/LCre cells. A total of 197 genes were expressed at significantly different levels between the groups (q value ≤ 0.05 and fold change ≥ 2; Supplementary Fig. S7A; Supplementary Table S3). Of these, 141 genes were downregulated, and 56 genes were upregulated by Huwe1 deletion. Among the downregulated genes, H19 expression was the most significantly altered, and this result was confirmed using real-time PCR (Fig. 4A; Supplementary Fig. S7B). H19 has been shown to enhance tumor properties during the initiation, progression, and metastasis stages in different cancers (42, 43). Furthermore, repressing H19 expression using H1.3 in human ovarian cancer cell lines inhibited cell proliferation and colony formation (44). To determine whether the downregulation of H19 would impair proliferation and colony formation in transformed MOSE cells, we investigated the effect of H19 knockdown in Huwe1-intact MOSE (Huwe1L/LCreER) cells. Doxycycline-induced H19 knockdown inhibited cell proliferation and colony formation in vitro (Fig. 4B and C; Supplementary Fig. S8) and tumor growth in vivo (Fig. 4D and E). Because H19 expression was inhibited by H1.3, we next sought to determine whether Huwe1 knockout would increase H1.3 protein expression. Because no specific antibodies for mouse H1.3 are currently available, we constructed a HA-tagged H1.3-expressing lentiviral vector. HA-H1.3 protein expression was increased in Huwe1L/LCreER MOSE cells by treatment with 4-OHT, which induced Huwe1 deletion (Fig. 5A). At the protein level, H19 expression decreased as H1.3 increased (Fig. 5B). Correspondingly, 4-OHT treatment inhibited cell proliferation and colony formation capacity in vitro (Fig. 5C and D). To confirm that the effect of H19 repression in Huwe1 knockout cells was mediated by H1.3 upregulation, Huwe1L/L CreER MOSE cells infected with a lentivirus expressing tet-on shH1.3 were treated with doxycycline and/or 4-OHT. The results of real-time PCR showed that histone H1.3 was efficiently silenced (Supplementary Fig. S9). Partial rescue of inhibition of cell proliferation by histone H1.3 knockdown was observed starting from passage 2 (Fig. 6A). The longer latency of the rescue effect in these cells may have been caused by the relatively longer half-life of histone proteins. Silencing histone H1.3 also partially rescued the effect of Huwe1 on H19 expression (Fig. 6B) and colony formation capacity (Fig. 6C). To investigate the interaction between Huwe1 and histone H1.3, HA-tagged mouse H1.3 sequence was introduced into MOSE cells through lentiviral infection, and coimmunoprecipitation was performed using anti-HA antibodies and Western blot analysis. The results demonstrated that Huwe1 binds to H1.3 in MOSE cells (Fig. 6D). Ubiquitinated H1.3 accumulated in response to MG-132 treatment, which inhibited its proteasome-mediated degradation, and Huwe1 deletion decreased the amount of ubiquitinated H1.3 (Fig. 6E). These data indicated that Huwe1 maintained transformation in MOSE cells and the induction of tumor formation by the Huwe1-H1.3-H19 cascade.
HUWE1 is overexpressed and associated with prognoses in human ovarian cancers (29). To confirm the relationship between HUWE1 and H1.3 in human ovarian cancer, the HUWE1 and H1.3 proteins were detected in 13 samples using Western blot analysis, and the results were quantified using Image J. HUWE1 expression was inversely correlated with H1.3 expression (Fig. 7A and B). To further explore the phenotype associated with HUWE1 inactivation in human cancer cells, SKOV-3 cells were infected with lentiviral particles that expressed tetracycline-inducible HUWE1 shRNA (SKOV-3/teton-shHUWE1). Treatment with doxycycline decreased HUWE1 expression (Supplementary Fig. S10) and cell proliferation (Supplementary Fig. S11). Immunocompromised mice were subcutaneously inoculated with SKOV-3/teton-shHUWE1 cells and then treated with doxycycline after xenograft formation to explore the role of HUWE1 in tumor development in vivo. Following treatment with doxycycline, the tumor growth rate and number of Ki67-positive cells decreased (Fig. 7C and D; Supplementary Fig. S12). HUWE1 silencing was confirmed in doxycycline-treated tumors (Supplementary Fig. S12).
To confirm that H1.3 is regulated by HUWE1 in human cells, SKOV-3/teton-shHUWE1 cells were infected with lentiviral particles that expressed HA-tagged human H1.3. H1.3 expression increased and H19 expression decreased in response to doxycycline-induced HUWE1 knockdown (Fig. 7E and F). Coimmunoprecipitation experiments demonstrated a binding interaction between the HUWE1 and H1.3 proteins (Supplementary Fig. S13A) and that H1.3 degradation is directed by HUWE1-mediated ubiquitination (Supplementary Fig. S13B), similar to what was observed in our experiments with mouse cells.
Unlike other cancer types, few mouse models accurately recapitulate the features of human ovarian cancer (45). Ovarian surface epithelial cells are currently regarded as the cells of origin of serous adenocarcinoma, the most common type of human ovarian cancer although this view has been challenged recently (30, 46). Considering the frequency of genetic events in human cancer, in this study, we took advantage of the capacity of mouse ovary surface epithelium to be transformed by oncogenic KRASG12D and MYC expression and TP53 inactivation because these are the most frequently implicated genetic alterations in human ovarian cancers (40). We efficiently transformed MOSE cells by using lentiviral infection to introduce oncogenic KRASG12D, MYC and tp53 shRNA. Our results suggest that the lentiviral technology is likely to be valuable for establishing ovarian cancer models.
The differential roles played by HUWE1 substrates in cancer, the finding that their expression and genetic elements are altered in human samples, and the results of studies using different mouse models suggest that HUWE1 may either promote or suppress tumors, depending on cellular context. In this study, we deleted the Huwe1 gene in transformed MOSE cells and found that this blocked tumor initiation and maintenance. Other recent studies have shown that HUWE1 expression is increased in 68% of human ovary samples and has an adverse effect on patient survival (29). These data, in combination with our data revealing its tumor-promoting role, indicate that HUWE1 promotes tumorigenesis in most human ovarian cancers.
We show that HUWE1 promotes ovarian tumorigenesis through a novel mechanism involving the inhibition of histone H1.3 expression and subsequent inhibition of H19 expression in MOSE cells. HUWE1 has been reported to ubiquitinate core histone proteins, including H1, H2A, H2B, H3 and H4, and is thought to play an essential role during spermiogenesis when proteins are replaced by protamines (5, 47). The effect of HUWE1 on histone proteins in other cells or tissues is not well-described. Here, we report a mechanism by which H1.3 expression is increased in Huwe1 knockout cells. Huwe1 binds to and ubiquitinates H1.3 to designate it for proteasome-dependent degradation in MOSE cells and a human ovarian cancer cell line. H19 is overexpressed in almost all cancers and involved in various cancer stages, including initiation, progression, and metastasis (42). Consistent with the results of studies using human ovarian cancer cell lines (44), we observed that decreasing H19 expression in MOSE cells led to reductions in cell proliferation and tumor growth. The p53 protein inhibits H19 expression by repressing H19 promoter activity and altering the epigenetic status of its upstream imprinting control region (48). In addition, the dysregulation of H1.3 caused by HUWE1 knockout is another important regulator of H19 expression in ovarian cancer. H1.3 knockdown rescued the HUWE1 knockout phenotype and H19 expression, indicating that the Huwe1-H1.3-H19 cascade is a key mediator of HUWE1 functions. Whether the Huwe1-H1.3-H19 cascade functions or plays essential roles in cancer in other cells or tissues remains to be determined.
In conclusion, the loss of HUWE1 inhibited cell proliferation and tumor formation in human ovarian cancer cells and transformed MOSE cells. The HUWE1-H1.3-H19 cascade is therefore an important mechanism that mediates the effects of HUWE1 in ovarian cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: D. Yang, A. Iavarone, F. Zou, X. Zhao
Development of methodology: D. Yang, B. Sun, L. Yan, X. Zhao
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D. Yang, B. Sun, X. Zhang, D. Cheng, L. Li, S. Zhang, X. Zhao
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): D. Yang, B. Sun, X. Yu, S. An, H. Jiang, A. Iavarone, F. Zou, X. Zhao
Writing, review, and/or revision of the manuscript: D. Yang, B. Sun, A. Lasorella, A. Iavarone, F. Zou, X. Zhao
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A. Lasorella, F. Zou, X. Zhao
Study supervision:F. Zou, X. Zhao
This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA 01040403), the National Natural Science Foundation of China (NSFC, 81171960), and the Top Talents Program of Yunnan Province, China (2012HA014 to X. Zhao).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.