ABSTRACT
Hepatitis B virus (HBV)-encoded X protein (HBx) plays a critical role in HBV-related hepatocarcinoma development. In this study, we demonstrate that HBx is specifically modified by NEDD8. We found that E3 ligase HDM2 promotes NEDDylation of HBx to enhance HBx stability by preventing its ubiquitination-mediated degradation. Consistently, analysis of 160 hepatocellular carcinoma patient specimens indicated that the amount of HDM2 protein correlates with HBx protein level. We identified that HBx K91 and K95 as the key HBx NEDDylation sites and observed that the NEDDylation-deficient HBx has shorter half-life. We generated Huh7 cell lines which ectopically express wild-type and NEDDylation-deficient HBx and found that NEDDylation-deficient HBx showed less chromatin localization and less DDB1 binding. Consistently, the expression of HBx-regulated genes (IL-8, MMP9, and YAP) and HBV transcription (the activity of HBV enhancer and the amount of pgRNA transcribed from cccDNA) were significantly higher in cells expressing wild-type (WT) HBx than that in cells expressing mutant HBx. In addition, HBx-expressing cells proliferated faster than control and mutant HBx-expressing cells. We also showed that the ability of WT HBx-expressing cells to form tumors in nude mice was significantly higher than that of mutant HBx-expressing cells. In conclusion, we revealed that E3 ligase HDM2 promotes NEDDylation of HBx to enhance HBx stability and chromatin localization, which in turn favors HBx-dependent transcriptional regulation, cell proliferation, and HBV-driven tumor growth.
IMPORTANCE Hepatitis B virus (HBV) HBx protein plays a critical role in viral replication and hepatocarcinogenesis. However, the regulation of HBx stability is not well understood. We found that HBx is modified by NEDD8 and that the HDM2 E3 ligase promotes HBx NEDDylation to enhance HBx stability by inhibiting its ubiquitination. We provide a new evidence to show the positive correlation between HDM2 and HBx in clinical hepatocellular carcinoma (HCC) samples. We also identified the major NEDDylation sites on HBx. Our studies indicate that the defective NEDDylation of HBx negatively affects its ability to activate the transcription of downstream genes and promote cell proliferation and tumor growth in vivo. Taken together, our findings reveal a novel posttranslational modification of HBx by HDM2 which regulates its stability, subcellular localization, and functions. These findings indicate that HDM2 is an important regulator on HBx and a potential diagnosis/therapeutic marker for HBV-associated HCC.
INTRODUCTION
Hepatitis B virus (HBV) infection remains a major global health threat. It has been reported that more than 350 million humans were chronically infected worldwide and 600,000 people died from chronic HBV infection (1). HBV chronic infection is a crucial risk factor for cirrhosis and HBV-related hepatocellular carcinoma (HCC) (2).
HBV genome is a partial double-stranded DNA that contains four overlapping open reading frames encoding virus envelope, core protein, virus polymerase, and a regulatory X protein (HBx) (3). HBx plays a critical role in regulating cell proliferation, apoptosis, and HCC development (4, 5). One of the important functions of HBx is that it acts as a transcriptional activator to modulate expression of target genes (4, 5), such as interleukin-8 (IL-8) (6, 7), MMP9 (8), and YAP (9), which are involved in HCC progression.
The half-life of HBx ranges from 15 min to 3 h, and the stability of HBx plays an important role in regulating its function (10). Ubiquitin and ubiquitin-like proteins such as NEDD8 (neural precursor cell expressed developmentally downregulated gene 8) and SUMO modification alter target a protein's half-life, subcellular localization, and patterns of interaction with other proteins (11, 12). NEDD8 is an 81-amino-acid polypeptide sharing 80% homology with ubiquitin. Similar to ubiquitin, NEDD8 is first activated by the E1 enzyme named NEDD8-activating enzyme (NAE) and transferred to an E2 enzyme (Ubc12), and then it is conjugated to target substrates mediated by specific E3 ligases (13). The best-characterized substrates of the NEDD8 are the cullin family of proteins (14), but non-cullin proteins such as p53, EGFR, E2F1, and transforming growth factor β II receptor have been shown to be NEDDylated as well (15–18). NEDDylation regulates numerous signaling pathways, such as apoptosis, DNA damage, and nucleolar stress signaling, by controlling the stability and activity of target proteins (19). For example, MDM2-mediated NEDDylation protects ribosomal protein L11 from degradation which is required for p53 stabilization during nucleolar stress (20).
In our study, we demonstrated that HBx can be specifically modified by NEDD8. We found that the NEDDylation modification of HBx by the E3 ligase HDM2 enhances HBx stability. By mutagenesis and mass spectrometry, we found that HBx K91 and K95 are the major NEDDylation sites and that NEDDylation-deficient HBx showed less chromatin localization. Our ex vivo and in vivo studies indicated that NEDDylation modification of HBx is important for HBx activity in transcriptional regulation, cell proliferation, and tumor growth.
RESULTS
HBx is NEDDylated by the E3 ligase HDM2.Ubiquitin and ubiquitin-like modifications play important roles in regulating the functions of target proteins. To determine whether HBx is modified by ubiquitin-like molecules, we transfected HBx-expressing plasmid with His-NEDD8 or His-SUMO2 into 293T cells. A His-pulldown assay showed that HBx is modified by NEDD8 but not by SUMO2 (Fig. 1A). We then examined whether HBx is modified by endogenous NEDD8 in coimmunoprecipitation assays. Consistently, the result indicated that HBx is modified by endogenous NEDD8. Importantly MLN4924, which is a specific inhibitor of NAE, completely prevents NEDDylation of HBx (Fig. 1B). Next, we found that HBx interacts with the Ubc12 NEDDylation E2-conjugating enzyme but not the Ubc9 SUMOylation E2 (Fig. 1C). Furthermore, we screened a series of NEDDylation E3 ligases, including SCCRO (21), c-Cbl (16), RBX1 (11), XIAP (22), HDM2 (15), TRIM40 (23), and RNF111 (24), to identify the HBx NEDDylation E3 ligase. As shown in Fig. 1D, E3 ligase HDM2 promotes the NEDDylation of HBx. We performed coimmunoprecipitation analysis and found that HBx interacts with both HDM2 and the HDM2-C464A mutant, which lacks E3 ligase activity (Fig. 1E), while the nonfunctional HDM2-C464A E3 no longer promoted NEDDylation of HBx (Fig. 1F). Furthermore, by using RNA interference, we examined how silencing HDM2 affects NEDDylation of HBx. Our data showed that HBx NEDDylation is significantly reduced upon HDM2 knockdown and restored by ectopic expression of HDM2 (Fig. 1G). Therefore, we confirmed that HDM2 is the major E3 ligase for HBx NEDDylation. Protein NEDDylation is a reversible process known as deNEDDylation by NEDD8 isopeptidases. Until now, CSN5 and NEDP1 have been reported as the well-characterized NEDD8 isopeptidase (25, 26). To determine which NEDD8 isopeptidase is the deNEDDylation of HBx, we coexpressed Myc-HBx and His-NEDD8 with FLAG-NEDP1 or FLAG-CSN5 and examined the intensity of NEDDylated HBx by His-pulldown assay. As shown in Fig. 1H, NEDP1 decreased the level of NEDDylated HBx, whereas CSN5 did not. Since NEDP1 shares the common features with other ubiquitin-like specific proteases including the active site of Cys-His-Asp triad (27), we constructed the protease-deficient NEDP1 C163S mutant and performed similar experiments. Protease-dead NEDP1 C163S did not reduce HBx NEDDylation (Fig. 1I), indicating that NEDP1 is the major deNEDDylase for HBx. Taken together, these data indicate that HBx is specifically NEDDylated by HDM2 and deNEDDylated by NEDP1.
HBx is specifically NEDDylated by HDM2. (A) 293T cells were cotransfected with pFLAG-CMV2-HBx and either pEF-His-NEDD8 or pEF-His-SUMO2. Cell lysates were harvested for His-pulldown assay. (B) 293T cells were transfected with pFLAG-CMV2-HBx for 24 h and then treated with MLN4924 (1 μΜ) for 24 h. The cell lysates were collected for immunoprecipitation assay. (C) 293T cells were cotransfected with pCMV-Myc-HBx and pFLAG-CMV2-ubc9 or pFLAG-CMV2-ubc12. The cell lysates were harvested for immunoprecipitation assay. (D) 293T cells were transfected with pCMV-Myc-RBX1, pCMV-Myc-TRIM40, pCMV-Myc-SCCRO, pCMV-Myc-XIAP, pCMV-Myc-HDM2, pCMV-Myc-c-Cbl, and pCMV-Myc-RNF111 expression plasmids, together with pEF-His-NEDD8. The cell lysates were harvested for His-pulldown assay. (E) 293T cells were transfected with pFLAG-CMV2-HBx and either pCMV-Myc-HDM2 or pCMV-Myc-HDM2-C464A for 48 h. The cell lysates were collected for immunoprecipitation assay with the indicated antibodies. (F) 293T cells were transfected with pFLAG-CMV2-HBx, pEF-His-NEDD8, and either pCMV-Myc-HDM2 or pCMV-Myc-HDM2-C464A for 48 h. The cell lysates were harvested for His-pulldown assay, followed by immunoblotting with anti-FLAG antibody. (G) 293T cells were transfected with pFLAG-CMV2-HBx, pEF-His-NEDD8 and si-control or si-HDM2 targeting endogenous HDM2 with or without pCMV-Myc-HDM2 cotransfection. The cell lysates were harvested for His-pulldown assay. (H) 293T cells were transfected with pCMV-Myc-HBx, pEF-His-NEDD8, and either pFLAG-CMV2-NEDP1 or pFLAG-CMV2-CSN5, and then the cell lysates were collected for His-pulldown assay. (I) 293T cells were transfected with pCMV-Myc-HBx, pEF-His-NEDD8, and either pFLAG-CMV2-NEDP1 or pFLAG-CMV2-NEDP1-C163S mutant. The cell lysates were harvested for His-pulldown assay. TCL, total cell lysate.
HDM2 promotes the stability of HBx.HDM2 stabilizes HuR protein through promoting its NEDDylation level (20). We wondered whether HDM2 also affected the stability of HBx. We transfected 293T cells with FLAG-HBx alone or with Myc-HDM2 or Myc-HDM2-C464A and then treated the cells with protein synthesis inhibitor cycloheximide for the indicated time periods. The cell lysates were harvested for examining the level of HBx by immunoblotting. As shown in Fig. 2A, HBx protein is much more stable in HDM2-overexpressing cells than in control cells or in HDM2-C464A-expressing cells. This finding suggests that HDM2 increases HBx stability in an E3 ligase-dependent manner. Next, we transfected 293T cells with HDM2 siRNAs and FLAG-HBx with or without Myc-HDM2 and treated the cells with cycloheximide. As shown in Fig. 2B, HBx protein appeared less stable in si-HDM2-transfected cells than that in control cells, whereas ectopic HDM2 expression restored the stability of HBx. It is known that HBx is ubiquitylated by E3 ligase Siah-1 to promote HBx degradation (28). Since HDM2 prolongs HBx half-life, we wondered whether HDM2-mediated NEDDylation prevents HBx ubiquitination. To address this question, we examined the levels of ubiquitylated HBx in the presence or absence of exogenous Myc-HDM2. As shown in Fig. 2C, HBx ubiquitination in Myc-HDM2-expressing cells was significantly lower than that in control and HDM2-C464A-expressing cells. In addition, HDM2 strongly inhibited the ubiquitination of HBx in the presence of Siah-1 (Fig. 2D). These data suggest that HDM2 increases the stability of HBx by inhibiting its ubiquitination-mediated degradation. Considering the important role of HBx in HCC development (4, 5) and the fact that HDM2 enhances HBx stability, we analyzed whether the expression of HBx correlates with HDM2 in HBV-associated HCC. A cohort of 160 HCC tumors was collected and subjected to immunohistochemistry (IHC) staining. The relative amounts of HDM2 and HBx were quantified. As shown in Fig. 2E, there was a positive correlation between the expression of HDM2 and HBx in HBV-related HCC samples. The representative IHC staining of HBx and HDM2 from three different patients are shown in Fig. 2F. Overall, these findings demonstrate that HDM2 enhances the stability of HBx, which implies that HDM2 may be involved in the development of HCC through acting as the NEDDylation E3 ligase of HBx.
HDM2 enhances HBx stability. (A) 293T cells were transfected with pFLAG-CMV2-HBx and either pCMV-Myc-HDM2 or pCMV-Myc-HDM2-C464A for 48 h and then treated with cycloheximide (100 ng/ml) for the indicated periods of time (1, 2, and 3 h). The cell lysates were collected for immunoblotting (left), and the relative protein levels were quantified (right) in three independent experiments. (B) 293T cells were transfected with pFLAG-CMV2-HBx and si-HDM2 with or without pCMV-Myc-HDM2 and then treated with cycloheximide for the indicated time. The cell lysates were harvested and subjected to immunoblotting (left), and the relative protein levels were quantified (right) in three independent experiments using ImageJ. (C) 293T cells were transfected with pFLAG-CMV2-HBx, pEF-His-Ub, and either pCMV-Myc-HDM2 or pCMV-Myc-HDM2-C464A for 48 h. The cell lysates were harvested for His-pulldown assay, followed by immunoblotting with anti-FLAG antibody. (D) 293T cells were transfected with pFLAG-CMV2-HBx and pFLAG-CMV2-Siah-1 with or without pCMV-Myc-HDM2, and then the cell lysates were harvested and subjected to His-pulldown assay. (E and F) Of the HBV-positive HCC liver tissues, 160 were collected and subjected to IHC staining with anti-HBx and anti-HDM2 antibodies. The correlation of HBx and HDM2 was evaluated (E), and representative images from immunohistochemical staining of HBx and HDM2 from the same HCC liver tissues are shown (F). The results are shown as means ± the standard deviations (SD). *, P < 0.05; **, P < 0.01 (two-tailed Student t test).
HBx Lys91 and Lys95 are the major NEDDylation sites.There are five conservative lysines on HBx among different HBV subtypes (Fig. 3A). To identify the NEDDylation site(s) on HBx, we generated HBx mutants (K91R, K95R, K113R, K118R, and K140R), transfected them into 293T cells with His-NEDD8, and then examined HBx NEDDylated by immunoblotting. We found that NEDDylation of HBx-K91R and HBx-K95R mutants was significantly lower than that of HBx-WT, while there was no difference on NEDDylation level between HBx-WT and other HBx mutants (Fig. 3B). Then, we transfected FLAG-HBx and His-NEDD8 into 293T cells and collected NEDDylated HBx for mass spectrometry. As shown in Fig. 3C (left), HBx-K95 is modified by NEDD8. Since we observed that NEDDylation of HBx-K91R was also reduced compared to that of HBx-WT, we then performed mass spectrometry using HBx-K95R mutant. The data indicated that HBx-K91 can also be NEDDylated (Fig. 3C, right). To further confirm that K91 and K95 are the NEDDylation sites, we generated stable Huh7 cell lines, which ectopically express GFP-tagged HBx-WT, -K91R, -K95R, or -K91/95R. We transfected the aforementioned cells with His-NEDD8 and then performed His-pulldown assays. In agreement with our mass spectrometry data, GFP-HBx-K91R and GFP-HBx-K95R showed remarkably reduced NEDDylation levels compared to GFP-HBx-WT, and the NEDDylation level of GFP-HBx-K91/95R double mutant was extremely low (Fig. 3D). Taken together, these data indicated that HBx K91 and K95 are the major NEDDylation sites. As HDM2 stabilizes HBx, we examined the stability of HBx and its mutants by treating the cells with cycloheximide. As expected, HBx-WT was more stable than the NEDDylation-deficient mutants (Fig. 3E).
HBx Lys91 and Lys95 are the major NEDDylation sites. (A) Schematic map of HBx. Five lysines (91, 95, 113, 118, and 140) on HBx are highly conserved in different HBV subtypes. (B) 293T cells were cotransfected with pEF-His-NEDD8 and either pFLAG-CMV2-HBx-WT, -K91R, -K95R, -K113R, -K118R, or -K140R, and the cell lysates were collected for His-pulldown assay, followed by immunoblotting with anti-FLAG antibody. (C) 293T cells were transfected with pFLAG-CMV2-HBx (left) or pFLAG-CMV2-HBx-K95R (right) and pEF-His-NEDD8 for 48 h. The cell lysates were then collected, and NEDDylated-HBx was enriched by Ni beads and subjected to mass spectrometry. The tandem mass spectra of the GlyGly-modified peptides VLHK(95)R and METTVNAHQVLPK(91) are shown. b and y ion designations are shown. (D) Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, Huh7-GFP-HBx-K91/95R, and Huh7-control cells were transfected with pEF-His-NEDD8. After 48 h, the cell lysates were collected and subjected to a His-pulldown assay. (E) 293T cells were transfected with pFLAG-CMV2-HBx-WT or with pFLAG-CMV2-HBx-K91R, -K95R, or -K91R/95R with pEF-His-NEDD8 for 48 h and then treated with cycloheximide for the indicated time. The cell lysates were harvested and subjected to an immunoblotting assay (left), and the relative protein levels were quantified (right) in three independent experiments.
NEDDylation of HBx favors its chromatin localization and transcriptional regulation activity.In order to know whether NEDDylation of HBx affects the localization of HBx, we examined the subcellular localization of HBx and its mutants. As shown in Fig. 4A, the level of WT HBx localized in the nucleus was higher than that of NEDDylation-deficient mutants. Consistently, the amount of wild-type (WT) HBx localized to chromatin was higher than that of HBx mutants (Fig. 4B). We examined whether NEDDylation of HBx affects the interaction of HBx and DDB1, which is known as an important HBx-associated transcriptional regulator. The data showed that the amount of NEDDylation-deficient HBx mutants bound DDB1 was lower than that of WT HBx-bound DDB1 (Fig. 4C). HBx is reported to be an important transcriptional regulator in HBV-infected cells and promotes cell proliferation and tumor growth by activating the transcription of target genes, such as IL-8, MMP9, and YAP. To examine whether the NEDDylation of HBx affects the transcription of IL-8, MMP9, and YAP, we first analyzed the mRNA levels of IL-8, MMP9, and YAP in Huh7 cells expressing green fluorescent protein (GFP)-tagged WT HBx and NEDDylation-deficient mutants, as well as control Huh7 cells, by real-time PCR. As shown in Fig. 4D, the mRNA levels of IL-8, MMP9, and YAP in WT HBx-expressing cells were higher than in NEDDylation-deficient HBx-expressing cells. We also performed chromatin immunoprecipitation (ChIP) assay to examine the amount of HBx occupying promoters of IL-8, MMP9, and YAP in these cells. The data showed that the amount of WT HBx associated with the promoters of IL-8, MMP9, and YAP was significantly higher than that of NEDDylation-deficient mutants (Fig. 4E). We then analyzed the effect of NEDDylation of HBx on viral transcription. First, we performed the luciferase assay with HBV enhancer reporter and found that the activity of HBV enhancer was higher in cells expressing WT HBx than that in cells expressing mutant HBx (Fig. 4F). Then, we examined the amount of pgRNA transcribed from cccDNA in these cells and observed a consistent result (Fig. 4G). These findings suggest that NEDDylation of HBx promotes its chromatin localization and its function in regulating DDB1 binding and the transcription of IL-8, MMP9, and YAP and HBV genes.
NEDDylation of HBx favors its chromatin localization and transcriptional regulation activity. (A) 293T cells were transfected with pFLAG-CMV2-HBx-WT, -K91R, -K95R, or -K91/95R for 48 h. The cells were then fixed and subjected to immunofluorescence with FLAG antibody. (B) Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, and Huh7-GFP-HBx-K91/95R cells were collected for subcellular fractionation. The total cell lysates and chromatin-bound proteins were extracted with appropriate lysis buffers and subjected to immunoblotting with the indicated antibodies. (C) 293T cells were transfected with pFLAG-CMV2-HBx-WT, -K91R, -K95R, or -K91/95R, respectively. The cell lysates were harvested for immunoprecipitation assay. (D) The total RNAs were extracted from Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, and Huh7-GFP-HBx-K91/95R cells, with Huh7-control cells used as a control, and subjected to real-time PCR with primers specific for IL-8, MMP9, or YAP mRNAs. (E) Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, and Huh7-GFP-HBx-K91/95R cells were collected and subjected to ChIP assay with anti-FLAG antibody (IgG as the negative control). Real-time PCR was performed to quantify expression of IL-8, MMP9, and YAP. (F) 293T cells were cotransfected with pHBV-Enhancer, pRL-TK, and either pFLAG-CMV2-HBx-WT, -K91R, -K95R, or -K91/95R, respectively. The cell lysates were harvested and subjected to a dual-luciferase assay. (G) 293T cells were cotransfected with prcccDNA, pCMV-Cre, and either pFLAG-CMV2-HBx-WT, -K91R, -K95R, or -K91/95R, respectively. The cells were harvested, and total RNA was extracted for real-time PCR to detect the amount of pgRNA. The results are shown as means ± the SD. *, P < 0.05; **, P < 0.01 (two-tailed Student t test).
NEDDylation of HBx facilitates its function in promoting cell proliferation and tumor growth.It has been reported that HBx promotes cell proliferation and accelerates the development of hepatocellular carcinoma. We sought to determine whether NEDDylation of HBx affects its impact on cellular viability and tumor growth. Therefore, we compared growth curves of Huh7 cells expressing HBx and its NEDDylation-deficient mutants. As shown in Fig. 5A, the cells expressing WT HBx grew faster than cells expressing NEDDylation-deficient HBx. Furthermore, we performed tumor growth assay in BALB/c nude mice with these cells. Our data indicated that WT HBx-expressing cells form larger tumors than NEDDylation-deficient HBx-expressing cells in vivo (Fig. 5B to E). These results suggest that NEDDylation of HBx could facilitate its function in promoting cell proliferation and tumor growth.
NEDDylation of HBx facilitates its function in promoting cell proliferation and tumor growth. (A) Huh7-control, Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, and Huh7-GFP-HBx-K91/95R cells (104) were plated to 96-well plates and collected at the indicated time points. Cell viability was determined by measuring the activity of CCK-8. (B to E) The cells were injected into the 6-week-old BALB/c nude mice (five animals per group). (B) Tumor sizes were measured at days 7, 10, 14, 17, and 21. (C) On day 21, the mice were photographed by in vivo imaging and then sacrificed. (D and E) The tumors were dissected (D), and the weights were measured (E). The results are shown as means ± the SD. *, P < 0.05; **, P < 0.01 (two-tailed Student t test).
DISCUSSION
Chronic HBV infection is the leading cause of HCC (29). The relative risk of HCC among HBV carriers is 10-fold higher than that of noncarriers (30). HBx is a transactivational regulator that makes a major contribution to HCC development (4, 5). The posttranslational modification of HBx could affect its function. It has been reported that HBx can be ubiquitylated by E3 ubiquitin ligase Siah-1, which promotes HBx proteasomal degradation to attenuate its transcriptional activity (28). In our study, we found that HBx is specifically modified by NEDD8. By showing that the overexpression of exogenous NEDD8 triggers NEDD8 conjugation through the NEDDylation machinery, we strictly followed the canonical criteria for identifying genuine NEDDylation substrates (19). Covalent attachment of NEDD8 to HBx was identified by using a His-pulldown assay. Further NEDDylation of HBx under homeostatic conditions was detectable and could be blocked by the NAE inhibitor MLN4924. Furthermore, we identified that HDM2 was the specific E3 enzyme for HBx NEDDylation and that NEDP1 was the deNEDDylase responsible for the HBx deNEDDylations. These results confirmed that HBx was a genuine NEDDylation substrate.
It has been reported that NEDDylation plays important roles in viral pathogenesis (31–33). For example, Hughes et al. demonstrated that NEDDylation is required for Kaposi's sarcoma-associated herpesvirus (KSHV) to replicate its genome and essential for the viability of KSHV-infected lymphoma cells. These researchers proposed that the inhibition of NEDDylation could be a novel approach for the treatment of KSHV-associated malignancies (31). Nekorchuk et al. found that NEDDylation plays an important role in HIV-1 and HIV-2 infection (32). These researchers suggested that NEDDylation could be a therapeutic target for HIV. However, in these studies the researchers did not find that any viral protein from KSHV and HIV had been NEDDylated. Recently, we showed that PB2 of influenza A virus can be NEDDylated, which inhibits viral replication (33). In the present study, we showed that the NEDDylation of HBx positively regulates its function. These findings indicate that the influences of NEDDylation on each virus are different.
It has been demonstrated that NEDDylation regulates diverse cellular processes by altering the substrate's conformation, stability, subcellular localization, or binding affinity to DNA or proteins. Dysregulation of NEDDylation is closely related to cancer, neurodegenerative disorders, and cardiac disease (34). For example, E3 ligase HDM2 promotes the NEDDylation of HuR and enhance its stability, which is related to the development of HCC and colon cancer (35). In our study, we found that HDM2 can increase the NEDDylation of HBx and stabilize its protein level by preventing its ubiquitination-mediated degradation. It has been reported that HDM2 can promote the ubiquitination-mediated proteolysis of the tumor suppressor protein p53 (22). It is known that HBx can interact with p53 and interfere with its transcriptional activity (36). Therefore, we postulated that HDM2 may regulate the function of p53 not only by affecting its protein level but also through stabilizing HBx to control its transcriptional activity in HBV-infected cells.
Previous studies showed that increased MDM2 (HDM2 in humans) expression has been observed in human HCC (37). In addition, a single-nucleotide polymorphism in the promoter region of MDM2 is associated with the risk of both HBV and HCV infection-related HCC (38, 39). Here, we found that there is a positive correlation between HDM2 and HBx expression in HBV-related HCC samples. It will be worthy to further evaluate whether HDM2 could serve as a prognostic marker for HBV-related HCC.
MLN4924 is a small molecule that can specifically block NAE. It has been reported that MLN4924 is a potent therapeutic agent for the treatment of cancers, such as osteosarcoma (40), gastric cancer (41), and bladder cancer (42). Regarding HCC, researchers reported that MLN4924 could trigger autophagy to suppress the outgrowth of liver cancer cells (43), but there was no study of MLN4924 treatment in HBV-associated HCC. Our study in NEDDylation modification of HBx suggested that MLN4924 could be a potent therapeutic agent for the treatment of HBV-associated HCC. Based on the fact that NEDDylation of HBx positively regulates the transcription of HBV, it will be also interesting to investigate whether MLN4924 can inhibit HBV replication.
In conclusion, we demonstrated that HBx is specifically modified by NEDD8. E3 ligase HDM2 can promote the NEDDylation of HBx and enhance its stability. Our study indicated that NEDDylation modification of HBx is important for its function in transcriptional regulation and in promoting cell proliferation and tumor growth.
MATERIALS AND METHODS
Patients and human specimens.HCC liver tissues from 160 patients were collected from 302 Hospital of PLA, Capital Medical University, and First Affiliated Hospital of PLA General Hospital. The patients were hospitalized during June 2013 to July 2015. The clinical characteristics of enrolled subjects were listed in Table 1. Written informed consent was provided by all study participants. Patient samples were assigned arbitrary identification based on the order of enrollment in our study. The study protocol was approved by the ethics committee of 302 Hospital of PLA, Capital Medical University, and First Affiliated Hospital of PLA General Hospital.
Clinical information of 160 HCC patients determined by IHC
Plasmids.HBx cDNA was obtained by PCR and cloned into pFLAG-CMV2 vector (Sigma, USA) and pCMV-Myc vector (Clontech, USA), respectively. pCMV-FLAG-HBx-K91R, -K95R, -K113R, -K118R, or -K140R and pCMV-Myc-HDM2-C464A were generated by site-directed mutagenesis. pCMV-HA-SCCRO and pCMV-HA-c-Cbl were kindly provided by Bhuvanesh Singh (Memorial Sloan-Kettering Cancer Center) and Vipul Chitalia (Boston University), respectively, and the cDNAs were then subcloned into pFLAG-CMV2 vector. pCMV-Cre and prcccDNA were kindly provided by Qiang Deng (Institute Pasteur of Shanghai, Chinese Academy of Sciences). pHBV-Enhancer luciferase reporter was kindly provided by Songdong Meng (Institute of Microbiology, Chinese Academy of Sciences).
Reagents and antibodies.The rabbit anti-HBx antibody was produced by immunizing rabbits with HBx protein. Mouse anti-FLAG antibody (M2) (Sigma, USA), rabbit anti-Myc antibody (9E10) (Sigma, USA), mouse anti-β-actin antibody (Santa Cruz, USA), rabbit anti-NEDD8 antibody (Sigma, USA), rabbit anti-HDM2 antibody (Bioworld, USA), mouse anti-DDB1 antibody (Santa Cruz, USA), horseradish peroxidase (HRP)-conjugated secondary antibodies (Jackson Laboratory, USA), anti-FLAG M2 beads (Sigma, USA), anti-Myc beads (Sigma, USA), and Ni-NTA beads (Qiagen, Germany) were purchased from indicated companies. Small interfering RNAs targeting the 3′ untranslated region of HDM2 were purchased from GenePharmaz (China).
Cell culture, transfection, and subcellular fractionation.Human hepatoma cell line Huh7 and human embryonic kidney cell line HEK293T (293T) were obtained from the American Type Culture Collection (Manassas, VA). Transfections were performed using Lipofectamine 2000 reagent (Invitrogen, USA). Subcellular fractionation was performed according to the procedure of a nuclear-cytosol extraction kit (Applygen, China).
His-pulldown assay.293T cells were transfected with pEF-His-NEDD8 and other plasmids as indicated for 48 h. The cells were then lysed in denaturing lysis buffer (6 M guanidinium-HCl, 0.1 M Na2HPO4/NaH2PO4, 0.01 M Tris-HCl [pH 8.0], 5 mM imidazole, and 10 mM β-mercaptoethanol) for 30 min at 25°C. The cell lysates were mixed with 80 μl of Ni-NTA beads, followed by incubation for 4 h at 25°C. The beads were washed with washing buffer, eluted with 30 μl of elution buffer (0.15 M Tris-HCl [pH 6.7], 30% glycerol, 5% SDS, and 0.5 M imidazole). The elutions were subjected to SDS-PAGE and immunoblotting analysis.
Immunohistochemistry (IHC).The tumor tissues were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 8 h at 4°C. The tissues were then embedded in paraffin and sliced in a microtome (Leica) to 5 μm and affixed onto the slides. The slides were treated with 3% H2O2 for 15 min at 25°C and retrieved antigen epitope with 0.01 M sodium citrate (pH 6.0). The adjacent slides were blocked with 10% goat serum for 1 h at 25°C and incubated with anti-HBx and anti-HDM2 antibodies, respectively, overnight at 4°C. After three washes, the slides were incubated with HRP-conjugated goat anti-rabbit antibody for 1 h at 25°C. The slides were counterstained with hematoxylin, and images were acquired by a Nikon Eclipse 80i. The IHC staining level was assessed with German semiquantitative scoring system (44). The score for each sample was obtained by multiplying the staining intensity (0, no staining; 1, weak; 2, moderate; and 3, strong) by the level based on the percentage of tumor cells (0, 0%; 1, 1 to 24%; 2, 25 to 49%; 3, 50 to 74%; and 4, 75 to 100%) at each intensity level, ranging from 0 (the minimum score) to 12 (the maximum score).
Generation of HBx ectopic-expressing cell lines.The lentiviral plasmid pLentiLox3.7 (pLL3.7 in short) and lentiviral packaging system were kindly provided by Wenjun Liu (Institute of Microbiology, Chinese Academy of Sciences). pLL3.7 GFP-HBx-WT, -K91R, -K95R, or -K91/95R plasmids were constructed and cotransfected with pCMV-VSV-G, pMDLg/pRRE, and pRSV/Rev into 293T cells. After 48 h, the supernatants were collected and infected Huh7 cells in the presence of Polybrene (8 μg/ml) for 48 h. The GFP-positive cells were sorted by using a BD FACS Aria II and expanded. The cells were named as Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, Huh7-GFP-HBx-K91/95R, or Huh7-control as the negative control.
Immunofluorescence.293T cells on slides were cotransfected with pFLAG-CMV2-HBx-WT, -K91R, -K95R, and -K91/95R for 48 h. The slides were fixed with 4% paraformaldehyde in PBS for 1 h at 25°C and permeabilized with 1% of Triton X-100 for 30 min at 25°C. The slides were then blocked with 10% goat serum for 1 h at 25°C, followed by incubation with mouse anti-FLAG antibody overnight at 4°C. After three washes in PBS-Tween, the slides were incubated with Alexa Fluor 594-conjugated goat anti-mouse IgG for 1 h at 25°C. The sections were counterstained with DAPI (4′,6′-diamidino-2-phenylindole), and the images were acquired with a Nikon Eclipse 80i instrument.
Dual-luciferase assay.293T cells were cotransfected with pHBV-Enhancer, pRL-TK, and pFLAG-CMV2-HBx-WT, -K91R, -K95R, or -K91/95R, respectively. The cell lysates were harvested for dual-luciferase assay according to the manufacturer's instructions (Promega, USA).
Analysis of cccDNA transcription.293T cells were cotransfected with prcccDNA, pCMV-Cre and pFLAG-CMV2-HBx-WT, -K91R, -K95R, or -K91/95R, respectively. The cells were harvested and subjected to total RNA extraction, reverse transcription, and real-time PCR for detecting the pgRNA level (45).
Real-time PCR.Total RNA was extracted from Huh7-GFP-HBx-WT, Huh7-GFP-HBx-K91R, Huh7-GFP-HBx-K95R, and Huh7-GFP-HBx-K91/95R cells and Huh7-control cells. Then, reverse transcription and real-time PCR was performed by using cDNA Synthesis SuperMix (TransGen Biotech, China) and SYBR green real-time PCR master mix (Toyobo, Japan). The relative expression was quantified by using the comparative threshold cycle (CT) method.
ChIP assay.A ChIP assay was performed as previously described (46). Briefly, the cells were lysed in lysis buffer, and the cell lysates were incubated with protein A-beads within anti-FLAG antibody or human IgG as control. Then, the beads were digested with DNase I. The bound DNA was subjected to real-time PCR with specific primers (YAP, forward [5′-CTTTTCGCTGCAAGTTGCTACAT-3′] and reverse [5′-CTTTTCGCTGCAAGTTGCTACAT-3′]; IL-8, forward [5′-GCTCCGGTGGTTTTTATATC-3′] and reverse [5′-GTCTTGCCTGACTTGGCAGT-3′]; and MMP9, forward [5′-TGACAGGCAAGTGCTGACTC-3′] and reverse [5′-CGTTCTCCGCAGACACCACAGTT-3′]).
Tumor growth assay.A total of 107 cells as indicated were suspended in PBS and injected subcutaneously into the left upper flank regions of 6-week-old BALB/c nude mice. The size of each tumor was measured at the indicated time point. At day 21, the mice were sacrificed, and the tumors were dissected. The sizes of tumors were measured and calculated as follows: tumor volume (mm3) = (L × W2)/2, where L is the long axis and W is the short axis. All animals received humane care, and the study of mice was permitted by the Research Ethics Committee of Institute of Microbiology (APZMCAS2016011).
Statistical analysis.Differences between each group were determined using the Student t test. A P value of <0.05 was considered significant. The correlation between variables was determined by Spearman's nonparametric correlation analysis.
ACKNOWLEDGMENTS
We thank Bhuvanesh Singh, Vipul Chitalia, Qiang Deng, and Songdong Meng for kindly providing plasmids. We also thank Tong Zhao for providing technical support for the fluorescence-activated cell sorting. We are grateful to G. Nalepa (Indiana University) for critical reading of the manuscript.
This study was supported by the Ministry of Science and Technology of China (2015CB910502 and 2016YFC1200304) and the National Natural Science Foundation of China (31600129 and 31471278). X.Y. is a principal investigator of the Innovative Research Group of the National Natural Science Foundation of China (81321063).
FOOTNOTES
- Received 1 March 2017.
- Accepted 1 June 2017.
- Accepted manuscript posted online 7 June 2017.
- Copyright © 2017 American Society for Microbiology.
