Cancer Biol Med 2021. doi: 10.20892/j.issn.2095-3941.2020.0468

ORIGINAL ARTICLE

CUL4B facilitates HBV replication by promoting HBx stabilization

Haixia Shan1,2*, Bo Wang1,3*, Xiaodong Zhang1*, Hui Song1, Xi Li4, Yongxin Zou4, Baichun Jiang4, Huili Hu4, Hao Dou4, Changshun Shao4, Lifen Gao1,5, Chunhong Ma1,5, Xiaoyun Yang6, Xiaohong Liang1,5, Yaoqin Gong4 1Key Laboratory for Experimental Teratology of the Ministry of Education, Key Laboratory of Infection and Immunity of Shandong Province and Department of Immunology, School of Basic Medical Sciences, Shandong University, Jinan 250012, China; 2Department of Laboratory Diagnosis, Cangzhou Combination of Traditional Chinese and Western Medicine Hospital of Hebei Province, Cangzhou 061000, China; 3Division of Gastroenterology and Hepatology, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai Institute of Digestive Disease, Shanghai 200120, China; 4Ministry of Education Key Laboratory of Experimental Teratology and Institute of Molecular Medicine and Genetics, School of Basic Medical Sciences, Shandong University, Jinan 250012, China; 5Collaborative Innovation Center of Technology and Equipment for Biological Diagnosis and Therapy in Universities of Shandong, Jinan 250012, China; 6Department of Gastroenterology, Qilu Hospital, Shandong University, Jinan 250012, China

ABSTRACT Objective: virus (HBV) infection is a major public health problem worldwide. However, the regulatory mechanisms underlying HBV replication remain unclear. 4B-RING ubiquitin E3 ligase (CRL4B) is involved in regulating diverse physiological and pathophysiological processes. In our study, we aimed to explain the role of CUL4B in HBV infection. Methods: Cul4b transgenic mice or conditional knockout mice, as well as liver cell lines with CUL4B overexpression or knockdown, were used to assess the role of CUL4B in HBV replication. Immunoprecipitation assays and immunofluorescence staining were performed to study the interaction between CUL4B and HBx. Cycloheximide chase assays and in vivo ubiquitination assays were performed to evaluate the half-life and the ubiquitination status of HBx. Results: The hydrodynamics-based hepatitis B model in Cul4b transgenic or conditional knockout mice indicated that CUL4B promoted HBV replication (P < 0.05). Moreover, the overexpression or knockdown system in human liver cell lines validated that CUL4B increased HBV replication in an HBx-dependent manner. Importantly, immunoprecipitation assays and immunofluorescence staining showed an interaction between CUL4B and HBx. Furthermore, CUL4B upregulated HBx levels by inhibiting HBx ubiquitination and proteasomal degradation (P < 0.05). Finally, a positive correlation between CUL4B expression and HBV pgRNA level was observed in liver tissues from HBV-positive patients and HBV transgenic mice. Conclusions: CUL4B enhances HBV replication by interacting with HBx and disrupting its ubiquitin-dependent proteasomal degradation. CUL4B may therefore be a potential target for anti-HBV therapy. KEYWORDS CUL4B; HBV; ubiquitination; HBx; HCC

Introduction vaccine, 257 million people worldwide are chronically infected with HBV and are at increased risk of developing liver cirrho- (HBV) infection remains a serious health sis and hepatocellular carcinoma (HCC). Each year, an esti- problem. Despite the availability of an effective preventive mated 600,000 people die from HBV-related liver diseases1. HBV is a hepatotropic, non-cytopathic, small enveloped

*These authors contributed equally to this work. DNA virus. The encapsidated viral genome consists of a 3.2 kb Correspondence to: Yaoqin Gong, Xiaohong Liang and Xiaoyun Yang partially double-stranded relaxed circular DNA (rcDNA)2. E-mail: [email protected], [email protected] After viral entry into hepatocytes, the dissociates, and and [email protected] ORCID ID: https://orcid.org/0000-0001-6871-1560, the rcDNA genome translocates into the nucleus and is con- https://orcid.org/0000-0002-7626-3478 and verted into a covalently closed circular DNA molecule. This https://orcid.org/0000-0001-7162-3756 molecule is then transcribed and produces pre-genomic RNA Received August 12, 2020; accepted December 24, 2020. (pgRNA) and other subgenomic viral RNAs. HBV polymerase Available at www.cancerbiomed.org ©2021 Cancer Biology & Medicine. Creative Commons (Pol) then recognizes the 5′-ε signal in pgRNA and facilitates Attribution-NonCommercial 4.0 International License the encapsidation of pgRNA into a newly formed HBV capsid, 2 Shan et al. CUL4B promotion of HBV replication wherein pgRNA is reverse transcribed into rcDNA. Mature which promotes viral replication and HIV pathogenesis19. To nucleocapsids with HBV rcDNA are then released from the date, no direct evidence has been reported regarding the role cells after being enveloped with HBs or recycled to of CRL4B in HBV replication. However, crystallographic and the closed circular DNA reservoir in the nucleus3. functional analyses have revealed an interaction of HBx with A variety of viral and host factors contribute to regulating UV-damaged DNA binding protein 1 (DDB1), an evolution- HBV replication. The ubiquitin- pathway of pro- arily conserved adaptor protein for CUL4-RING E3 ubiquitin tein degradation is also involved in modulating the life cycle of ligases20 that is required for HBx protein stability21 and has HBV. Bandi et al.4 have reported that the proteasome inhibitor biological roles in viral promoter activation and bortezomib (Velcade) dose-dependently inhibits HBV repli- dysregulation22,23. Moreover, CRL4 is recruited by HBx and cation as well as viral RNA and protein expression. However, functions in the ubiquitination and degradation of the struc- another proteasome inhibitor, MLN-273, has only a minor tural maintenance of (SMC) complex proteins effect on wild-type HBV transgenic mice, but substantially SMC5/6, thus restricting HBV replication by inhibiting HBV enhances HBV replication in HBx-deficient HBV transgenic expression24. These findings suggest that CUL4B-RING mice5. These findings indicate that the proteasome system E3 ligase might be involved in HBV replication. might play complicated roles in HBV replication. However, Here, we provide the first reported evidence that CUL4B similarly to other viral proteins, HBV viral proteins interact promotes HBV replication in an HBx-dependent ­manner. with or otherwise modulate the ubiquitin-proteasome activity, Interestingly, CUL4B protects HBx against proteasomal thus manipulating the cellular environment to the advantage ­degradation. Moreover, CUL4B expression is positively of the virus and/or resulting in physiopathological changes ­correlated with the HBV replication level in HBV-positive liver in host cells6. The HBV X protein (HBx), well-known for its tissues. Our findings thus reveal a novel mechanism under- pleiotropic roles in HBV replication and its numerous bind- lying the involvement of CUL4B E3 in HBV ing partners in both the cytoplasm and nucleus7, has been replication and shed light on a potential therapeutic target for shown to interact with several proteasome subunits, includ- HBV elimination. ing PSMA7 and PSMC1, and to inhibit proteasome activation or the degradation of ubiquitinated proteins, thus facilitating Materials and methods viral persistence8,9. CUL4B, a member of the Cullin 4 proteins, assembles Clinical samples Cullin 4B-Ring E3 ligases (CRL4B), together with DDB1 and ROC110. Unlike CUL4A and other Cullin proteins, CUL4B HBV-positive liver tissues from patients with hepatic cav- mainly localizes to the nucleus, as directed by its unique ernous hemangioma or distant nontumor normal liver tis- N-terminal nuclear sequence, and mediates either polyubi­ sues from HBV-positive HCC patients were collected at Qilu quitination for proteasomal degradation of substrates or H2A Hospital and Shandong Provincial Hospital affiliated with monoubiquitination for epigenetic regulation11. Loss-of- Shandong University (Shandong, China) between October 30 function mutations in human CUL4B lead to X-linked men- 2012 and August 31 2015. All patients were negative for HCV tal retardation, and Cul4b knockout is embryonically lethal in or HIV, had no history of heavy drinking, and did not receive mice12, thus indicating the important biological roles of the any therapy before surgery. Informed consent was obtained Cul4b gene. Accumulating data indicate that CRL4B displays from all patients before the study was performed, with the pleiotropic functions in both physiologic (e.g., DNA replica- approval of Shandong University Medical Ethics Committee tion licensing13, cell cycle regulation14, and metabolic home- in accordance with the Declaration of Helsinki (approval No. ostasis15) and pathologic (carcinogenesis11,16,17 and obesity18) 2011023). All tissues were stored at −80 °C for analysis of HBV milieux. In viral infection, CRL4B is hijacked by viral proteins, pgRNA levels and CUL4B expression. thus causing polyubiquitination and proteasomal degradation of host proteins. For example, HIV and bind with Plasmids and cell lines CRL4B, thus triggering the degradation of the DNA repair protein uracil-N-glycosylase 2 (UNG2), the anti-viral pro- pcDNA3-HBV1.1, carrying 110% of HBV genotype C, tein SAMHD1, and the human silencing hub complex HUSH, ­pcDNA3-HBx-HA, and pcDNA3-Pol, containing C-terminal Cancer Biol Med Vol xx, No x Month 2021 3

HA-tagged HBx and a polymerase gene fragment, were as and approved by the Animal Use Committee, Shandong described previously25. HBx-null HBV plasmids (ΔHBx) University School of Medicine (approval No. 2011008). with a stop codon at position 7 of the HBx open reading frame, and polymerase-null HBV plasmids (ΔPol) with a Immunoprecipitation frame-shift mutation involving deletion of the T of the second ATG, and a point mutation of the first ATG to HEK293 cells that had been transfected with pcDNA3-HBx-HA ACG in the polymerase open reading frame, which did not for 48 h were washed with cold PBS; lysed with lysis buffer alter the encoded , were constructed with a PCR- containing 1% NP-40, 50 mM Tris-HCl pH 7.4, 50 mM EDTA, based specific mutagenesis kit (TOYOBO, Shanghai, China). 150 mM NaCl and a protease inhibitor cocktail for 30 min The human HCC cell lines HepG2 and SMMC7721, human at 4 °C; and then centrifugated for 20 min at 12,000 g. The embryonic kidney 293 (HEK293) cells, and the human cer- supernatants were collected and incubated with 2 µg of anti- vical cancer cell line HeLa were purchased from the Shanghai CUL4B, anti-HA or normal rabbit/mouse immunoglobulin Cell Collection (Chinese Academy of Sciences). These cells G (IgG) at 4 °C overnight. Protein A/G agarose beads (Santa were maintained in DMEM (Gibco, Invitrogen) containing Cruz Biotechnology) were added to the mixture and incubated 10% fetal bovine serum, 100 units/mL penicillin, and 100 for 6 h, then washed 6 times with cold PBS. The beads were µg/mL streptomycin (Invitrogen, Beijing, China). Cells were eluted with 1% SDS, boiled 5 min, and subjected to SDS-PAGE grown under 5% CO2 at 37 °C. HEK293 or HeLa cell lines followed by immunoblotting with specific primary and cor- stably infected with CUL4B shRNA were generated responding secondary antibodies. The immunodetection was as described previously13. Transfections were performed with performed with ECL. Lipofectamine 2000 (Invitrogen) according to the manufac- turer’s instructions. In vivo ubiquitination assays

Mice and hydrodynamic injection model HEK293 cells were cotransfected with HA-tagged HBx and CUL4B siRNA or negative control siRNA. Forty-eight hours HBV complete genome (ayw subtype) transgenic Balb/c mice later, cells were treated with MG132 for 6 h. The cell lysates were expressing high levels of HBV antigens and HBV DNA were immunoprecipitated with anti-HA antibody and then subjected purchased from the Transgenic Animal Central Laboratory to Western blot with antibodies specific to Ub and HA. (458 Hospital, Guang Zhou, China). The Cul4b transgenic CD1 mice were generated with the pEGFP-CUL4B construct. The Statistical analysis Cul4b floxed mice were produced as described previously11. To generate mice with inducible Cul4b deletion, we crossed Data are reported as mean values ± SEM. Cell experiments Cul4b-floxed mice with Mx1-Cre or Alb-Cre transgenic mice were performed in triplicate, and a minimum of 3 independ- (The Jackson Laboratory)26. In the Mx1-Cre; Cul4bflox/y mice, ent experiments were evaluated. GraphPad Prism (GraphPad the ablation of Cul4b was achieved through 6 injections of Software, San Diego, CA, USA) was used for data analysis. poly(I:C) i.p. at 48 h intervals. The statistical significance of differences between groups was Six- to seven-week-old male Cul4b transgenic mice, Mx1- determined with Student’s t test. Spearman correlation analy- Cre; Cul4bflox/y mice, Alb-Cre; Cul4bflox/y mice or the corre- sis was performed between CUL4B expression and HBV rep- sponding control mice were hydrodynamically injected with lication markers. P values < 0.05 were considered significant. pcDNA3-HBV1.127. In brief, 50 µg of pcDNA3-HBV1.1 in phosphate-buffered saline (PBS) was intravenously injected Results into the anesthetized mice at a volume equivalent to 8% of the body weight within 5–8 s. Twenty-four hours later, the liver CUL4B promotes HBV replication both in vivo tissues were collected, and HBV pgRNA levels were examined. and in vitro All mice were housed in the Department of Genetics, Shandong University under pathogen-free conditions. All ani- To understand the role of CUL4B in regulating HBV repli- mal procedures were in compliance with national regulations cation, we generated hydrodynamic-based HBV infection 4 Shan et al. CUL4B promotion of HBV replication models in both Cul4b transgenic mice (Supplementary Figure Pol-null HBV replication to similar levels (Figure 2B and 2C), S1A) and Cul4b conditional knockout mice (Supplementary this regulatory effect significantly decreased for the HBx-null Figure S1B). After hydrodynamic tail-vein injection of pcD- HBV replication, thus resulting in comparable levels of pgRNA NA3-HBV1.1, Cul4b transgenic mice displayed higher HBV and secreted HBV antigens to those in control cells; in con- pgRNA (P < 0.01) and serum HBsAg (P < 0.001) and HBeAg trast, the ectopic HBx expression plasmid rescued the CUL4B- (P < 0.05) levels than those in wild type mice (Figure 1A). mediated promotion of HBx-null HBV replication (Figure 2D Accordingly, Mx1-Cre; Cul4bflox/Y mice treated with poly- and 2E) (P < 0.01). Together, these results suggest that HBx is inosinic-polycytidylic acid to induce ablated CUL4B expres- required for CUL4B-mediated regulation of HBV replication. sion in the liver (Supplementary Figure S1B) had lower lev- els of pgRNA (P < 0.01) and serum HBV antigen (P < 0.01) HBx is physically associated with CUL4B- than wild type mice (Figure 1B). In line with these findings, RING E3 ligase lower serum HBV DNA levels were also observed in Alb-Cre; Cul4bflox/Y mice than in wild type mice (Figure 1C) (P < 0.01). DDB1 has been found to be essential to the role of HBx in HBV These data revealed that CUL4B promotes HBV replication in replication, through interaction with HBx21. Interestingly, vivo. DDB1 functions as an evolutionarily conserved adaptor pro- To further validate CUL4B-mediated promotion of HBV tein for CUL4-RING E3 ubiquitin ligases20. Thus, we hypoth- replication, we next performed overexpression and loss-of esized that HBx might be physically associated with CUL4B- function experiments in HCC cell lines. CUL4B overexpres- RING E3 ligase. To address this possibility, we subjected total sion, as compared with the control, significantly upregulated proteins from HEK293 cells transfected with HA-tagged HBx the levels of HBV pgRNA (P < 0.01) and secreted HBV antigens to coimmunoprecipitation experiments. CUL4B efficiently (P < 0.01) in HepG2 cells (Supplementary Figure S2A, Figure coimmunoprecipitated with HBx as well as DDB1 and ROC1 1D). Accordingly, CUL4B overexpression significantly upreg- (Figure 3A), and vice versa (Figure 3B). To provide further ulated the levels of HBV DNA (P < 0.01) and secreted HBV support for the association between CUL4B and HBx, we sub- antigens (P < 0.0001) in Huh7 cells (Figure 1E). In contrast, jected HepG2 cells transfected with HA-tagged HBx expres- a specific miRNA construct against CUL4B in HepG2 cells sion plasmid to fixing and staining with antibodies recogniz- led to markedly decreased levels of HBV replication markers ing CUL4B and HA tag. As shown in Figure 3C, the signals (Supplementary Figure S2B, Figure 1F) (P < 0.01). Similar representing CUL4B were predominantly distributed in the results were further verified in BEL7402 and HepG2.2.15 nuclei in HepG2 cells and colocalized with the HBx signal. We cells (Supplementary Figure S3). Collectively, our in vivo also observed colocalization of CUL4B and HBx in human and in vitro data clearly showed that CUL4B increases HBV liver tissues (Supplementary Figure S4). Collectively, these replication. results support the physical association of HBx with CUL4B- RING E3 ligase. HBx is responsible for optimal CUL4B- promoted HBV replication CUL4B promotes the accumulation of HBx protein To explore the mechanisms through which CUL4B promotes HBV replication, we first investigated whether HBV viral pro- To further explore the functional connection between teins might be involved in this regulatory process. For this pur- CUL4B and HBx, we next investigated the potential regu- pose, HBx-null or Pol-null HBV plasmids were cotransfected lation of CUL4B in HBx expression. HA-tagged HBx and with CUL4B expression vector into HepG2 cells, and the effects CUL4B expression plasmid or empty vector were cotrans- of CUL4B on the replication of these different HBV replicons fected into HEK293 cells. Although CUL4B showed no were analyzed. As expected, HBx-null or Pol-null HBV plas- detectable effect on HBx mRNA levels, the protein level of mids resulted in a marked decrease in virus replication, which HBx was dramatically upregulated in HEK293 cells with was restored to wild type levels by the corresponding HBx CUL4B overexpression (Figure 4A and Supplementary or polymerase expression plasmids (Figure 2A) (P < 0.01). Figure S5A). In contrast, CUL4B knockdown significantly Interestingly, although CUL4B upregulated wild type and decreased HBx protein (Figure 4B, Supplementary Figure Cancer Biol Med Vol xx, No x Month 2021 5

A CUL4B Tg (n = 10) 250 3 *** ** WT (n = 10) l 200 .450) .D 2 150 *

100 1

lative pgRNA leve 50 Re Serum HBV Ag (O 0 0 CUL4B Tg WT HBsAg HBeAg B CUL4BΔ/Y (n = 10) C 50 WT (n = 10) 2.0 1.8 ** * l 40 * * .450) 1.5 .D 30 1.2 1.0 20

HBV DN A 0.6 lative pgRNA leve 10 0.5 Re Serum HBV Ag (O 0 0.0 0.0 CUL4BΔ/Y WT HBsAg HBeAg CUL4BΔ/Y WT

D Flag-empty 0.4 2.0 ** Flag-CUL4B l **

.450) 0.3 1.5 .D

1.0 0.2 ** HBV Ag (O

lative pgRNA leve 0.5 0.1 Re 0.0 0.0 Flag-empty Flag-CUL4B HBsAg HBeAg E l ** **** 2.5 0.8 Flag-empty Flag-CUL4B 2.0 0.6 .450)

1.5 .D

HBV DNA leve 0.4 1.0 **** 0.2 0.5 HBV Ag (O

0.0 0.0

Fold change of Flag-empty Flag-CUL4B HBsAg HBeAg F Neg-mi 2.0 CUL4B-mi 1.2 * l * 1.5

0.8 .450) .D 1.0

0.4 *

lative pgRNA leve 0.5 HBV Ag (O Re

0.0 0.0 Neg-mi CUL4B-mi HBsAg HBeAg

Figure 1 CUL4B enhances HBV replication both in vivo and in vitro. (A) pcDNA3-HBV1.1 plasmid was hydrodynamically injected into CUL4B Tg mice or WT control mice. Twenty-four hours later, the mice were sacrificed, and the serum or liver tissues were used for monitoring of HBV replication. The pgRNA levels (left panel) in liver tissues were detected by qPCR. Serum HBV antigen levels (right panel) were detected by ELISA. (B) Mx1-Cre; CUL4Bflox/Y mice (CUL4BΔ/Y) or Mx1-Cre; CUL4B+/Y mice (WT) were peritoneally injected with PIPC (300 µg per mouse) at 48 h intervals for 6 times. Five days after the last PIPC injection, pcDNA3-HBV1.1 plasmid was hydrodynamically injected into CUL4BΔ/Y mice or control mice. Twenty-four hours later, the mice were sacrificed, and the pgRNA (left panel) in the liver (middle panel) and the HBsAg/HBeAg level in the serum (right panel) were detected. (C) Alb-Cre; CUL4Bflox/Y mice (CUL4BΔ/Y) or Alb-Cre; CUL4B+/Y mice (WT) were hydrodynamically injected with pcDNA3-HBV1.1 plasmid. Twenty-four hours later, the mice were sacrificed, and HBV DNA in liver tissues was detected by qPCR. (D) pcDNA3-HBV1.1 plus Flag-CUL4B or Flag-empty were transfected into HepG2 cells; 48 h later, the pgRNA levels (left panel) in HepG2 cells were detected by qPCR. HBV antigen levels in cell supernatants (right panel) were detected by ELISA. (E) pcDNA3-HBV1.1 plus Flag-CUL4B or 6 Shan et al. CUL4B promotion of HBV replication

Flag-empty were transfected into Huh7 cells; 48 h later, HBV DNA levels (left panel) in Huh7 cells were detected by qPCR. HBV antigen levels in cell supernatants (right panel) were detected by ELISA. (F) pcDNA3-HBV1.1 plus CUL4B-miRNA or Neg-miRNA was transfected into HepG2 cells; 48 h later, pgRNA levels (left panel) in HepG2 cells were detected by qPCR. HBV antigen levels in cell supernatants (right panel) were detected by ELISA. The data in panels A–C are the mean ± SEM, with each sample value from an individual mouse. The data in panels D and E are the mean ± SD of 3 independent experiments, and Student’s t-test was performed (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

A B C ol) ol) pgRNA 8 HBsAg

** l 2.5 1.2 ** l 2.0 6 vs . contr vs . contr 0.8 1.5 ssion f HBsAg level ssion f pgRNA leve 4

xp re 1.0 exp re

0.4 re 2 lative pgRNA leve 0.5 Re Fold change o 0.0 Fold change o 0.0 0 l (CUL4B over (CUL4B ove o l o l o l WT WT ΔP ΔP ΔP ΔHBx HBV1. 1 HBV1. 1 ol + Po ΔP ΔHBx + HBx

D pcDNA-HBV1.1 E pcDNA-HBV1.1 (ΔHBx) pcDNA-HBV1.1 (ΔHBx) + HBx

l 2.0 * 15 *** * 1.5 10 ** f CUL4 B f CUL4 B

1.0 ** *** 5 0.5 Fold change o Fold change o egulation on pgRNA leve egulation on HBV antigen upr 0.0 upr 0 HBsAg HBeAg

Figure 2 HBx is required for CUL4B-mediated enhancement of HBV replication. (A) pcDNA3-HBV1.1, pcDNA3-HBV1.1(ΔHBx), pcD- NA3-HBV1.1(ΔHBx) plus pcDNA3-HBx, pcDNA3-HBV1.1(ΔPol), or pcDNA3-HBV1.1(ΔPol) plus pcDNA3-Pol was transfected into HepG2 cells. Forty-eight hours later, pgRNA levels were detected by qPCR. (B, C) HepG2 cells were cotransfected with pcDNA3-HBV1.1 or pcD- NA3-HBV1.1(ΔPol) together with Flag-CUL4B or Flag-empty. pgRNA levels were detected by qPCR (B), and HBsAg was detected by ELISA (C). (D, E) pcDNA3-HBV1.1, pcDNA3-HBV1.1(ΔHBx), and pcDNA3-HBV1.1(ΔHBx) plus pcDNA3-HBx were cotransfected into HepG2 cells together with Flag-CUL4B or Flag-empty. After transfection for 48 h, pgRNA (D) and HBV antigen levels in cell supernatant (E) were detected. The fold change in CUL4B upregulation was calculated for each group relative to Flag-empty controls. All data are shown as the mean ± SD of 3 inde- pendent experiments, and Student’s t-test was performed (*P < 0.05; **P < 0.01; ***P < 0.001).

S5B). Moreover, similar results were found in 3 HCC cell in a dose-dependent manner (P < 0.05). These results sup- lines, HepG2 (Figure 4C and Supplementary Figure S5C), ported the argument that CUL4B maintains HBx protein SMMC7721 (Figure 4D and Supplementary Figure S5D), expression. and HepG2.2.15 (Supplementary Figure S5E), thus validat- To test the possibility that CUL4B might regulate HBx ing that CUL4B promotes the accumulation of HBx protein. ­degradation, we measured the half-life of HBx protein. HEK293 To further confirm this effect of CUL4B on HBx, we tran- cells stably expressing Neg-miRNA or CUL4B-miRNA were siently transfected different doses of Flag-tagged, full-length transfected with HBx and treated with cycloheximide (CHX) RNAi-insensitive CUL4B expression vector into CUL4B- for the indicated intervals. Silencing of CUL4B resulted in miRNA stably transfected HeLa cells. As shown in Figure a significant decrease in the half-life of HBx (Figure 4F) 4E, the decreased HBx protein in CUL4B knockdown cells (P < 0.00001), thus further supporting that CUL4B promotes was rescued by the RNAi-resistant CUL4B expression vector the accumulation of HBx. Cancer Biol Med Vol xx, No x Month 2021 7

A IP B IP Input Ig G CUL4 B Input Ig G HA HA-HBx CUL4B

ROC1 ROC1 IB IB DDB1 DDB1

CUL4B HA-HBx

C CUL4B HBx CUL4B/HBx CUL4B/HA/DAPI

10 µm 10 µm 10 µm 10 µm

Figure 3 HBx interacts with the CUL4B-DDB1-Roc1 complex. (A, B) pcDNA3-HBx-HA was transfected into HEK293 cells; 24 h later, whole cell lysates were immunoprecipitated with antibodies against CUL4B (A) or HA (B). Immunocomplexes were then immunoblotted with anti- bodies against the indicated proteins. IgG served as the negative control. (C) HepG2 cells were transfected with pcDNA3-HBx-HA for 48 h. Immunofluorescence staining was used to detect the interaction between HBx and CUL4B.

CUL4B inhibits the ubiquitination higher in cells with CUL4B knockdown (Figure 5C, lane 3) and proteasomal degradation of HBx than in control cells. Collectively, these results further sup- ported our hypothesis that CUL4B protects HBx expression HBx is a protein with rapid turnover, which is finely regu- from ubiquitination and proteasomal degradation. lated by proteasome-mediated degradation through ubiquit- ination8. We therefore sought to determine whether CUL4B CUL4B expression is positively correlated with might affect the ubiquitination and proteasomal degradation HBV replication in human and mouse livers of HBx. To test this possibility, we transfected CUL4B knock- down or control HepG2 cells, with HA-tagged HBx expression To further demonstrate CUL4B regulation of HBV replication, vector, then treated them with MG132, an inhibitor of the we analyzed the relationship between CUL4B expression and 26S proteasome. Although CUL4B knockdown led to a severe HBV pgRNA level in liver tissues from patients with chronic decrease in HBx protein in the DMSO-treated group, MG132 HBV and from HBV transgenic mice. CUL4B mRNA in HBV- treatment substantially mitigated this inhibitory effect (Figure positive human liver tissues positively correlated with pgRNA 5A and Supplementary Figure S6). Similar results were levels (Figure 6A) (P < 0.01). Moreover, a positive correlation obtained in HEK293 cells (Figure 5B and Supplementary between CUL4B mRNA and HBV pgRNA levels was found in Figure S6). These data support that CUL4B regulates HBx liver tissues from HBV transgenic mice (Figure 6B) (P < 0.01). expression in a proteasomal-degradation dependent manner. Collectively, these data further suggested that CUL4B is indeed We next investigated whether CUL4B might also affect involved in regulating HBV replication. HBx ubiquitination. HEK293 cells were transfected with a HA-tagged HBx expression construct together with a con- Discussion struct for CUL4B knockdown. The cell lysates were immu- noprecipitated with anti-HA antibody and immunoblotted A wide array of viral and host factors contribute to the regu- with anti-Ub antibody to detect ubiquitinated HBx proteins. lation of HBV replication. Exploration of the novel regulatory As shown in Figure 5C, HBx ubiquitination was significantly mechanisms and development of the potential therapeutic 8 Shan et al. CUL4B promotion of HBV replication

A HEK293 HA-HBx B HEK293 HA-HBx A A A A CUL4B- miRN CUL4B- miRN Neg - miRN Empty Flag - CUL4 B Empty Flag - CUL4 B Neg - miRN CUL4B CUL4B

HBx HBx

Actin Actin

C HepG2 HA-HBx D SMMC7721 HA-HBx A A A A CUL4B- miRN Neg - miRN CUL4B- miRN Empty Flag - CUL4 B Neg - miRN Empty Flag - CUL4 B CUL4B CUL4B

HBx HBx

Actin Actin

E Flag- Neg-mi CUL4BR – –00.2 0.5 0.8 1 µg CUL4B-mi CUL4BR 0.2 µg CUL4B **CUL4BR 0.5 µg – R -miRNA +++ + + + ** ** CUL4B 0.8 µg *** *** R l 1.0 *** *** CUL4B 1.0 µg HA-HBx + +++ + + + 0.8 CUL4B 0.6 ession leve

HBx 0.4 0.2 lative expr

Actin Re 0.0 CUL4B HBx

F Neg-mi CUL4B-mi HA-HBx + +++++++ 1.0 Neg-mi CHX CUL4B-mi

0 0.5 12 0 0.5 12 x Time (h) 0.8

mainin g 0.6

CUL4B re 0.4 lative HB **** Re HBx otein 0.2 pr 0.0 0 0.5 12 Actin Time after CHX treatment (h)

Figure 4 CUL4B upregulates the HBx protein level. (A, B) pcDNA3-HBx was transfected into HEK293 cells together with Flag-CUL4B, or CUL4B-miRNA, or control plasmid; 24 h later, CUL4B or HBx expression was detected by RT-PCR or Western blot. (C, D) HepG2 or SMMC7721 cells were cotransfected with pcDNA3-HBx plus Flag-CUL4B, or CUL4B-miRNA, or control plasmid. CUL4B or HBx expression was detected by Western blot. (E) HepG2.2.15 cells were transfected with Neg-miRNA or CUL4B-miRNA; 48 h later, CUL4B or HBx expression was detected by Western blot. (F) HeLa cells stably expressing Neg-miRNA or CUL4B-miRNA were transfected with HBx alone, or HBx plus CUL4B-miRNA rescue plasmid Flag-CUL4BR; 24 h later, CUL4B or HBx expression was detected by Western blot. (G) A representative CHX chase analysis of HBx protein degradation in HEK293 cells. HEK293 cells stably expressing Neg-miRNA or CUL4B-miRNA were transfected with HBx for 24 h. Then, the cells were treated with CHX for the indicated period, and the HBx expression was detected by Western blot (left panel). Relative HBx protein remaining was calculated according to the band density of HBx; β-actin served as an internal control (right panel). Student’s t-test was performed (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001). targets for HBV-related diseases are urgently needed. In this potent antiviral effects. Chronic HBV infection is a high- report, we first identified the effect of CUL4B E3 ligase on risk factor for human HCC28. Thus, CUL4B’s promotion of HBV replication and found that suppression of CUL4B had HBV replication might also be involved in the development Cancer Biol Med Vol xx, No x Month 2021 9

ACDMSO MG132 IP: anti-HA ol ol A A A A Neg - siRN CUL4B- siRN Neg - siRN CUL4B- siRN RNAi Contr Contr CUL4 B

HBx HA-HBx – + +

MG132 + + + CUL4B

Actin

HepG2 ) Ub (n

B DMSO MG132 A A A A

Neg- siRN CUL4B- siRN Neg- siRN CUL4B- siRN HA HBx Input

CUL4B CUL4B

Actin HA HEK293

Figure 5 CUL4B inhibits the ubiquitination and proteasomal degradation of HBx. (A, B) HepG2 or HEK293 cells were cotransfected with pcDNA3-HBx plus CUL4B-siRNA or negative control siRNA; 18 h after transfection, cells were exposed to 50 µg/mL MG132 or DMSO. Then, 6 h later, CUL4B or HBx expression was detected by Western blot. (C) HEK293 cells stably expressing Neg-miRNA or CUL4B-miRNA were trans- fected with HBx for 16 h. Then the cells were treated with 50 µg/mL MG132 for another 6 h. Cell lysates were immunoprecipitated with anti-HA antibody and immunoblotted with anti-Ub antibody to detect the ubiquitinated HBx protein.

ABHuman liver tissues HBV Tg mice 10 Spearman r = 0.42 p = 0.012, n = 34 –3 Spearman r = 0.3759 p = 0.032, n = 25

) –6 0 ) –86–4–20 –9

–10 pgRNA (–Δ Ct pgRNA (–Δ Ct –12

–15 –20 –11 –10 –9 –8 –7 –6 CUL4B mRNA (–Δ Ct) CUL4B mRNA (–Δ Ct)

Figure 6 The positive correlation of CUL4B mRNA with HBV replication in human and mouse livers. (A, B) HBV pgRNA and CUL4B mRNA in HBV-positive human liver tissues or in HBV Tg mouse liver tissues were detected with real-time PCR. The correlation of HBV pgRNA levels with CUL4B mRNA expression in HBV-positive human liver tissues (A) or in HBV Tg mouse liver tissues (B) was statistically analyzed. of HBV-associated HCC. Indeed, our previous study has by effectively suppressing both HBV replication and HCC shown that CUL4B also promotes the malignancy of HCC by malignancy. upregulating Wnt/β-catenin signaling29. In mice, overexpres- CUL4B has recently been reported to have an interchange- sion of CUL4B strongly promotes spontaneous and DEN- able role with that of CUL4A in HIV-induced cell cycle arrest induced hepatocarcinogenesis30. Thus, inhibition of CUL4B and depletion of the anti- SAMHD119. Although might be a potential therapeutic strategy for HCC treatment several studies have revealed that DDB1, the only known 10 Shan et al. CUL4B promotion of HBV replication

on HBx-null HBV replication. Immunoprecipitation assays Hepatocyte and immunofluorescence staining further indicated the inter- action of CUL4B with HBx. Given that DDB1 is a conserved CUL4B Proteasome 20 ROC1 binding partner of HBx , CUL4B-RING ligase together with DDB1 HBx might plausibly form a functional complex, and the integrity of this complex may be important for the efficient

HBX replication of HBV.

HBX An important question is what biological functions are exe- cuted by the CRL4B ligase and HBx complex. We first analyzed HBX the effect of CUL4B on HBx and found that CUL4B upregu- lates HBx protein expression by protecting it from uniquitina- tion and proteasomal degradation. These results suggest that CUL4B determines the accumulation of HBx, thereby facilitat- 31-33 HBV ing maximal HBV replication . Nevertheless, the detailed replication mechanisms underlying how CUL4B inhibits HBx ubiquitina- tion and proteasomal degradation remain to be determined. Several molecules have been reported to control or regulate Figure 7 CUL4B promotes HBV replication via inhibition of HBV the degradation of HBx. Siah-1 is one known E3 ligase that ubiquitination and degradation. Cullin 4B-RING ubiquitin E3 ligase directly mediates HBx poly-ubiquitination and proteasomal interacts with HBx and inhibits its ubiquitin-dependent proteaso- degradation34. In addition, both Hdj1, a human Hsp40/DnaJ mal degradation, thus facilitating HBV replication. chaperone protein, and Id-1, a member of the HLH protein family, have been found to facilitate the proteasomal degra- adaptor for CUL4 E3 ubiquitin ligases20, is required for max- dation of HBx35,36. Whether CUL4B E3 ligase disrupts HBx imal HBV replication through association with HBx, little is ubiquitination by competing for HBx binding or by modulat- known about the exact role of the scaffold protein CUL4B in ing the function of these regulatory molecules must be studied this process. In this work, results from both Cul4b transgenic further. or conditional knockout mice and HBV-infected cell lines verified that CUL4B promotes HBV replication. Moreover, Conclusions we demonstrated the potential correlation between CUL4B expression and pgRNA levels in liver tissues from HBV-positive In summary, our study revealed that CUL4B enhances HBV patients and HBV transgenic mouse liver tissues. These find- replication by interacting with HBx and disrupting its ubiq- ings extend understanding of the relationship between CRL4B uitin-dependent proteasomal degradation, thus providing a and HBV. In addition, although CUL4B and CUL4A share sev- molecular basis for the interplay between HBV and the host eral functional similarities in certain biological milieux (e.g., ubiquitin-proteasome system (Figure 7). Our data indicate HIV infection), owing to their 80% amino acid sequence iden- that CUL4B stimulates HBV replication, thus supporting the tity10, the differences in the N-terminal protein sequences and potential of CUL4B as a target in anti-viral therapy. subcellular distribution confer functional specificity. Whether CUL4A and CUL4B play redundant or complementary roles Grant support in HBV regulation is worthy of further study. HBx, a regulatory viral protein, is well conserved among This work was supported by grants from the National mammalian hepadnaviruses, thus suggesting an important Natural Science Foundation of China (Grant Nos. 81970508, biological function. Several studies have found that HBx pro- 81672425, 818300178, 8197148, 81670520, and 31671427), tein is essential for the viral replication process31,32, mainly the National Key Research and Development Program (Grant through promoting viral gene expression33. Here, we validated No. 2018YFE0126500), and the Key Research & Development that HBx, but not polymerase, is indispensable for CUL4B Plan of Shandong Province (Grant No. 2018YFJH0503). We regulation of HBV replication. CUL4B had no clear effects acknowledge support from the Collaborative Innovation Cancer Biol Med Vol xx, No x Month 2021 11

Center of Technology and Equipment for Biological Diagnosis 16. Qian Y, Yuan J, Hu H, Yang Q, Li J, Zhang S, et al. The CUL4B/ and Therapy in Universities of Shandong. AKT/beta-catenin axis restricts the accumulation of myeloid- derived suppressor cells to prohibit the establishment of a tumor- permissive microenvironment. Cancer Res. 2015; 75: 5070-83. Conflict of interest statement 17. Yang Y, Liu R, Qiu R, Zheng Y, Huang W, Hu H, et al. CRL4B promotes tumorigenesis by coordinating with SUV39H1/HP1/ No potential conflicts of interest are disclosed. DNMT3A in DNA -based epigenetic silencing. . 2013; 34: 104-18. References 18. Li P, Song Y, Zan W, Qin L, Han S, Jiang B, et al. Lack of CUL4B in adipocytes promotes PPARgamma-mediated adipose tissue 1. Ward JW, Hinman AR. What is needed to eliminate hepatitis expansion and insulin sensitivity. Diabetes. 2017; 66: 300-13. B virus and virus as global health threats. 19. Sharifi HJ, Furuya AK, Jellinger RM, Nekorchuk MD, de Noronha Gastroenterology. 2019; 156: 297-310. CM. Cullin4A and cullin4B are interchangeable for HIV Vpr and 2. Tsai KN, Kuo CF, Ou JJ. Mechanisms of hepatitis B virus Vpx action through the CRL4 ubiquitin ligase complex. J Virol. persistence. Trends Microbiol. 2018; 26: 33-42. 2014; 88: 6944-58. 3. Glebe D, Bremer CM. The molecular virology of hepatitis B virus. 20. Li T, Robert EI, van Breugel PC, Strubin M, Zheng N. A Semin Liver Dis. 2013; 33: 103-12. promiscuous alpha-helical motif anchors viral hijackers and 4. Bandi P, Garcia ML, Booth CJ, Chisari FV, Robek MD. Bortezomib substrate receptors to the CUL4-DDB1 ubiquitin ligase machinery. inhibits hepatitis B virus replication in transgenic mice. Antimicrob Nat Struct Mol Biol. 2010; 17: 105-11. Agents Chemother. 2010; 54: 749-56. 21. Bergametti F, Sitterlin D, Transy C. Turnover of hepatitis B virus 5. Zhang Z, Sun E, Ou JH, Liang TJ. Inhibition of cellular proteasome X protein is regulated by damaged DNA-binding complex. J Virol. activities mediates HBX-independent hepatitis B virus replication 2002; 76: 6495-501. in vivo. J Virol. 2010; 84: 9326-31. 22. Martin-Lluesma S, Schaeffer C, Robert EI, van Breugel PC, Leupin 6. Blanchette P, Branton PE. Manipulation of the ubiquitin- O, Hantz O, et al. Hepatitis B virus X protein affects S phase proteasome pathway by small DNA tumor viruses. Virology. 2009; progression leading to segregation defects by binding 384: 317-23. to damaged DNA binding protein 1. Hepatology. 2008; 48: 1467-76. 7. Bouchard MJ, Schneider RJ. The enigmatic X gene of hepatitis B 23. van Breugel PC, Robert EI, Mueller H, Decorsiere A, Zoulim virus. J Virol. 2004; 78: 12725-34. F, Hantz O, et al. Hepatitis B virus X protein stimulates gene 8. Hu Z, Zhang Z, Doo E, Coux O, Goldberg AL, Liang TJ. Hepatitis B expression selectively from extrachromosomal DNA templates. virus X protein is both a substrate and a potential inhibitor of the Hepatology. 2012; 56: 2116-24. proteasome complex. J Virol. 1999; 73: 7231-40. 24. Murphy CM, Xu Y, Li F, Nio K, Reszka-Blanco N, Li X, et al. 9. Zhang Z, Torii N, Furusaka A, Malayaman N, Hu Z, Liang TJ. Hepatitis B virus X protein promotes degradation of SMC5/6 to Structural and functional characterization of interaction between enhance HBV replication. Cell Rep. 2016; 16: 2846-54. hepatitis B virus X protein and the proteasome complex. J Biol 25. Liang X, Liu Y, Zhang Q, Gao L, Han L, Ma C, et al. Hepatitis B Chem. 2000; 275: 15157-65. virus sensitizes hepatocytes to TRAIL-induced through 10. Jackson S, Xiong Y. CRL4s: the CUL4-RING E3 ubiquitin ligases. Bax. J Immunol. 2007; 178: 503-10. Trends Biochem Sci. 2009; 34: 562-70. 26. Yuan J, Han B, Hu H, Qian Y, Liu Z, Wei Z, et al. CUL4B activates 11. Hu H, Yang Y, Ji Q, Zhao W, Jiang B, Liu R, et al. CRL4B catalyzes Wnt/beta-catenin signalling in hepatocellular carcinoma by H2AK119 monoubiquitination and coordinates with PRC2 to repressing Wnt antagonists. J Pathol. 2015; 235: 784-95. promote tumorigenesis. Cancer Cell. 2012; 22: 781-95. 27. Yang PL, Althage A, Chung J, Chisari FV. Hydrodynamic injection 12. Jiang B, Zhao W, Yuan J, Qian Y, Sun W, Zou Y, et al. Lack of Cul4b, of viral DNA: a mouse model of acute hepatitis B virus infection. an E3 ubiquitin ligase component, leads to embryonic lethality and Proc Natl Acad Sci U S A. 2002; 99: 13825-30. abnormal placental development. PLoS One. 2012; 7: e37070. 28. Chen Y, Tian Z. HBV-induced immune imbalance in the 13. Zou Y, Mi J, Wang W, Lu J, Zhao W, Liu Z, et al. CUL4B promotes development of HCC. Front Immunol. 2019; 10: 2048. replication licensing by up-regulating the CDK2-CDC6 cascade. 29. Yuan J, Han B, Hu H, Qian Y, Liu Z, Wei Z, et al. CUL4B activates J Cell Biol. 2013; 200: 743-56. Wnt/β-catenin signalling in hepatocellular carcinoma by repressing 14. Zou Y, Mi J, Cui J, Lu D, Zhang X, Guo C, et al. Characterization Wnt antagonists. J Pathol. 2015; 235: 784-95. of nuclear localization signal in the N terminus of CUL4B and its 30. Yuan J, Jiang B, Zhang A, Qian Y, Tan H, Gao J, et al. Accelerated essential role in cyclin E degradation and cell cycle progression. hepatocellular carcinoma development in CUL4B transgenic mice. J Biol Chem. 2009; 284: 33320-32. Oncotarget. 2015; 6: 15209-21. 15. Li Q, Cui M, Yang F, Li N, Jiang B, Yu Z, et al. A cullin 4B-RING 31. Keasler VV, Hodgson AJ, Madden CR, Slagle BL. Enhancement of E3 ligase complex fine-tunes pancreatic delta cell paracrine hepatitis B virus replication by the regulatory X protein in vitro interactions. J Clin Invest. 2017; 127: 2631-46. and in vivo. J Virol. 2007; 81: 2656-62. 12 Shan et al. CUL4B promotion of HBV replication

32. Tsuge M, Hiraga N, Akiyama R, Tanaka S, Matsushita M, Mitsui F, 35. Ling MT, Chiu YT, Lee TK, Leung SC, Fung MK, Wang X, et al. Id-1 et al. HBx protein is indispensable for development of viraemia in induces proteasome-dependent degradation of the HBX protein. J human hepatocyte chimeric mice. J Gen Virol. 2010; 91: 1854-64. Mol Biol. 2008; 382: 34-43. 33. Lucifora J, Arzberger S, Durantel D, Belloni L, Strubin M, Levrero 36. Sohn SY, Kim JH, Baek KW, Ryu WS, Ahn BY. Turnover of hepatitis M, et al. Hepatitis B virus X protein is essential to initiate and B virus X protein is facilitated by Hdj1, a human Hsp40/DnaJ maintain virus replication after infection. J Hepatol. 2011; 55: protein. Biochem Biophys Res Commun. 2006; 347: 764-8. 996-1003. 34. Zhao J, Wang C, Wang J, Yang X, Diao N, Li Q, et al. E3 ubiquitin ligase Siah-1 facilitates poly-ubiquitylation and proteasomal Cite this article as: Shan H, Wang B, Zhang X, Song H, Li X, Zou Y, et al. CUL4B degradation of the hepatitis B viral X protein. FEBS Lett. 2011; 585: facilitates HBV replication by promoting HBx stabilization. Cancer Biol Med. 2943-50. 2021; x: xx-xx. doi: 10.20892/j.issn.2095-3941.2020.0468 Cancer Biol Med Vol xx, No x Month 2021 13

Supplementary materials Detection of HBV antigens

The concentrations of HBsAg and HBeAg in cell culture Materials and methods supernatant or mouse sera were measured by using ELISA kits (Lizhu Co, Shenzhen, China) according to the manufactur- Plasmids and siRNAs er’s instructions. The absorbance at a 450-nm wavelength was measured by a micro-ELISA reader. A plasmid carrying 110% of hepatitis B virus (HBV) genome (pcDNA3-HBV1.1, subtype adr) and pcDNA3-HBx-HA, con- RNA isolation and PCR taining HA-tagged HBx gene, were described previously25. HBx-null HBV construct with a stop codon for amino acid Total RNA was extracted from transfected cells or liver tissues 7 of HBx [pcDNA3-HBV1.1(ΔHBx)] and polymerase-null using TRIzol reagent (Invitrogen, Shanghai, China) accord- HBV construct with a point mutation at the initiation codon ing to the manufacturer’s instructions. The RNA was treated (ATG-ACG) [pcDNA3-HBV1.1(ΔPol)] were generated by with RNase-free DNase (Thermo Fisher Scientific, Shanghai, PCR amplification using the KOD-Plus-Mutagenesis Kit China) and reversely transcribed into cDNA with RevertAid (TOYOBO, Shanghai, China) with the primers described in M-MuLV-RT (Thermo Fisher Scientific, Shanghai, China). The Supplementary Table S1. HA-tagged polymerase gene frag- expression of CUL4B, pgRNA, and HBx was determined by real- ment was amplified with pcDNA3-HBV1.1 using the primers time PCR with SYBR Green Master Mix (TOYOBO, Shanghai, in Supplementary Table S1 and cloned into pcDNA3 with China) or by regular PCR with Taq PCR Mastermix (Tangen, the restriction Kpn I and EcoR I (pcDNA3-Pol-HA). Beijing, China) according to the manufacturer’s instructions, The pCMV-Tag2B-CUL4B, pCMV-Tag2B-CUL4BR (RNAi- and the human or mouse β-actin was used as an internal con- resistant CUL4B vector), and CUL4B-specific RNAi expres- trol. All primer sequences are listed in Supplementary Table S1. sion vector miCUL4B as well as their corresponding control plasmids were described previously13. Western blot The siRNAs were synthesized by (GenePharma, Shanghai, China) Co. Ltd and the targeted sequences were (sense The protein samples were lysed from transfected cells with sequences): CUL4B: 5′-CCACCCAGAAGTCATTA ATTT- CelLytic™ Cell Lysis Reagent (Sigma, Shanghai, China) sup- 3′; DDB1: 5′-GCGATAATAAAGAACTCAATT-3′; ROC1: plemented with a protease inhibitor phenylmethylsulfonyl 5′-GAAGCGCT TTGAAGTGAAATT-3′. fluoride (PMSF) (Sigma, Shanghai, China). The cell lysates

Table S1 Primers used in vector construction and PCR assay Forward Reverse pcDNA3-HBV1.1(ΔHBx) 5′-CTAACTGGATCCTGCGCGGGACGTCCTTTG-3′ 5′-CAGCACACCCGAGCAGCCATGGAAAGGAGG-3′ pcDNA3-HBV1.1(ΔPol) 5′-ACGCCCCTATCTTATCAACACTTCCG-3′ 5′-TTGGTGGTCTGTAAGCAGGAGGAGTG-3′ pcDNA3-Pol-HA 5′-CCGGGTACCATGCCCCTATCTTATCAACACTTC-3′ 5′-CCGGAATTCTAAGCGTAGTCTGGTACGTCGTAAGGG TACGGTGGTCTCCATGC-3′ pgRNA 5′-CTCAATCTCGGGAATCTCAATGT-3′ 5′-AGGATAGAACCTAGCAGGCATAAT-3′

HBx 5′-TCCTTTGTCTACGTCCCG-3′ 5′-TAATCTCCTCCCCCAACTCCTC-3′

Human CUL4B 5′-GCAACTGGAATAGAGGATGGA-3′ 5′-TCTTTCTGTAGTGCTTGCTTGT-3′

Mouse CUL4B 5′-AGTGTCTGCCCAGGCACTT-3′ 5′-GATTCCTCAGCCATTTTCGTAT-3′

Human β-actin 5′-AGTTGCGTTACACCCTTTC-3′ 5′-CCTTCACCGTTCCAGTTT-3′

Mouse β-actin 5′-TGCGTGACATCAAAGAGAAG-3′ 5′-TCCATACCCAAGAAGGAAGG-3′ 14 Shan et al. CUL4B promotion of HBV replication

AB

6

WT CUL4B WT CUL4B

ssion 4 #1 Tg #1 #2 Tg #2 GFP-CUL4B

exp re 2 CUL4B lative CUL4B mRNA

Re GAPDH 0 CUL4B Tg WT CD *** 1.4

ssion F/Y re WT Mx1-Cre/Cul4b #1 #2 #3 #4 #5 #1 #2 #3 0.7

in liver CUL4B

GAPDH lative CUL4B exp

Re 0.0 CUL4B∆/Y Control

Figure S1 Related to Figure 1: CUL4B enhances hepatitis B virus (HBV) replication both in vivo and in vitro. (A–D) pcDNA3-HBV1.1 plasmid was hydrodynamically injected into CUL4B Tg mice (A, B), Mx1-Cre; CUL4Bflox/Y mice (CUL4BΔ/Y) (pretreated with 300 µg PIPC per mouse) (C, D), or WT control mice. Twenty-four hours later, mice were sacrificed, Quantitative PCR (A, C) and Western blot (B, D) were performed to detect CUL4B mRNA and protein expression in liver. β-Actin was used as an internal control. Error bars represent mean ± SD of three independent experiments. (***P < 0.001).

AB* 3 **** 1.2 ession ession 2 0.8

1 0.4 lative CUL4B expr lative CUL4B expr Re Re 0 0.0 Flag-empty Flag-CUL4B Neg-mi CUL4B-mi

Figure S2 Related to Figure 1: CUL4B enhances hepatitis B virus (HBV) replication both in vivo and in vitro. (A, B) pcDNA3-HBV1.1 plus Flag-CUL4B (A), CUL4B-miRNA (B), or control plasmids were transfected into HepG2 cells. Forty-eight hours later, CUL4B mRNA expression in HepG2 cells was detected by qPCR. Error bars represent mean ± SD of three independent experiments. (*P < 0.05; ****P < 0.0001). were incubated in ice for 30 min, and then centrifugated for antibodies against CUL4B (Sigma, Shanghai, China), Ub 20 min at 12,000 rpm. The supernatants were collected, sep- (BD Pharmingen™, Shanghai, China), anti-HA Tag (Abcam, arated by sodium sulfate polyacrylamide gel electrophoresis Shanghai, China), anti-DDB1 (Santa Cruz Biotechnology, (SDS-PAGE), and transferred on polyvinylidene difluoride Dallas, Texas, USA), anti-ROC1 (Abcam, Shanghai, China), (PVDF) membranes. After being blocked with 5% fat-free and β-actin (Sigma, Shanghai, China), respectively. The dry milk, the membranes were incubated with the specific bound antibodies were detected with horseradish peroxidase Cancer Biol Med Vol xx, No x Month 2021 15

ABBEL7402 HepG2.2.15

Neg-mi ** l 1.2 Neg-mi 1.5 *** Cul4B-mi 1.6 CUL4B-mi ** 0.9 1.2 1.0 450 450

0.6 HBV DNA leve D. D. 0.8

O. ** O. 0.5 0.3 * 0.4

0.0 0.0 HBsAg HBeAg 0.0 Fold change of HBsAg HBeAg Neg-mi CUL4B-mi

Neg-mi CUL4B-mi Neg-mi CUL4B-mi

CUL4B CUL4B

Actin Actin

Figure S3 Related to Figure 1: CUL4B enhances hepatitis B virus (HBV) replication both in vivo and in vitro. BEL7402 cells were cotrans- fected with pcDNA3-HBV1.1 plus CUL4B-miRNA or Neg-miRNA. HepG2.2.15 cells, harboring four copies of HBV DNA and expressing all viral proteins, were transfected with CUL4B-miRNA or Neg-miRNA. Forty-eight hours later, HBV antigen levels in cell supernatant from BEL7402 (A) or HepG2.2.15 cells (B) were detected by ELISA (upper panel). HBV DNA level in HepG2.2.15 cells was detected by qPCR. (B) CUL4B pro- tein expression was detected by Western blot (lower panel). Error bars represent mean ± SD of three independent experiments. (*P < 0.05; **P < 0.01; ***P < 0.001).

(HRP)-conjugated secondary antibodies and visualized using for 10 min. After blocking with 10% goat serum in phos- Enhanced Western Lightning Chemiluminescence Reagent phate-buffered saline (PBS) dilution for 15 min at 37 °C, cells (Amersham Biosciences, Piscataway, NJ, USA). The relative were incubated with the mouse anti-HA (Abcam, Shanghai, levels of target proteins to the control β-actin were determined China) and the rabbit anti-CUL4B (Sigma, Shanghai, China) by densitometry analysis using ImageJ software (National antibodies at 4 °C overnight. Cells were washed with TBS Institutes of Health, Bethesda, Maryland, USA). and incubated for 60 min at 37 °C with rhodamine-conju- gated goat anti-rabbit and fluorescein isothiocyanate (FITC)- Immunofluorescence conjugated rat anti-mouse secondary antibody (Jackson ImmunoResearch Laboratories, Inc. West Baltimore Pike West HeLa cells were transfected with pcDNA3-HBx-HA. Twenty- Grove, Pennsylvania, USA). After washing with TBS 4 times, four hours later, cells were washed with Tris-buffered saline cells were stained with DAPI (Sigma, Shanghai, China) for 15 (TBS), fixed with 4% paraformaldehyde for 10 min at room min at 37 °C and further observed using fluorescence micro- temperature, and permeabilized with 0.2% Triton X-100 scopy (Olympus, Shinjuku, Japan). 16 Shan et al. CUL4B promotion of HBV replication

CUL4B HBx

CUL4B/HBx CUL4B/HBx/DAPI

Figure S4 Related to Figure 3: HBx is physically associated with CUL4B in human liver tissue. The expression of CUL4B and HBx was detected by immunofluorescence staining in human liver tissues. Cancer Biol Med Vol xx, No x Month 2021 17

ABHEK293 cells Flag-empty HEK293 cells Neg-mi 6 Flag-CUL4B 1.5 CUL4B-mi l *** l *** *** *** *** 4 1.0 ssion leve

** ssion leve re re

2 0.5 lative exp lative exp Re Re 0 0.0 CUL4B HBx CUL4B HBx CUL4B HBx CUL4B HBx mRNA Protein mRNA Protein

CUL4B C D HepG2 CUL4B SMMC7721 HBx 4 HBx 4 *** l l *** ** ** 3 3 ession leve ession leve 2 2 ** ** *** ** 1 1 lative expr lative expr Re Re 0 0 i

Neg-mi Neg-m Neg-mi Neg-mi CUL4B-mi CUL4B-mi CUL4B-mi CUL4B-mi Flag-emptyFlag-CUL4B Flag-emptyFlag-CUL4B Flag-emptyFlag-CUL4B Flag-emptyFlag-CUL4B

HepG2.2.15 E A A Neg- miRN CUL4B- miRN

CUL4B

HBx

Actin

Figure S5 Related to Figure 4: CUL4B promotes the accumulation of HBx protein. Relative CUL4B or HBx mRNA or protein level was calcu- lated as the ratio of band density relative to β-actin as an internal control in HEK293 (A, B), HepG2 (C), or SMMC7721 cells (D). (E) HepG2.2.15 cells were transfected with Neg-miRNA or CUL4B-miRNA and the CUL4B and HBx expression were detected by Western blot. Error bars ­represent mean ± SD of three independent experiments. (**P < 0.01; ***P < 0.001). 18 Shan et al. CUL4B promotion of HBV replication

n 1.0 DMSO otei 0.8 MG132 CUL4 B 0.6

0.4 down on HBx pr 0.2 Fold change of knock 0.0 3

HepG2 HEK29

Figure S6 Related to Figure 5: CUL4B inhibits the ubiquitina- tion and proteasomal degradation of HBx. The band intensity in Western blot assay in Figure 4A–D by gray scanning analysis. Fold change of CUL4B knockdown on HBx protein level was calculated in DMSO- or MG132-treated group relative to Neg-miRNA controls in HepG2 and HEK293 cells. Error bars represent mean ± SD of three independent experiments.