PJA1 Coordinates with the SMC5/6 Complex To Restrict DNA Viruses and Episomal Genes in an Interferon-Independent Manner

DNA viruses, including hepatitis B virus and herpes simplex virus, induce a series of immune responses in the host and lead to human public health concerns worldwide. In addition to cytokines in the cytoplasm, restriction of viral DNA in the nucleus is an important approach of host immunity. However, the mechanism of foreign DNA recognition and restriction in the cell nucleus is largely unknown. This work demonstrates that an important cellular factor (PJA1) suppresses DNA viruses and transfected plasmids independent of type I and II interferon (IFN) pathways. Instead, PJA1 interacts with the chromosome maintenance complex (SMC5/6), facilitates the complex to recognize and bind viral and episomal DNAs, and recruits DNA topoisomerases to restrict the foreign molecules. These results reveal a distinct mechanism underlying the silencing of viral and episomal invaders in the cell nuclei and suggest that PJA1 acts as a potential agent to prevent infectious and inflammatory diseases.

I nfections by DNA viruses, including hepatitis B virus (HBV) and herpes simplex virus 1 (HSV-1), lead to human public health concerns worldwide. HBV is the leading cause of chronic hepatitis B (CHB), liver cirrhosis (LC), and hepatocellular carcinoma (HCC) (1). The viral genome is a partially double-stranded relaxed circular DNA (rcDNA) replicating by the viral reverse transcriptase from an RNA intermediate (2). HSV-1 is a doublestranded DNA (dsDNA) virus and one of the most common human pathogens, infecting 70 to 90% of the population (3). It may cause diseases ranging from mild conditions to mucocutaneous lesions in lips and skin, herpes keratitis, and herpes encephalitis (4). DNA viruses deliver their genomes into the cell nucleus for transcription and replication. Viral transcription exploits cellular factors but is also repressed by nuclear defense proteins.
PJA1 (Praja1 or RNF70) is a RING-H2 domain E3 ubiquitin ligase expressed mainly in brain, liver, kidney, and embryo (5,6) and related to liver development and nervous system function (7). Deletion of PJA1 was observed in patients with craniofrontonasal syndrome and associated with mild learning disabilities (8), whereas overexpression of PJA1 facilitates skeletal myogenesis and contributes to neural precursor development (9,10). Several targets of PJA1-mediated polyubiquitination were identified, including Dlxin-1, Smad3, and PRC2 (11)(12)(13). However, the role of PJA1 in the regulation of viral and extrachromosomal DNAs has not been reported.
DNA topoisomerases (Tops) are highly conserved enzymes that control DNA supercoiling through transiently breaking and rejoining DNA strands and thus play central roles in maintaining DNA integrity for many vital cellular processes (26)(27)(28). Topological entrapment of DNA is essential for the function of SMC family complexes, which have been reported to bind DNA molecules and stimulate Top2-dependent catenation of plasmids (29). Functional cooperation between the SMC5/6 complex and Top1 was suggested for the maintenance of topologically challenged chromosomes (30). More recently, it was reported that Top inhibitors regulate SAMHD1 and human immunodeficiency virus type 1 (HIV-1) permissiveness at a post-reverse-transcription step (31).
This study demonstrates that PJA1 restricts DNA viruses but not RNA viruses and represses transfected reporters but not endogenous or chromosome-integrated genes independent of type I and II interferons (IFNs). Instead, PJA1 interacts with the SMC5/6 complex and diverts the complex to recognize viral and episomal DNAs. Interestingly, Top1 or Top2 is involved in PJA1-mediated repression of episomal DNA. Thus, we reveal a distinct mechanism underlying the restriction of DNA viruses and foreign DNAs in the nucleus, in which PJA1 coordinates with the SMC5/6 complex to silence viral and episomal DNAs in an IFN-independent manner.

RESULTS
PJA1 represses the transcription and replication of HBV and HSV-1. We initially evaluated 16 candidate genes that may be involved in the regulation of HBV replication. Enzyme-linked immunosorbent assay (ELISA) results showed that among the candidates, PJA1 significantly represses hepatitis B virus e antigen (HBeAg) and hepatitis B virus s antigen (HBsAg) production in pHBV1.3-transfected cells (Fig. 1A). Further analyses revealed that HBeAg, HBsAg, and HBV pregenomic RNA (pgRNA) were downregulated by PJA1 ( Fig. 1B and C), suggesting that PJA1 acts as a potential antiviral factor against HBV. PJA1 encodes two isoforms, PJA1 and PJA1B, with a 55-amino-acid difference (6), and the N-terminus-lacking isoform, PJA1B, was reported to function consistently with full-length PJA1 in nerve growth factor-induced differentiation of rat However, HBV pgRNA and HBV DNA replication intermediates were upregulated in HepG2-NTCP cells treated with short hairpin RNA (shRNA) targeting PJA1 (sh-PJA1) and infected with HBV (Fig. 1H), indicating that PJA1B knockdown upregulates HBV infection. Moreover, the activities of pre-S1, pre-S2, core, and X promoters of HBV were repressed by PJA1B in HepG2 cells (Fig. 1I) and Huh7 cells (Fig. 1J). Thus, we demonstrate that PJA1 represses HBV promoter activation and gene transcription and thereby attenuates HBV replication and infection.
We further determined whether PJA1 has any effect on the replication of HSV-1 containing a liner double-stranded DNA genome. The viral US11 and ICP27 mRNAs were significantly attenuated in HepG2 cells stably expressing PJA1B and infected with HSV-1 (Fig. 1K), suggesting that PJA1B overexpression represses HSV-1 gene transcription. However, US11 and ICP27 mRNAs were significantly upregulated in HepG2 cells treated with sh-PJA1B and infected with HSV-1 (Fig. 1L), indicating that PJA1B knockdown facilitates HSV-1 gene transcription. Moreover, the viral titer was significantly reduced in the supernatant of Vero cells transfected with pHA-PJA1B and infected with HSV-1 (Fig. 1M), revealing that PJA1B attenuates HSV-1 replication. Taken together, these results demonstrate that PJA1 represses the transcription and replication of the DNA viruses HBV and HSV-1.
PJA1 represses DNA viruses and episomal plasmids independent of type I and II IFNs. The host immune system utilizes pattern recognition receptors to sense pathogen-associated molecular patterns or damage-associated molecular patterns, leading to immune responses. Viral or cellular DNA has the potential to activate immune responses through different pathways, and the best-characterized one is the activation of interferon regulatory factors (IRFs) and IFNs (32). Since PJA1 attenuates DNA virus replication, we assumed that PJA1 may play a role in the activation of IFN signaling. However, in HEK293T (293T) cells, PJA1B did not induce endogenous type I and II IFN (IFN-␣, IFN-␤, and IFN-␥) expression ( Fig. 2A), while in HepG2 cells, PJA1B slightly attenuated endogenous IFN-␣ and IFN-␤ expression and had no effect on IFN-␥ expression (Fig. 2B), indicating that PJA1 is not associated with IFN signaling. Similarly, the endogenous interferon-stimulated genes (ISGs) PKR (Fig. 2C), OAS1 (Fig. 2D), and MX1 ( Fig. 2E) induced by recombinant human IFN-␣ (rhIFN-␣), rhIFN-␤, and rhIFN-␥ were relatively unaffected by PJA1 in 293T cells, revealing that PJA1 is not associated with IFN signaling. Additionally, endogenous PKR expression induced by rhIFN-␣, rhIFN-␤, and rhIFN-␥ was relatively unaffected by PJA1 in HepG2 cells (Fig. 2F), confirming that PJA1 is not associated with IFN signaling. Moreover, endogenous PJA1 mRNA was not induced by rhIFN-␣, rhIFN-␤, and rhIFN-␥ in both 293T cells (Fig. 2G) and HepG2 cells (Fig. 2H), suggesting that type I and II IFNs are not required for PJA1 expression. Thus, we reveal that PJA1 restricts DNA virus replication independent of the IFN pathways but through a different mechanism.
Meanwhile, the roles of PJA1 in regulating different signaling pathways were evaluated. Interestingly, luciferase (Luc) driven by promoters of several irrelevant pathways, including IFN-␤-Luc, interferon-stimulated response element (ISRE)-Luc, NF-B-Luc, Tp53-Luc, and cytomegalovirus (CMV)-Luc, was significantly repressed by PJA1B (Fig. 2I). Basal IFN-␤-Luc and ISRE-Luc expressions were repressed by PJA1B, suggesting that inhibition was not through disruption of stimulus-induced recruitment of specific transcription factors. CMV-Luc activity was downregulated by both PJA1 and PJA1B (Fig. 2J), and CMV-Luc activity and mRNA were attenuated by PJA1B in dosedependent manners ( Fig. 2K and L). However, transfected CMV-Luc DNA was not affected by PJA1 (Fig. 2M). These results demonstrate that PJA1B suppresses CMV-Luc mRNA transcription but not CMV-Luc DNA plasmid copy numbers. The RING domain of PJA1 is responsible for its enzyme activity. We demonstrated that the RING domain deletion mutant PJA1BΔR failed to inhibit ISRE-Luc and CMV-Luc ( Fig. 2M and O), suggesting that this domain is required for PJA1-mediated episomal plasmid repression. Similarly, HSV-TK Renilla-Luc activity was inhibited by PJA1B and slightly induced by PJA1BΔR (Fig. 2P). Moreover, enhanced green fluorescent protein (EGFP) production ( Fig. 2Q) and intensity (Fig. 2R) were attenuated by PJA1B but not by PJA1BΔR. Thus, we demonstrate that PJA1 represses transfected plasmids regardless of the DNA sequence features of their enhancers, promoters, and coding sequences (CDSs).
PJA1 has no effect on chromosome-integrated genes and RNA viruses. Since PJA1 restricts DNA viruses and extrachromosomal plasmids independent of IFN pathways, we assumed that PJA1 may recognize extrachromosomal DNAs by their distinct structure features from cell chromosomes. We determined whether PJA1 also represses genome-integrated plasmids and RNA viruses. Initially, we generated two stable cell lines, 293T-Luc and 293T-EGFP, in which the CMV promoter driving the Luc gene or the EGFP gene was integrated into cell chromosomes by a lentiviral system. 293T was chosen for maximum efficiency of the following transfections. Luciferase activity in 293T-Luc stable cells and EGFP production in 293T-GFP stable cells were not affected by PJA1 or PJA1ΔR ( Fig. 3A and B). Interestingly, plasmid-expressed interleukin 1 receptor-associated kinase 1 (IRAK1) was significantly attenuated by PJA1 and not by PJA1ΔR, but endogenous IRAK1 was not affected by PJA1 or PJA1ΔR ( Fig. 3C and D). Similarly, plasmid-expressed interferon alpha and beta receptor subunit 1 (IFNAR1) was significantly downregulated by PJA1 and not by PJA1ΔR, but endogenous IFNAR1 was not affected by PJA1 or PJA1ΔR ( Fig. 3E and F). We then investigated the role of PJA1 in the replication of the RNA viruses enterovirus 71 (EV71), containing a single-stranded and positive-sense RNA genome, and vesicular stomatitis virus (VSV), carrying a single-stranded and negative-sense RNA genome. The EV71 VP1 protein was not affected by PJA1B in rhabdomyosarcoma (RD) cells transfected with pHA-PJA1B and infected with EV71 (Fig. 3G), revealing that PJA1 has no effect on EV71 replication. Green fluorescent protein (GFP) was relatively unchanged by PJA1 in 293T cells transfected with pHA-PJA1B and infected with a recombinant VSV (VSV-GFP) (Fig. 3H), suggesting that PJA1 has no effect on VSV replication. Taken together, these results demonstrate that PJA1 restricts DNA viruses and transfected plasmids but not chromosome-integrated plasmids, endogenous genes, or RNA viruses.  (Fig. 4C), revealing that PJA1 knockout does not affect cell proliferation. 293T control cells and 293T(PJA1-KO) cells were transfected with pHBV-Enhancer 1 (Enh1)-Luc and pTp53-Luc for 24 h, respectively. The activities and mRNA levels of luciferase driven by HBV Enh1 were significantly higher in 293T(PJA1-KO) cells than in 293T control cells (Fig. 4D). Similarly, the activities and mRNA levels of luciferase driven by Tp53 transcriptional recognition sequences were significantly higher in 293T(PJA1-KO) cells than in 293T control cells (Fig. 4E). Endogenous Tp53 protein was not affected in both 293T(PJA1-KO) cells and 293T control cells (Fig. 4F), indicating that knockout of PJA1 has no effect on Tp53 expression. Luciferase activities driven by HBV Enh1 were activated in 293T(PJA1-KO) cells transfected with HBV Enh1-Luc but significantly attenuated by PJA1B in 293T(PJA1-KO) cells cotransfected with HBV Enh1-Luc and pHA-PJA1B (Fig. 4G), demonstrating that PJA1 represses HBV Enh1 activation. Additionally, 293T control cells and 293T(PJA1-KO) cells were transfected with pHBV-Enh1-Luc. The binding of NSE4 with episomal Enh1-Luc DNA or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) DNA as a negative control was monitored by chromatin immunoprecipitation (ChIP) assays. The binding of NSE4 to Enh1-Luc DNA was significantly reduced in 293T(PJA1-KO) cells compared to that in 293T control cells (Fig. 4H), indicating that PJA1 facilitates the binding of NSE4 to Enh1-Luc. Moreover, 293T control cells and 293T(PJA1-KO) cells were infected with HSV-1. The levels of HSV-1 US11 mRNA (Fig. 4I) and ICP27 mRNA (Fig. 4J) were significantly higher in 293T(PJA1-KO) cells than in 293T control cells at 10 h postinfection, suggesting that PJA1 knockout upregulates HSV-1 gene transcription. Taken together, these data reveal that PJA1 represses the transcription of ectopic plasmids and DNA viruses.
PJA1 interacts with the SMC5/6 complex in the nucleus. The mechanism by which PJA1 restricts DNA viruses and episomal plasmids was investigated. Based on a large interactome screening, it was predicted that PJA1 may interact with the SMC5/6 complex (33), which consists of SMC5 and SMC6 along with the NSE1, -2, -3, and -4 proteins (Fig. 5A). Since knockdown of SMC5/6 rescues the replication of HBx-deficient HBV and enhances the expression of the transfected episomal HBV Enh1-driven firefly luciferase reporter as well as the episomal Renilla luciferase construct driven by the CMV promoter (23), we speculated that PJA1 may silence extrachromosomal DNA through interacting with the complex. Coimmunoprecipitation (co-IP) results showed that PJA1ΔR (the RING domain deletion mutant was used to avoid inhibition of cotransfected plasmids by PJA1) was coprecipitated with SMC6, NSE3, and NSE4 but not with NSE1 (Fig. 5B). Endogenous co-IP confirmed that endogenous PJA1 interacted with endogenous NSE4 (Fig. 5C, left) and that NSE4 interacted with endogenous PJA1 (Fig.  5C, right). A glutathione S-transferase (GST) pulldown assay using fusion proteins overexpressed and purified from Escherichia coli revealed that maltose-binding protein (MBP)-PJA1BΔR strongly interacted with GST-NSE3 and weakly interacted with GST-NSE4 (Fig. 5D). The distribution and association of PJA1 with the components of the SMC5/6 complex were examined by confocal microscopy. In the presence of a single protein, PJA1 was located in the nucleus (Fig. 5Ea to d), SMC5 was mainly distributed in the nucleus (Fig. 5Ee to h), and NSE4 was present in both the cytoplasm and nucleus ( Fig. 5Ei to p). Interestingly, in the presence of two proteins, PJA1 colocalized with NSE3 ( Fig. 5Fa to d), NSE4 (Fig. 5Fe to h), and SMC5 (Fig. 5Fi to l) to form distinctive spots in the nucleus but was not colocalized with an unrelated protein, DDB1 (Fig. 5Fm to p). Taken together, our data reveal that PJA1 interacts with the SMC5/6 complex in the nucleus independent of its RING domain (Fig. 5G).
PJA1 and SMC5/6 Restrict DNA Viruses and Episomal DNA Journal of Virology cooperated with the complex in restricting viral and extrachromosomal DNAs. The SMC5/6 ring structure entraps DNA, and the NSE1/3/4 subcomplex binds to dsDNA and loads DNA onto the ring (34). Here, ChIP assays revealed that the interaction of NSE4 with CMV-Luc DNA was not affected by NSE1 (a SMC5/6 complex protein containing a RING domain analogous to that of PJA1) (35) but was significantly enhanced by PJA1B ( Fig. 6A and B). In addition, PJA1ΔR attenuated the interaction of NSE4 with NSE1 but not interactions of NSE4 with SMC5 or NSE3 and downregulated the interaction of NSE1 with NSE4 in a dose-dependent manner ( Fig. 6C and D), revealing that PJA1 may compete with NSE1 in interacting with the SMC5/6 complex. Moreover, PJA1B significantly enhanced the binding of the SMC5/6 complex to HSV-1 VP16 promoter DNA and UL36 gene DNA in HepG2 cells stably expressing PJA1B and infected with HSV-1 ( Fig.  6E and F), demonstrating that PJA1 promotes the binding of the SMC5/6 complex to HSV-1 genomic DNA. Therefore, we reveal that PJA1 collaborates with the SMC5/6 complex in binding viral and extrachromosomal DNAs. DNA topoisomerases are involved in PJA1-mediated restriction of viral and episomal DNA. The functional correlation between PJA1 and the SMC5/6 complex was further verified. Plasmids expressing shRNAs targeting SMC6 (sh-SMC6) and PJA1B (sh-PJA1B) were generated and then transfected into 293T cells. Results showed that sh-SMC6 significantly attenuated SMC6 and NSE4 protein levels, slightly reduced NSE1 production, and had no effect on PJA1 and GAPDH proteins (Fig. 7A, left), which was consistent with a previous report showing that knockdown of any protein of the SMC5/6 complex leads to degradation of other components and disruption of complex formation (36). In addition, sh-PJA1B significantly attenuated PJA1 production but not SMC6, NSE4, NSE1, or GAPDH expression (Fig. 7A, middle), suggesting that knockdown of PJA1 has no effect on the production of SMC5/6 complex components. Moreover, the productions of SMC6, NSE4, and NSE1 proteins were not affected in 293T(PJA1-KO) cells compared to those in 293T cells (Fig. 7A, right), confirming that knockout of PJA1 has no effect on the production of SMC5/6 complex components. 293T cells were cotransfected with pCMV-Luc, sh-SMC6, and pHA-PJA1B. Luciferase activity was repressed by PJA1B in the presence or absence of sh-SMC6 (Fig. 7B), indicating that knockdown of SMC6 has no effect on PJA1-mediated repression of CMV-Luc. To confirm this observation, we generated two stable cell lines in which sh-NC and sh-SMC6 were stably expressed in HepG2 cells. The stable cells were cotransfected with pCMV-Luc and/or pHA-PJA1B. CMV-Luc activity was significantly higher in the presence of sh-SMC6 than that in the absence of sh-SMC6 (Fig. 7C), indicating that knockdown of SMC6 upregulates CMV-Luc. Interestingly, CMV-Luc activity was significantly reduced by PJA1B in the presence or absence of sh-SMC6 (Fig. 7C), suggesting that knockdown of SMC6 has no effect on PJA1-mediated repression of CMV-Luc. Thus, these results suggest that PJA1 functions downstream of the SMC5/6 complex.
The mechanism by which PJA1 represses episomal plasmids was evaluated. Initially, we determined the effects of 8 potential inhibitors on PJA1B-mediated repression of the HBV Enh1-Luc reporter. Among the inhibitors, topotecan (a specific inhibitor of topoisomerase 1) and idarubicin (a specific inhibitor of topoisomerase 2) significantly suppressed PJA1B-mediated repression of HBV Enh1-Luc activity (Fig. 7D), indicating that Top1 and Top2 participated in the repression of HBV Enh1-Luc activity mediated by PJA1. The results confirmed that topotecan and idarubicin recovered PJA1Bmediated repression of the HBV Enh1-Luc reporter in dose-dependent fashions (Fig. 7E to G).
To further confirm the role of topoisomerases in PJA1-mediated repression of extrachromosomal and viral DNAs, we constructed small interfering RNAs (siRNAs) targeting Top1 (siR-Top1), Top2a (siR-Top2a), and Top2b (siR-Top2b). The specificity and efficiency of the siRNAs were verified and confirmed in 293T cells (Fig. 7H). Like the inhibitors, siR-Top1 and siR-Top2a significantly attenuated PJA1B-mediated repression of HBV Enh1-Luc (Fig. 7I). These results demonstrate that DNA topoisomerases are involved in the restriction of viral and extrachromosomal DNAs mediated by PJA1. Therefore, we propose that PJA1 coordinates with the SMC5/6 complex to restrict viral and episomal DNAs through topoisomerases (Fig. 7J and K).

DISCUSSION
Multiple DNA sensors and adaptors detecting pathogen-derived nucleic acids in the cytoplasm have been identified, including the cytosolic DNA sensor DAI (ZBP1), AIM2, and cGAS (32,37,38). The molecular mechanism by which host cells sense microbial and episomal DNAs to eliminate foreign invaders in the nucleus remains largely unknown. One primary concern is how cells distinguish the viral genome from the host genome in the same compartment. This study reveals a distinct mechanism underlying the silencing of foreign DNA in the nucleus mediated by PJA1. PJA1 restricts DNA viruses and extrachromosomal DNA but not RNA viruses or chromosome-integrated plasmids. Notably, PJA1 has no effect on the production of type I and II IFNs and antiviral proteins, and IFNs have no effect on the expression of PJA1, demonstrating that PJA1 restricts foreign DNA independent of type I and II IFN pathways but through different mechanisms.
Interestingly, we reveal that PJA1 plays an important role in the restriction of DNA viruses and extrachromosomal DNAs coordinating with the SMC5/6 complex, which was identified previously as a host restriction factor against HBV (23)(24)(25). Additionally, a recent report has shown that SMC5/6 restricts HBV infection in primary human hepatocytes (PHHs) without inducing an IFN response (39). We further demonstrate that PJA1 interacts with the SMC5/6 complex in the nucleus and facilitates the binding of NSE4 to viral and episomal DNAs. The interaction of NSE4 with episomal DNA is enhanced by PJA1 but not by NSE1, whereas PJA1 attenuates the interaction of NSE4 with NSE1. These results potentially suggest that PJA1 competes with NSE1 in interacting with the SMC5/6 complex and may replace NSE1 to facilitate the interaction of NSE4 with episomal DNA. Both PJA1 and NSE1 share a RING domain that commonly indicates E3 ubiquitin ligase activity (11,12). The RING domain of NSE1 supports the SMC5/6 complex in genome stability maintenance, and our work reveals that the RING domain is required for PJA1 in the restriction of viral and plasmid DNAs. Multiple RING domain proteins were identified as being binding partners of melanoma-associated antigen (MAGE) proteins (40)(41)(42). It is reasonable to speculate that similarly to how NSE1 interacts with NSE4 to form the NSE1 (RING)/NSE4 (MAGE) subcomplex, PJA1 may also interact with NSE4 to form a PJA1 (RING)/NSE4 (MAGE) subcomplex to establish specific functions. Thus, we propose that NSE1/3/4 ensures the normal function of the SMC5/6 complex in host chromosome maintenance, whereas PJA1/NSE3/4 converts the function of the SMC5/6 complex to viral DNA restriction, although the underlying physiological significance of the alternative RING-MAGE heterodimers needs to be further investigated.
More interestingly, treatment with inhibitors of DNA topoisomerases and knockdown of the enzymes result in attenuation of PJA1-mediated restriction of HBV Enh1-Luc, demonstrating an important role of Tops in this process. The SMC5/6 complex is loaded onto chromosomes at collapsed replication forks and at sites of DNA damage where chromosome DNA shows abnormal structures and possibly binds through topological entrapment of DNA strands (43,44). Thus, we suggest that instead of the IFN pathways, PJA1 coordinates with the SMC5/6 complex to recognize viral and episomal DNA structures and recruit DNA Tops to restrict the foreign DNA molecules. This work reveals a distinct mechanism of silencing of foreign DNA invaders in the nuclei and suggests that PJA1 may act as a potential agent for the prevention and treatment of infectious and inflammatory diseases.

MATERIALS AND METHODS
Cell cultures and generation of stable cell lines. African green monkey kidney (Vero) cells and human hepatoma (HepG2) cells were obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). Human hepatocarcinoma (Huh7) cells, human embryonic kidney (HEK293T) cells, and human rhabdomyosarcoma (RD) cells were obtained from the China Center for Type Culture Collection (CCTCC) (Wuhan, China). Cells were grown in Dulbecco's modified Eagle medium (DMEM) (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco), 100 U/ml penicillin, and 100 g/ml streptomycin sulfate at 37°C in 5% CO 2 . Cells were transfected with Lipofectamine 2000 (Invitrogen, IL, USA) according to the manufacturer's instructions.
To generate the stable cell lines HepG2-3Flag, HepG2-PJA1B, 293T-Luc, 293T-GFP, HepG2-sh-NC, HepG2-sh-PJA1, and HepG2-sh-SMC6, lentiviruses carrying the encoding gene or target interfering shRNA sequences were produced by cotransfecting the corresponding constructs, the psPAX2 and pMD2.G plasmids, into HEK293T cells. Cells stably expressing the genes of interest were selected with 2.5 g/ml puromycin for a week. The protein level for the target gene was verified by Western blotting. (Note that the titer of PJA1B-overexpressing lentiviral particles was generally dozens of times lower than that of control lentiviral particles.) Generation of PJA1 knockout cells. PJA1 knockout (PJA1-KO) cells were generated from HEK293T cells with the CRISPR-Cas9-2hitKO system according to the manufacturer's instructions (Beijing Pitoop Bioscientific Inc., Beijing, China). Target guide RNAs (target 1, 5=-CACCGTCTCCCCTGCCACATCGGTT-3=; target 2, 5=-CACCGTTCCACTACTCGTCGTAGTT-3=) were predicted online (Atum). 293T cells (5 ϫ 10 5 ) were transfected with 2 g of a CRISPR-Cas9 vector carrying two guide RNA-expressing cassettes in six-well tissue culture plates for 2 days. The cells were transferred to 10-cm dishes and cultured with medium containing puromycin (1.5 g/ml) for a week. The culture medium was changed every 2 to 3 days. Individual cell colonies were isolated by limiting dilution. After 1 to 2 weeks, the cells were observed under a microscope, and cells from those wells containing only one cell colony were selected and allowed to expand from a 96-well plate to a 6-well plate. Knockout efficiency was assessed by Western blotting and verified by genomic DNA sequencing.
Viruses and infection. For HBV infection, the HBV inoculum was concentrated 100-fold from the supernatants of HepaAD38 cells (provided by Ying Zhu of Wuhan University, China) by ultracentrifugation at 100,000 ϫ g for 5 h at 4°C. For infection, HepG2-NTCP cells (provided by Ying Zhu, Wuhan University, China) were plated on collagen I-coated plates in DMEM overnight, and the medium was then changed to PMM (Williams' E medium supplemented with insulin, transferrin, and sodium selenite (ITS), 2 mM L-glutamine, 10 ng/ml of human epidermal growth factor [EGF], 18 g/ml of hydrocortisone, 40 ng/ml of dexamethasone, 2% dimethyl sulfoxide [DMSO], 100 U/ml of penicillin, and 100 g/ml of streptomycin) for 6 h. Cells were then infected with 1,000 genome equivalents (GE) per cell of HBV in PMM containing 4% (wt/vol) polyethylene glycol 8000 (PEG 8000) for 16 h. Virus-containing medium was then removed, and the cells were washed five times and further incubated in PMM with 2% serum (45).
HSV-1 strain F (provided by Ying Zhu, Wuhan University, China) was propagated in Vero cells as previously described (46). For infection, cells were exposed to HSV-1 at the indicated multiplicity of infection (MOI) for 1 h at 37°C, washed twice with a phosphate-buffered saline (PBS) solution, and overlaid with fresh DMEM containing 10% FBS. Infected cells were incubated at 37°C for the indicated lengths of time. HSV-1 progeny were titrated on Vero cells as previously described (46). In brief, Vero cells were plated in six-well plates to high confluence (Ͼ80%), exposed to HSV-1 for 1 h, and maintained in DMEM-agarose for 72 h. The cells were fixed with 4% (wt/vol) paraformaldehyde for 30 min, washed three times with PBS, and stained with crystal violet dyes.
The recombinant enhanced green fluorescent protein (EGFP)-expressing vesicular stomatitis virus strain (VSV-GFP) (provided by Mingzhou Chen of Wuhan University) was amplified in Vero cells, and the titer was determined on Vero cells by fluorescence microscopy. Cells were incubated with VSV-GFP at the indicated MOI for 2 h, and unbound virus was washed away later.
The enterovirus 71 Xiangyang strain (GenBank accession no. JN230523.1) was isolated previously by our group (47). Cells were infected with EV71 at the indicated MOI after serum starvation overnight, and unbound virus was washed away 2 h later (48).
Luciferase assay. The determination of reporter luciferase activity was performed as described previously (49). In brief, cells were transfected with the indicated plasmids and a luciferase reporter plasmid for 24 h in 293T cells or for 48 h in HepG2 cells and then washed twice with ice-cold PBS. One hundred microliters of luciferase lysis buffer (Promega, Madison, WI, USA) was added to each well of a 24-well plate. Cells were lysed for 10 min at room temperature, and 50 l of each sample was then transferred to a new centrifuge tube and mixed with 15 l luciferase assay substrate (Promega, Madison, WI, USA). Luciferase activity was typically measured for 10 s after a 3-s delay by using a luminometer (TD-20/20; Turner Designs, Sunnyvale, CA, USA). All assays were performed in triplicate, and data are expressed as means Ϯ standard deviations (SD).