UHRF1 Deficiency Inhibits Alphaherpesvirus through Inducing RIG-I-IRF3-Mediated Interferon Production

ABSTRACT

Type I interferon (IFN-I) response plays a prominent role in innate immunity, which is frequently modulated during viral infection. Here, we report DNA methylation regulator UHRF1 as a potent negative regulator of IFN-I induction during alphaherpesvirus infection, whereas the viruses in turn regulates the transcriptional expression of UHRF1. Knockdown of UHRF1 in cells significantly increases interferon-β (IFN-β)-mediated gene transcription and viral inhibition against herpes simplex virus 1 (HSV1) and pseudorabies virus (PRV). Mechanistically, UHRF1 deficiency promotes IFN-I production by triggering dsRNA-sensing receptor RIG-I and activating IRF3 phosphorylation. Knockdown of UHRF1 in cells upregulates the accumulation of double-stranded RNA (dsRNA), including host endogenous retroviral sequence (ERV) transcripts, while the treatment of RNase III, known to specifically digest dsRNA, prevents IFN-β induction by siUHRF1. Furthermore, the double-knockdown assay of UHRF1 and DNA methyltransferase DNMT1 suggests that siUHRF1-mediated DNA demethylation may play an important role in dsRNA accumulation and subsequently IFN induction. These findings establish the essential role of UHRF1 in IFN-I-induced antiviral immunity and reveal UHRF1 as a potential antivrial target.

IMPORTANCE Alphaherpesviruses can establish lifelong infections and cause many diseases in humans and animals, which rely partly on their interaction with IFN-mediated innate immune response. Using alphaherpesviruses PRV and HSV-1 as models, we identified an essential role of DNA methylation regulator UHRF1 in IFN-mediated immunity against virus replication, which unravels a novel mechanism employed by epigenetic factor to control IFN-mediated antiviral immune response and highlight UHRF1, which might be a potential target for antiviral drug development.

INTRODUCTION

Type I Interferon (IFN-I) plays a critical role in defending against viral infection and regulating the immune response. Upon viral infection, mammalian cells trigger rapid IFN-I response through recognizing the pathogen-associated molecular patterns (PAMPs) by pattern recognition receptors (PRRs) (1). Viral RNA is recognized by TLRs and RIG-I-like receptors, while viral DNA is mainly recognized by the cyclic GMP-AMP synthase (cGAS) (2 to 4). These sensors further interact with distinct adaptor proteins, such as stimulator of interferon genes (STING) and mitochondrial antiviral signaling (MAVS), to activate transcription factors IRF3, IRF7, and NF-κB, leading to IFN-I production (5). The produced IFN-I induces the expression of IFN-stimulated genes (ISGs) via the JAK-STAT signaling pathway and functions to clear viral pathogens (6, 7). Additionally, IFN signaling is also subject to extensive regulation, and an increasing number of studies have found that additional coregulators are required to modulate the production of IFN and the transcription of ISGs. For instance, the Bcl-2-associated transcription factor 1 (Bclaf1) enhances the IFN-mediated antiviral gene transcription through maintaining STAT1/STAT2 phosphorylation and IFN-stimulated response elements (ISRE) activation (8). The methyltransferase SETD2 promotes IFN-α-dependent antiviral immunity via catalyzing STAT1 methylation on K525 (9). Thus, identifying novel regulators involved in IFN signaling will provide a deeper understanding of IFN immune response and new targets for antiviral strategies.

UHRF1 (ubiquitin-like containing PHD and RING finger domain 1) is an important DNA methylation regulator that recruits DNA-methyltransferase 1 (DNMT1) to hemimethylated DNA and maintains DNA methylation during mitosis (10). Studies have increasingly reported that UHRF1 participated in diverse biological processes, including immune cell development, differentiation, and antitumor immunity (11 to 14). Recently, a role for UHRF1 in cellular antiviral defense is emerging, and UHRF1 was identified as an IFN-I regulating protein during influenza virus infection. More strikingly, a single-nucleotide methylation site in the IFN-β promoter region was found as the target of UHRF1 to disrupt IRF3 recruitment (15). However, whether UHRF1 is also involved in DNA virus infection and the molecular mechanism responsible is not known.

Herpesviridae is a family of large double-stranded DNA viruses with an ability to establish persistent infection and cause various diseases in hosts. Herpes simplex virus 1 (HSV-1) and pseudorabies virus (PRV) belong to the alphaherpesvirus subfamily and are often used as model viruses to study alphaherpesvirus biology in different species (16). Previous studies indicated that HSV-1 and PRV had evolved multiple strategies to combat host IFN antiviral response and establish persistent infection, while the molecules that participate in the regulation of IFN signaling could be potential targets of herpesviruses (8, 17 to 20). In the present study, we identified an essential role of UHRF1 in IFN-mediated immunity against PRV/HSV-1 replication and found that the expression of UHRF1 was transcriptionally regulated during PRV and HSV-1 infection. More importantly, further mechanistical studies revealed UHRF1 deficiency inhibited alphaherpesvirus through activating RIG-I-IRF3-mediated IFN-I production, which may be related to DNA demethylation-induced dsRNA accumulation, including host endogenous retroviral sequence (ERV) transcripts. These results reveal a critical mechanism by which DNA methylation regulator UHRF1 controls IFN-mediated immune response during alphaherpesvirus infection.

RESULTS

HSV-1 and PRV affect UHRF1 expression.

To examine the effect of alphaherpesvirus infection on UHRF1, we infected human cells with HSV-1 and porcine cells with PRV (MOI = 1 or 0.1) and harvested the cells for Western blotting and qRT-PCR analysis. As shown in Fig. 1, a dramatic decrease in UHRF1 protein was observed in all the cells examined at the time points when substantial viral proteins (VP5, US3, or ICP4) were expressed, including human Hep2 and HeLa cells (Fig. 1A to C) and porcine kidney PK15 and porcine alveolar macrophage 3D4/31 (CRL-2844) cells (Fig. 1D to F). However, a significantly upregulated UHRF1 expression was observed at 6 and 12 h postinfection when the cells were infected with a low MOI of 0.1 (Fig. 1C and F). Consistent with the protein levels, the mRNA levels of UHRF1 were significantly reduced in a time-dependent manner when the cells were infected with an MOI of 1 (Fig. 1G and H) but were temporally increased followed by a downregulation at MOI of 0.1 (Fig. 1I and J). These results suggest that both PRV and HSV-1 are able to regulate UHRF1 transcription, while UHRF1 discrepancy changes may be affected by virus loads.

FIG 1.

HSV-1 and PRV affect UHRF1 expression. Human Hep2 cells and HeLa cells were infected with HSV-1 (MOI = 1 or 0.1). At different time points postinfection, as indicated, the cells were harvested and lysed for UHRF1 and virus protein (VP5, US3, or ICP4) detection by Western blotting (A to C), and the supernatants of Hep2 cells were collected for virus titers analysis by plaque assay (G and I). Porcine PK15 cells and 3D4/31 cells were infected with PRV (MOI = 1 or 0.1). At different time points postinfection, as indicated, the cells were harvested and lysed for UHRF1 and virus protein (VP5 and US3) detection by Western blotting (D to F), and the supernatants were collected for virus titer analysis by plaque assay (H and J). Data are shown as mean ± SD of three independent experiments. Statistical analysis was performed by a two-way ANOVA test. **, P < 0.01; ***, P < 0.001.

UHRF1 deficiency inhibits HSV-1 and PRV replication.

The up- and-downregulation of UHRF1 upon HSV-1/PRV infection suggests that UHRF1 may play an important role in the host antiviral immune system. To explore the role of UHRF1 in alphaherpesvirus infection, we first designed two siRNA targeting different domains of the UHRF1-encoding region (siUHRF1-1 and siUHRF1-2) and transfected them into human Hep2 and A549 cells, respectively. Twenty-four hours after transfection, the cells were infected with HSV-1 and the viral proteins US3 and VP5 in HSV-1-infected cells were examined by Western blotting (Fig. 2A), and infectious viral particles in the culture medium of Hep2 cells were measured by plaque assay (Fig. 2B). The results showed that transfection with siUHRF1 efficiently restricted expression of UHRF1 compared to the nonsilencing control (siCtrl), and disrupting UHRF1 caused a significant reduction in the expression of viral proteins (Fig. 2A) and in viral loads (Fig. 2B). Furthermore, qPCR analysis was performed to assess the copies of HSV-1 viral genome with specific primers corresponding to the HSV-1 immediately-early gene ICP27-encoding region (Fig. 2C) and the mRNA levels of HSV-1 viral genes with primers specific to ICP27, ICP8, and VP16 (Fig. 2D). In agreement with the above observations, the copies of viral genomic DNA (Fig. 2C) and the transcriptional expression of viral genes (Fig. 2D) were also significantly reduced in UHRF1-silencing cells compared with that in the nonsilencing control cells, demonstrating that knockdown of UHRF1-inhibited HSV-1 replication.

FIG 2.

UHRF1 deficiency inhibits HSV-1 and PRV replication. (A) Human Hep2 cells and A549 cells were transfected with nonsilencing control (siCtrl) or UHRF1 siRNA (siUHRF1 number 1 or siUHRF1 number 2), and then infected with HSV-1 (MOI = 0.1). At different time points postinfection, as indicated, the cells were harvested and lysed for virus protein detection by Western blotting. (B) Hep2 cells were transfected with siCtrl or siUHRF1 and then infected with HSV-1 (MOI= 0.1 or 1). At different time points postinfection, as indicated, the supernatants of Hep2 were collected for virus titer analysis by plaque assay. (C and D) Hep2 cells were transfected with siCtrl or siUHRF1 and then were infected with HSV-1 (MOI = 0.1) for the indicated time. Genomic DNA was isolated and subjected to qPCR analysis with primers specific to the HSV-1 ICP27 coding region. The relative copy of the viral genome was normalized to GAPDH in each sample (C). Total RNA was isolated and subjected to qRT-PCR analysis with the primers specific to HSV-1 viral genes ICP27, ICP8, and VP16. The relative mRNA levels of ICP27, ICP8, and VP16 were normalized to 18S rRNA in each sample (D). (E) Porcine 3D4/31 cells and PK15 cells, as well as Vero cells, were transfected with siCtrl or siUHRF1 and then infected with PRV (MOI = 0.1). At different time points postinfection, as indicated, the cells were harvested and lysed for virus protein detection by Western blotting. (F) PK15 cells were transfected with siCtrl or siUHRF1 and then infected with PRV (MOI= 0.1 or 1). At different time points postinfection, as indicated, the supernatants of PK15 were collected for virus titer analysis by plaque assay (F). Data are shown as mean ± SD of three independent experiments. Statistical analysis was performed by two-way ANOVA test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

To further verify the effect of disrupting UHRF1 on alphaherpesviruses, porcine 3D4/31 and PK15 cells, as well as Vero cells, were transfected with siUHRF1 and then infected with PRV. Consistently, a significantly lower level of viral proteins (US3 and VP5) (Fig. 2E) and decreased viral loads (Fig. 2F) of PRV were observed in UHRF1-silencing porcine cells compared with that in nonsilencing control (siCtrl) cells, emphasizing the inhibitory effect of UHRF1 deficiency on alphaherpesvirus replication. Notably, the apparent siUHRF1-induced decrease in PRV production was only observed in early infection of MOI 1 (Fig. 2F, lower panel, 6 hours postinfection [hpi]), which was paralleled by the rapid downregulation of UHRF1 in infection of high MOI in PK15 cells (Fig. 1D). Therefore, we speculate that the rapid and substantial degradation of UHRF1 by PRV in late infection of MOI 1 may result in the diminished influence of siUHRF1 in virus replication.

Additionally, we also observed that the knockdown of UHRF1 in Vero cells had no influence in PRV replication in infection of MOI 0.1 (Fig. 2E). Since Vero cells have been reported as a cell line with a deficient interferon-mediated antiviral response, we thus speculated that UHRF1 might regulate antiviral response through interferon signaling pathways.

UHRF1 deficiency inhibits alphaherpesvirus through specifically upregulating IFN-I production.

To determine whether UHRF1 was involved in interferon-mediated antiviral response, we depleted UHRF1 using siRNAs in Hep2 and HIEC-6 cells and then analyzed the transcription of IFN-I and the activation of interferon-stimulated genes (ISG15 and ISG56) by qPCR (Fig. 3A). As expected, the transcription levels of IFN-β, ISG15, and ISG56 in UHRF1 knockdown cells were significantly increased compared with that in control cells. Correspondingly, UHRF1 deficiency in Hep2 cells significantly enhanced the expressions of the critical signaling components involved in the JAK-STAT pathway in response to IFN-α treatment (Fig. 3B), confirming the regulation of UHRF1 to the type-I IFN signaling pathway.

FIG 3.

UHRF1 deficiency inhibits alphaherpesvirus through specifically upregulating IFN-I production. (A and B) Hep2 cells or HIEC-6 cells transfected with siCtrl or siUHRF1 were treated with and without IFN-α for 12 h. The cells were then collected to examine, by Western blotting, the mRNA levels of IFN-β, ISG15, and ISG56 by qRT-PCR (A) and the protein levels of relevant components of IFN-I signaling (JAK1, TYK2, P-STAT1, P-STAT2, STAT1, STAT2, and IRF9) and downstream ISGs in Hep2 cells (B). (C to E) HeLa wild-type (WT) or IRF9-KO cells transfected with siCtrl or siUHRF1 were infected with HSV-1 (MOI = 0.1) for the indicated hours. Cells were then collected to examine the mRNA levels of UHRF1, type-I IFN, and ISGs by qRT-PCR (C). (D) The protein levels of VP5, US3, IRF9, P-STAT1, and ISG15 were detected by Western blotting. The relative protein levels of UHRF1 were quantified by densitometry and normalized to the levels of Tubulin. (E) The supernatants of HeLa WT or IRF9-KO cells were collected for virus titer analysis by plaque assay. Data are shown as mean ± SD of three independent experiments. Statistical analysis was performed by two-way ANOVA test. **, P < 0.01; ***, P < 0.001.

To further determine the essential role of UHRF1 in promoting IFN production during virus infection, we constructed a IRF9-knockout HeLa cell line (IRF9-KO) to disrupt the JAK-STAT pathway, and then examined the effect of UHRF1 depletion on the mRNA levels of IFN-I and ISGs (Fig. 3C), the expression of viral proteins (Fig. 3D), and the infectious viral particles (Fig. 3E) in IRF9-KO and wild-type (WT) cells infected with HSV-1. Compared with their respective controls, UHRF1 deficiency by RNAi significantly increased the transcriptional activation of IFN-ɑ and IFN-β in both IRF9-KO and WT cells, whereas it only upregulated the ISG15 and ISG56 transcription in WT cells but not in IRF9-KO cells (Fig. 3C). Meanwhile, compared with the nonsilencing control cells, UHRF1 depletion led to significantly lower levels of viral proteins (US3 and VP5) (Fig. 3D) and a remarkable decrease in viral loads (Fig. 3E) of HSV-1 in HeLa WT cells, but not in IRF9-KO cells, indicating that UHRF1 deficiency inhibits alphaherpesvirus through specifically upregulating IFN-I production. Notably, as shown in Fig. 3D, knocking out IRF9 remarkably increased UHRF1 expression, suggesting IRF9 might be a negative regulator of UHRF1. To confirm the efficiency of UHRF1 knocking down in WT and IRF9-KO cells, the relative protein levels of UHRF1 were quantified by densitometry (Fig. 3D). The same low expression levels of UHRF1 in two cell lines at 24 hpi verified that the increased virus replication in IRF9-KO cells was due to the disruption of the JAK-STAT pathway instead of the increase of UHRF1.

Regulation of IFN-I by UHRF1 is dependent on IRF3-mediated signal transduction.

To determine the target of UHRF1 in regulating IFN production during virus infection, Hep2 cells were transfected with siUHRF1 or siCtrl followed by HSV-1 infection (MOI = 0.1) or nucleic acid analogues Poly(dA:dT) stimulation, and then the proteins of IRF3, P-IRF3, P-TBK1, P65, and P-P65, as well as the downstream signaling P-STAT1/2 and ISGs, were examined by Western blotting. The results found that depleting UHRF1 effectively induced the phosphorylation of TBK-1/IRF3 but did not affect P65 expression and its phosphorylation in HSV-1-infected cells (Fig. 4A). Consistently, the protein levels of P-IRF3 were also remarkably upregulated in UHRF1 knockdown cells responding to Poly(dA:dT) (Fig. 4B). To further verify whether UHRF1 regulates IFN-I induction by targeting IRF3 phosphorylation, a siRNA targeting IRF3 was designed and cotransfected with siUHRF1 into HeLa cells. Western blot analysis showed that the knockdown of IRF3 abrogated siUHRF1-induced STAT1/2 phosphorylation and ISG15 expression (Fig. 4C). Moreover, qRT-PCR analysis revealed that the siUHRF1-induced transcriptional activation of IFN-β and ISGs (ISG15 and ISG56) was severely impaired in cells knocking down IRF3 (Fig. 4D). Collectively, these data demonstrate that UHRF1 regulates IFN-I induction depending on IRF3-mediated signal transduction.

FIG 4.

Regulation of IFN-I by UHRF1 is dependent on IRF3-mediated signal transduction. (A) Hep2 cells transfected with siCtrl or siUHRF1 were infected with HSV-1 (MOI = 0.1) for the indicated hours. The cells were then collected to examine the protein levels of relevant components of IFN-I signaling (IRF3, P-IRF3, P-TBK1, P65, P-P65, P-STAT1, P-STAT2, STAT1, STAT2) and downstream ISGs by Western blotting. (B) Hep2 cells transfected with siCtrl or siUHRF1 were treated with or without Poly(dA:dT) for 12 h. Cells were then collected for western blots analysis of the relevant components of IFN-I signaling (P-IRF3 and ISG15). (C and D) HeLa cells were transfected with nonsilencing control (siCtrl) or siRNA specifically targeting UHRF1 and IRF3 for 24 h. The cells were then collected and lysed for Western blot analysis of relevant components of IFN-I signaling (P-STAT1, P-STAT2, IRF3, IRF7) and downstream ISG15 (C), and for qRT-PCR analysis of IRF3, IFN-β, and ISGs mRNA (D). Data are shown as mean ± SD of three independent experiments. Statistical analysis was performed by two-way ANOVA test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

UHRF1 deficiency induces IFN-I response by triggering RIG-I-MAVS pathway.

To further elucidate the mechanism by which UHRF1 regulates IFN transactivation, the key cytosolic sensors and adaptor responsible for initiating IFN induction, including RIG-I, MDA5, and MAVS, were screened using RNA interference (RNAi) in HeLa and Hep2 cells to determine the target of UHRF1. Western blot analysis showed that the knockdown of RIG-I and MAVS, but not MDA5, could effectively suppress the siUHRF1-induced STAT1/2 phosphorylation and ISG15 expression (Fig. 5A). Furthermore, the knockdown of adaptor protein MAVS was also sufficient to inhibit the upregulated expression of P-TBK1, P-STAT1/2, and ISGs (Fig. 5B), as well as the transcriptional activation of IFN-β and ISGs (Fig. 5C) induced by UHRF1 depletion. Accordingly, the suppressed virus titers of HSV-1 by siUHRF1 were also restored in either siRIG-I or siMAVS cells (Fig. 5D), suggesting that UHRF1 might regulate IFN response by triggering the RIG-I-MAVS pathway.

FIG 5.

UHRF1 deficiency activates RIG-I-MAVS-mediated interferon transactivation. (A) HeLa cells were transfected with nonsilencing control (siCtrl) or siRNA specifically targeting UHRF1 and a cytosolic sensor of innate immune system (RIG-I, MDAS, cGAS, TLR3, TLR7, TLR9, or MAVS) for 24h. The cells were then collected and lysed for Western blot analysis of relevant components of IFN-I signaling (P-STAT1, P-STAT2, IRF7) and downstream ISGs (left panels). The efficiency of UHRF1, RIG-I, MDA5, and MAVS knockdown were verified by qPCR (right panels). (B to D) Hep2 cells were transfected with nonsilencing control (siCtrl) or siRNA specifically targeting UHRF1 or/and MAVS for 24h. The cells were then collected and lysed for Western blot analysis of relevant components of IFN-I signaling (P-TBK1, IRF7, P-STAT1, P-STAT2) and downstream ISGs (B), and for qRT-PCR analysis of relevant mRNA (UHRF1, MAVS, IFN-β, ISG15, and ISG56) (C). 2KD represents siUHRF1 plus siMAVS. The supernatants of Hep2 transfected with siCtrl, siUHRF1, siUHRF1+siMAVS, or siUHRF1+siRIG-I were collected for virus titer analysis by plaque assay (D). Data are shown as mean ± SD of three independent experiments. Statistical analysis was performed by two-way ANOVA test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

UHRF1 regulates interferon signaling through DNA demethylation-mediated dsRNA activation.

RIG-I is known as a cytosolic sensor that triggers innate immune activation through recognizing short double-stranded RNA (dsRNA) generated during viral replication. How UHRF1 deficiency triggers the RIG-I-mediated sensing pathway remains obscure. Endogenous retroviral sequences (ERVs) were found in the human genome, and the increased ERV transcripts, much like the pathogen-associated molecular patterns (PAMPs) of exogenous viruses, could be recognized by cytosolic RNA sensors (21). Increasing evidence has shown that the DNA-demethylating therapy could upregulate ERV transcripts and activate interferon response in human cancer cell lines (22, 23). Since UHRF1 is an important epigenetic regulator that plays an essential role in DNA methylation, we thus determined whether UHRF1 depletion upregulated ERV transcripts. Several ERV transcripts (including MER57BP1, MLT1C49, MLT1B, and MER4D) were examined by qRT-PCR in HeLa cells transfected with siUHRF1 or siCtrl, and the results showed that UHRF1 deficiency led to a significantly increased expression of specific ERV transcripts (Fig. 6A). Next, we applied the J2 antibody, which specifically recognizes dsRNA to determine whether UHRF1 depletion leads to the accumulation of dsRNA. The significantly increased dsRNA level was observed in UHRF1-depleted cells compared to the siCtrl cells (Fig. 6B). To determine whether dsRNA generated in cells transfected with siUHRF1 is involved in IFN induction, RNAs extracted from siUHRF1-transfected cells were treated with RNase III, which is known to specifically digest dsRNA (24), and then transfected into HeLa cells for qPCR analysis with primers specific to UHRF1, ERVs, IFN-β, and ISGs (Fig. 6C), or transfected into HEK293 cells for an IFN-β-specific luciferase-based reporter assay (Fig. 6D). When the RNA from siUHRF1-transfected cells was treated with RNase III, it lost its ability to induce IFN-β and ISG15 transcription (Fig. 6C), as well as to trigger IFN-β activation (Fig. 6D), compared with that from siCtrl-transfected cells, indicating that the increased IFN-β production by UHRF1 deficiency is a consequence of dsRNA accumulation. As known, UHRF1 functions as an essential partner in maintaining DNA methylation with DNA methyltransferase DNMT1, which directly methylates the newly synthesized daughter strand. To further test whether an association exists between DNA demethylation and UHRF1-mediated interferon regulation, we knocked down DNMT1 by siRNA (siDNMT1) in HeLa cells with/without UHRF1 deficiency. As shown in Fig. 6D, knocking down DNMT1 could individually increase the transcripts of ERVs, IFN-β, and the downstream genes ISG15/56, but did not form a synergistic enhancement in their transcriptions with siUHRF1, indicating that DNA demethylation may play an important role in siUHRF1-mediated increase of multiple ERV transcripts and induction of IFN. Meanwhile, the siDNMT1 also exhibited a similar effect with siUHRF1 in promoting the expression of interferon signals (P-STAT1 and ISG15) in HeLa cells (Fig. 6E) and in inhibiting virus replication (Fig. 6F). These data suggest that the IFN induction in UHRF1-deficient cells may be activated via DNA demethylation-induced dsRNA, including endogenous retroviral sequences.

FIG 6.

UHRF1 deficiency induces interferon signaling through DNA demethylation-mediated dsRNA activation. (A) Disrupting UHRF1induces specific human endogenous retrovirus transcripts (ERVs). HeLa cells were transfected with siCtrl or siUHRF1 and/or siDNMT1, and then collected for qPCR analysis with primers specific to ERV transcripts, IFN-β, and ISGs. (B) HeLa cells were transfected with siCtrl or siUHRF1 and then were visualized by immunofluorescence with dsRNA-specific J2 antibody (green). The nucleus was stained with DAPI (blue) (bar = 10 μm) (left panels). The dsRNA intensity was quantified by Image J software (right panels). (C) RNAs extracted from siCtrl- or siUHRF1-transfected HeLa cells were treated with 4U RNase III and then were transfected into HeLa cells for qPCR analysis with primers specific to UHRF1, ERVs, IFN-β, and ISGs. (D) RNAs extracted from siCtrl- or siUHRF1-transfected HeLa cells were digested with 4U RNase III and then transfected into HEK293 cells for an IFN-β-specific luciferase-based reporter assay. (E and F) DNMT1i induces interferon signaling. HeLa cells were transfected with siCtrl or siUHRF1 and/or siDNMT1, then collected for qPCR analysis with primers specific to ERV transcripts, IFN-β, and ISGs (E) and for Western blotting with antibodies against P-STAT1 and downstream ISG15. (F and G) Hep2 cells were transfected with siCtrl or siUHRF1and/or siDNMT1, and then infected with HSV-1 (MOI= 0.1). Twenty-four hours postinfection, the supernatants of Hep2 cells were collected for virus titer analysis by plaque assay. Data are shown as mean ± SD of three independent experiments. Statistical analysis was performed by Student’s t tests (B and D) or one-way or two-way ANOVA test (A, C, E, G). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Overall, these data provide evidence that UHRF1 deficiency promotes IFN-I production by triggering dsRNA-sensing receptor RIG-I and activating IRF3 phosphorylation, leading to inhibition of alphaherpesvirus replication.

DISCUSSION

The IFN response is critical in the control of viral infection and is often modulated during various virus infection. Most current studies emphasize the known signaling transduction and transcription factors in the IFN pathway, but epigenetic factors less. Here, we revealed the mechanism of DNA methylation regulator UHRF1 in promoting IFN production during alphaherpesvirus PRV and HSV-1 infection, and determined the effects of PRV/HSV-1 on UHRF1 expression.

DNA methylation is one of the most intensely studied epigenetic modifications in mammals. In normal cells, DNA methylation plays a pivotal role in histone modifications, X chromosome inactivation, genomic imprinting, and transposable element repression (25), while in virus infection and tumor microenvironments, it is significantly correlated with immune infiltration (13, 15, 26 to 28). As a key regulator of DNA methylation, the role of UHRF1 in innate immunity remains poorly investigated. A recent study demonstrated that UHRF1 deficiency could significantly upregulate IFN-β transcription by removing the methylation modification in the IFN-β promoter region and promoting the IRF3 recruitment (15). In agreement with this finding, we identified UHRF1 as a negative regulator of IFN-I production (Fig. 3) and demonstrated its antiviral function during virus infection (Fig. 2). However, inconsistent with the previous study, which suggested that UHRF1 does not affect canonical antiviral signal transduction of IFN-I (including NF-κB, MAPK, and IRF3 activation) in VSV- or IAV-infected cells (15), our data revealed that the activation of the RIG-I/MAVS/IRF3 pathway by dsRNAs is the main mechanism responsible for the antiviral effects of UHRF1 during HSV-1 infection (Fig. 4 and 5). The relationship between DNA methylation and gene expression is complex because the modification occurs in different regions and specific cell types; thus, the specific mechanisms by which DNA methylation regulator UHRF1 affects antiviral innate immunity may depend on different virus species and cell types.

Herpesviruses can establish long-term latent infections in which immune evasion is a pivotal step, and the molecules that modulate IFN production are usually the key targets of viral immune evasion. To determine whether alphaherpesviruses affect UHRF1, we examined the protein and mRNA levels of UHRF1 in PRV/HSV-1-infected cells with an MOI of either 1 or 0.1 (Fig. 1). On one hand, as expected, the expression of UHRF1 was significantly upregulated by PRV/HSV-1 at the early stage of infection with a low concentration of virus (MOI = 0.1), which could be an immune-evasion strategy of herpesviruses to maintain DNA methylation and antagonize IFN antiviral response. There are at least two potential mechanisms by which HSV-1/PRV stimulates UHRF1 expression within 12 to 24 hpi at a low MOI. As shown in Fig. 3D, knocking out IFN regulation factor IRF9 could remarkably increase UHRF1 expression, suggesting IRF9 as a negative regulator of UHRF1. Our recent study has identified PRV early protein EP0 as an inhibitor of IRF9 during PRV infection (29); we thus speculate that the strong anti-IRF9 function of EP0 at the early stage of virus infection may lead to the upregulation of UHRF1. The preliminary data have proved that the EP0-deficient PRV (ΔEP0) could not trigger a temporal increase in UHRF1 early in infection at MOI 0.1 (unpublished data), but how IRF9 regulates UHRF1 still needs further investigation. Alternatively, considering that most cells will not be infected at early time points postinoculation under the condition of MOI 0.1, an alternative explanation for the temporal increase in UHRF1 could be that virus does not directly lead to UHRF1 upregulation in the infected cells but perhaps may indirectly lead to UHRF1 upregulation in noninfected bystander cells early in infection, which should be further demonstrated by viral visual experiments.

On the other hand, surprisingly, there was a quite rapid and dramatic decrease in UHRF1 expression in cells with virus infection at MOI = 1 or at the late stage of MOI = 0.1, especially with PRV, which is seemingly contradictory to the immune-evasive nature of herpesviruses. The molecular mechanism responsible is still unknown, but we speculate that it may be related to the virus-mediated host transcriptional shutoff, because alphaherpesviruses like PRV and HSV-1 have been reported to trigger broad inhibition of host gene transcription (30, 31). We also found that the inactivated viruses and nucleic acid analogues did not exhibit a similar inhibitory influence in UHRF1 (data not shown), suggesting the transcriptional inhibition activity of viruses may be necessary for the downregulation of UHRF1. Also, the half-life of endogenous UHRF1 protein in porcine and human cells was very short, especially in porcine cells (data not shown), which would allow a rapid downregulation of UHRF1 protein levels upon host transcriptional shutoff during PRV/HSV-1 infection, and thereby release of the suppression of IFN-I expression. All these findings suggest that the downregulation of UHRF1 may be due to the host-driven sensitive and fine regulation under the condition of viral transcriptional inhibition, but the specific mechanism responsible needs further studies.

DNA methyltransferase inhibitors (DNMTis) such as 5-AZA and Dac are effective cancer therapies in hematologic neoplasms (32, 33). 5-AZA targets cancer cells by upregulating ERVs via DNA demethylation and then inducing the IFN-I response (34). In addition to treatment with a methylation inhibitor, our data show that UHRF1-mediated demethylation can also upregulate IFN-I production and further inhibit virus replication, suggesting that UHRF1 may be a good target for antiherpesvirus, even antitumor, drug development.

MATERIALS AND METHODS

Cell culture and viruses.

Hep2 cells (human laryngeal cancer cells), A549 cells (human lung adenocarcinoma cells), HeLa cells (human cervical carcinoma cells), HIEC-6 cells (human normal embryonic intestinal epithelial cells), Vero cells (monkey kidney epithelial cells), PK15 cells (porcine kidney epithelial cells), and 3D4/31 cells (porcine alveolar macrophage cells, also named CRL-2844) were cultured in Dulbecco's minimum essential medium (DMEM) supplemented with 10% (vol/vol) fetal bovine serum (FBS) and penicillin (100U/mL)-streptomycin (100 μg/mL). The cell culture medium, serum, and antibiotics were purchased from Invitrogen. All cells were maintained at 37°C in 5% CO₂.

The PRV (Bartha-K61) and HSV-1 (KOS strain) were described previously (32, 33).

Antibodies and reagents.

The antibodies against PRV VP5, PRV US3, HSV-1 VP5, and HSV-1 US3 were described previously (8, 17). Rabbit anti-UHRF1 and rabbit anti-IRF3 were purchased from Abcam. Rabbit anti-JAK1, rabbit anti-TYK2, rabbit anti-STAT1, rabbit anti-STAT2, rabbit antiphosphorylated STAT1 (P-STAT1), rabbit antiphosphorylated STAT2 (P-STAT2), rabbit antiphosphorylated TBK1 (P-TBK1), rabbit anti-P65, rabbit anti-IRF9, and rabbit anti-α-Tubulin antibodies were purchased from Cell Signaling Technology. Mouse anti-ISG15 and horseradish peroxidase (HRP)-conjugated goat anti-mouse and anti-rabbit secondary antibodies were purchased from Santa Cruz Biotechnology. J2 antibody against double-stranded RNA was from Scicons Biotechnology. The FITC-conjugated goat anti-rabbit secondary antibody and DAPI (4-,6-diamidino-2-phenylindole) were from DingGuoChangSheng Biotech Co., Ltd., Beijing.

Recombinant porcine IFN-α was a gift from Wenjun Liu (Chinese Academy of Sciences, Beijing, China). Recombinant human IFN-α was purchased from Peprotech. Poly(dAdT) was purchased from InvivoGen. The RNase III was from Biosystems, and cycloheximide (CHX) was from Amresco.

RNA interference and plasmid construction.

UHRF1 and DNMT1 RNAi sequences were found from the Genetic Perturbation Platform (https://portals.broadinstitute.org/gpp/public/) and synthesized from GenePharma (Shang-hai, China). The siRNA sequences are listed below:

siUHRF1:number 1, ATGTGGGATGAGACGGAATTG

siUHRF1:number 2, GCUGACCAUGCAGUAUCCATT

siDNMT1:number 1, CGAGUCUGGUUUGAGAGUTT

siDNMT1:number 2, GGAAUGGCAGAUGCCAACAGCTT

porcine-siUHRF1:ACATGGGACGAGACGGAGTTG

siIRF3:CCCUUCAUUGUAGAUCUGATT

siRIG-I:AUCACGGAUUAGCGACAAA

siMDA5:GUAUCGUGUUAUUGGAUU

siMAVS:UAGUUGAUCUCGCGGACGA.

The full-length UHRF1 was cloned from the human cDNA library and subcloned into the pRK5 vector with an N-terminal FLAG tag. The primers used for UHRF1-related gene amplification are listed in Table 1.

TABLE 1.

Primers used for UHRF1-related gene cloning

Primer	Sequence (5′–3′)
UHRF1 (WT)-F	GACGATGACAAGGGATCCATGTGGATCCAGGTTCGG
UHRF1 (WT)-R	GCCATGGCGGCCAAGCTTTCACCGGCCATTGCCGTA
UHRF1 (anti-siRNA)-F1	GACGATGACAAGGGATCCATGTGGATCCAGGTTCGG
UHRF1 (anti-siRNA)-R1	TACAGCCCTA GCTCTGTTTCGTCCCACATGTCCTCA
UHRF1 (anti-siRNA)-F2	TGAGGACATGTGGGACGAAACAGAGCTAGGGCTGTA
UHRF1 (anti-siRNA)-R2	GCCATGGCGGCCAAGCTTTCACCGGCCATTGCCGTA

Western blot analysis.

Whole-cell lysates were collected and lysed in lysis buffer (50 mM Tris-Cl [pH 8.0], 150 mM NaCl, 1.0% Triton X-100, 10% glycerol, 20 mM NaF, 1 mM DTT [dithiothreitol], and 1× complete protease mixture). The lysates were subjected to SDS-PAGE and transferred to nitrocellulose membranes. The membranes were blocked with 5% skim milk in PBST (PBS containing 0.5% Tween 20) for 2 h at room temperature and then incubated with specific primary antibodies overnight at 4°C, followed by secondary antibodies for 45 min at room temperature. The reactive protein bands were visualized using an enhanced chemiluminescence reagent with a Tanon-5200 luminescent imaging workstation.

Real-time quantitative PCR and RT-PCR.

Genomic DNA was isolated from HSV-1-infected Vero cells. Real-time quantitative PCR (qPCR) was performed to measure HSV-1 viral DNA. Total RNA isolation, cDNA synthesis, and quantitative RT-PCR (qRT-PCR) were used to measure HSV-1 viral gene expression. Primers corresponding to the ICP27 gene were used to measure the HSV-1 viral genome; the relative genome copy numbers were calculated based on the normalization with the housekeeping gene GAPDH.

Total RNA was extracted with TRIzol reagent (Invitrogen). A reverse transcription system (Qiagen) was used to synthesize cDNA. SYBR green PCR mix (CWBiotech) and ViiATM7 real-time PCR system (Applied Biosystems) were used for quantitative RT-PCR (qRT-PCR). 18S rRNA (for HSV-1 viral gene expression) or GAPDH was used as a housekeeping gene to normalize the target genes. Gene-specific primers for qPCR are listed in Table 2.

TABLE 2.

Primers used for quantitative RT-PCR

Primer	Sequence (5′–3′)
HSV-1-ICP8-F	ACAGCTGCAGATCGAGGACT
HSV-1-ICP8-R	CCATCATCTCCTCGCTTAGG
HSV-1-ICP27-F	TCCGACAGCGATCTGGAC
HSV-1-ICP27-R	TCCGACGAGGAACACTCC
HSV-1-VP16-F	GCGCTCTCTCGTTTCTTCC
HSV-1-VP16-R	GGCCAACACGGTTCGATA
GAPDH-F	AACGACCCCTTCATTGACCT
GAPDH-R	ATGTTAGTGGGGTCTCGCTC
UHRF1-F	CGACGGAGCGTACTCCCTAG
UHRF1-R	TCATTGATGGGAGCAAAGCA
DNMT1-F	CGTGGTGGTGGATGACAAG
DNMT1-R	GGCTCCCCGTTGTAGGAGAT
IFN-α-F	AATGACAGAATTCATGAAAGCGT
IFN-α-R	GGAGGTTGTCAGAGCAGA
IFN-β-F	GCCATCAGTCACTTAAACAGC
IFN-β-R	GAAACTGAAGATCTCCTAGCCT
ISG15-F	CAGATCACCCAGAAGATCG
ISG15-R	CCCTTGTTATTCCTCACCAG
ISG56-F	ACACCTGAAAGGCCAGAATGAGGA
ISG56-R	TGCCAGTCTGCCCATGTGGTAATA
IRF3-F	CGGAAGCTTCTGAAGCGGCTGTTGGTG
IRF3-R	GTGCTCGAGACCATGAGGAGCGAGGGC
RIG-I-F	CTGGACCCTACCTACATCCTG
RIG-I-R	GGCATCCAAAAAGCCACGG
MDA5-F	GCCCGCTACATGAACCCTG
MDA5-R	CAGCAATCCGGTTTCTGTCTT
MAVS-F	ATGCCGTTTGCTGAAGAC
MAVS-R	CTAGTGCAGACGCCGCCG
MLT1C49-F	TATTGCCGTACTGTGGGCTG
MLT1C49-R	TGGAACAGAGCCCTTCCTTG
MER57BP1-F	CCTCCTGAGCCAGAGTAGGT
MER57BP1-R	ACCAGTCTGGCTGTTTCTGT
MTL2B4-F	GGAGAAGCTGATGGTGCAGA
MTL2B4-R	ACCAACCTTCCCAAGCAAGA
MER34-F	GAATTCAGTGCCACTAAGCAGAC
MER34-R	TCGGTATATCCAAGACATGATCC
18S-F	CGGCTACCACATCCAAGGAA
18S-R	GCTGGAATTACCGCGGCT

Immunofluorescence staining.

Cells were fixed with 2% of paraformaldehyde for 15 min, permeabilized with 0.2% Triton X-100 for 15 min on ice, and blocked with 1% BSA (bovine serum albumin) in PBS for 30 min. The cells were then incubated with primary antibody against double-stranded RNA (J2, 1:200) and FITC-conjugated goat anti-rabbit secondary antibody (1:1,000). Nuclei were stained with DAPI for 3 to 5 min. Images were captured using a Nikon Eclipse Ni-E microscope. The captured images were processed and analyzed using SPOT software (Nikon).

Luciferase reporter assay.

A luciferase reporter assay was performed as described previously (19). Briefly, HEK293 cells were seeded in 24-well plates and transfected with 100 ng of luciferase reporter plasmid pGL3-IFN-β-Luc and 10 ng of pRL-TK (renilla) by using JetPRIME DNA transfection reagent (Polyplus-Transfection SA). Twelve hours after transfection, cells were transfected with the RNA extracted from siCtrl- or siUHRF1-transfected HeLa cells for 12 h, and were harvested and analyzed for luciferase activities using a Dual luciferase reporter assay kit (Promega), according to the manufacturer’s instruction. Data shown are representative of three independent experiments done in duplicate.

Virus infection and plaque assay.

PRV or HSV-1 was propagated and tittered in Vero cells. To infect, the cells were incubated with virus for 1 h, then washed with PBS and incubated in DMEM supplemented with 5% FBS until the time indicated. The viral yields of PRV or HSV-1 were determined by plaque assay in Vero cells. Briefly, the collected supernatants from virus-infected PK15 cells were cleared of cell debris by centrifugation and then used to infect Vero cells in triplicate with serial dilutions for 1 h in serum-free DMEM. After washes with PBS, the cells were overlaid with 1× DMEM–1% agarose and incubated at 37°C until plaque formation was observed (72 to 96 h). The cells were stained with 0.5% neutral red for 4 to 6 h at 37°C, and the plaques were counted.

Statistical analysis.

Statistical analysis was performed using GraphPad Prism software to perform analysis of Student's t test or variance (ANOVA) on at least three independent replicates. P values of <0.05 were considered statistically significant for each test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.