Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jan 15.
Published in final edited form as: Virology. 2019 Nov 11;540:75–87. doi: 10.1016/j.virol.2019.11.011

Protein inhibitor of activated STAT1 (PIAS1) inhibits IRF8 activation of Epstein-Barr virus lytic gene expression

Kun Zhang a,b,#, Dong-Wen Lv a,b,*,#, Renfeng Li a,b,c,d,#
PMCID: PMC6957754  NIHMSID: NIHMS1543283  PMID: 31743858

Abstract

Epstein-Barr virus (EBV), a major human oncogenic pathogen, establishes life-long persistent infections. In latently infected B lymphocytes, the virus persists as an episome in the nucleus. Periodic reactivation of latent virus is controlled by both viral and cellular factors. Our recent studies showed that interferon regulatory factor 8 (IRF8) is required for EBV lytic reactivation while protein inhibitor of activated STAT1 (PIAS1) functions as an EBV restriction factor to block viral reactivation. Here, we show that IRF8 directly binds to the EBV genome and regulates EBV lytic gene expression together with PU.1 and EBV transactivator RTA. Furthermore, our study reveals that PIAS1 antagonizes IRF8/PU.1-mediated lytic gene activation through binding to and inhibiting IRF8. Together, our study establishes IRF8 as a transcriptional activator in promoting EBV reactivation and defines PIAS1 as an inhibitor of IRF8 to limit lytic gene expression.

Keywords: Epstein-Barr virus (EBV), IRF8, PU.1, PIAS1, RTA, BGLF2, Gene expression, Latency, Reactivation

Introduction

As a major human pathogen, Epstein-Barr virus (EBV) is associated with 200,000 new cases of cancer and 140,000 deaths annually (Cohen et al., 2011). EBV encodes about 80 genes and the life cycle of EBV can be divided into latent or lytic phase. The lytic phase of EBV is a cascade of regulated gene expression: immediate-early, early and late (Young et al., 2007). ZTA (also called ZEBRA, EB1, BZLF1 or Z) and RTA (also known as BRLF1 or R) are two immediate-early gene products, which trigger the expression of EBV early and late genes. The early gene products are required for viral DNA replication, whereas the late gene products are viral structural proteins for packaging of infectious virions. In addition to viral transcription factor ZTA and RTA, cellular factors also play key roles in lytic gene expression (Kenney and Mertz, 2014).

Our recent study showed that interferon regulatory factor 8 (IRF8), also known as interferon consensus sequence-binding protein (ICSBP), plays an important role in caspase activation upon lytic induction and caspase activation facilitates the degradation of cellular factors to promote EBV lytic replication (Lv et al., 2018). IRF8 is a transcription factor that contains a DNA binding domain (DBD) and an IRF association domain (IAD). IRF8 associates with other transcription factors PU.1, IRF1, IRF2 or IRF4 through IAD. In addition, IRF8 undergoes tyrosine phosphorylation (Huang et al., 2007; Huang et al., 2008; Kautz et al., 2001; Unlu et al., 2007), SUMOylation (Chang et al., 2012) and ubiquitination (Kim and Ozato, 2009; Kong et al., 2007) and the transcription-regulatory activity of IRF8 is regulated by these post-translational modifications. One of the IRF8 binding partners PU.1, encoded by SPI1 (also known as Sfpi1), is a transcription factor of the E26-transformation specific (ETS) family and is involved in B cell development. PU.1 interacts with IRF8 or IRF4 to regulate gene expression by binding to composite DNA sequences known as ETS/interferon consensus elements (EICE) (Pang et al., 2016). PU.1 also interacts with CRE-binding protein (CBP) or p300 and functions as a coactivator for several other transcription factors (Yamamoto et al., 1999).

Although interferon consensus sequences are mostly located within the promoter and enhancer regions of host anti-viral genes, recent studies showed that viruses can integrate these sequences into their genomes for optimal viral gene expression (Lace et al., 2009; O’Flaherty et al., 2014). Given the fact that IRF8 is critical for EBV lytic reactivation in B cells (Lv et al., 2018), we investigated whether IRF8 directly regulates EBV lytic gene expression. We demonstrated that IRF8 binds to the promoter regions of a series of EBV lytic genes and, in conjugation with PU.1 and EBV RTA, regulates the expression of an EBV late gene BGLF2. BGLF2 is a tegument protein capable of triggering EBV reactivation through p38 pathway (Liu and Cohen, 2016) and is important for de novo viral infection (Konishi et al., 2018).

We recently demonstrated that protein inhibitor of activated STAT1 (PIAS1) is an EBV restriction factor by inhibiting viral and cellular transcription factors (Zhang et al., 2017). In this study, we further demonstrated that PIAS1 interacts with IRF8 and inhibits IRF8-mediated lytic gene activation.

Overall, our findings provide novel insights into the role of a conserved interferon consensus sequence within the viral genome in IRF8/PU.1-medaited EBV lytic gene expression and the negative regulation of IRF8 by PIAS1.

Results

IRF8 binds to the EBV genome

IRF4 has been shown to bind to an interferon consensus sequence within the promoter of viral M1 gene to promote murine γ-herpesvirsus-68 (MHV68) replication (O’Flaherty et al., 2014). In addition to regulating cellular gene CASP1 (encoding caspase-1) (Lv et al., 2018), we hypothesize that IRF8, the closest IRF4 homolog (Escalante et al., 2002), directly binds to the viral genome to promote EBV lytic gene expression. Previous studies showed that IRF8 binds to two different motifs (Marquis et al., 2011; Shin et al., 2011): TTTCNNTTCC (Motif-1) and TTTCNNTTTC (Motif-2). The Motif-1 is an IRF8/PU.1 composite EICE motif, while Motif-2 is an interferon-stimulated response element (ISRE)-like motif independent of PU.1. To identify potential IRF8 binding sites in the EBV genome, we used FIMO (http://meme-suite.org/tools/fimo) to extract sequences from Akata and B95–8 EBV genomes based on the two IRF8 binding motifs. We identified five Motif −1 and six Motif −2 sequences in the EBV genomes (Table 1 and Fig. 1A). Six of the eleven potential binding sites fell within 800 bp upstream of the corresponding EBV genes. Including LMP1, all putative IRF8 binding sites were located at lytic gene promoters.

Table 1.

The predicted IRF8 binding sites in the EBV genome

EBV strain Matched Sequence Strand Start End Downstream gene name Distance to CDS*
Akata Motif-1 (TTTCNNTTCC)
TTTCGCTTCC − 114614 114623 BGLF2 −484
TTTCCCTTCC + 958 967 BNRF1 −778
TTTCCCTTCC − 105381 105390 BBRF3 −1023
TTTCCTTTCC − 158262 158271 BARF0 −1292
TTTCTATTCC + 65633 65642 BaRF1 −800
Motif-2 (TTTCNNTTTC)
TTTCGTTTTC + 55787 55796 BORF1 −6984
TTTCGTTTTC + 132042 132051 BXRF1 −80
TTTCGTTTTC − 50020 50029 BFLF1 −3890
TTTCCATTTC − 76764 76773 BLLF3 −766
TTTCCATTTC − 168800 168809 LMP1 −296
TTTCAATTTC − 42801 42810 BHLF1 −2382
B95.8 Motif-1 (TTTCNNTTCC)
TTTCGCTTCC − 115063 115072 BGLF2 −484
TTTCCCTTCC + 958 967 BNRF1 −778
TTTCCCTTCC − 105826 105835 BBRF3 −1023
TTTCCTTTCC − 158720 158729 BARF0 −1292
TTTCTATTCC + 65812 65821 BaRF1 −800
Motif-2 (TTTCNNTTTC)
TTTCGTTTTC + 55885 55894 BORF1 −7065
TTTCGTTTTC + 132491 132500 BXRF1 −80
TTTCGTTTTC − 50118 50127 BFLF1 −3890
TTTCCATTTC − 76943 76952 BLLF3 −766
TTTCCATTTC − 169303 169312 LMP1 −257
TTTCAATTTC − 42902 42911 BHLF1 −2642
Open in a new tab
*

CDS: Coding sequence.

Fig 1. IRF8 binds to the EBV genome.

Fig 1.

Open in a new tab

A. Schematic representation of predicted IRF8 binding sites on the EBV genome (Akata strain). Two binding motifs were labeled as indicated.

B. Akata (EBV+) cells were either not treated (Latent) or treated with anti-IgG for 24 hrs to induced lytic replication. ChIP-qPCR analysis showing IRF8 binding to the EBV gene promoters. ChIP by a nonspecific IgG was included as negative controls.

C. P3HR1 (EBV+) cells were either not treated (Latent) or treated with TPA/NaBu for 24 hrs to induced lytic replication. ChIP-qPCR was performed as in panel B.

Representative results from three biological replicates are presented. Error bars indicate the standard deviation. * p<0.05; ** p<0.01; ns, not significant; nd, not detected.

To test whether IRF8 binds to these promoters during latency or reactivation, we performed chromatin immunoprecipitation (ChIP) experiments using chromatin prepared from EBV+ Akata and P3HR1 cells. Our results showed that IRF8 binds to all 11 predicted binding sites in Akata (EBV+) cells during latency and upon lytic reactivation induced by IgG cross-linking (Fig. 1B). Similarly, IRF8 binds to the majority of the predicted sites in P3HR1 (EBV+) cells without or with lytic induction (Fig. 1C).

IRF8/PU.1 regulates BGLF2 promoter activity

To test whether IRF8 regulates the activity of these EBV gene promoters, we selected five promoter regions with putative IRF8 binding sites within 800 bp upstream of coding sequence (CDS) to construct luciferase reporters, namely BGLF2p, LMP1p, BXRF1p, BNRF1p and BLLF3p. These reporters contain the predicted IRF8, PU.1 (Johannsen et al., 1995) or IRF8/PU.1 binding sites (Fig. 2). We then performed luciferase reporter assays using these promoters together with IRF8 and PU.1 overexpression. Transfection of PU.1 itself activated BGLF2p and LMP1p and the addition of IRF8 further enhanced PU.1-mediated BGLF2p activation (Fig. 2). In contrast, IRF8, PU.1 or IRF8/PU.1 failed to activate BXRF1p, BNRF1p and BLLF3p (Fig. 2). These results suggested that IRF8 synergizes with PU.1 in BGLF2p activation.

Fig 2. IRF8/PU.1 regulates BGLF2 promoter activity.

Fig 2.

Open in a new tab

The relative positions of IRF8 or IRF8/PU.1 binding sites on the constructed luciferase reporters were labeled as indicated. 293T cells were transfected with BGLF2p, LMP1p, BXRF1p, BNRF1p or BLLF3p luciferase reporters and effector plasmid DNA expressing PU.1 and IRF8 as indicated. The total amount of effector plasmid DNA used in each transfection was normalized by adding empty vector DNA. The transactivation of these promoters was presented by the relative luciferase activity. The value of cells transfected with empty vectors was set as 1. Representative results from three biological replicates are presented. Error bars indicate the standard deviation. ** p<0.01.

By comparing the predicted IRF8 binding sites with various EBV strains and an EBV-related herpesvirus that infects rhesus macaque (rhEBV; also called rhesus lymphocryptovirus, RLV) (Rivailler et al., 2002), we found that only the IRF8/PU.1 binding site in the BGLF2p is completely conserved across all virus strains (Fig. 3A), suggesting that IRF8/PU.1 binding to this region is evolutionarily conserved. Because the BGLF2p downstream gene product BGLF2 protein is important for EBV de novo infection and reactivation (Konishi et al., 2018; Liu and Cohen, 2016), we further explored the regulation of BGLF2p by IRF8/PU.1.

Fig 3. IRF8/PU.1 binding motif on BGLF2 promoter is evolutionally conserved.

Fig 3.

Open in a new tab

A. Sequence alignment showing the conservation of IRF8/PU.1 binding site on BGLF2 promoter (boxed) across different EBV strains and an EBV-related rhesus macaque herpesvirus (rhesus lymphocryptovirus, RLV). Reverse complement sequences were shown to follow the left-to-right order of viral genomes.

B. PU.1 and IRF4 binding profiles were extracted from the ChIP-seq analyses of EBV-immortalized LCL cells (strain B95.8) (Arvey et al., 2013; Arvey et al., 2012). The overlapped peaks for PU.1 and IRF4 on BGLF2 promoter were enlarged in the lower panel.

C. Relative mRNA expression level for IRF8, SPI1 (encoding PU.1), and IRF4. The mRNA expression levels of IRF8, SPI1, and IRF4 in LCLs were extracted from RNA-Seq analyses of 147 LCL cell lines by Arvey et al (Arvey et al., 2012) (upper panel). The mRNA expression levels of IRF8, SPI1, and IRF4 in Akata (EBV+) cells were extracted from our previous RNA-Seq analyses (Lv et al., 2018) (lower panel). RPKM, Reads Per Kilobase of transcript per Million mapped reads. The results were presented as mean ± standard deviation.

D. Schematic representation of the proposed binding of IRF8/PU.1 or IRF4/PU.1 to the promoter of BGLF2.

By analyzing the ChIP coupled with next-generation sequencing (ChIP-seq) data obtained from EBV-immortalized lymphoblastoid cell lines (LCLs), the Lieberman group extracted the binding profile of a large number of transcription factors on the EBV genome, including PU.1 and IRF4 (Arvey et al., 2013; Arvey et al., 2012). The PU.1 ChIP-seq results revealed two major PU.1 binding peaks in the EBV genome, one of which is overlapped with the IRF4 peak (Fig. 3B, upper panel). Further examination of the binding sequence located the PU.1/IRF4 peak at the promoter of BGLF2 (within BGLF1 gene locus) (Arvey et al., 2013) (Fig. 3B, lower panel). One binding sequence within the peak matched the position of our predicted IRF8/PU.1 binding site within BGLF2p. IRF8 and IRF4 are structurally similar and both could co-localize with PU.1 at the EICE composite site (Escalante et al., 2002; Pang et al., 2016), suggesting that BGLF2 may be regulated by IRF8/PU.1 and/or IRF4/PU.1.

We analyzed the expression profile for IRF8, SPI1 (encoding PU.1) and IRF4 using RNA-seq data obtained from 147 LCLs (Arvey et al., 2012). We found that both IRF8 and IRF4 are highly expressed in LCLs (Fig. 3C, upper panel). In contrast, our previous RNA-seq results showed that the expression level of IRF4 is very low in Akata (EBV+) cells (Lv et al., 2018), which was also reported in other studies (Ersing et al., 2017; Wang et al., 2014), whereas the expressions of IRF8 and SPI1 are relative high (Fig. 3C, lower panel). Thus, we reasoned that IRF8 cooperates with PU.1 to regulate the activation of BGLF2p in both LCLs and Akata (EBV+) cells while IRF4/PU.1 may also contribute to BGLF2p regulation in LCLs (Fig. 3D).

To test whether the promoter binding is required for IRF8/PU.1-mediated BGLF2p activation (Fig. 4A), we first deleted the binding site and found that IRF8/PU.1 could not activate the luciferase activity driven by the truncated BGLF2p (Fig. 4B). To further test whether the predicted binding site is required, we mutated the IRF8/PU.1 binding site and found that the mutations blocked IRF8/PU.1-mediated BGLF2p activation (Fig. 4C).

Fig 4. IRF8 cooperates with PU.1 and RTA to activate the BGLF2 promoter.

Fig 4.

Open in a new tab

A. IRF8/PU.1 and the RTA consensus binding sites in EBV BGLF2 promoter are highlighted in green and cyan, respectively. The ATG of BGLF2 is highlighted in red.

B. Two 5’ BGLF2p deletion luciferase constructs were co-transfected into 293T cells with either vector control or IRF8 and PU.1 expression vectors and luciferase assays were performed 36 hrs post-transfection.

C. The pGL2-BGLF2p619 constructs (with or without IRF8/PU.1 consensus site mutated) were co-transfected into 293T cells with either vector control or IRF8 and PU.1 expression vectors and luciferase assays were performed 36 hrs post-transfection.

D. The pGL2-BGLF2p619 construct was co-transfected into 293T cells with either vector control or IRF8, PU.1 and RTA expression vectors and luciferase assays were performed 36 hrs post-transfection. The value of cells transfected with empty vectors was set as 1.

E-F. IRF8 interacts with EBV transactivator RTA. 293T cells were co-transfected with HA-RTA and Flag IRF8 indicated. Western blot analyses showing that IRF8 is co-immunoprecipitated (co-IPed) with RTA (E) and RTA is co-IPed with IRF8 (G). IP, immunoprecipitation; β-actin blot was included as loading controls.

G. DNA binding partially contributes to the interaction between IRF8 and RTA. 293T cells were co-transfected with HA-RTA and Flag-IRF8. The cell lysates were either not treated or treated with DNase and then Flag-IRF8 was IPed by anti-Flag antibody. Western blot analyses showing that RTA is co-IPed with IRF8 without or with DNase treatment. IgG IP was included as a negative control. 2% input was loaded as a positive control.

Representative results from three biological replicates are presented. Error bars indicate the standard deviation. ** p<0.01.

IRF8 binds to PU.1 and EBV RTA to enhance BGLF2 promoter activation

IRF4 has been shown to synergize with MHV68 RTA to promote viral M1 gene (O’Flaherty et al., 2014). A previous study also showed that there is a RTA responsive element (RRE) in the BGLF2p (Fig. 4A) (McKenzie et al., 2016). Our recent study also demonstrated that RTA can significantly enhance BGLF2p activation (Zhang et al., 2017). To test whether IRF8/PU.1 can cooperate with RTA to further activate BGLF2p, we co-transfected IRF8/PU.1/RTA with the BGLF2p-luciferase construct and we found that co-transfection of these three plasmids further enhanced BGLF2p activation (Fig. 4D). To test whether IRF8 interacts with RTA, we performed co-immunoprecipitation (co-IP) experiments using transfected 293T cells and we found that Flag-IRF8 is co-immunoprecipitated (co-IPed) with HA-RTA (Fig. 4E). To further confirm these results, we performed a reciprocal IP and found that Flag-IRF8 co-IPs HA-RTA (Fig 4F). We also included an IgG control to show that only Flag-IRF8 co-IPs HA-RTA and DNase treatment partially reduces the interaction between IRF8 and RTA, suggesting that DNA binding partially contributes to their interaction (Fig 4G). Together, these results suggested that IRF8 binds to PU.1 and RTA to promote EBV BGLF2 expression.

Nuclear localization and phosphorylation are required for IRF8/PU.1-dependent activation of BGLF2 promoter

Nuclear localization (Salem et al., 2014), phosphorylation (Huang et al., 2007; Huang et al., 2008; Kautz et al., 2001; Unlu et al., 2007) and SUMOylation (Chang et al., 2012) of IRF8 have been implicated in its transcription regulatory activities. To test whether nuclear localization and post-translational modifications of IRF8 regulate IRF8/PU.1-mediated BGLF2p activation, we constructed a series of IRF8 mutants (Fig. 5A and 5B). Compared with wild-type IRF8, the nuclear localization-deficient (K108E) and phospho-deficient (Y107F/Y110F) mutations abrogated IRF8/PU.1-dependent BGLF2p activation (Fig 5C). However, the SUMOylation-deficient (K312R) and two other phospho-deficient (Y221F and Y373F) mutants still had the ability to activate BGLF2p (Fig 5C). These results suggested that nuclear localization and phosphorylation at Y107 and Y110 are required for IRF8/PU.1-dependent BGLF2 gene expression.

Fig 5. Nuclear localization and phosphorylation are necessary for IRF8-mediated activation of BGLF2 promoter.

Fig 5.

Open in a new tab

A. Schematic representation of IRF8 domains. The relative positions of amino acids in each domain are labeled as indicated. DBD, DNA binding domain; IAD, IRF association domain. B. Western blot showing the relative equal expression of various IRF8 mutants.

C. The pGL2-BGLF2p619 construct was co-transfected into 293T cells with either vector control, wild-type (WT) or various IRF8 mutants (K108E, Y107F/Y110F, Y221F, Y373F, and K312R) and PU.1 expression vectors and luciferase assays were performed 36 hrs post-transfection. The value of cells transfected with empty vectors was set as 1. Representative results from three biological replicates are presented. Error bars indicate the standard deviation. ** p<0.01; ns, not significant.

IRF8/PU.1 induces EBV reactivation

To further demonstrate whether IRF8/PU.1 can trigger EBV reactivation, we used an EBV+ Hela cells (Hela-Akata) to overexpress IRF8 and PU.1. These cells do not express IRF8 and PU.1, which provides an advantage to investigate the role of IRF8 and PU.1 without the interference of endogenous proteins. Our results showed that overexpression of IRF8/PU.1 activated EBV BGLF2 gene expression as well as other immediate-early, early and late genes (Fig. 6). Because BGLF2 protein is important for EBV de novo infection and reactivation (Konishi et al., 2018; Liu and Cohen, 2016), our results suggest that IRF8/PU.1 contributes to EBV lytic replication partially through BGLF2 induction.

Fig 6. IRF8/PU.1 induces EBV reactivation.

Fig 6.

Open in a new tab

EBV-positive Hela-Akata cells were transfected with empty vector or IRF8 and PU.1 plasmids. The cell pellets were harvested 48 hrs post-transfection and RNAs were extracted. RT-qPCR analyses showing that IRF8/PU.1 triggers the expression of EBV immediate-early (IE), early and late genes. The value of cells transfected with empty vectors was set as 1. The results are presented as mean ± standard deviation of triplicate assays. ** p<0.01.

PIAS1 suppresses IRF8/PU.1-mediated EBV lytic gene expression

Although IRF8 and PU.1 are highly expressed in latently infected B cells, EBV lytic gene expression is suppressed to maintain viral latency. We reasoned that there are factors capable of blocking IRF8 activity. Our recent study showed that PIAS1 inhibits EBV lytic replication through inhibiting both viral and cellular transcription factors (Zhang et al., 2017). PIAS1 has also been shown to inhibit IRF3- and IRF7-mediated gene activation through protein-protein interaction and SUMOylation, respectively (Chang et al., 2009; Li et al., 2013). We hypothesize that PIAS1 interacts with IRF8 and inhibits IRF8-mediated lytic gene activation. To test our hypothesis, we first performed co-IP experiments using transfected 293T cells and we found that Flag-IRF8 is strongly co-IPed with V5-PIAS1 (Fig. 7A) and that V5-PIAS1 is strongly co-IPed with Flag-IRF8 (Fig. 7B and 7C). To further demonstrate the association of IRF8 and PIAS1 in vivo, we performed co-IP experiments using an anti-PIAS1 antibody to enrich endogenous PIAS1 protein from Akata (EBV+) cells and we found that IRF8 is detected in the anti-PIAS1 precipitated complex (Fig. 7D).

Fig 7. PIAS1 inhibits IRF8/PU.1-mediated lytic gene expression.

Fig 7.

Open in a new tab

A-B. PIAS1 interacts with IRF8 in transfected cells. 293T cells were co-transfected with V5-PIAS1 and Flag-IRF8 plasmids as indicated. The cell pellets were harvested 48 hrs post-transfection and protein were extracted for immunoprecipitation by anti-V5 antibody (A) or anti-Flag antibody (B). Western blot analysis showing that IRF8 is co-IPed with V5-PIAS1 (A) and PIAS1 is co-IPed with Flag-IRF8 (B).

C. 293T cells were co-transfected with V5-PIAS1and Flag-IRF8. Flag-IRF8 was IPed by anti-Flag antibody. Western blot analyses showing that PIAS1 is co-IPed with IRF8. IgG IP was included as a negative control. 2% input was loaded as a positive control.

D. PIAS1 interacts with IRF8 in Akata (EBV+) cells. Co-IP showing that PIAS1 interacts with IRF8 in vivo. IgG IP was included as a negative control. 2% input was loaded as a positive control.

E. PIAS1 suppresses IRF8/PU.1-mediated BGLF2 promoter activation. 293T cells were transfected with 250 ng of plasmid DNA encoding BGLF2p-Luc (pGL2-BGLF2p619), and effector plasmid DNA expressing IRF8/PU.1 and increasing amount (250 and 750 ng) of wild-type (WT) and E3 ligase dead (C351S) PIAS1 as indicated. The total amount of effector plasmid DNA used in each transfection was normalized by adding vector DNA. The value of cells transfected with empty vectors (lane 1) was set as 1. Representative results from three biological replicates are presented. Error bars indicate the standard deviation. * p<0.05; ** p<0.01.

Subsequently, we checked the role of PIAS1 in IRF8/PU.1-mediated gene activation and found that PIAS1 suppresses IRF8/PU.1-mediated activation of BGLF2p (Fig. 7E, lanes 2 vs 3–4). Because the E3 SUMO ligase activity of PIAS1 has been implicated in suppressing IRF7 transactivation (Chang et al., 2009), we tested the E3 ligase-dead mutant PIAS1 (C351S) in IRF8/PU.1-mediated BGLF2p activation. Although the expression level of PIAS1 (C351S) was lower than wild-type PIAS1, this mutant still suppressed BGLF2p activation at a higher dosage (Fig. 7E, lanes 2 vs 6), suggesting that PIAS1 suppresses IRF8/PU.1-mediated BGLF2p activation in a SUMO ligase activity-independent manner.

To define the domains within PIAS1 responsible for binding to IRF8, we utilized a series of PIAS1 truncation mutants (Fig. 8A). We found that both the N- and C-terminal parts of PIAS1 interacted with IRF8 (Fig. 8B, lanes 6 and 9). To further probe the domains required for PIAS1-mediated repression, we tested the role of both N- and C-terminal PIAS1 in IRF8/PU.1-mediated activation of BGLF2p. Interestingly, both N- and C-terminal PIAS1 suppressed IRF8/PU.1 activity at a higher dosage but full-length PIAS1 showed the highest suppression (Fig. 8C, lanes 4 vs 6, 8 and 10), suggesting that both N- and C-terminal PIAS1 are required for maximal inhibition of IRF8/PU.1.

Fig 8. Both N- and C-terminal parts of PIAS1are required for inhibiting IRF8/PU.1-mediated lytic gene expression.

Fig 8.

Open in a new tab

A. The schematic representation of full length PIAS1 (1–651) or PIAS1 truncation mutants. The interaction of PIAS1 with IRF8 was summarized in the right part.

B. The N- or C-terminal part of PIAS1 interacts with IRF8. 293T cells were co-transfected with Flag-IRF8 and V5-tagged full length PIAS1 (aa 1–651) or PIAS1 truncation mutants as indicated. Western blot analyses showing that IRF8 is co-IPed with the N- and C-terminal part of PIAS1. IP, immunoprecipitation; β-actin blot was included as loading controls.

C. The Both N- and C-terminal PIAS1 mutants partially suppress IRF8/PU.1-mediated BGLF2 promoter activation. 293T cells were transfected with 250 ng of plasmid DNA encoding BGLF2p619-Luc, and effector plasmid DNA expressing IRF8/PU.1 and increasing amount (250 and 750 ng) of full-length (1–651) or truncated PIAS1 (1–433, 1–415 and 409–651) as indicated. The relative luciferase activity was measured using the Dual-luciferase assay kit as described in the method. The expression levels of IRF8, PU.1 and PIAS1 were monitored by Western blot. Β-actin blot was included as loading controls. Representative results from three biological replicates are presented. Error bars indicate the standard deviation. ∗ p < 0.05.

D. PIAS1 (1–100) truncation mutant does not interact with IRF8. Halo-V5-tagged PIAS1 was used to construct PIAS1 (1–100) truncation to ensure the expression of this short truncated PIAS1. 293T cells were co-transfected with Flag-IRF8 and Halo-V5-tagged full length PIAS1 (1–651) or PIAS1 (1–100) truncation mutant as indicated. Western blot analyses showing that IRF8 is co-IPed with the full-length PIAS1 but not PIAS1 (1–100) mutant. IP, immunoprecipitation; β-actin blot was included as loading controls.

E. PIAS1 (1–100) truncation mutant fails to suppress IRF8/PU.1-mediated BGLF2 promoter activation. 293T cells were transfected with 250 ng of plasmid DNA encoding BGLF2p619-Luc, and effector plasmid DNA expressing IRF8/PU.1 and increasing amount (250 and 750 ng) of Halo-V5-PIAS1 (1–651) or Halo-V5-PIAS1 (1–100) as indicated. The relative luciferase activity was measured using the Dual-luciferase assay kit as described in the method. The expression levels of IRF8, PU.1 and PIAS1 were monitored by Western blot. β-actin blot was included as loading controls. Representative results from three biological replicates are presented. Error bars indicate the standard deviation. ∗p < 0.05.

The above experiments showed that all PIAS1 constructs equally immunoprecipitated with IRF8 (Fig. 8B). To demonstrate the specificity for PIAS1 and IRF8 interaction, we generated a shorter truncated PIAS1 (aa 1–100) construct using Halo-V5-PIAS1 (Fig 8D). The addition of Halo tag to V5-PIAS1 is to stabilize the shorter PIAS1 truncation mutant (aa 1–100). We found that PIAS1 (1–100) does not bind to IRF8 (Fig 8D). Together, our results indicated that the N-terminal (aa 101–205) and C-terminal (aa 409–651) parts of PIAS1 bind to IRF8.

We then used the wild-type and shorter version of PIAS1 to test their function in IRF8/PU.1 activation of BGLF2p. We found that wild-type Halo-V5-PIAS1 strongly suppresses BGLF2p activity as expected while the PIAS1 (1–100) truncation mutant barely affects BGLF2p activity even at a higher concentration (Fig 8E). These results further suggested that PIAS1 association with IRF8 is required for suppressing BGLF2p activity.

PIAS1 binds to BGLF2 promoter but does not affect the recruitment of IRF8

PIAS1 was reported to directly associate with cellular and EBV ZTA and RTA promoters (Liu et al., 2010; Liu et al., 2014; Toropainen et al., 2015; Zhang et al., 2017). We reasoned that PIAS1 may also interact with BGLF2p in EBV-infected cells. To test this hypothesis, we performed chromatin immunoprecipitation (ChIP) experiments on PIAS1-knockout (PIAS1-KO) and parental cells using an anti-PIAS1 antibody and examined the enrichment of PIAS1 on BGLF2p by qPCR using specific primers. As expected, we found PIAS1 is significantly enriched in the BGLF2p in parental cells compared to the PIAS1-KO cells (Fig. 9A).

Fig 9. PIAS1 associates with EBV BGLF2 promoter and does not affect IRF8 recruitment.

Fig 9.

Open in a new tab

A. ChIP-qPCR analysis performed on wild type (WT) Akata (EBV+) cells showing PIAS1 binding to BGLF2 promoter. PIAS1 knockout (KO) Akata (EBV+) cells were included as controls. ChIP by a nonspecific IgG was included as a negative control.

B. PIAS1 does not affect IRF8 recruitment to BGLF2 promoter. The WT and PIAS1-KO Akata (EBV+) cells were untreated (Latent, 0 h) or treated with anti-IgG to induce lytic reactivation (Lytic, 24 h). ChIP-qPCR analysis performed on these cells showing that IRF8, but not histone H3, is recruited to BGLF2 promoter following lytic induction while PIAS1-KO does not affect the IRF8 recruitment. The values of untreated group (Latent, 0 h) were set as 1.

C. The C-terminal part of IRF8 interacts with PIAS1. 293T cells were co-transfected with V5-PIAS1 and Flag-tagged full length IRF8 (aa 1–427) or IRF8 truncation mutants as indicated. Western blot analyses showing that C-terminal IRF8 is co-IPed with PIAS1. Arrows indicate the co-IPed full length and C-terminal IRF8. IgG light chain was labeled as indicated. β-actin blot was included as loading controls. IP, immunoprecipitation.

Representative results from three biological replicates are presented. Error bars indicate the standard deviation. * p < 0.05; ** p < 0.01.

To explore whether PIAS1 suppresses the recruitment of IRF8 to BGLF2p, we performed IRF8 ChIP assays using the parental and PIAS1-KO Akata (EBV+) cells. We found that PIAS1 knockout does not affect the IRF8 recruitment to BGLF2p upon lytic induction (Fig. 9B), suggesting PIAS1 does not interfere with the DNA binding of IRF8. Because the N-terminal DBD domain of IRF8 is responsible for DNA binding, we reasoned that PIAS1 does not bind to the DBD domain of IRF8. To prove this, we performed co-IP experiments using IRF8 truncation mutants and full length PIAS1. As expected, we found that the C-terminal part of IRF8, but not the DBD containing N-terminal part, binds to PIAS1 (Fig. 9C). These results together suggested that PIAS1 could inhibit IRF8 activation through protein-protein interactions but not through blocking the DNA binding of IRF8.

Discussion

As a unique member of the IRF family, IRF8 is primarily expressed in hematopoietic cells and has important roles in granulocyte development (Netherby et al., 2017). By binding to interferon consensus sequences, IRF8 regulates a variety of cellular genes critical for B cell development and host immunity (Adams et al., 2018; Clement et al., 2018; Karki et al., 2018; Kurotaki et al., 2018; Shin et al., 2011). Our previous study showed that IRF8 plays a critical role in EBV reactivation through regulating caspase-1 expression (Lv et al., 2018). In this study, we discovered that IRF8 binds to multiple sites on the EBV genome and IRF8 and PU.1 regulate the expression of a lytic gene BGLF2 through targeting an evolutionarily conserved site. We further demonstrated that the up-regulation of BGLF2 by IRF8/PU.1 is antagonized by PIAS1, which likely contributes EBV latency maintenance even in the presence of IRF8/PU.1 (Fig. 10).

Fig 10.

Fig 10.

Open in a new tab

Hypothesized model by which IRF8/PU.1 and PIAS1 contribute to EBV latency and lytic replication through regulation of viral gene expression.

IRF8 acts either as a transcriptional activator or repressor depending on its interacting factors and target DNA elements (Kanno et al., 2005; Tamura and Ozato, 2002). The IRF8-PU.1 heterodimer can activate genes with the EICE in their promoter regions (Kurotaki et al., 2013; Tamura et al., 2008). IRF8 can also associate with other IRFs (IRF1 or IRF2) to activate or repress gene expression (Kuwata et al., 2002; Tamura et al., 2008). As a transcription factor, we hypothesized that IRF8 may also bind to EBV gene promoter to promote viral gene expression. Following this hypothesis, we identified a series of IRF8 and IRF8/PU.1 binding sites within the EBV genome (Fig. 1 and Table 1). By using luciferase reporter assays, we further demonstrated that IRF8/PU.1 regulates the promoter activity of the EBV lytic gene BGLF2 (Fig. 2).

The IRF8/PU.1 regulation of BGLF2 promoter is likely conserved during evolution because the binding site is shared across different EBV strains and an EBV-related virus that infects rhesus macaque (Fig. 3). We found that IRF8 and PU.1, together with EBV RTA, activate lytic gene expression (Fig. 4). A similar observation was reported for IRF4 and MHV68 RTA in viral M1 gene expression (O’Flaherty et al., 2014), highlighting subversion of IRF family members as a common strategy by different viruses.

Our luciferase assay results indicated that nuclear localization and the tyrosine phosphorylation of IRF8 within DBD are required for IRF8/PU.1-mediated BGLF2 promoter activation (Fig. 5), suggesting that IRF8 by phosphorylation may contribute to EBV reactivation upon lytic induction.

BGLF2 is a late lytic gene product and previous study showed that it can trigger EBV lytic reactivation by regulation of ZTA/BZLF1. BGLF2 activates p38 and JNK to induce AP-1 binding to EBV ZTA/BZLF1 promoter. During reactivation, BGLF2 expression would act in a positive-feedback manner to enhance the expression of ZTA/BZLF1 (Liu and Cohen, 2016). During primary infection, BGLF2 is delivered into host cell as part of the tegument proteins and therefore would enhance ZTA/BZLF1 expression necessary for EBV replication and survival (Konishi et al., 2018). Our study demonstrated that IRF8/PU.1 overexpression triggers EBV lytic gene expression in EBV-positive epithelial cells (Fig. 6), suggesting that IRF8/PU.1-mediated BGLF2 expression likely contributes to EBV reactivation in B cells where IRF8 and PU.1 are physiologically expressed in high levels.

Our recent study demonstrated that PIAS1 is a key EBV restriction factor and PIAS1 can antagonize RTA-mediated gene activation (Zhang et al., 2017). In this study, we for the first time showed that PIAS1 interacts with IRF8 and inhibits IRF8/PU.1-mediated lytic gene activation (Fig. 7). The PIAS1 inhibition of IRF8 is independent of its SUMO ligase activity but requires both N- and C-terminal domains of PIAS1 (Figs. 7 and 8).

PIAS1 is a multifunctional protein with the capability to bind to a group of cellular and viral factors (Liu and Shuai, 2008; Shuai and Liu, 2005). PIAS1 inhibition of transcription factors can be achieved through blocking their DNA binding or through promoting protein SUMOylation (Chang et al., 2009; Li et al., 2013; Liu et al., 1998; Liu et al., 2004; Liu et al., 2005; Toropainen et al., 2015; Zhang et al., 2017). Our study showed that protein-protein interaction, but not IRF8 SUMOylation, is required for PIAS1 inhibition of IRF8 (Figs. 7–9), illustrating a new role for PIAS1 in inhibition of transcription factors. Upon lytic induction, PIAS1 is destabilized by caspase-mediated cleavage (Zhang et al., 2017), which may allow IRF8/PU.1 activation of EBV lytic gene BGLF2 (Fig. 10).

In LCLs, IRF4 gene is highly expressed and IRF4 protein binds to EBV BGLF2 promoter (Fig. 3) (Arvey et al., 2012; Jiang et al., 2014). Future study is required to test whether IRF4 and PU.1 also contribute to EBV gene expression. In addition to PIAS1, it will be interesting to investigate whether other PIAS family members (Liu and Shuai, 2008; Liu et al., 2005) regulate IRF8 and EBV replication in the future.

In summary, our study demonstrated that IRF8, in complex with PU.1, promotes EBV lytic gene expression while PIAS1 inhibits IRF8/PU.1 transcriptional activity to suppress lytic gene expression.

Materials and Methods

Cell culture

Akata (EBV+) cells (a gift from Dr. Diane Hayward, Johns Hopkins University) and P3HR1 cells (ATCC, HTB-62) were grown in RPMI 1640 media supplemented with 10% FBS (Cat# 26140079, Thermo Fisher Scientific) in 5% CO2 at 37°C (Li et al., 2015; Lv et al., 2017; Lv et al., 2018; Zhang et al., 2017). 293T cells (a gift from Dr. Diane Hayward, Johns Hopkins University) and EBV-positive Hela-Akata cells (a gift from Dr. Shannon Kenney) were grown in DMEM media supplemented with 10% FBS in 5% CO2 at 37°C. The PIAS1-knockout (KO) cell line, derived from Akata (EBV+) cells were established by using CRISPR/Cas9 technology (Zhang et al., 2017).

Plasmids, cloning, and site-directed mutagenesis

Plasmid DNA was purified using Miniprep or Midiprep columns according to the manufacturer’s protocol (Qiagen). The original PIAS1 plasmid was a gift from Dr. Heng Zhu (Cox et al., 2017; Uzoma et al., 2018). V5-PIAS1 (full length, aa 101–433, aa 1–415, aa 409–651, C351S, aa 1–433 stop) were constructed previously (Zhang et al., 2017). pCMV3-N-FLAG (vector), pCMV3-N-FLAG-IRF8 and pGEM-SPI1 (encoding PU.1) were obtained from Sino biological. Plasmids pCMV3-N-FLAG-IRF8(K108E) was constructed by QuikChange II site-directed mutagenesis kit previously (Lv et al., 2018) [dummy_incomplete para]

PIAS1 truncation mutants (aa 1–205, aa 1–301), V5-PU.1, and Flag-IRF8 truncation mutants (aa 1–204, aa 205–427) plasmids were constructed using Gibson assembly as previously described (Zhang et al., 2017). Briefly, the DNA fragments were prepared by PCR amplification with about 40 bp overlap to the adjacent sequence using specific primers and the Q5 High-Fidelity DNA polymerase (New England Biolabs, Cat# M0491S). The Gibson assembly reaction were carried out by using the 2x Gibson Assembly Master Mix (405 μl Isothermal Start Mix, 25 μl 1M DTT, 20 μl 25 mM dNTPs, 50 μl NAD+, 1 μl T5 exonuclease, 31.25 μl Phusion High Fidelity DNA Polymerase, 250 μl Taq Ligase, 467.75 μl H2O). Aliquots of 10 μl were prepared and then used for a single Gibson reaction. Each assembly reaction contained approximately 100 ng of each insert and 50 ng of the linearized vector backbone and added to the 10 μl master mix in a 20 μl total volume reaction mixture. The reaction was incubated at 50°C for 60 min. After the assembly reactions, the reaction mixture was transformed into DH5α competent cells.

The V5-PIAS1 mutants and V5-PU.1 (encoded by SPI1) were constructed into pEFIRESpuro vector based on pEFIRESpuro-6His-SUMO2 plasmid (a gift from Dr. Ronald Hay, University of Dundee) (Tammsalu et al., 2014). Flag-IRF8 mutants were constructed using pCMV3-N-FLAG-IRF8. The primer sets used are: V5-PIAS1–1-205: Set 1, forward (5’-GGTAAGCCTATCCCTAACCCTCTCCTCGGTCTCGATTCTACGCTCGAGATGGCGGACAGTGCGGAACTA-3’) and reverse (5’-ATCTATGCGGCCGCGGATCCTCATGATAAACAAAACCTTAACTG-3’); Set 2, forward (5’-CAGTTAAGGTTTTGTTTATCATGAGGATCCGCGGCCGCATAGAT-3’) and reverse (5’CGTAGAATCGAGACCGAGGAGAGGGTTAGGGATAGGCTTACCGCTAGCATGATGATGATGATGATGCAT-3’); V5-PIAS1–1-301: Set 1, forward (same as V5-PIAS1–1-205-Set 1 forward primer) and reverse (5’-ATCTATGCGGCCGCGGATCCTCATCCCTTTGCTCGTAACCTCTG-3’); Set 2, forward (5’-CAGAGGTTACGAGCAAAGGGATGAGGATCCGCGGCCGCATAGAT-3’) and reverse (same as V5-PIAS1–1-205-Set2 reverse primer); V5-SPI1: Set 1, forward (5’-GGTAAGCCTATCCCTAACCCTCTCCTCGGTCTCGATTCTACGCTCGAGATGTTACAGGCGTGCAAAATG-3’) and reverse (5’-ATCTATGCGGCCGCGGATCCTCAGTGGGGCGGGTGGCGCC-3’); Set 2, forward (5’GGCGCCACCCGCCCCACTGAGGATCCGCGGCCGCATAGAT-3’) and reverse (5’-CGTAGAATCGAGACCGAGGAGAGGGTTAGGGATAGGCTTACCGCTAGCATGATGATGATGATGATGCAT-3’); Flag-IRF8–1-204: Set 1, forward (5’-ATGGATTACAAGGATGACGACGATAAGGGTGGAGGCGGTAGCATGTGTGACCGGAATGGTGGTCGGCGG-3’) and reverse (5’-AATTCGGCGGCCGCTCTAGATTACTGGGAGAATGCTGAATGGT-3’); Set 2, forward (5’ACCATTCAGCATTCTCCCAGTAATCTAGAGCGGCCGCCGAATT-3’) and reverse (5’GCTACCGCCTCCACCCTTATCGTCGTCATCCTTGTAATCCAT-3’); Flag-IRF8–205-427: Set 1, forward (5’-ATGGATTACAAGGATGACGACGATAAGGGTGGAGGCGGTAGCATGGTGATCAGCTTCTACTATGGGGGC-3’) and reverse (5’-AATTCGGCGGCCGCTCTAGATTAGACGGTGATCTGTTGGT-3’); Set 2, forward (5’ACCAACAGATCACCGTCTAATCTAGAGCGGCCGCCGAATT-3’) and reverse (5’-GCTACCGCCTCCACCCTTATCGTCGTCATCCTTGTAATCCAT-3’).

EBV HA-RTA (pGL196) expression vector was a gift from Dr. S. Diane Hayward (Johns Hopkins). A second PU.1 expression plasmid (V5-His-PU.1) was cloned by PCR amplifying SPI1 from pGEM-SPI1 and inserting it into pcDNA3.1-V5-His (Invitrogen) by using BamHI and XhoI restriction sites and the following primer sets: forward (5’-CGCGGATCCATGTTACAGGCGTGCAAAATGG-3’) and reverse (5’-CCGCTCGAGCGGTGGGGCGGGTGGCG-3’). The pGL2-BGLF2p619 luciferase reporter plasmid was constructed as previously described (Zhang et al., 2017). The pGL2-BGLF2p699, pGL2-BGLF2p382 luciferase reporter plasmids contain the Akata strain sequence from the −699 to −1 and −382 to −1, respectively (relative to the BGLF2 ORF) inserted upstream of the luciferase gene in pGL2-basic (Promega) using XhoI and HindIII restriction sites. The primer sets used are: pGL2-BGLF2p699 forward (5’-CCGCTCGAGTGCCTGTTTCTTTCTCTAAGACC-3’) and reverse (5’-CCCAAGCTTGACGTGCCAGAATATATCCCCT-3’); pGL2-BGLF2p382 forward (5’-CCGCTCGAGCCCACCTCCTACTCCCGTAT-3’) and reverse (5’-CCCAAGCTTGACGTGCCAGAATATATCCCCT-3’). Similarly, The pGL2-LMP1p577, pGL2-BXRF1p79, pGL2-BNRF1p922 and pGL2-BLLF3p981 luciferase reporter plasmids contain the Akata strain sequences from −577 to −2, −791 to +8, −922 to −1 and −981 to −1 (relative to their ORFs), respectively. These DNA sequences were inserted upstream of the luciferase gene in pGL2-basic using XhoI and HindIII restriction sites. The primer sets used are: pGL2-LMP1p577 forward (5’-CCGCTCGAGACAGCCCACACCCTTTTC-3’) and reverse (5’CCCAAGCTTGTCAGGGTAGTGTGGCAGGA-3’); pGL2-BXRF1p791 forward (5’-CCGCTCGAGCGGTTCAGGGACTACAATGG-3’) and reverse (5’-CCCAAGCTTGGATCCATGGCTGGATTTC-3’); pGL2-BNRF1p922 forward (5’-CCGCTCGAGGCCTTTTTGCCACATGTTTT-3’) and reverse (5’-CCCAAGCTTCCTTGCACGTCAAGTTACACG-3’) and pGL2-BLLF3p981 forward (5’-CCGCTCGAGATGATGACGCCGAGACCTAT-3’) and reverse (5’-CCCAAGCTTGGTGATGATCTAACAGACAGG-3’).

The Halo-V5-PIAS1 was created as described previously (Zhang et al., 2017). The Halo-V5-PIAS1 truncation mutant (aa 1–100) was generated via introducing a stop codon by using the QuickChange II site Mutagenesis Kit (Stratagene) according to the manufacturer’s instruction. The primer sets used are: Halo-V5-PIAS1 (1–100 stop) forward (5’-CACAACTCACTTACTAGGGTCACCCTGCATCATCGCC-3’) and reverse (5’-CAGGGTGACCCTAGTAAGTGAGTTGTGGAATGGTAG-3’).

Plasmids pCMV3-N-FLAG-IRF8(Y107F/Y110F), pCMV3-N-FLAG-IRF8(Y221F), pCMV3-N-FLAG-IRF8(Y373F), pCMV3-N-FLAG-IRF8(K312R), and pGL2-BGLF2p619-mut (IRF8/PU.1 binding site mutation) were constructed by using QuikChange II site-directed mutagenesis kit and the following primer sets:

IRF8 (Y107F/Y110F) forward (5’-ACTGGACATTTCCGAGCCATTCAAAGTTTTCCGAATTGTTCCTG-3’) and reverse (5’-CAGGAACAATTCGGAAAACTTTGAATGGCTCGGAAATGTCCAGT-3’); IRF8 (Y221F) forward (5’-AGATGGTGATCAGCTTCTACTTTGGGGGCAAGC-3’) and reverse (5’-GCTTGCCCCCAAAGTAGAAGCTGATCACCATCT-3’); IRF8 (Y373F) forward (5’-TCGTGCAGATTGAGCAGCTGTTTGTCCGGCAAC-3’) and reverse (5’-GTTGCCGGACAAACAGCTGCTCAATCTGCACGA-3’); IRF8 (K312R) forward (5’-AAAGGCAGGCCCAACAGGCTGGAGCGTG-3’) and reverse (5’-CACGCTCCAGCCTGTTGGGCCTGCCTTT-3’); BGLF2p619-mut forward (5’-CCAGAGAGGGAGTTTCGCTGGATCATTTCAGTGGCCCTCAG-3’) and reverse (5’-CTGAGGGCCACTGAAATGATCCAGCGAAACTCCCTCTCTGG-3’). DNA sequences in all these plasmids were authenticated by Sanger sequencing.

Chromatin-immunoprecipitation (ChIP) assay

ChIP assay was performed using a SimpleChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology) (Lv et al., 2018; Zhang et al., 2017). Briefly, 2×107 Akata (EBV+) cells or P3HR-1 cells were cross-linked with 1% formaldehyde and digested with micrococcal nuclease to achieve DNA fragments of 150–900 bp. 2% of the chromatin was reserved as input sample. DNA-protein complexes were immunoprecipitated with mouse anti-IRF8 antibody (Santa Cruz, Cat # sc-365042x) or mouse IgG control (Santa Cruz, Cat # sc-2025), rabbit anti-PIAS1 antibody (Abcam, Cat# 77231) or rabbit IgG control (Cell Signaling Technology, Cat# 2729) and Histone H3 (D2B12) (ChIP Formulated, Cell Signaling Tech Cat# 4720).

After being reverse crosslinked, DNA samples were purified and quantified by quantitative PCR (qPCR) with the specific primers listed below: BNRF1 promoter: forward (5’-CAGGACTAACCATGCCATCTC-3’) and reverse (5’-GCCAAGTTAGAGCTGCGATT-3’); BHLF1 promoter: forward (5’-GCTGGCCATATCTACAATTGGG-3’) and reverse (5’-GTGCTTACACACTTCCCGTT-3’); BFLF1 promoter: forward (5’-CTGAGGGAGTGTTCCACAGT-3’) and reverse (5’-GCATGTATTACCCGCCATCC-3’); BORF1 promoter: forward (5’-GGAGAGGCGGGAGAGATG-3’) and reverse (5’-CTGGACGACATCGAGGCAAT-3’); BaRF1 promoter: forward (5’-GTGGTTTCACTGGCACGATT-3’) and reverse (5’-GGCATCAGGGCGATAAACTG-3’); BLLF3 promoter: forward (5’-GTTTGGCGTCTCAGGCTATG-3’) and reverse (5’-TGAATGCGGAGGGTCAGATT-3’); BBRF3 promoter: forward (5’-CCCGTCCCACATAATGGATG-3’) and reverse (5’-GGCTCAGATTTCCAGCCCTA-3’); BGLF2 promoter: forward (5’-TGGGATTTCTTGTTGTGCTG-3’) and reverse (5’-AGGGCAGACTGCACAGGAT-3’); BXRF1 promoter: forward (5’-CACCTGGTAGAGTCCGTCAT-3’) and reverse (5’-CAACCGGCAGTGACTCAG-3’); BARF0 promoter: forward (5’-ACTGGATGTCCGAGGAGAAG-3’) and reverse (5’-CAGCTTTCCTTTCCGAGTCTG-3’); LMP1 promoter: forward (5’-GAATGCGGTGGTAAAGCGTA-3’) and reverse (5’-CCACAACACTGCTCACTCC-3’).

Reverse transcription and quantitative PCR (RT-qPCR)

The Total RNA was extracted from cells following instruction using Isolate II RNA Mini Kit (Bioline). RNA was reverse transcribed to cDNA using the High Capacity cDNA Reverse Transcription Kit (Invitrogen). qPCR was performed using Brilliant SYBR Green qPCR master mix with specific primers, as described (Lv et al., 2018; Zhang et al., 2017). The relative expression of mRNA was normalized to β-actin expression using the comparative Ct method. The specificity of amplification of targets with high Ct values was confirmed by analysis of the temperature dissociation curves. Primers used for measuring gene transcriptional level: ZTA, RTA, BGLF4, BHRF1, BALF1, BORF1, BPLF1, BOLF1, BVRF1, BGLF2, BLLF1 and β-actin primers were described previously (Zhang et al., 2017).

Lytic induction

To induce the EBV lytic cycle, Akata (EBV+) cells were treated with anti-human IgG antibody (1:200 dilution, MP Biomedicals Cat# 55087) for 0 to 24 hr. EBV lytic replication was reactivated in P3HR-1 cells by addition of phorbol-12-myristate-13-acetate (TPA, Fisher Scientific Cat# NC9325685) (32 nM) and sodium butyrate (NaBu, Millipore Cat# 19–137) (3 mM) for 24 hr (Zhang et al., 2017; Zhang et al., 2019). Akata (EBV+) cells or P3HR-1 cells were then treated for ChIP assays or harvest for IP and western blot analysis.

Luciferase reporter assay

Luciferase assay was performed as previously described (Huang et al., 2006; Lv et al., 2018; Zhang et al., 2017). Breifly, 293T cells were co-transfected with the firefly luciferase reporter vectors along with IRF8, PU.1, or RTA and Renilla expression plasmids using Lipofectamine 2000 reagent (Cat# 11668019, Life Technologies). Thirty-six hours after transfection, cell extracts were prepared and assayed with the dual-luciferase assay kit from Promega (Cat # E1960). Each condition was performed in triplicate.

Immunoprecipitation assay

293T cells were co-transfected with indicated plasmids using Lipofectamine 2000. The cells were harvested 48 hrs post-transfection using RIPA lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 1% NP40, 1% deoxycholate, 0.1% SDS and 1 mM EDTA) containing protease inhibitors and phosphatase cocktail I and II. For some of the experiments, the cell lysates were treated with RQ1 RNase-Free DNase (200 U, Cat# M6101, Promega) for 30 min at room temperature to remove genomic DNA. The immunoprecipitation was carried out as previously described (Li et al., 2011; Zhang et al., 2017). For immunoprecipitation from Akata (EBV+) cells, the cells were harvested using the cell lysis buffer (Cell Signaling Tech, Cat# 9803). The cell pellets were removed by centrifugation at 14000 g for 10 min. The lysates were incubated with Rabbit anti-PIAS1 antibody (Abcam, Cat # 77231) or Rabbit IgG control (Cell Signaling Tech, Cat # 2729) overnight, followed by incubation with 25 μL Protein A/G Plus-Agarose (Santa Cruz, SC-2003) for 3 hrs. After 4 times wash using cell lysis buffer, the samples were eluted in 2x SDS-PAGE buffer for immunoblot analysis.

Immunoblot analysis

Cell lysates were harvested in lysis buffer including protease inhibitors (Roche) as described previously (Lv et al., 2017). Protein concentration was determined using the Bradford assay (Biorad), and proteins were separated in SDS 4–20% polyacrylamide gels and then transferred onto a PVDF membrane. Membranes were blocked in TBS containing 5% milk, and 0.1% Tween 20 solution. Membranes were then incubated in the following primary antibodies: mouse anti-RTA (Argene, Cat# 11–008, Discontinued), anti-β-actin (Sigma, Cat # A5441), rabbit anti-PU.1 antibody (Cell Signaling Tech, Cat # 2258), rabbit anti-IRF8 antibody (Cell Signaling Tech, Cat # 5628), rabbit anti-PIAS1 antibody (Abcam, Cat # 109388), anti-HA-HRP (Cell Signaling Tech, Cat# 14031), anti-V5-HRP antibody (Thermo Fisher, Cat # R961–25) and anti-Flag-HRP antibody (Cell Signaling Tech, Cat# 86861 and Cat # 2044). The secondary antibodies (1:5,000 dilutions) used were horseradish peroxidase (HRP)-labeled anti-rabbit antibody (Jackson ImmunoResearch, Cat # 111–035-144), HRP-labeled anti-mouse antibody (Jackson ImmunoResearch, Cat # 111–035-166), HRP-labeled anti-rabbit antibody, light chain specific (Jackson ImmunoResearch, Cat # 211–032-171).

Statistical analysis

All numerical data were presented as mean ± standard deviation of triplicate assays. The statistical significances were determined using Student’s two-tail t-test, where p<0.05 was considered statistically significant.

Acknowledgments

We thank Dr. S. Diane Hayward (Johns Hopkins University) for comments and suggestions and for providing various cell lines and constructs. We thank Dr. Shannon Kenney (University of Wisconsin-Madison) for providing the Hela-Akata cell line. We also thank Dr. Heng Zhu (Johns Hopkins University) for PIAS1 construct and Dr. Ronald Hay (University of Dundee) for providing the pEFIRESpuro-6His-SUMO2 plasmid. This work was in part supported by grants from the National Institute of Allergy and Infectious Diseases (AI104828 and AI141410; https://grants.nih.gov/grants/oer.htm). The work was also supported by Institutional Research Grant IRG-14–192-40 from the American Cancer Society. R.L. received support from the VCU Philips Institute for Oral Health Research, the VCU NCI Designated Massey Cancer Center (NIH grant P30 CA016059) (https://grants.nih.gov/grants/oer.htm), and the VCU Presidential Quest for Distinction Award. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  1. Adams NM, Lau CM, Fan X, Rapp M, Geary CD, Weizman OE, Diaz-Salazar C, Sun JC, 2018. Transcription Factor IRF8 Orchestrates the Adaptive Natural Killer Cell Response. Immunity 48, 1172–1182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arvey A, Tempera I, Lieberman PM, 2013. Interpreting the Epstein-Barr Virus (EBV) epigenome using high-throughput data. Viruses 5, 1042–1054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arvey A, Tempera I, Tsai K, Chen HS, Tikhmyanova N, Klichinsky M, Leslie C, Lieberman PM, 2012. An atlas of the Epstein-Barr virus transcriptome and epigenome reveals host-virus regulatory interactions. Cell Host Microbe 12, 233–245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chang T-H, Xu S, Tailor P, Kanno T, Ozato K, 2012. The small ubiquitin-like modifier-deconjugating enzyme sentrin-specific peptidase 1 switches IFN regulatory factor 8 from a repressor to an activator during macrophage activation. The Journal of Immunology 189, 3548–3556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chang TH, Kubota T, Matsuoka M, Jones S, Bradfute SB, Bray M, Ozato K, 2009. Ebola Zaire virus blocks type I interferon production by exploiting the host SUMO modification machinery. PLoS Pathog 5, e1000493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Clement M, Haddad Y, Raffort J, Lareyre F, Newland SA, Master L, Harrison J, Ozsvar-Kozma M, Bruneval P, Binder CJ, Taleb S, Mallat Z, 2018. Deletion of IRF8 (Interferon Regulatory Factor 8)-Dependent Dendritic Cells Abrogates Proatherogenic Adaptive Immunity. Circulation research 122, 813–820. [DOI] [PubMed] [Google Scholar]
  7. Cohen JI, Fauci AS, Varmus H, Nabel GJ, 2011. Epstein-Barr virus: an important vaccine target for cancer prevention. Sci Transl Med 3, 107fs107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cox E, Hwang W, Uzoma I, Hu J, Guzzo CM, Jeong J, Matunis MJ, Qian J, Zhu H, Blackshaw S, 2017. Global Analysis of SUMO-Binding Proteins Identifies SUMOylation as a Key Regulator of the INO80 Chromatin Remodeling Complex. Mol Cell Proteomics 16, 812–823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ersing I, Nobre L, Wang LW, Soday L, Ma Y, Paulo JA, Narita Y, Ashbaugh CW, Jiang C, Grayson NE, Kieff E, Gygi SP, Weekes MP, Gewurz BE, 2017. A Temporal Proteomic Map of Epstein-Barr Virus Lytic Replication in B Cells. Cell Rep 19, 1479–1493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Escalante CR, Brass AL, Pongubala JM, Shatova E, Shen L, Singh H, Aggarwal AK, 2002. Crystal structure of PU.1/IRF-4/DNA ternary complex. Mol Cell 10, 1097–1105. [DOI] [PubMed] [Google Scholar]
  11. Huang J, Liao G, Chen H, Wu FY, Hutt-Fletcher L, Hayward GS, Hayward SD, 2006. Contribution of C/EBP proteins to Epstein-Barr virus lytic gene expression and replication in epithelial cells. J Virol 80, 1098–1109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Huang W, Horvath E, Eklund EA, 2007. PU. 1, interferon regulatory factor (IRF) 2, and the interferon consensus sequence-binding protein (ICSBP/IRF8) cooperate to activate NF1 transcription in differentiating myeloid cells. Journal of Biological Chemistry 282, 6629–6643. [DOI] [PubMed] [Google Scholar]
  13. Huang W, Zhu C, Wang H, Horvath E, Eklund EA, 2008. The interferon consensus sequence-binding protein (ICSBP/IRF8) represses PTPN13 gene transcription in differentiating myeloid cells. Journal of Biological Chemistry 283, 7921–7935. [DOI] [PubMed] [Google Scholar]
  14. Jiang S, Willox B, Zhou H, Holthaus AM, Wang A, Shi TT, Maruo S, Kharchenko PV, Johannsen EC, Kieff E, Zhao B, 2014. Epstein-Barr virus nuclear antigen 3C binds to BATF/IRF4 or SPI1/IRF4 composite sites and recruits Sin3A to repress CDKN2A. Proc Natl Acad Sci U S A 111, 421–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Johannsen E, Koh E, Mosialos G, Tong X, Kieff E, Grossman SR, 1995. Epstein-Barr virus nuclear protein 2 transactivation of the latent membrane protein 1 promoter is mediated by J kappa and PU.1. J Virol 69, 253–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kanno Y, Levi B-Z, Tamura T, Ozato K, 2005. Immune cell-specific amplification of interferon signaling by the IRF-4/8-PU. 1 complex. Journal of interferon & cytokine research 25, 770–779. [DOI] [PubMed] [Google Scholar]
  17. Karki R, Lee E, Place D, Samir P, Mavuluri J, Sharma BR, Balakrishnan A, Malireddi RKS, Geiger R, Zhu Q, Neale G, Kanneganti TD, 2018. IRF8 Regulates Transcription of Naips for NLRC4 Inflammasome Activation. Cell 173, 920–933 e913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kautz B, Kakar R, David E, Eklund EA, 2001. SHP1 Protein-tyrosine Phosphatase Inhibits gp91PHOXand p67PHOX Expression by Inhibiting Interaction of PU. 1, IRF1, Interferon Consensus Sequence-binding Protein, and CREB-binding Protein with Homologous Cis Elements in the CYBB andNCF2 Genes. Journal of Biological Chemistry 276, 37868–37878. [DOI] [PubMed] [Google Scholar]
  19. Kenney SC, Mertz JE, 2014. Regulation of the latent-lytic switch in Epstein-Barr virus. Semin Cancer Biol 26, 60–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kim JY, Ozato K, 2009. The sequestosome 1/p62 attenuates cytokine gene expression in activated macrophages by inhibiting IFN regulatory factor 8 and TNF receptor-associated factor 6/NF-κB activity. The Journal of Immunology 182, 2131–2140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kong HJ, Anderson DE, Lee CH, Jang MK, Tamura T, Tailor P, Cho HK, Cheong J, Xiong H, Morse HC, 2007. Cutting edge: autoantigen Ro52 is an interferon inducible E3 ligase that ubiquitinates IRF-8 and enhances cytokine expression in macrophages. The Journal of Immunology 179, 26–30. [DOI] [PubMed] [Google Scholar]
  22. Konishi N, Narita Y, Hijioka F, Masud H, Sato Y, Kimura H, Murata T, 2018. BGLF2 Increases Infectivity of Epstein-Barr Virus by Activating AP-1 upon De Novo Infection. mSphere 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kurotaki D, Nakabayashi J, Nishiyama A, Sasaki H, Kawase W, Kaneko N, Ochiai K, Igarashi K, Ozato K, Suzuki Y, Tamura T, 2018. Transcription Factor IRF8 Governs Enhancer Landscape Dynamics in Mononuclear Phagocyte Progenitors. Cell Rep 22, 2628–2641. [DOI] [PubMed] [Google Scholar]
  24. Kurotaki D, Osato N, Nishiyama A, Yamamoto M, Ban T, Sato H, Nakabayashi J, Umehara M, Miyake N, Matsumoto N, 2013. Essential role of the IRF8-KLF4 transcription factor cascade in murine monocyte differentiation. Blood 121, 1839–1849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kuwata T, Gongora C, Kanno Y, Sakaguchi K, Tamura T, Kanno T, Basrur V, Martinez R, Appella E, Golub T, 2002. Gamma interferon triggers interaction between ICSBP (IRF-8) and TEL, recruiting the histone deacetylase HDAC3 to the interferon-responsive element. Molecular and cellular biology 22, 7439–7448. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lace MJ, Anson JR, Klingelhutz AJ, Harada H, Taniguchi T, Bossler AD, Haugen TH, Turek LP, 2009. Interferon-beta treatment increases human papillomavirus early gene transcription and viral plasmid genome replication by activating interferon regulatory factor (IRF)-1. Carcinogenesis 30, 1336–1344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li R, Liao G, Nirujogi RS, Pinto SM, Shaw PG, Huang T-C, Wan J, Qian J, Gowda H, Wu X, 2015. Phosphoproteomic profiling reveals Epstein-Barr virus protein kinase integration of DNA damage response and mitotic signaling. PLoS Pathog 11, e1005346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Li R, Pan Y, Shi DD, Zhang Y, Zhang J, 2013. PIAS1 negatively modulates virus triggered type I IFN signaling by blocking the DNA binding activity of IRF3. Antiviral Res 100, 546–554. [DOI] [PubMed] [Google Scholar]
  29. Li R, Zhu J, Xie Z, Liao G, Liu J, Chen MR, Hu S, Woodard C, Lin J, Taverna SD, Desai P, Ambinder RF, Hayward GS, Qian J, Zhu H, Hayward SD, 2011. Conserved herpesvirus kinases target the DNA damage response pathway and TIP60 histone acetyltransferase to promote virus replication. Cell Host Microbe 10, 390–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu B, Liao J, Rao X, Kushner SA, Chung CD, Chang DD, Shuai K, 1998. Inhibition of Stat1-mediated gene activation by PIAS1. Proc Natl Acad Sci U S A 95, 10626–10631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Liu B, Mink S, Wong KA, Stein N, Getman C, Dempsey PW, Wu H, Shuai K, 2004. PIAS1 selectively inhibits interferon-inducible genes and is important in innate immunity. Nat Immunol 5, 891–898. [DOI] [PubMed] [Google Scholar]
  32. Liu B, Shuai K, 2008. Targeting the PIAS1 SUMO ligase pathway to control inflammation. Trends Pharmacol Sci 29, 505–509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Liu B, Tahk S, Yee KM, Fan G, Shuai K, 2010. The ligase PIAS1 restricts natural regulatory T cell differentiation by epigenetic repression. Science 330, 521–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Liu B, Yang R, Wong KA, Getman C, Stein N, Teitell MA, Cheng G, Wu H, Shuai K, 2005. Negative regulation of NF-kappaB signaling by PIAS1. Mol Cell Biol 25, 1113–1123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Liu B, Yee KM, Tahk S, Mackie R, Hsu C, Shuai K, 2014. PIAS1 SUMO ligase regulates the self-renewal and differentiation of hematopoietic stem cells. EMBO J 33, 101–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Liu X, Cohen JI, 2016. Epstein-Barr Virus (EBV) Tegument Protein BGLF2 Promotes EBV Reactivation through Activation of the p38 Mitogen-Activated Protein Kinase. Journal of virology 90, 1129–1138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lv D-W, Zhong J, Zhang K, Pandey A, Li R, 2017. Understanding Epstein-Barr Virus Life Cycle with Proteomics: A Temporal Analysis of Ubiquitination During Virus Reactivation. OMICS: A Journal of Integrative Biology 21, 27–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lv DW, Zhang K, Li R, 2018. Interferon regulatory factor 8 regulates caspase-1 expression to facilitate Epstein-Barr virus reactivation in response to B cell receptor stimulation and chemical induction. PLoS Pathog 14, e1006868. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Marquis J-F, Kapoustina O, Langlais D, Ruddy R, Dufour CR, Kim B-H, MacMicking JD, Giguère V, Gros P, 2011. Interferon regulatory factor 8 regulates pathways for antigen presentation in myeloid cells and during tuberculosis. PLoS Genet 7, e1002097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. McKenzie J, Lopez-Giraldez F, Delecluse H-J, Walsh A, El-Guindy A, 2016. The Epstein-Barr Virus Immunoevasins BCRF1 and BPLF1 Are Expressed by a Mechanism Independent of the Canonical Late Pre-initiation Complex. PLoS Pathogens 12, e1006008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Netherby CS, Messmer MN, Burkard-Mandel L, Colligan S, Miller A, Cortes Gomez E, Wang J, Nemeth MJ, Abrams SI, 2017. The Granulocyte Progenitor Stage Is a Key Target of IRF8-Mediated Regulation of MyeloidDerived Suppressor Cell Production. J Immunol 198, 4129–4139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. O’Flaherty BM, Soni T, Wakeman BS, Speck SH, 2014. The murine gammaherpesvirus immediate-early Rta synergizes with IRF4, targeting expression of the viral M1 superantigen to plasma cells. PLoS Pathog 10, e1004302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Pang SH, Minnich M, Gangatirkar P, Zheng Z, Ebert A, Song G, Dickins RA, Corcoran LM, Mullighan CG, Busslinger M, Huntington ND, Nutt SL, Carotta S, 2016. PU.1 cooperates with IRF4 and IRF8 to suppress pre-B-cell leukemia. Leukemia 30, 1375–1387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rivailler P, Jiang H, Cho YG, Quink C, Wang F, 2002. Complete nucleotide sequence of the rhesus lymphocryptovirus: genetic validation for an Epstein-Barr virus animal model. J Virol 76, 421–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Salem S, Langlais D, Lefebvre F, Bourque G, Bigley V, Haniffa M, Casanova JL, Burk D, Berghuis A, Butler KM, Leahy TR, Hambleton S, Gros P, 2014. Functional characterization of the human dendritic cell immunodeficiency associated with the IRF8(K108E) mutation. Blood 124, 1894–1904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Shin DM, Lee CH, Morse HC 3rd, 2011. IRF8 governs expression of genes involved in innate and adaptive immunity in human and mouse germinal center B cells. PLoS One 6, e27384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shuai K, Liu B, 2005. Regulation of gene-activation pathways by PIAS proteins in the immune system. Nat Rev Immunol 5, 593–605. [DOI] [PubMed] [Google Scholar]
  48. Tammsalu T, Matic I, Jaffray EG, Ibrahim AFM, Tatham MH, Hay RT, 2014. Proteome-wide identification of SUMO2 modification sites. Science signaling 7, rs2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tamura T, Ozato K, 2002. ICSBP/IRF-8: its regulatory roles in the development of myeloid cells. Journal of interferon & cytokine research 22, 145–152. [DOI] [PubMed] [Google Scholar]
  50. Tamura T, Yanai H, Savitsky D, Taniguchi T, 2008. The IRF family transcription factors in immunity and oncogenesis. Annu. Rev. Immunol 26, 535–584. [DOI] [PubMed] [Google Scholar]
  51. Toropainen S, Malinen M, Kaikkonen S, Rytinki M, Jaaskelainen T, Sahu B, Janne OA, Palvimo JJ, 2015. SUMO ligase PIAS1 functions as a target gene selective androgen receptor coregulator on prostate cancer cell chromatin. Nucleic Acids Res 43, 848–861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Unlu S, Kumar A, Waterman WR, Tsukada J, Wang KZ, Galson DL, Auron PE, 2007. Phosphorylation of IRF8 in a pre-associated complex with Spi-1/PU. 1 and non-phosphorylated Stat1 is critical for LPS induction of the IL1B gene. Molecular immunology 44, 3364–3379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Uzoma I, Hu J, Cox E, Xia S, Zhou J, Rho HS, Guzzo C, Paul C, Ajala O, Goodwin CR, Jeong J, Moore C, Zhang H, Meluh P, Blackshaw S, Matunis M, Qian J, Zhu H, 2018. Global Identification of Small Ubiquitin-related Modifier (SUMO) Substrates Reveals Crosstalk between SUMOylation and Phosphorylation Promotes Cell Migration. Mol Cell Proteomics 17, 871–888. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wang L, Yao ZQ, Moorman JP, Xu Y, Ning S, 2014. Gene expression profiling identifies IRF4-associated molecular signatures in hematological malignancies. PloS one 9, e106788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Yamamoto H, Kihara-Negishi F, Yamada T, Hashimoto Y, Oikawa T, 1999. Physical and functional interactions between the transcription factor PU. 1 and the coactivator CBP. Oncogene 18, 1495–1501. [DOI] [PubMed] [Google Scholar]
  56. Young LS, Arrand JR, Murray PG, 2007. EBV gene expression and regulation. [PubMed] [Google Scholar]
  57. Zhang K, Lv D-W, Li R, 2017. B Cell Receptor Activation and Chemical Induction Trigger Caspase-Mediated Cleavage of PIAS1 to Facilitate Epstein-Barr Virus Reactivation. Cell Rep 21: 3445–3457. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zhang K, Lv DW, Li R, 2019. Conserved Herpesvirus Protein Kinases Target SAMHD1 to Facilitate Virus Replication. Cell Rep 28, 449–459 e445. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES