NS1-mediated enhancement of MVC transcription and replication promoted by KAT5/H4K12ac

ABSTRACT

Histone modifications function in both cellular and viral gene expression. However, the roles of acetyltransferases and histone acetylation in parvoviral infection remain poorly understood. In the current study, we found the histone deacetylase (HDAC) inhibitor, trichostatin A (TSA), promoted the replication and transcription of parvovirus minute virus of canines (MVC). Notably, the expression of host acetyltransferases KAT5, GTF3C4, and KAT2A was increased in MVC infection, as well as H4 acetylation (H4K12ac). KAT5 is not only responsible for H4K12ac but also crucial for viral replication and transcription. The viral nonstructural protein NS1 interacted with KAT5 and enhanced its expression. Further study showed that Y44 in KAT5, which may be tyrosine-phosphorylated, is indispensable for NS1-mediated enhancement of KAT5 and efficient MVC replication. The data demonstrated that NS1 interacted with KAT5, which resulted in an enhanced H4K12ac level to promote viral replication and transcription, implying the epigenetic addition of H4K12ac in viral chromatin-like structure by KAT5 is vital for MVC replication.

IMPORTANCE

Parvoviral genomes are chromatinized with host histones. Therefore, histone acetylation and related acetyltransferases are required for the virus to modify histones and open densely packed chromatin structures. This study illustrated that histone acetylation status is important for MVC replication and transcription and revealed a novel mechanism that the viral nonstructural protein NS1 hijacks the host acetyltransferase KAT5 to enhance histone acetylation of H4K12ac, which relies on a potential tyrosine phosphorylation site, Y44 in KAT5. Other parvoviruses share a similar genome organization and coding potential and may adapt a similar strategy for efficient viral replication and transcription.

INTRODUCTION

Minute virus of canines (MVC) was first isolated from canine feces in 1970 (1). It causes respiratory diseases and gastroenteritis with severe diarrhea in puppies (2, 3) and is often associated with other viruses (4). MVC belongs to the family Parvoviridae and is a small, nonenveloped, single-stranded DNA virus containing a 5,402 nucleotide (nt) linear genome flanked by terminal palindromes (5, 6). Similar to other bocavirus members, MVC generates various mRNA transcripts from a single promoter (P6) located at the 5′ end of the genome (5, 7, 8), which encode four viral proteins, namely, two nonstructural proteins, NS1 and NP1, and two structural proteins, VP1 and VP2 (7, 9–12). Among them, NS1 contains DNA-binding and endonuclease domains within the N-terminus, ATPase and helicase domains in the center, and a transactivation domain at the C-terminus, which play an essential role in viral replication upon interaction with various cellular proteins (13–17). NP1 regulates alternative polyadenylation and alternative splicing of MVC pre-mRNA and controls the expression of structural proteins. It promotes the read-through of the proximal polyadenylation cleavage site, p(A)p, and splicing of the upstream intron. However, it also facilitates viral replication by interacting with host factors (18–21).

Histone modifications and their chromatin modifiers play important roles in regulating gene expression (22). Chromatin which consists of DNA and histones is a transcriptionally silent structure. The histone modifications at certain regions of chromatin, including phosphorylation, ubiquitination, acetylation, and methylation, regulate chromatin structure and transcriptional status and further affect host gene expression (23). Among these modifications, histone acetylation, a hallmark of active transcription (24), is a reversible process catalyzed by histone acetyltransferases (HATs), such as GNAT, p300/CBP, and MYST, and removed by histone deacetylases (HDACs). Histone acetyltransferase KAT5 (TIP60) is a catalytic subunit of the NuA4 complex (nucleosomal acetyltransferase of histone H4) identified in yeast and is involved in the transcriptional activation of select genes, principally through the acetylation of histones H4 and H2A (25). KAT5 and its complex partners, especially P400, interact with H4K12ac modifications in heterochromatin and euchromatin, thus allowing for the maintenance of chromatin stability and integrity (26). Additionally, the acetyltransferase activity of KAT5 is controlled by posttranslational modifications, such as SUMO, phosphorylation, and ubiquitination (27, 28).

The histone acetylation and host acetyltransferases are also important factors in various DNA virus infections (29–33). The entering viral DNAs in host cells are wrapped around host histones and densely compacted into repressive heterochromatin, which is not accessible to the transcription machinery, resulting in the silencing of viral DNA (34, 35). Thus, efficient DNA virus infection requires host chromatin modifiers and remodelers to modify viral DNA-associated histones (36–39). A recent study has shown that the replicating DNA of parvoviruses is chromatinized with host histones in a specific nuclear structure called parvovirus-associated replication bodies (PAR-bodies) during infection (40, 41). Thus, as many other DNA viruses (29–33), MVC needs to recruit the host epigenetic complex into PAR-bodies to modify the histones and overcome heterochromatin-induced gene silencing. Indeed, enriched histone acetylation in the promoter region during active viral replication is observed in canine parvoviruses for the efficient completion of viral DNA (40). However, the types of modifiers recruited to the parvovirus DNA and the viral proteins involved remain unknown.

In the current study, we characterize the role of host histone acetylation and acetyltransferase in MVC replication and identified KAT5/H4K12ac as an essential factor during MVC infection. The viral protein NS1 interacted with host acetyltransferase KAT5 to promote histone acetylation of H4K12ac, which enhances viral replication and transcription. Our data reveal that KAT5/H4K12ac is an important factor for MVC replication and transcription.

RESULTS

Histone acetylation is involved in MVC replication and transcription

The MVC DNA forms a chromatin-like structure during replication, which wraps around the histone proteins (40, 41). These viral DNA-associated histones need to be modified to an open chromatin structure for efficient viral gene expression (36–39). To investigate whether acetylation of host histones is involved in MVC infection, the levels of viral protein, RNA, and DNA were assessed in MVC-infected Walter Reed Canine cell/3873D (WRD) cells treated with the HDAC inhibitor, trichostatin A (TSA) (Fig. 1A). The major protein isoform of NS1 (66KD) was slightly increased by TSA treatment, whereas the expression of NP1 and VP2 remained unchanged (Fig. 1B). Viral transcripts were also significantly upregulated (Fig. 1C). As the 5,000 bp band represented unspliced mRNA terminated at the p(A)d site (5), the enhanced RNA transcription might the reason of the increased protein expression of NS1 (Fig. 1A and B). Additionally, MVC replication was measured using Southern blotting as described previously (42). TSA treatment resulted in a significant increase in the replicative form (RF) DNA and the single-stranded DNA genome (ssDNA) (Fig. 1D). These data suggest that TSA promotes viral replication and transcription.

Fig 1.

Positive effects of trichostatin A (TSA) on viral DNA replication, protein expression, and RNA alternative splicing and polyadenylation in MVC infection. (A) WRD cells were treated with DMSO or 2 µM TSA and were then infected with MVC, as shown in the diagram. (B) Western blot analysis of viral protein expression in MVC-infected cells with or without TSA treatment 48 h post-infection; GAPDH was used as an internal control. (C) Total RNA was extracted from MVC-infected cells with or without TSA treatment and subjected to Northern blot analysis 48 h post-infection. The MVC full-length probe targeting the 1–5,402 region was used for hybridization. Ethidium bromide (EB) staining of the 18S RNA was used as an internal control. (D) Hirt DNA was extracted from the infected cells and subjected to Southern blot analysis using the same probe used in the Northern blot at 48 h post-infection. A linear DNA fragment of 5,402 bp length excised from pIMVC (5) was used as a marker (lane 1). EB staining of Hirt DNA was used as an internal control. RF, replicative form of MVC DNA; dRF, dimer of RF; single-strand (ss) DNA genome of MVC. (E) The top panel illustrates a diagram of the MVC genomic structure, including the P6 promoter, four splice donor sites (1D, 1D’, 2D, and 3D), three acceptor sites (1A, 2A, and 3A), the proximal polyadenylation cleavage site [p(A)p], and the distal polyadenylation cleavage site [p(A)d]. The bottom panel shows the positions and sizes of the probes used in the RNase protection assay (RPA). The size of the expected RPA products from un-spliced RNA (unspl), spliced RNA (spl), and RNA utilizing the p(A)p site or p(A)d site is given below each probe. (F–I) RNase protection assay (RPA) of total RNA extracted from MVC-infected cells with or without TSA treatment at 48 h post-infection was performed using 1D-probe (F), 2D-probe (G), 3D-probe (H), and p(A)p-probe (I) targeting the indicated donor sites and p(A)p, respectively. Radioactive probes were loaded into each experiment and served as size controls. The read-through (RT) product shown in panel H represents the RNA transcript that extends through the p(A)p and is terminated by the p(A)d site.

MVC transcription is initiated at a single unique promoter (P6) and coupled with alternative RNA splicing and polyadenylation (Fig. 1E). To determine whether TSA treatment regulated RNA processing, we performed an RNase protection assay (RPA) using four probes targeting donor sites 1D, 2D, 3D, and the p(A)p cleavage site (Fig. 1E). Consistent with the Northern blotting results (Fig. 1C), more RPA products were generated in each TSA-treated sample (Fig. 1F through I). The ratio of unspliced and spliced products was increased to those of the untreated samples (Fig. 1F through H). In particular, more than two-fold of viral RNAs were spliced at the 1D site in TSA-treated samples (Fig. 1F). In addition, consistent with the observation of the previous report (19, 43), the RT product, which represented the RNAs read-through p(A)p site and terminated by the p(A)d site, was more than the p(A)d products (Fig. 1I), indicating a higher usage of the p(A)d site. Moreover, more MVC RNAs were polyadenylated at the p(A)d site upon TSA treatment (Fig. 1I). Taken together, these data suggest that the global histone acetylation status is important for viral replication, transcription, and viral RNA processing.

KAT5 and H4K12ac are essential for MVC replication and transcription

To determine whether MVC infection influenced histone acetylation, we evaluated several types of histone acetylation in MVC-infected cells (Fig. 2A; Fig. S1A). The elevated levels of pan-acetylation of histone H3 (H3ac), H4 (H4ac), and histone H4 acetyl K12 (H4K12ac) were observed in MVC infection (Fig. 2A), while the level of histone H3 acetyl K9 (H3K9ac) was not changed. Furthermore, TSA treatment synergized with MVC infection to increase H4ac but not H3ac (Fig. 2A; Fig. S1A). Histone acetylation usually favors gene transcription (44); therefore, the effects of TSA treatment on viral transcription may be mediated by H4ac.

Fig 2.

Identification of histone acetyltransferases (HATs) responsible for H4K12ac modification in MVC infection. (A) TSA synergization with MVC infection increased H4ac and H4K12ac. WRD cells were treated with DMSO or 2 µM TSA and were then infected with MVC as described in Fig. 1. The levels of histone deacetylases (HDACs) and epigenetic modifications of histone H3 and H4 in MVC-infected cells with or without TSA treatment were measured using Western blot analysis at 48 h post-infection. While the same samples in Fig. 1B were used to assess the histone modifications, the image of GAPDH control was reused in this panel. (B) Screening of HAT and HDAC expressions altered by MVC infection. The levels of HATs in MVC-infected cells were measured using Western blot analysis at 48 h post-infection; GAPDH was used as an internal control. (C–E) Verification of knockdown efficiency. The expression of HATs and the modified histone acetylation was examined using Western blot in shRNA-expressing cells after puromycin selection shown in B and C. GAPDH or actin was used as an internal control. The relative level of histone acetylation or the relative expression of acetyltransferases shown under each protein was calculated by band intensity after being normalized to their internal control. (F–G) Knockdown of HATs decreased MVC DNA replication determined using Southern blot (F) and RNA expression using Northern blot (G). WRD cells were transfected with two shRNA-expressing vectors targeting KAT5 (Lanes 2 and 3), GTF3C4 (Lanes 4 and 5), and KAT2A (Lanes 6 and 7); non-targeting shRNA was used as the negative control (shNC). The shRNA-expressing cells were selected under puromycin for 5–7 days and infected with MVC. (F) Hirt DNA was extracted at 48 h post-infection. Southern blot analysis was performed as described in Fig. 1C. EB staining of Hirt DNA served as an internal control. RF, replicative form of MVC DNA; dRF, dimer of RF; single-strand DNA genome of MVC. (G) Total RNA was extracted at 48 h post-infection. Northern blot analysis was performed as described in Fig. 1B. EB staining of 18S RNA served as an internal control.

We next checked whether enhanced histone acetylation was the consequence of aberrant HAT expression during MVC infection. By screening six well-characterized HATs and two HDACs, we observed a 2.8-fold increase in KAT5, a 1.6-fold increase in GTF3C4, and a 1.8-fold increase in KAT2A in MVC-infected samples (Fig. 2B). The loss of KAT5 is linked to hypoacetylated H4K12 in the promoter region (45). Consistently, knockdown of KAT5 by shRNAs was associated with decreased H4K12ac (Fig. 2C), whereas knockdown of GTF3C4 or KAT2A was associated with decreased H3ac (Fig. 2D and E). Furthermore, the depletion of all three HATs reduced viral replication (Fig. 2F), whereas the depletion of KAT5 and GTF3C4 reduced viral transcription (Fig. 2G). KAT2A knockdown, coupled with the deletion of H3ac, showed little effect on MVC transcription (Fig. 2E and G), indicating that H3ac was not an essential factor for viral transcription, but possibly for viral replication (Fig. 2F). In addition, knockdowns of KAT5, GTF3C4, and KAT2A did not affect cell viability, indicating that the impact of these knockdowns on viral replication or transcription was not because of an altered cell cycle (Fig. S1B through E). Taken together, these results indicate that KAT5 plays a key role in viral replication and transcription by regulating the level of H4K12ac.

Enhanced acetylation of H4K12 subunits was mediated by NS1 protein in MVC infection

Since activated histone acetylation is often regulated by viral proteins during the virus infection (46–48) and histone acetylation of H4K12ac is important for MVC replication and transcription, we determined the nonstructural viral proteins that are involved in the regulation of histone acetylation. MVC encodes two nonstructural proteins, NS1 and NP1 (3), which are required for viral replication and RNA processing (Fig. 3A, top). Thus, the start codons of NS1 or NP1 were mutated in an MVC infectious clone to block their expression (Fig. 3A, bottom). The level of H4K12ac was significantly reduced in the NS1-mt mutant and was restored by overexpression of Flag-tagged NS1 (Fig. 3B, left). Moreover, Flag-NS1 expression was sufficient to increase H4K12ac levels in WRD cells without viral infection (Fig. 3B, right). In contrast, NP1 did not affect histone acetylation (Fig. 3C). Taken together, acetylation of H4K12 subunits was enhanced by the NS1 protein during MVC infection.

Fig 3.

MVC infection modulated histone H4 epigenetic modification, H4K12ac, through viral protein NS1. (A) ORFs and transcripts of viral protein NS1 and NP1 are shown under the MVC genomic structure. Long isoforms read-through p(A)p to p(A)d are not shown. At least two protein isoforms of NS1 were expressed upon alternative splicing. Four pairs of primers for CHIP-qPCR analysis are shown below the MVC genome. The NS1 or NP1 knockout infection clone, NS1-mt or NP1-mt, respectively, was generated by the elimination of the start codon in each ORF as shown at the bottom. (B and C) NS1, but not NP1, regulates acetylation of H4, but not H3, during MVC infection. Left panel: WRD cells were transfected with wild-type (WT), NS1-mt, or NP1-mt of MVC infection clones as shown in B, respectively. Right panel: WRD cells were transfected with Flag-NS1 (B) or Flag-NP1 (C); cells transfected with an empty Flag vector were used as a control. The levels of H3 and H4 acetylation were measured using Western blot analysis at 48 h post-transfection. Detection of NS1 and NP1 indicated efficient transfection, and actin was used as an internal control. (D) ChIP-qPCR was performed with an anti-H4ac antibody with chromatin preps from MVC-infected cells with or without TSA treatment using four pairs of primers shown in A. ***, P < 0.001 denotes statistical significance by two-tailed Student’s t test.

To further investigate whether H4ac was present on the MVC genome, the relative levels of the H4ac-associated MVC genome in the presence or absence of TSA were quantified by CHIP-qPCR analysis using four pairs of primers (Fig. 3A). These data suggested the MVC genome was associated with H4ac, although the level of genome-associated H4ac was not altered by TSA treatment (Fig. 3D). These data also supported that MVC DNA was chromatinized with host histones, especially acetylated H4.

NS1 activated KAT5/H4K12ac by tyrosine phosphorylation, which promotes viral replication

Given that KAT5 is necessary for the increased acetylation of H4K12 subunits, MVC replication, and transcription, and that NS1 is required to increase H4K12ac levels, we speculated that they might be in the same replication complex. To this end, we performed co-immunoprecipitation (co-IP) assays and found that KAT5, but not GTF3C4 or KAT2A, co-immunoprecipitated with an anti-Flag antibody against Flag-NS1, regardless of MVC replication (Fig. 4A). However, none of the three HATs was associated with NP1 (Fig. 4B). Although five KAT RNA isoforms were predicted in the latest NCBI Canis lupus familiaris annotation (Fig. S2A), only one isoform (XM_038563907.1) was detected and cloned from the WRD cells (Fig. S2B). By co-expressing NS1 and KAT5 in WRD cells, we found that KAT5 expression was promoted by NS1 in a dose-dependent manner, which was also associated with enhanced tyrosine phosphorylation (Fig. 4C). The most likely phosphorylated tyrosine residue in KAT5 was predicted using GPS (http://gps.biocuckoo.cn/). Tyrosine 470 (Y470) had a much higher score than other ones (Fig. S2C). Thus, mutations were introduced at Y470 and another reported position, Y44. The increase in the level of the Y44F mutant by NS1 was much lower than that of WT and Y470F (Fig. 4D). Moreover, the Y44F mutant decreased MVC replication (Fig. 4E) compared to that of the Y470F mutant, indicating that Y44 is indispensable for KAT5-mediated MVC replication. To map the interaction domains of KAT5 and NS1, we constructed an N-terminus (aa 1–424) and a C-terminus (aa 425–576) of the NS1 major isoform (66KD) and an N-terminus (aa 1–235) and a C-terminus (aa 236–522) of KAT5 for co-IP analyses. As shown in Fig. 5A and B, their interaction was dependent on the C terminus of each other. Therefore, NS1 interacts with KAT5 through the C-terminal domain to promote its expression and tyrosine phosphorylation for MVC replication.

Fig 4.

NS1 interacted with KAT5 and promoted its expression. (A) Only KAT5 in three MVC-related HATs interacted with NS1 regardless of viral infection. WRD cells were transfected with pFlag-NS1 (Lane 1) or co-transfected with pFlag-NS1 and NS1-mt infectious clone (Lane 2). IP was performed at 24 h post-transfection with an anti-Flag antibody, and normal IgG was used as a control (top panel). Three endogenous MVC-related HATs and NS1 in both IP and input samples were immunoblotted with specific antibodies; actin was used as a loading control. (B) None of the three MVC-related HATs interacted with NP1. Transfection, IP, and Western blot were performed as shown in A. (C) NS1 promoted the expression and tyrosine phosphorylation of KAT5. WRD cells were co-transfected with the indicated dose of pFlag-KAT5 and pHA-NS1. The expression of KAT5 and NS1 was measured using Western blot at 48 h post-transfection (bottom). Anti-Flag IP was performed with extracts isolated from the samples in Lanes 1 and 3. KAT5 and NS1 expression and the total level of tyrosine phosphorylation of KAT5 were measured using Western blot in IP samples. (D) The expressions of KAT5 and NS1 in WRD cells co-transfected with pHA-NS1 and empty Flag vector (Lane 1), pHA-NS1 and wild-type pFlag-KAT5 (Lane 2), pHA-NS1 and Tyr44 mutated pFlag-KAT5 (Lane 3), or pHA-NS1 and Tyr470 mutated pFlag-KAT5 (Lane 4) were measured using Western blot. (E) Tyr44 mutant of KAT5 failed to promote MVC replication. MVC DNA replications in viral-infected WRD cells transfected with an empty Flag vector (Lane 2), wild-type pFlag-KAT5 (Lane 3), Tyr44 mutated pFlag-KAT5 (Lane 4), or Tyr470 mutated pFlag-KAT5 (Lane 5) were determined by Southern blot, as described in Fig. 1C. A linear DNA fragment of 5,402 bp length from the plasmid was used as a marker (Lane 1). Ethidium bromide (EB) staining of Hirt DNA was used as an internal control. RF, replicative form of MVC DNA; dRF, dimer of RF; single-strand (ss) DNA genome of MVC.

Fig 5.

NS1 interacted with KAT5 through the C-terminal domain. (A) WRD cells were co-transfected with pFlag-KAT5 and pHA-NS1, pHA-NS1-N, or pHA-NS1-C. IP was performed at 48 h post-transfection with an anti-Flag antibody, and normal IgG was used as a control. NS1 and KAT5 in both IP and input samples were immunoblotted with anti-Flag or anti-HA. (B) WRD cells were co-transfected with pHA-NS1 and pFlag-KAT5, pFlag-KAT5-N, or pFlag-KAT5-C. IP and immunoblotting were performed as described in A. LC, light chain of IgG. (C) Proposed model of how MVC hijacks the host acetyltransferase KAT5 to enhance histone acetylation of H4K12ac through the viral nonstructural protein NS1.

DISCUSSION

Although parvoviral genomes are chromatinized in host cells, data on histone acetylation and acetyltransferases required to overcome epigenetic silencing remain unclear. In this study, MVC-enhanced H4K12ac was found to play an important role in viral replication and transcription, which are mediated by the host acetyltransferase KAT5 and the viral protein NS1.

Interplay with the host epigenetic complex is important for various viruses to establish efficient infections, especially for DNA viruses to overcome heterochromatin-induced gene silencing (29–33). Our study showed that the expression of both H3ac and H4ac was increased in MVC infection. H4ac might be indispensable for viral replication and transcription, which is supported by two lines of evidence: (1) TSA treatment only boosting H4ac enhanced both viral replication and transcription and (2) reduction of H3ac by depleting KAT2A did not alter viral replication. Furthermore, we identified H4K12ac and KAT5 as essential by screening a few types of H4 acetylation, acetyltransferases, and deacetylases. The colocalization of H4K12ac and KAT5 on euchromatin and heterochromatin in different types of breast tumors suggests their potential interaction (26). KAT5 also catalyzes histone H4K12 acetylation and promotes genetic instability (45). KAT5 knockdown was accompanied by a decreased acetylation of H4K12 subunits and a reduction in MVC replication and transcription; therefore, we propose that this specific host acetylation system is required for MVC infection, which most likely assists the virus in overcoming epigenetic silencing.

Activated histone acetylation is regulated by viral proteins during infection. For example, the herpes simplex virus (HSV) protein VP16 recruits the histone acetyltransferase required for H3K9ac acetylation to viral DNA and activates the transcription of immediate early genes (IEGs) (49), and similar mechanisms are proposed in other DNA viruses (46–48). Although H3K27ac has been shown to colocalize with NS1 and is enriched in the promoter region of canine parvovirus (40), how the parvovirus protein recruits host acetyltransferases to modify the histones is still poorly understood. We demonstrated NS1 participated in this process: (1) knockdown of NS1 in a mutated virus failed to stimulate H4K12ac and its acetyltransferase KAT5 and (2) the increased expression of KAT5 was dependent on the interaction with NS1 through the C-terminal domain.

During viral replication, NS1 is localized in PAR-bodies and binds to the viral replication origin to initiate viral DNA replication and transactivate the viral P6 promoter (50) and some host genes (51, 52). Given that the C-terminus of NS1 contains a putative transactivation domain, the interaction with KAT5 through this domain may be necessary for its trans-activity. Thus, KAT5 may be recruited by NS1 to chromatinized viral DNA and maintain the H4K12ac level, which may help the virus to unwind the highly packed DNA structure for accessibility to other important replication or transcription factors. In addition, NS1 stabilizes KAT5 via posttranslational modifications (Fig. 5C). Since tyrosine phosphorylation of KAT5, but not of other modifications, was not altered by MVC infection (data not shown), mutations were introduced at tyrosines 44 and 470, according to a previous study and a GPS program based on a verified KAT isoform in WRD cells. KAT5 phosphorylation at Y44, but not Y470, is required for NS1-dependent stabilization, and the mutant protein failed to enhance MVC replication.

Since parvoviruses share similar genome organization and coding potentials along with replication and transcription strategies (43, 53), they might use a similar histone acetylation system during infection. Further studies are warranted to confirm the presence of epigenetic modifications in other parvoviruses. Therefore, antiviral drugs may be developed to target this conserved histone acetylation system to combat multiple parvoviruses in the future. In summary, the recruitment of host acetyltransferases to acetylate histones is necessary for the MVC life cycle. Our study indicates that targeting the histone acetylation system KAT5/H4K12ac is a promising strategy for preventing MVC infection.

MATERIALS AND METHODS

MVC virus and cells

The original strain of MVC (GA3) and Walter Reed canine cell/3873D (WRD) cells were kindly gifted by Dr. Jianming Qiu of the University of Kansas Medical Center. HEK293T (ATCC, CRL-11268) and WRD cells were cultured in Dulbecco’s modified Eagle medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS) at 37°C with 5% CO₂.

Virus amplification and quantification

WRD cells were seeded in 10-cm plates 1 day before and infected with MVC virus at 80% confluence with a multiplicity of infection (MOI) of 1 × 10⁴ genomic copies/cell. Infected cells were cultured in DMEM containing 2% FBS for 3 days. After freezing and thawing thrice, the cell debris was removed via low-speed centrifugation. The collected culture supernatant containing released viruses was further filtered with a 0.45-µm filter and stored at –80°C.

Absolute MVC genomic copies were quantified using qPCR as previously described (54). Briefly, viral DNA was extracted using Hirt solution (10 mM Tris, 10 mM EDTA pH 7.5, 0.6% SDS), and qPCR was performed using the SYBR Green Master Mix kit (Yeasen Biotech Co., Shanghai, China) with a forward primer of 5′-AGG ACC ATC GCT TGG ATA CATT-3′ and a reverse primer of 5′-TAC TGG TCC GAG GGC TTG TT-3′. Serial dilutions of the pIMVC plasmid were used to generate a standard curve for absolute quantification.

Plasmid construction

Plasmids Flag-NS1, Flag-NP1, HA-NS1, and Flag-KAT5 were constructed by inserting the coding sequences of NS1, NP1, and KAT5 into pXJ40-Flag, pXJ40-HA, and p3XFlag-CMV-14 (Sigma-Aldrich, St. Louis, MO, USA), respectively. Plasmids Flag-KAT5-Tyr44 and Flag-KAT5-Tyr470 were mutated by substituting tyrosine at sites 44 and 470 in KAT5 with phenylalanine. NS1-mt and NP1-mt were constructed by mutating the start codons of NS1 or NP1, respectively, in the MVC infectious clone pIMVC. The primers used for plasmid construction are listed in Table S1.

Short hairpin RNA (shRNA) knockdown and transfection

DNA fragments containing gene-specific shRNA sequences were cloned into the pLKO.1-TRC vector (Plasmid 10878; Addgene, Cambridge, MA, USA). Lentivirus-expressing shRNAs were packaged via co-transfection with psPAX2 and pMD2.G into HEK293T cells. Stable knockdown cell lines were generated by lentiviral infection assisted with polybrene (0.75 µg/mL) and selected with puromycin at 1 µg/mL. The shRNAs used in the study were as follows: KAT5 (shKAT5-1:5-GCA AGG GTA CCA TCT CTT TCT-3, shKAT5-2:5-GCT GAT CGA GTT CAG CTA TGA-3), KAT2A (shKAT2A-1:5- GCT GAA CTT TGT GCA GTA CAA-3, shKAT2A-2:5-GGC TAC CTA CAA GGT CAA TTA-3), and GTF3C4 (shGTF3C4-1:5-CCA TCT CTT CAT GCA ACA CAA-3, sh GTF3C4-2:5-CCT GCC AGA GTT TGA TAT ATA-3).

Plasmid transfection of WRD or 293T cells was performed using the Lipofectamine 2000 reagent (Invitrogen, 11668–019), according to the manufacturer’s instructions.

Southern blot analysis

Low-molecular-weight (Hirt) DNA was isolated to detect the replicating forms of MVC, as described previously (42, 55). After transfection of infectious clones or MVC virus infection, WRD cells were washed twice with phosphate-buffered saline (PBS) and lysed with a Hirt extraction solution (10 mM Tris, 10 mM EDTA pH 7.5, 0.6% SDS) followed by proteinase K (0.5 mg/mL) treatment, and the DNA was extracted using the phenol:chloroform method. The purified DNA was separated on 1% agarose gel and transferred to the Hybond N + membrane, followed by ultraviolet cross-linking. The membranes were hybridized with an MVC genome probe (nt 1–5,402), which was generated using the DIG-High Prime DNA Labeling and Detection Starter Kit II (Roche) according to the manufacturer’s instructions. Signals were detected using a ChemiDoc MP imaging system (Bio-Rad).

RNA isolation and Northern blot

Total RNA from MVC-infected cells was extracted using the TRIzol reagent (Ambion) and run in 1% agarose gel containing 2.2 M formaldehyde for 9 h at 28 V. The separated RNA in agarose gel was transferred to a Hybond-N +membrane, followed by ultraviolet crosslinking. Signals were detected using the same probe described above for Southern blotting.

TSA treatment, Western blot, and co-immunoprecipitation

The WRD cells seeded in the 6-well plates were treated with DMSO or 2 µM TSA 2 hours before MVC infection. The expressions of viral proteins, HDACs, and epigenetic modifications of histone H3 and H4 were assessed by Western blot at 48 h post-infection as follows. WRD cells were lysed in SDS protein sample buffer 48 h post-transfection or infection. The samples were heat-denatured, separated using 12% SDS-PAGE gels, and transferred to a nitrocellulose membrane. Proteins were detected using primary monoclonal antibody against GAPDH (60004–1-lg, Proteintech, Rosemont, IL, USA); mouse monoclonal antibody against beta-actin (sc47778, Santa Cruz Biotechnology, Dallas, TX, USA); rabbit polyclonal antibody against KAT5 (GTX112197), anti-KAT7 (GTX102041), anti-GTF3C4 (GTX118874), anti-KAT2A (GTX114428), and anti-HAT1 (GTX110643) from GeneTex; anti-H3ac (39140), H3K9ac (39586), H4ac (39967), and H4K12ac (39166) from Active Motif; anti-HDAC1 (10197–1-AP, Proteintech); anti-HDAC2 (12922–3-AP, Proteintech) anti-Flag (F1804-1 MG, Sigma-Aldrich); anti-HA (66006–1-Ig, Proteintech); three rabbit polyclonal antibodies against MVC NP1, NS1, and VP2 generated in rabbit (5). Secondary antibodies goat anti-mouse IgG and goat anti-rabbit IgG were purchased from AntiGene Biotech GmbH (Stuttgart, Germany). Luminescence signals were detected using a ChemiDoc MP imaging system (Bio-Rad).

For immunoprecipitation (IP), IgG or an anti-FLAG antibody was mixed with supernatants of cell lysates in IP lysis buffer (50 mM Tris-HCl pH 7.5, 1 mM EGTA, 1 mM EDTA, 1% Triton X-100, 150 mM NaCl, 2 mM DTT) supplemented with protease inhibitors (DI101-02, TransGen) for 2 h at 4°C and incubated with protein G agarose overnight at 4°C. Proteins were denatured at 95°C for 10 min and subjected to Western blotting.

RNase protection assay (RPA)

RPA was performed as previously described (8). Plasmids containing RPA probe sequences were constructed by cloning nt 250–500, nt 2,200–2,380, nt 2,400–2,548, and nt 3,107–3,333 of the MVC genome into the pBluescript KS (+) vector. The ³²P-labeled probes were transcribed from the EcoRI- or Xho1-digested templates using the T7 MEGAshortscript kit (AM1354, Thermo Fisher Scientific) in the presence of [a-³²P]-GTP (BLU006H500UC, PerkinElmer), according to the manufacturer’s instructions. Ten micrograms of total RNAs isolated from infected cells was denatured at 95°C for 5 min and incubated with the RPA probe in RPA buffer (80% formamide, 40 mM pipes pH 6.4, 400 mM NaCl, 1 mM EDTA) at 51°C overnight, followed by treatment with RNase A and RNase T1 (Thermo Fisher Scientific) for 1 h at 30°C and digestion with proteinase K(P9460, Solarbio) for 30 min at 37°C. The RPA products were precipitated by adding two volumes of 100% ethanol for 2 h at –20°C. The RNA samples were separated on 6% urea PAGE gel. The signals were detected under a Cyclone℗ Plus (PerkinElmer) and analyzed using OptiQuant software.

Chromatin immunoprecipitation (ChIP)

The ChIP-IT High-Sensitivity (HS) Kit from Active Motif (53040) was used for CHIP. The MVC-infected cells with or without TSA treatment were prepared as described above, and then the CHIP experiments were performed according to the manufacturer’s instruction. Briefly, the cells were fixed with a complete cell fixation solution for 15 min at RT. The stop solution was added to stop the fixation reaction for 5 min. The fixed cells were resuspended in ice-cold PBS and washed once, and then chromatin prep buffer with fresh protease inhibitor cocktail (PIC) and PMSF was added into the cells for 10-min incubation on ice. To disrupt the cell membranes, the cells were passed through a 25-gauge needle for 15 times, and nuclear extract was resuspended in ChIP buffer in the presence of PIC and PMSF. The fragmentation of chromatin was done by sonication. The input DNA was prepared, and the size of the DNA fragments was confirmed by running an agarose gel. The sheared chromatin was immunoprecipitated by anti-H4ac antibody (39043, Active Motif) overnight at 4°C and then mixed with protein G agarose. DNA was recovered and purified from the protein G agarose. The primers used for RT-qPCRwere listed in Table S6. At least three samples in each qPCR analysis wereprepared, and three independent experiments were performed.Four pairs of qPCR primers, P1 (5′- GAA GAA GAC ATA ACA GGT GA −3′) and P2 (5′- AAC AGT GGA GGA CGA TTG −3′), P3 (5′- CTA CGA GAC ATA TGA GCA AG −3′) and P4 (5′- CAT TTC TCT ACA TTG ATC CCA −3′), P5 (5′-CCA TTC AAT CCA CTA GAT AA-3′) and P6 (5′-TTG CGC CCT ATC TTG TCT A-3′), and P7 (5′-AGA CGC TAC TTC GCT ACA C-3′) and P8 (5′-TAC TGG ACT GAC ATC ATA A-3′) that span the MVC genome were used to measure the enriched DNA by H4ac antibody.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jvi.01695-23.

Supplemental material. jvi.01695-23-s0001.pdf.

Table S1; Fig. S1 and S2.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.