Conserved Herpesvirus Kinase ORF36 Activates B2 Retrotransposons during Murine Gammaherpesvirus Infection

MHV68 infection induces B2 SINEs in physiologically relevant antigen-presenting cell types.

Our previous work established that MHV68 infection of murine fibroblasts results in robust activation of B2 SINEs (21). While fibroblasts are commonly used to study MHV68 infection in vitro, two of the most physiologically relevant cell types for the in vivo MHV68 life cycle and establishment of lifelong latency are B cells and macrophages (35). We therefore sought to determine whether B2 SINE induction is also a feature of MHV68 infection in these key cell types.

Although B cells are the main viral reservoir in vivo, they are highly resistant to de novo MHV68 infection in cell culture (36). The only latently infected B cell line isolated from an MHV68-infected mouse tumor, S11, reactivates to very low frequency, making study of lytic cell populations impractical (37). However, a B cell line has been generated (A20-HE-RIT) that is latently infected with MHV68 and contains a doxycycline (dox)-inducible version of the viral lytic transcriptional activator gene RTA. Treatment of these cells with Dox and phorbol ester (PMA) enables the switch from latency to lytic replication in approximately 80% of the cells (38, 39). Induction of the lytic cycle by dox and PMA treatment of the A20-HE-RIT cells caused a marked increase in B2 SINE levels as measured by primer extension, with levels peaking at 24 to 32 h poststimulation (Fig. 1A). Importantly, B2 RNA induction was not seen in the uninfected A20 parental cells subjected to the same dox and PMA treatment. Furthermore, the induction observed in infected cells was specific to B2 SINEs, as levels of another RNAPIII transcript, 7SK, remained unchanged. Further, reverse transcriptase quantitative PCR (RT-qPCR) analysis confirms that 7SK plus several other RNAPIII transcripts are not significantly elevated upon reactivation of A20-HE-RIT (Fig. 1B), similar to our previous findings during MHV68 lytic infection (21). Additionally, phosphonoacetic acid (PAA) treatment to block viral DNA replication did not prevent B2 SINE induction during reactivation in A20-HE-RIT cells, although the levels were modestly reduced (Fig. 1C). Thus, upon lytic reactivation of latently infected B cells, B2 SINEs are induced early in the viral lytic cycle and continue to accumulate as infection progresses.

We next examined the potential for B2 SINE induction upon MHV68 infection of primary bone marrow-derived macrophages (BMDMs). Unlike fibroblasts, which are highly susceptible to MHV68, the highest level of infection we achieved in WT BMDMs was ∼20%, which occurred with a multiplicity of infection (MOI) of 20 and did not increase upon addition of more virus (A. M. Schaller, unpublished observation). Despite the lower infection efficiency, primer extension reactions demonstrated that in MHV68-infected primary BMDMs, B2 SINE induction began at 30 h postinfection (hpi) and reached their highest relative levels by 40 to 48 hpi (Fig. 1D). These induction kinetics were slower than what we observed in NIH 3T3 cells (Fig. 1E), likely due to overall slower replication kinetics of MHV68 in the BMDMs, as previously documented (40). In summary, B2 SINE RNA induction occurs during lytic MHV68 infection of multiple primary and immortalized cell types.

B2 SINE RNAs are not induced in uninfected cells by paracrine signaling.

We were struck by the robust B2 upregulation in primary BMDMs, given that at most 20% of these cells were infected by MHV68. We therefore considered the possibility that infected cells produce paracrine signals that cause B2 upregulation in neighboring uninfected cells as well. We first tested this possibility using 3T3 cells, as their susceptibility to infection should yield a higher concentration of relevant paracrine signaling molecules. We performed a supernatant transfer assay, in which uninfected cells were incubated for 1 h or 24 h with cell supernatants from infected NIH 3T3 cells, either in crude form or after 0.1 μm filtration to remove viral particles. B2 SINE levels were then measured 24 h posttransfer using primer extension. We observed no B2 SINE induction in cells incubated in filtered supernatants, suggesting that paracrine signals derived from infected 3T3 cells are not sufficient to stimulate B2 induction in uninfected cells. In contrast, there was robust B2 SINE induction in cells incubated with crude supernatants, as expected since these supernatants contain MHV68 virions to initiate a de novo infection (Fig. 2A). This experiment was repeated in BMDMs, where filtered or crude supernatants were taken from infected 3T3 cells and incubated with plated BMDMs for 1 h or 24 h before removal. BMDMs were harvested 48 h after the beginning of incubation with 3T3 supernatants. These data were identical to those observed with 3T3s, in which paracrine signals contained within filtered supernatant were insufficient for B2 induction (Fig. 2B). While it is theoretically possible that some filterable component, and not viral infection, could be responsible for B2 SINE induction, these results strongly suggest that paracrine or cell-to-cell signaling through the supernatant is not sufficient to induce this phenotype.

B2 SINE induction is RNAPIII-dependent but does not involve the RNAPIII regulator Maf1.

We previously showed that treatment of 3T3 cells with an RNAPIII inhibitor or B2-directed antisense oligonucleotides (ASOs) reduced the B2 RNA levels upon MHV68 infection, strongly suggesting that RNAPIII activity was required for their induction (21). However, given that small molecule inhibitors can have off-target effects and B2 ASOs will also target mRNAs containing embedded SINE elements, we sought to independently validate that the B2 SINE transcriptional induction is RNAPIII-dependent. We chose the strategy of depleting Brf1, a critical component of the TFIIIB transcription factor complex needed for RNAPIII transcription of type II (e.g., SINE) promoters using small interfering RNA (siRNA)-mediated knockdown (8). Knockdown of Brf1 was robust through 48 h posttransfection (Fig. 3A). In both BMDMs (Fig. 3B) and 3T3 cells (Fig. 3C), depletion of Brf1 completely abrogated B2 expression as measured by primer extension throughout the time course of infection. Notably, the levels of 7SK were not affected by Brf1 knockdown, as this RNAPIII transcript has a type III promoter that does not require Brf1 (41). Thus, these results confirm that RNAPIII is required for MHV68-induced B2 SINE activation.

We next considered the possibility that MHV68 infection alters the regulation of RNAPIII to increase its activity on B2 promoters. A master regulator of RNAPIII is Maf1, which acts by binding free RNAPIII at its clamp domain, thereby impairing RNAPIII binding to the TFIIIB-promoter complex and preventing RNAPIII transcription initiation (9, 42). To test the hypothesis that release of Maf1-mediated repression of RNAPIII transcription is responsible for B2 SINE induction, we derived primary BMDMs from Maf1^–/– mice (43). Surprisingly, we observed no increase in B2 SINE RNA in uninfected Maf1^–/– BMDMs compared to WT BMDMs, suggesting that Maf1 is not required for the normal silencing of B2 loci (Fig. 3D). We did observe somewhat more of an increase in B2 levels at 24 hpi with MHV68 in the Maf1^–/– cells relative to WT cells, although this difference was not sustained at 48 hpi (Fig. 3D). We therefore conclude that the primary mechanism of B2 induction by MHV68 is not through interference with the RNAPIII repressor Maf1.

B2 SINE induction is independent of canonical innate immune signaling pathways.

Due to their activation during herpesvirus infection and broadly acting signaling cascades, we considered that innate immune signaling may be involved upstream of B2 SINE induction. Pattern recognition receptors, namely, the toll-like receptors 2, 3, 7, and 9, RIG-I-like receptors, and AIM2, can become activated during lytic herpesvirus infection (44–47). To examine the possible upstream involvement of infection-induced innate immune signaling in the induction of B2 SINE transcription, we quantified B2 SINE levels in primary BMDMs derived from WT B6 mice versus mice lacking several canonical innate immune signaling pathways. These included mutants in Toll-like receptor signaling (MyD88/TRIF^–/–), cytoplasmic RNA recognition signaling (MAVS^–/–), or type I interferon (IFN) receptor-mediated signaling through the type I IFN receptor (IFNAR^–/–) (Fig. 4A), as well as cGAS/STING-mediated DNA sensing using the golden ticket (gt/gt) mutant (48), which contains a missense mutation in exon 6 of the mouse STING gene, rendering STING inactive (Fig. 4B). In each case, primer extension experiments showed equivalent or greater B2 SINE RNA induction upon infection of the mutant BMDMs compared to the WT BMDMs. Thus, none of these innate immune components is individually required for SINE activation during MHV68 infection.

To control for the possibility that multiple innate immune sensors could be activated in a redundant manner to induce B2 SINEs, we also tested primary BMDMs derived from mice lacking the downstream transcription factors interferon-regulatory factor 3 (IRF3) and interferon regulatory factor 7 (IRF7). All pattern recognition receptor signaling pathways converge on IRF3 and IRF7, which activate transcription of interferon-stimulated genes (ISGs) and inflammatory cytokines (49). In agreement with the data from BMDMs lacking the upstream innate immune sensors, MHV68 infection still caused robust B2 SINE induction in IRF3^–/– and IRF3/7^–/– BMDMs (Fig. 4C). Thus, innate immune signaling does not activate B2 SINE transcription during MHV68 infection.

We noted that the infection-induced B2 levels were even more pronounced in each of the single and double knockout BMDMs than in WT cells (Fig. 4A to C). We hypothesize that this is a result of increased MHV68 infection under conditions of impaired immune restriction, as we noted that the knockout BMDMs routinely achieved higher MHV68 infection rates (as measured by green fluorescent protein [GFP] positivity) than WT BMDMs (Schaller, unpublished).

The conserved herpesvirus kinase ORF36 is sufficient to induce B2 SINE transcriptional upregulation.

To search for viral factors involved in B2 SINE induction, we obtained and resequenced a partial MHV68 open reading frame (ORF) library previously generated by Ren Sun, which contained 47 full-length MHV68 ORF plasmids (50) (Table 1). The ORFs were first screened by cotransfection of 3T3 cells with 3 to 5 plasmids that were grouped based on similar temporal class and/or proposed or known function (Fig. 5A; Table 1) (51, 52). Only the group that contained ORFs 33, 35, and 36 showed B2 SINE induction above that of the control GFP-expressing plasmid as measured by primer extension (Fig. 5B). We then tested each of these ORFs individually for the ability to induce B2 SINEs, revealing that only MHV68 ORF36 expression was sufficient to upregulate B2 SINEs both as an untagged construct as well as with an N-terminal FLAG-tag (Fig. 5B).

MHV68 ORF36 is a conserved herpesvirus serine/threonine kinase with a variety of reported kinase-dependent and -independent roles relating to the DNA damage response, inhibition of histone deacetylation, and inhibiting IRF3-driven ISG production (33, 34, 53–55). To determine whether ORF36 kinase activity was required for B2 SINE upregulation, we compared the activity of WT ORF36 to an ORF36 kinase-null mutant (K107Q) (55). Primer extension of RNA from transfected 3T3 cells showed that only WT ORF36 but not K107Q induced B2 SINEs (Fig. 5C). To determine the contribution of ORF36 toward B2 induction in the context of infection, we obtained versions of MHV68 either lacking ORF36 (36S) or containing a kinase-null version of ORF36 (36KN) (54). Notably, infection of primary BMDMs with these viruses revealed a reduction in MHV68-induced B2 SINE RNA upon loss or kinase inactivation of ORF36 compared to infection with the repaired WT virus (Fig. 5D). We observed similar defects in B2 induction upon infection of 3T3 cells with 36S and KN viruses compared to WT, across a range of MOI (Fig. 5E). The fact that some residual B2 induction remained in BMDM and 3T3 cells infected with the ORF36 mutant viruses indicates that other viral factors also contribute to SINE induction. However, ORF36 expression is sufficient to activate B2 SINEs when expressed alone and is required for WT levels of B2 SINE induction in the context of MHV68 infection.

Induction of B2 SINE transcription is conserved among ORF36 CHPK homologs.

ORF36 homologs are found in all subfamilies of herpesviruses, where they are collectively referred to as the conserved herpesvirus protein kinases (CHPKs). Several examples exist of shared CHPK functions and shared substrate specificity (54, 56, 57). We therefore examined whether other CHPKs were able to induce B2 SINE RNA. We transfected NIH 3T3 cells with plasmids expressing HA- or FLAG-tagged CHPKs from KSHV (ORF36), varicella zoster virus (VZV) (ORF47), human cytomegalovirus (HCMV) (UL97), EBV (BGLF4), and MHV68 (ORF36) and measured B2 SINE RNA using primer extension (Fig. 6A). MHV68 ORF36 produced the most robust induction, followed by the other gammaherpesvirus CHPKs, KSHV ORF36, and EBV BGLF4. The alpha- and betaherpesvirus protein kinases, VZV ORF47 and HCMV UL97, induced B2 SINEs to a minimal degree, although they were expressed to similar (albeit low) levels as MHV68 ORF36 (Fig. 6B). Attempts at optimizing transfection conditions to achieve equivalent CHPK expression among homologs were unsuccessful, likely due to inherent properties of the expressed proteins in NIH 3T3s. However, these data suggest that B2 SINE induction is conserved among the CHPKs, with MHV68 ORF36 being highly potent for inducing B2 expression.

Derepression of the chromatin landscape allows for B2 SINE induction.

Previous studies of features linked to SINE repression in uninfected cells indicated the importance of the repressive histone H3 lysine 9 tri-methylation (H3K9me3) mark and, to a lesser degree, DNA methylation at CpG sites (11, 12, 58, 59). These marks are deposited and maintained by the histone methyltransferases SU(VAR)3-9 and the DNMT family of DNA methyltransferases, respectively. Furthermore, ORF36 has been shown to inhibit histone deacetylases 1 and 2 (HDACs 1/2) (33), although whether HDACs are involved in repression of SINE loci is unknown.

To test the role of each of these factors in B2 induction, we treated NIH 3T3 cells with inhibitors of HDACs 1/2 (ACY-957), DNMTs (5-azacytidine), and SU(VAR)3-9 (chaetocin) or a cocktail composed of ACY-957 and chaetocin together (Fig. 7A). We observed induction of B2 SINEs following treatment with ACY-957 and chaetocin and an additive effect when using both inhibitors together (Fig. 7A, lane 5). Treatment of cells with 5-azacytidine yielded no increase in levels of B2 RNA, in agreement with previous work (11).

Given that the strongest effects on B2 induction were observed upon inhibition of histone methyltransferases combined with HDAC inhibition, we next tested whether treatment with these inhibitors during infection was sufficient to rescue B2 levels in ORF36 KN and S infection to ORF36 WT infection levels. We observed that, in the context of infection, treatment with ACY-957 and chaetocin restored the levels of B2 ncRNA in the ORF36 S- and KN-infected cells to those observed during WT MHV68 infection (Fig. 7B), showing that chromatin derepression induced B2 ncRNA accumulation in an additive manner. Taken together, these data show that keeping an actively repressed chromatin state, primarily through maintenance of H3K9me3, is important for preventing constitutive B2 SINE induction.

Cells.

NIH 3T3 mouse fibroblasts were obtained from the UC Berkeley Cell Culture Facility and maintained in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen) with 10% fetal calf serum (FBS; Seradigm). A20 B cells were maintained in RPMI (Gibco), 10% fetal bovine serum (FBS; VWR), 2 mM l-glutamin, 100 U/ml penicillin,100 mg/ml streptomycin, and 50 mM β-mercaptoethanol (BME). A20-HE-RIT cell lines (kindly provided by Laurie Krug) (39) were maintained under the same conditions as A20 cells, with the addition of 300 μg/ml hygromycin B, 300 μg/ml G418 and 2 μg/ml puromycin. To reactivate A20-HE-RIT, cells were cultured in medium without antibiotic selection for 24 h and then seeded at a cell density of 1e6 cells/ml in the presence of 5 μg/ml doxycycline and 20 ng/μl phorbol myristate acetate (PMA) for the indicated time. To block viral DNA replication, PAA was used at a concentration of 200 μg/ml and was added at the start of reactivation. Bone marrow-derived macrophages (BMDMs) containing knockouts for innate immune pathway components (48, 77–79) were kindly provided by the lab of Gregory Barton (UC Berkeley, Department of Immunology). Wild-type and Maf1 knockout BMDMs were differentiated as follows: femurs and tibias from C57BL/6J (B6) mice (43) aged 3 to 6 months were flushed with bone marrow medium plus antibiotics (BMM+A; high glucose DMEM + 10% FBS, + 10% MCSF + 1% PenStrep) using a 3-ml syringe with attached 23-gauge needle. Cell-containing medium was filtered through a 70-μM filter to remove debris. Cells were pelleted at 280 × g in an Allegra X-15R Beckman Coulter centrifuge for 5 minutes. Supernatant was removed by aspiration, and cells resuspended in BMM+A. Cells were counted using a hemocytometer and plated in non-tissue culture (TC)-treated 15-cm petri dishes (ref. no. 351058; Falcon) at a concentration of 10e6 cells/25 ml BMM+A/plate. On day 3 of differentiation, 5 ml BMM+A was added to each plate to feed cells. On day 7 of differentiation, BMM+A was aspirated and replaced with 10 ml cold Dulbecco’s phosphate-buffered saline (DPBS; Invitrogen) per plate and placed at 4°C for 10 min. Cells were then lightly scraped from each plate and collected, pelleted as previously mentioned, and resuspended in bone marrow medium without antibiotics (BMM) containing 10% DMSO at a concentration of 10e6/ml. Then, 1.5-ml CryoTube vials containing 1 ml/10e6 BMDMs were frozen at –80°C for 24 h before being stored in liquid nitrogen for the duration. Subsequently, thawed vials of BMDMs were maintained in BMM except during infections. Experiments involving mice were performed under a protocol approved by the Institutional Animal Care and Use Committee (IACUC) of the Albert Einstein College of Medicine.

Transfections.

Transfection of BMDMs with Brf1 or control siRNA was completed as follows: BMDM cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen) with 10% fetal calf serum (FBS; Seradigm). Prior to siRNA transfection, BMDMs were grown for 3 days in 15-cm TC-treated plates to 90% confluence. Cells were removed and washed in DPBS twice by spinning at 475 × g for 5 minutes each time. Transfection of siRNA was done using the Neon transfection system (Thermo Fisher) as follows: for each condition, 2e6 cells were resuspended in buffer R at a concentration of 1e6 cells/100 μl. To this, 200 nM (assuming a final culture volume of 2 ml medium) siRNA was added from 100 μM stock or either control (ON-TARGETplus nontargeting control pool; Dharmacon) or Brf1 siRNA (SMARTpool; ON-TARGETplus mouse Brf1 siRNA; Dharmacon). Then, 100 μl Neon transfection system tips were used to transfect siRNA into cells as follows: parameters set for BMDMs were 1,680/20/1 (pulse/length/width). Cells were then quickly removed to 2 ml total medium and plated at 2e6 cells/plate in 60-mm TC-treated plates, before being placed at 37°C to incubate. Transfection of NIH 3T3 cells with MHV68 ORF library plasmids and ORF36 WT and mutant plasmids was completed as follows: NIH 3T3s were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen) with 10% fetal calf serum (FBS; Seradigm). Prior to transfection, cells were maintained at parameters suggested in the PolyJet in vitro DNA transfection reagent protocol (http://signagen.com/DataSheet/SL100688.pdf). Guidelines for the advanced protocol for transfecting hard-to-transfect mammalian cells was followed strictly. Briefly, 2.7 × 10⁶ cells were transfected with 5 μg of plasmid DNA for each sample. In cases of multiple ORF transfections, total plasmid DNA levels were not to exceed 25 μg.

Plasmids and cloning.

MHV68 ORF library plasmids were generously provided by the lab of Ren Sun (University of California Los Angeles), and their construction was previously described (50). For generation of the ORF36 kinase-null mutant, the K107Q mutation was introduced using QuikChange PCR with the following primers: 5′-GTGCTGTCAATTTTGGGATATACTGTATGCAGAGCGTGTCATCTGAT-3′ and 5′-ATCAGATGACACGCTCTGCATACAGTATATCCCAAAATTGACAGCAC-3′. Plasmids for conserved herpesvirus protein kinase homologs of ORF36 were purchased through Addgene from the laboratory of Robert Kalejta (https://www.addgene.org/Robert_Kalejta/) (56).

Virus preparation and infections.

MHV68 containing a stop mutation or kinase-null mutation in ORF36 as well as the corresponding mutant rescue virus were generously provided by Vera Tarakanova (Medical College of Wisconsin) (54). MHV68 was amplified in NIH 3T12 fibroblast cells, and the viral 50% tissue culture infective dose (TCID₅₀) was measured on NIH 3T3 fibroblasts by limiting dilution. NIH 3T3 fibroblasts were infected at the indicated multiplicity of infection (MOI) by adding the required volume of virus to cells in 1 ml total volume (for each well of a 6-well plate), 2 ml total volume (for 6-cm plates), or 5 ml (for 10-cm plates). Infection was allowed to proceed for 45 min prior to removal of virus medium and replacement with DMEM plus 10% FBS. BMDMs were infected with the minimal volume of MHV68 required to achieve maximum infection (20% to 30%), as determined by titration experiments with GFP-marked MHV68 followed by flow cytometry for GFP. For infection of BMDMs, virus was added to cells in serum-free DMEM for 4 h in non-TC-treated plates. Virus-containing medium was then aspirated and replaced with macrophage medium without antibiotics.

RT-qPCR and primer extension.

Total RNA was extracted from cells using TRIzol reagent (Invitrogen). For RT-qPCR analysis, RNA was treated with Turbo DNase (Ambion) and reverse transcribed with avian myeloblastosis virus reverse transcriptase (AMV RT) (Promega) primed with random 9-mers. RT-qPCR was performed with iTaq universal SYBR green Supermix (Bio-Rad) using the following primers: (7SK_F: CCCCTGCTAGAACCTCCAAAC, 7SK_R: CACATGCAGCGCCTCATTT, U6_F: CGCTTCGGCAGCACATATAC, U6_R: AAAATATGGAACGCTTCACGA, 5S_F: TCTCGTCTGATCTCGGAAGC, 5S_R: AGCCTACAGCACCCGGTATT, 7SL_F: ATCGGGTGTCCGCACTAAGTT, 7SL_R: CAGCACGGGAGTTTTGACCT). The fold change was calculated using the ΔΔCT method. Primer extension was performed on 10 to 15 μg of total RNA using a 5′ fluorescein labeled oligo specific for B2 SINEs or 7SK. RNA was ethanol precipitated in 1 ml 100% EtOH, washed in 70% ethyl alcohol (EtOH), and pelleted at 21,130 × g and 4°C for 10 min. Pellets were resuspended in 9 μl annealing buffer (10 mM Tris-HCl, pH 7.5, 0.3 M KCl, 1 mM EDTA) containing 1 μl of (10 pmol/μl) 5′-fluorescein labeled primer (B2 SINE: TACACTGTAGCTGTCTTCAGACA, 7SK: GAGCTTGTTTGGAGGTTCT; Integrated DNA Technologies). Samples were heated briefly to 95°C for 2 min, followed by annealing for 1 h at 55°C. Then, 40 μl of extension buffer (10 mM Tris-HCl, pH 8.8, 5 mM MgCl₂, 5 mM DTT, 1 mM deoxynucleoside triphosphate [dNTP]), and 1 μl AMV reverse transcriptase (Promega) were added, and extension was carried out for 1 h at 42°C. Samples were EtOH precipitated, and then pellets were briefly air dried and resuspended in 20 μl 1× RNA loading dye (47.5% formamide, 0.01% SDS, 0.01% bromophenol blue, 0.005% xylene cyanol, and 0.5 mM EDTA). Then, 10 μl of each sample was run on an 8% urea-PAGE gel for 1 h at 250 V. Gels were imaged on a Bio-Rad Chemidoc with fluoroscein imaging capability.

Protein extraction and analysis.

Cells were washed with cold DPBS once before being lysed with RIPA lysis buffer (50 mM Tris HCl, 150 mM NaCl, 1.0% [vol/vol] NP-40, 0.5% [wt/vol] sodium deoxycholate, 1.0 mM EDTA, and 0.1% [wt/vol] SDS). Cell lysates were vortexed briefly, rotated at 4°C for 1 h, and then spun at 18,000 × g in a tabletop centrifuge at 4°C for 12 min to remove debris.

For Western blot analyses, 30 μg of whole-cell lysate was resolved with 4 to 15% mini-PROTEAN TGX gels (Bio-Rad). Transfers to polyvinylidene difluoride (PVDF) membranes were done with the Trans-Blot Turbo transfer system (Bio-Rad). Blots were incubated in 5% milk/TBS plus 0.1% Tween 20 (TBST) to block, followed by incubation with primary antibodies against FLAG (Sigma F1804, 1:1,000), Brf1 (Bethyl a301-228a, 1:1,000), hemagglutinin (HA) (Sigma H9658, 1:1,000), TUBA1A (abcam ab729, 1:1,000), or GAPDH (glyceraldehyde-3-phosphate dehydrogenase) (Abcam ab8245, 1:1,000) in 5% milk/TBST. Washes were carried out in TBST. Blots were then incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (Southern Biotechnology, 1:5,000). Washed blots were incubated with Clarity Western ECL substrate (Rio-Rad) for 5 min and visualized with a Bio-Rad ChemiDoc imager.

Inhibitor treatment.

Cells were plated 12 h before inhibitor treatment to achieve 70% confluence at time of treatment. ACY-957 (MedChemExpress HY-104008), 5-azacytidine (Sigma A2385), and chaetocin (Cayman Chemicals 13156), were resuspended with DMSO prior to treatment. Inhibitors were diluted to working concentrations in warmed DMEM plus 10% FBS before addition to cells. Pretreatment of cells with inhibitor-containing medium preceded infection with MHV68 by 1 h. Upon removal of virus-containing medium, inhibitor-containing medium was replaced onto cells.