The IκB Kinases Restrict Human Cytomegalovirus Infection

Loss of IKKα and IKKβ attenuates HCMV sensitivity to TNF-α treatment.

The IKKs play central roles in the signal transduction cascades of canonical and noncanonical NF-κB signaling (Fig. 1A). We employed CRISPR to generate monoclonal cell lines with knockouts of either IKKα or IKKβ, to determine the contributions that the IKKs make to HCMV infection. Frameshift mutations were introduced into both alleles of either the conserved helix-loop-helix ubiquitous kinase (CHUK) gene encoding IKKα (position −1 and −37 deletions) or the IKBKB gene encoding IKKβ (position +1 and +1 insertions) (Fig. 1B). In both cases, we observed total loss of the targeted IKK protein in each cell line. There was an observable decrease in the accumulation of the nontargeted IKK protein (e.g., reduced IKKα was evident upon knockout of IKKβ) (Fig. 1C). Cell viability of the IKKα and IKKβ knockout cell lines was assessed via the trypan blue exclusion cell viability assay, and we observed ∼2-fold growth reduction in cells lacking IKKα expression, whereas no discernible defect was observed in IKKβ^−/− cells.

RelA and RelB are prominent transcription factors that are strongly associated with canonical and noncanonical NF-κB signaling, respectively. To analyze how loss of the IKKs perturbs the expression of downstream NF-κB pathway constituents, we assessed the accumulation of these transcription factors in our CRISPR control and IKK knockout cell lines. We found that loss of IKKα strongly limited the accumulation of RelA and RelB in both mock-infected and HCMV-infected cells, whereas both RelA and RelB still accumulated to WT levels in the absence of IKKβ (Fig. 2A). These observations suggest that IKKα is required for optimal expression of the RelA and RelB NF-κB transcription factors.

TNF-α pretreatment of cells has been shown to negatively influence the ability of HCMV to establish initial lytic infection and to form plaques in human fibroblasts (20). To determine the extent to which the IKKs contribute to this viral sensitivity to TNF-α treatment, we assessed the impact of TNF-α treatment on viral plaque formation efficiencies in our CRISPR control and IKK knockout cell lines. We observed that treatment of CRISPR control cells with10 ng/ml TNF-α reduced viral plaque formation to 40% of the rate seen in nontreated cells (Fig. 2B). In IKKα^−/− cells, the rate of plaque formation increased to 60% of that in nontreated controls. Further, this ratio increased to 80% in IKKβ^−/− cells (Fig. 2B). These results suggest that both IKKα and IKKβ contribute to the antiviral effects of TNF-α treatment. However, the plaque formation efficiency in both IKK knockout cell lines after TNF-α challenge was still reduced, relative to non-TNF-α-treated cells (Fig. 2B), suggesting that, in each case, the presence of the reciprocal IKK could contribute to the TNF-α-induced antiviral state. Interferon gamma (IFN-γ) has also been shown to inhibit the establishment of HCMV lytic infection in human fibroblasts, putatively through JAK/STAT signaling (21). In contrast to the findings for TNF-α challenge, we found that knockout of each IKK had little impact on the plaque formation efficiency of HCMV on cells pretreated with 5 U/ml IFN-γ (Fig. 2C). Collectively, these data indicate that the IKKs are important for HCMV-associated sensitivity to TNF-α treatment but are largely dispensable for the antiviral signaling associated with IFN-γ treatment.

IKKα and IKKβ restrict HCMV viral replication.

The relationship between NF-κB signaling and the HCMV life cycle is complex. HCMV induces IKK activity, which is important for viral replication in certain contexts (11, 12, 22). However, NF-κB activation has been found to inhibit HCMV infection in other contexts (21, 23, 24). Therefore, we wanted to explore how IKKα and IKKβ contribute to HCMV viral growth. We found that knockout of either IKK increased low-multiplicity of infection (MOI) viral replication by ∼1 log unit, compared to growth in CRISPR control cells, as assessed by 50% tissue culture infective dose (TCID₅₀) analysis (Fig. 3A). Analysis of viral DNA accumulation during low-MOI infection showed an ∼4-fold increase in viral DNA replication in the absence of IKKα (Fig. 3B). Upon inactivation of IKKβ, a much smaller but still statistically significant increase in viral DNA abundance was also evident (Fig. 3B). The increased viral DNA abundance observed in the absence of IKKα correlated with an increase in the accumulation of viral gene products observed in these cells (Fig. 3C), most notably in the true late protein pp28. In contrast, IKKβ-deficient cells did not display a substantial change in the accumulation of viral immediate early or late proteins (Fig. 3D). Notably, this difference in viral gene expression did not affect the production of infectious viral progeny, as the loss of either IKKβ or IKKα increased production of infectious virions to similar extents, relative to CRISPR control cells. These results demonstrate that IKKα and IKKβ each play a role in restricting HCMV infection and reducing the viable progeny produced during viral replication but each kinase affects viral gene expression differently.

Previously, we found that the U_L26 protein is necessary and sufficient to attenuate TNF-α-induced NF-κB activity and that U_L26-deficient HCMV (ΔU_L26) is more susceptible to challenge with TNF-α treatment (13). To determine whether either of the IKKs was functioning in a U_L26-specific manner during viral infection, we assessed viral growth of a U_L26-deficient green fluorescent protein (GFP)-positive HCMV mutant in our CRISPR IKK knockout cell lines (Fig. 4A). We found that loss of either IKK increased ΔU_L26 infectious virion production by approximately 1 log unit at 6 days postinfection (dpi) (Fig. 4A), which was similar to the effect observed for GFP-positive WT (WT-GFP) HCMV (Fig. 3A). The difference in infectious virion production between IKK-deficient and control cells was reduced by day 8 (Fig. 4A). Viral DNA accumulation was also affected somewhat similarly by the loss of the IKKs, with an ∼9-fold increase in the amount of viral DNA present in IKKα^−/− cells and an almost 2-fold increase in IKKβ^−/− cells. Concomitantly, during ΔU_L26 infection of the IKKα^−/− cells, we observed significantly more production of the late viral protein pp28 (Fig. 4C). Levels of the pp28 protein were also elevated in IKKβ knockout cells during ΔU_L26 infection (Fig. 4D), albeit to a lesser extent. The accumulation of the immediate early viral protein IE2 is enhanced in IKKα^−/− cells but is unaffected in IKKβ-deficient cells (Fig. 4C and D). The elevated expression of pp28 at later time points during ΔU_L26 infection in the IKKβ^−/− cells differs from WT infection of the same cells (Fig. 3D), i.e., IKKβ inactivation did not substantially increase pp28 accumulation. This suggests the possibility that, in the absence of U_L26, IKKβ may be limiting DNA replication and the expression of viral late genes. However, the observation that the loss of the IKKs increased WT and ΔU_L26 viral replication to similar extents (∼10-fold at 6 dpi) suggests that the IKKs influence overall viral replication in a largely U_L26-independent manner.

The U_L26 domains required for high-titer viral replication are dispensable for inhibition of TNF-α-induced NF-κB activation.

Previously, we generated a panel of HCMV mutant strains encoding truncated forms of the U_L26 protein and examined the contributions of the various U_L26 domains to infection (Fig. 5A) (19). To determine whether specific U_L26 domains are required for both NF-κB modulation and high-titer viral replication, we analyzed these mutants with respect to their ability to inhibit NF-κB signaling. In contrast to ΔU_L26-infected cells, cells infected with recombinant viruses encoding N-terminal- and C-terminal-truncated U_L26 isoforms were sufficient to inhibit TNF-α-induced phosphorylation of the IKK complex during infection (Fig. 5B). Similarly, in the absence of infection, transient expression of either the N-terminal or the C-terminal U_L26 truncation mutant was sufficient to inhibit TNF-α-induced NF-κB-dependent luciferase activity (Fig. 5C and D). Collectively, these data suggest that the N- and C-terminal domains of U_L26 are not necessary for attenuation of TNF-α-induced NF-κB activation. Further, given that the U_L26ΔC mutant inhibits NF-κB signaling but has a replication defect similar in magnitude to that of the ΔU_L26 null mutant (19), these data indicate that the U_L26 sequences necessary for high-titer replication are separable from those necessary for prevention of TNF-α-induced NF-κB activation.

IKKα and IKKβ restrict initiation of infection and cell-to-cell spread.

To further explore the roles the IKKs play during viral infection, we analyzed how the loss of IKKα or IKKβ activity affected various plaque phenotypes. We found that knockout of either IKKα or IKKβ resulted in an ∼2-fold increase in the incidence of WT plaque formation (Fig. 6A). Similarly, treatment with 2-[(aminocarbonyl)amino]-5-(4-fluorophenyl)-3-thiophenecarboxamide (TPCA-1), a pan-IKK inhibitor at the concentrations employed (25), also induced WT plaque formation by ∼2-fold (Fig. 6B). Plaque formation of the ΔU_L26 virus was affected differentially in response to individual IKK deletions. First, knockout of IKKβ induced a 3-fold increase in ΔU_L26 plaque formation efficiency, relative to control cells; this was higher than the level observed for the WT virus (Fig. 6A and C), suggesting that, in the absence of U_L26, IKKβ restricts plaque formation to a greater extent. A more notable U_L26-specific difference was that the knockout of IKKα did not increase the plaque formation of the ΔU_L26 virus (Fig. 6C), which suggests that, in this context, U_L26 is necessary for HCMV to benefit from the absence of IKKα. When both IKKs were inhibited via TPCA-1 treatment, the plaque formation efficiency of the ΔU_L26 virus increased 2-fold (Fig. 6D), similar to WT infection.

In addition to measuring the impact on plaque formation efficiency, we measured the role that the IKKs play in cell-to-cell spread via an analysis of plaque size. WT HCMV plaques increased in size in the absence of the IKKs but to differing degrees. The size of WT viral plaques was enhanced ∼2-fold in the IKKα knockout cell line, while the IKKβ knockout cell line exhibited a plaque size increase of almost 5-fold (Fig. 6E). Similar increases in ΔU_L26 plaque size were observed in both IKKα and IKKβ knockout cells, which exhibited ∼2-fold and ∼5-fold increases, respectively (Fig. 6F). These results suggest that, regardless of the presence of U_L26, IKKβ restricts HCMV cell-to-cell spread to a greater extent than does IKKα and their individual contributions to plaque initiation can largely be separated from their contributions to cell-to-cell spread. In contrast to the knockout of either IKK alone, treatment with TPCA-1, which inhibits both IKKs, did not affect WT cell-to-cell spread, as evidenced by equal plaque sizes (Fig. 6G). Notably, ΔU_L26 plaque sizes increased by ∼2.5-fold in the presence of TPCA-1 (Fig. 6H). These results suggest the possibility that, during WT infection, the potential benefits associated with losing one IKK are masked by the simultaneous loss of the other IKK. This is not seen in the context of ΔU_L26 infection, however, where dual inhibition of the IKKs in the absence of the NF-κB regulator U_L26 still increases the extent of cell-to-cell spread.

To rule out the possibility that the increased plaque formation and size phenotypes observed in the original IKKβ^−/− cell line were a result of a clonal artifacts, we assessed these phenotypes in a polyclonal pool of newly generated IKKβ clones in which we combined six individual monoclonal IKKβ^−/− clones in equal numbers. Upon assessment of WT and ΔU_L26 plaque formation (Fig. 6I and J), we observed similar increases in plaque formation efficiency, with plaque formation increases of 1.5-fold and 2.5-fold for HCMV WT and ΔU_L26 viruses, respectively (Fig. 6I and J). Similarly, cell-to-cell spread for both WT and ΔU_L26 viruses was also increased ∼3- to 4-fold in this pooled population, only a slight reduction in comparison with the ∼4- to 5-fold increases observed in the monoclonal IKKβ^−/− cell line (Fig. 6K and L). These findings imply that clonal selection effects are not a likely explanation for the IKKβ-related plaque formation phenotypes we observe. Collectively, our results indicate that the IKKs restrict the initiation of infection and restrict cell-to-cell spread.

Contributions of IKKα and IKKβ to ΔU_L26 innate immune phenotypes.

The ΔU_L26 virus is more vulnerable to TNF-α challenge than WT HCMV (13). To assess the extent to which the IKKs contribute to viral TNF-α sensitivity in a U_L26-dependent manner, we assessed viral plaque formation in our IKK knockout cell lines in the face of TNF-α or IFN-γ challenge (Fig. 7A and B). We found that loss of either of the IKKs modulated ΔU_L26 sensitivity to TNF-α treatment, with the IKKβ knockout having a larger effect than the IKKα knockout and exhibiting approximately twice the number of plaques formed upon TNF-α treatment, relative to the control cells (Fig. 7A). In contrast, knockout of either IKKα or IKKβ had little effect on the plaque formation efficiency after treatment with IFN-γ (Fig. 7B). The increase in plaque formation efficiency observed in the IKKα or IKKβ knockout cells, relative to control cells, was slightly higher in the context of ΔU_L26 infection, compared to WT HCMV infection (Fig. 7C). However, the effect was very modest and not statistically significant, suggesting that IKK activity does not have a substantially larger inhibitory effect on plaque formation efficiency in the absence of U_L26.

Previously, we found that ΔU_L26 infection induces noncanonical NF-κB activation, including inducing the nuclear translocation of RelB, which is typically mediated by IKKα (Fig. 1A) (reviewed in reference 26). In addition, expression and secretion of the antiviral cytokine interleukin 6 (IL-6), a canonical NF-κB target gene, are induced during ΔU_L26 infection (Fig. 7D) (13). To elucidate how IKKα and IKKβ contribute to ΔU_L26-mediated induction of IL-6, we measured the accumulation of IL-6 mRNA during WT and ΔU_L26 infection. IKKα knockout cells showed notably higher levels of IL-6 transcription during mock infection. This large induction of IL-6 transcription was preserved during infection in the absence of U_L26 but not WT infection. Loss of IKKβ, however, entirely ablated the transcription of IL-6 across all mock infection and infection conditions (Fig. 7D). To corroborate these findings, we utilized the IKK inhibitor TPCA-1 to impede IKK signaling during infection with either WT or ΔU_L26 HCMV (MOI of 3.0), and we measured the IL-6 mRNA abundance at 48 h postinfection (hpi). We observed that TPCA-1 inhibition had little effect on the transcription of IL-6 during WT infection but significantly reduced IL-6 mRNA abundance, almost 2-fold, during ΔU_L26 infection (Fig. 7E). These results indicate that the IKKs both regulate the accumulation of the downstream NF-κB target gene IL-6 and that IKKβ in particular is required for the upregulation of IL-6 during infection in the absence of U_L26.

Cell culture, viruses, and viral infection.

MRC5 fibroblasts (ATCC CCL-171), human embryonic kidney (HEK) 293T cells, and telomerase-expressing BJ fibroblasts (generated via lentiviral transduction, as described previously [34]) were cultured at 37°C in Dulbecco’s modified Eagle serum (DMEM) (Invitrogen) supplemented with 10% (vol/vol) fetal bovine serum (FBS), 4.5 g/liter glucose, and 1% penicillin-streptomycin (Life Technologies), in a 5% (vol/vol) CO₂ atmosphere. For seeding and cell growth assays, cell viability was determined via the trypan blue exclusion method, with a TC10 automated cell counter (Bio-Rad). The WT strain of HCMV used in this work was BADwt, a bacterial artificial chromosome (BAC) clone of AD169 (35). The ΔU_L26-DBmetΔ (referred to here as ΔU_L26), U_L26ΔN, and U_L26ΔC recombinant viruses were previously generated from this BAC as described (19). GFP-expressing AD169 (BADsubUL21.5, referred to as WT-GFP) and ΔU_L26 with a transposon insertion (ΔU_L26-TI, referred to as ΔU_L26-GFP) were used for plaque formation efficiency and plaque size experiments and have been described previously (19, 36). Unless otherwise indicated, viral infection was performed as follows: cells were grown to a confluence of ∼3.2 × 10⁴ cells per cm² in a 6-well plate and were serum starved for 24 h prior to infection, after which viral inocula were added for an adsorption period of 1.5 h and then replaced with fresh serum-free medium. Cells were infected at a MOI of 3.0 unless otherwise indicated. Culturing in serum-free medium prior to infection was employed to minimize the effects of serum on cellular signaling, as well as to synchronize and to control for cell cycle progression in the cell population. Viral stocks were propagated in MRC5 fibroblasts. For titering of viral stocks, as well as experiments involving the assessment of viral replication in CRISPR cell lines, viral titers were determined via TCID₅₀ analysis.

Lentiviral plasmids.

CRISPR-targeted IKK knockout in BJ/hTert cells was mediated by the following guide RNAs: IKKα, 5′-ACAGACGTTCCCGAAGCCGC-3′; IKKβ, 5′-ACCACCGCTCTCGGTTCCGG-3′. These sequences were determined by inserting early segments of the CHUK (IKKα) and IKBKB (IKKβ) open reading frames into the Zhang Lab online generator (http://crispr.mit.edu). Guide RNAs were appended with flanking homology arms and cloned into the LentiCRISPRv2 Cas9 expression platform according to the Zhang Lab single guide RNA protocol (37, 38). CRISPR control cells were generated via lentiviral transduction of the unmodified LentiCRISPRv2 plasmid.

Lentiviral transduction and cell line generation.

Pseudotyped lentivirus was produced in 293T cells seeded in 10-cm dishes at a density of 2 × 10⁶ cells per cm² and grown for 24 h prior to transfection with 2.6 μg lentiviral vector, 2.4 μg PAX2, and 0.25 μg vesicular stomatitis virus G glycoprotein expression plasmid, using Fugene 6 (Promega). The medium was aspirated after 24 h, and 4 ml fresh medium was added to the plate. After another 24 h, the supernatant was filtered through a 0.45-μm filter and applied to fibroblast cells in the presence of 5 μg/ml Polybrene (Millipore), and the cells were incubated for 24 h. The medium was refreshed and the cells were allowed to recover for 72 h prior to selection with 1 μg/ml puromycin (VWR). Following transduction with guide RNA-expressing lentivirions and subsequent puromycin selection, BJ/hTert fibroblasts were clonally selected by seeding individual cells in 96-well plates. CRISPR control cells transduced with non-guide RNA-containing LentiCRISPRv2 were retained as heterogeneous polyclonal populations. Genomic DNA was harvested from monoclonal lines, and the guide RNA-targeted region of each IKK was amplified by PCR using the following primers: IKKα forward primer, 5′-GGGGACTTAAAGAGCGGACC-3′; IKKα reverse primer, 5′-CCCCACTGATATCATATGGCCT-3′; IKKβ forward primer, 5′-TGGATGGATAGAAATGAGGTGGG-3′; IKKβ reverse primer, 5′-TAGAAGCAGCAGAGTCACCGT-3′. To verify the presence of indels in the IKKα/IKKβ reading frames of both alleles in clonally isolated cell lines, the amplified PCR product was gel purified, ligated into the pCR2.1-TOPO TA vector using a TA cloning kit (Invitrogen), according to manufacturer’s instructions, and transformed into DH10B competent cells. Bacterial colonies were grown and color selected on LB plates containing 50 μg/ml kanamycin and supplemented with 40 μl each of 100 mM isopropyl-β-d-thiogalactopyranoside (IPTG) and 40 mg/ml 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal), spread over the surface of the plate. Plasmids were isolated from colorless colonies containing the PCR inserts and sequenced using M13 forward (position −20) and M13 reverse primers. CRISPR-mediated indel generation in the genomic IKKα and IKKβ reading frames was also confirmed via TIDE analysis (39). In addition to the monoclonal IKKα and IKKβ cell lines, an IKKβ^−/− pooled cell line was created by combining equivalent cell numbers of multiple IKKβ^−/− clones with the following frameshift deletions: positions −7 and +1, positions +1 and +10, and positions +1 and +1.

Immunoblotting.

Cells were harvested in disruption buffer (50 mM Tris [pH 7.0], 2% SDS, 5% 2-mercaptoethanol, and 2.75% sucrose), and the proteins were solubilized via sonication and boiling and separated by 10% SDS-PAGE. Samples were transferred to a nitrocellulose membrane in Tris-glycine transfer buffer and stained with Ponceau S to ensure uniform protein loading, and blots were blocked with 5% milk solution in Tris-buffered saline-Tween 20 (TBST). Primary and secondary antibody incubations were followed by development with an enhanced chemiluminescence (ECL) kit (Bio-Rad) and imaging using a Molecular Imager Gel Doc XR+ system (Bio-Rad). Primary antibodies were specific for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (D16H11; Cell Signaling), phospho-IKKα(Ser176)/IKKβ(Ser177) (16A6; Cell Signaling), IKKα (2682; Cell Signaling), IKKβ (2684; Cell Signaling), RelA (C-20; Santa Cruz Biotechnology), RelB (C-19; Santa Cruz Biotechnology), α-tubulin (11H10; Cell Signaling), IKKα/IKKβ (H-470; Santa Cruz Biotechnology), pp28 (40), IE2 (41), and U_L26 (7H19) (18, 40). Secondary antibodies were rabbit polyclonal (Santa Cruz Biotechnology) and mouse monoclonal (Abcam) anti-IgG antibodies.

Analysis of DNA and transcripts during infection.

Viral DNA accumulation was assessed by real-time quantitative PCR, as described previously (42). Briefly, CRISPR control, IKKα^−/−, and IKKβ^−/− cells were infected with WT or ΔU_L26 virus at a MOI of 0.1 and then, at various time points, scraped in medium, washed with phosphate-buffered saline, and resuspended in lysis buffer (100 mM NaCl, 100 mM Tris [pH 8.0], 25 mM EDTA, 0.5% SDS, 0.1 mg/ml proteinase K, and 40 μg/ml RNase A). Cells were lysed overnight at 55°C, extracted with phenol/chloroform/isoamyl alcohol (25:24:1), extracted again with chloroform/isoamyl alcohol (24:1), precipitated with ethanol, and resuspended in water. The DNA abundance was assessed via quantitative PCR using Fast SYBR green master mix (Applied Biosystems), a model 7500 Fast real-time PCR system (Applied Biosystems), and Fast 7500 software (Applied Biosystems), according to the manufacturer's instructions. Viral DNA abundance was determined with a primer pair targeting the viral IE1 gene, i.e., 5′-CCATGTCCACTCGAACCTTAAT-3′ (forward) and 5′-TGAACAAGTGACCGAGGATTG-3′ (reverse). Host cell DNA was assessed using the following primers targeting GAPDH: 5′-CATGTTCGTCATGGGTGTGAACCA-3′ (forward) and 5′-ATGGCATGGACTGTGGTCATGAGT-3′ (reverse). IE1 and GAPDH gene equivalent values were determined using the 2^−ΔΔCT method, and normalized viral DNA accumulation values were obtained by dividing the abundance of IE1 gene equivalents by the abundance of GAPDH gene equivalents. To measure IL-6 transcription during infection, RNA was isolated from infected cells via TRIzol (Invitrogen) extraction, according to manufacturer’s instructions, and used to synthesize cDNA with random hexamer primers and the Superscript II reverse transcriptase system (Invitrogen). Relative quantities of gene expression were measured and normalized to GAPDH levels via the 2^−ΔΔCT method, using the following primers: IL-6, 5′-AAATTCGGTACATCCTCGACGGCA-3′ (forward) and 5′-AGTGCCTCTTTGCTGCTTTCACAC-3′ (reverse); GAPDH, 5′-CATGTTCGTCATGGGTGTGAACCA-3′ (forward) and 5′-ATGGCATGGACTGTGGTCATGAGT-3′ (reverse).

NF-κB luciferase assay.

The NF-κB reporter construct pNF-κB-FF was induced by TNF-α treatment, as described previously (43). Briefly, calcium phosphate (Stratagene) was used to cotransfect 50% confluent 293T cells in 96-well plates with 0.25 μg pNFκB-FF, 0.25 μg of a Renilla luciferase expression plasmid with an SV40 promoter (pRL-SV40), and 0.25 μg of either pCDNA3.1(+) empty vector (Invitrogen) or one of a panel of pCDNA vectors encoding U_L26 mutants (pCDNA-U_L26WT, pCDNA-U_L26ΔN, and pCDNA-U_L26ΔC). The following oligonucleotides were used to amplify portions of the U_L26 open reading frame and to insert the sequences into the EcoRI-NotI region of the pCDNA3.1(+) vector via Gibson assembly: pCDNAU_L26 short isoform, 5′-CGGATCCACTAGTCCAGTGTGGTGGAATTATGACGAGTAGGCGCGCACCCGACGGCGG-3′ (forward) and 5′-GTTTAAACGGGCCCTCTAGACTCGAGCGGCCTTACGGCAACAGCGCTGATGGCACGTTGC-3′ (reverse); pCDNAU_L26ΔC isoform, 5′-CGGATCCACTAGTCCAGTGTGGTGGAATTATGTACGCCGTTTTCGGCCTCACGAGGTCG-3′ (forward) and 5′-GTTTAAACGGGCCCTCTAGACTCGAGCGGCCCTAGCGGATGACCTGGCCGTCGGCGTCGC-3′ (reverse). The U_L26 WT insert was amplified using the pCDNAU_L26ΔC isoform forward primer and the pCDNAU_L26 short isoform reverse primer. Cells were allowed to recover from transfection for 24 h and then exposed to 10 ng/ml TNF-α treatment for 24 h prior to analysis. Luciferase activity was measured using the Dual-Glo luciferase assay system (Promega), according to the manufacturer’s instructions, and a Synergy 2 plate reader (BioTek). NF-κB reporter activity was normalized to Renilla luciferase activity as a transfection efficiency control.

Plaque formation and size assays.

Serum-starved confluent cells were infected for 1.5 h with a known number of PFU from freshly thawed stocks of indicated GFP-positive viruses. A standard agarose gel overlay was placed over the cells, and plaques were allowed to develop for 10 days before the number of plaques was counted and plaque sizes were measured using ImageJ software. The criterion for identification of a plaque was a locus of multiple GFP-positive cells displaying a cytopathic effect. For experiments involving plaque formation during cytokine treatment, cells were exposed to 10 ng/ml TNF-α (Sigma) or 5 U/ml IFN-γ (PBL Assay Science) for 4 h prior to infection. For plaque formation and size experiments using the IKKα/IKKβ inhibitor TPCA-1, cells were pretreated with 10 μM TPCA-1 (Sigma) for 2 h prior to infection.

Statistical analysis.

All Western blot analyses are representative blots from at least two independent experiments. Statistical significance was determined using a nonpaired one-tailed homoscedastic Student's t test unless otherwise indicated. A probability value of <0.05 was considered statistically significant. For the growth curves in Fig. 3 and Fig. 4, the values produced for a given biological replicate experiment were normalized to the average amount of infectious virus produced in the control cells for the biological replicate in the same experiment before statistical comparisons between the infectious virions produced by control and IKK knockout cells were performed. This normalization serves to reduce the noise associated with the variability in the kinetics of peak viral titers between experiments, which are independent of any potential IKKα/IKKβ contributions to infectious virus production.