Murine gamma-herpesvirus 68 hijacks MAVS and IKKbeta to initiate lytic replication

Abstract

Upon viral infection, the mitochondrial antiviral signaling (MAVS)-IKKbeta pathway is activated to restrict viral replication. Manipulation of immune signaling events by pathogens has been an outstanding theme of host-pathogen interaction. Here we report that the loss of MAVS or IKKbeta impaired the lytic replication of gamma-herpesvirus 68 (gammaHV68), a model herpesvirus for human Kaposi's sarcoma-associated herpesvirus and Epstein-Barr virus. gammaHV68 infection activated IKKbeta in a MAVS-dependent manner; however, IKKbeta phosphorylated and promoted the transcriptional activation of the gammaHV68 replication and transcription activator (RTA). Mutational analyses identified IKKbeta phosphorylation sites, through which RTA-mediated transcription was increased by IKKbeta, within the transactivation domain of RTA. Moreover, the lytic replication of recombinant gammaHV68 carrying mutations within the IKKbeta phosphorylation sites was greatly impaired. These findings support the conclusion that gammaHV68 hijacks the antiviral MAVS-IKKbeta pathway to promote viral transcription and lytic infection, representing an example whereby viral replication is coupled to host immune activation.

Figure 1. MAVS deficiency reduces γHV68 lytic replication.

MAVS wild-type (MAVS^+/+) or knockout (MAVS^−/−) mice were intranasally (i.n.) infected with 40 PFU γHV68. The lungs were harvested at 4, 7, 10, 13, and 16 days post-infection (d.p.i.) and viral titer was determined by a plaque assay. (B) Mouse embryonic fibroblasts (MEFs) were infected with a GFP-marked γHV68 K3/GFP at multiplicities of infection (MOI) of 0.01. Viral replication in MAVS^+/+ and MAVS^−/− MEFs were photographed. (C) MEFs were infected with γHV68 K3/GFP as in (B) and viral titers were determined by a plaque assay. Data represent three independent experiments and error bars denote standard error of the mean (SEM). (D) MAVS^+/+ and MAVS^−/− MEFs were infected with 120 PFU γHV68 K3/GFP or 5 PFU vesicular stomatitis virus (VSV), and plaques were counted. Data represent the mean ± SEM of three independent experiments. (E to G) MAVS^−/− MEFs were respectively infected with control lentivirus (Vec) or lentivirus containing the Flag-tagged human MAVS (MAVS), and selected with puromycin. (E) MAVS expression was confirmed by immunoprecipitation and immunoblot with anti-Flag antibody. (F) γHV68 plaque assays were performed as in (D). (G) Reconstituted MAVS^−/− MEFs as indicated were infected with γHV68 K3/GFP (MOI = 0.01), and viral multi-step growth was determined by a plaque assay. Statistical significance in (A), (D), and (F): *, P<0.05; **, P<0.02; ***, P<0.005.

Figure 2. The MAVS-IKKβ pathway is necessary for efficient γHV68 lytic replication ex vivo.

(A) Two known pathways, the IKKα/β/γ-NFκB and TBK-1/IKKε-IRF pathways, downstream of MAVS. (B) The initiation of γHV68 lytic replication in wild-type (WT) MEFs and MAVS^−/−, IKKα^−/−, IKKβ^−/−, IKKγ^−/−, TRAF6^−/−, IFNAR^−/−, and IRF3^−/−IRF7^−/− (double knockout) MEFs was assessed by a plaque assay. Data represent the mean ± SEM of three independent experiments. (C) Multi-step growth properties of γHV68 (MOI = 0.01) in wild-type MEFs and IKKβ^−/−, IKKγ^−/−, and IKKα^−/− MEFs were examined by plaque assays. Data represents three independent experiments. (D to F) Wild-type, MAVS^−/−, and IKKβ^−/− MEFs were respectively infected with control lentivirus (Vec) or lentivirus containing the Flag-tagged IKKβ (IKKβ), and selected with puromycin. (D) IKKβ expression was confirmed by immunoprecipitation and immunoblot with anti-Flag antibody (top). γHV68 plaque assays were performed as in (B). (E) Reconstituted MEFs of indicated genotypes were used for γHV68 plaque assays as in (B) with increasing doses of γHV68. Data represent the mean ± SEM of three independent experiments. (F) Reconstituted IKKβ^−/− MEFs as indicated were infected with γHV68 K3/GFP (MOI = 0.01), and viral multi-step growth was determined by a plaque assay. Statistical significance (P-value) in (B), (D), and (E) was calculated with two-tailed unpaired Student's t-test: *, P<0.05; **, P<0.02; ***, P<0.005.

Figure 3. γHV68 infection activates IKKβ in a MAVS-dependent manner.

(A) Wild-type MEFs were treated with the IKKβ inhibitor, Bay11-7082, for 30 min at 0.5 h before infection or 7 h post-infection (h.p.i.) with γHV68. Cells were washed with medium and incubated for plaque formation. Plaques formed at 6 d.p.i. were counted. Data represent the mean ± SEM. (B) MEFs were infected with γHV68 (MOI = 10) and whole cell lysates of MEFs at indicated time points after γHV68 infection were precipitated with anti-IKKβ antibody. One half of IKKβ was used for an in vitro kinase assay with GST-IκBNT (amino terminal 50 amino acids of IκBα) (top) or analyzed by immunoblot (middle). Relative intensity of phosphorylated GST-IκBNT was normalized to IKKβ protein (bottom). (C) γHV68 infection was carried out as in (B) and whole cell lysates were analyzed by immunoblot with anti-IκBα (top) and β-actin (bottom). (D and E) Equal amount of live (MOI = 10) or UV-inactivated (UV) γHV68 was used to infect wild-type MEFs. The IKKβ kinase activity was assessed as in (B) and whole cell lysates were analyzed by immunoblot as in (C) for IκBα and β-actin. Graphs at the bottom show normalized IKKβ kinase activity (D) and IκBα protein (E).

Figure 4. The MAVS-IKKβ pathway is important for γHV68 mRNA production.

(A and B) MAVS^+/+ and MAVS^−/− MEFs were infected with γHV68 (MOI = 0.01) and cells were harvested at two hours post-infection. Total DNA was extracted and viral genomes were analyzed by PCR and agarose gel electrophoresis (A) or quantitative real-time PCR (qRT-PCR) (B). (C and D) MEFs were infected with γHV68 as in (A). Total RNA was extracted and levels of γHV68 mRNA transcripts were examined by reverse transcription and PCR (C) or qRT-PCR (D) using primers specific for viral genes as indicated. I.E., immediate early; E, early. (E and F) 293T cells were transfected with the γHV68 BAC and a plasmid containing TRAF6. At 28 h post-transfection, levels of the γHV68 genome were determined by qRT-PCR (E) and levels of various γHV68 gene transcripts as indicated were determined by reverse transcription and qRT-PCR analyses (F). Data represent the mean ± SEM.

Figure 5. IKKβ phosphorylates and potentiates the transcription activity of γHV68 RTA.

(A) IKKβ or IKKβΔKD purified from 293T cells (left) were incubated with [³²P]γATP and bacterial GST fusion proteins containing RTA fragments (middle), and examined by autoradiography (right). (B) MEFs of indicated genotype were infected with γHV68 (MOI = 2) for 4 h, labeled with [³²P]-orthophosphoric acid for 8 h. Whole cell lysates were precipitated with anti-RTA antibody and analyzed by autoradiography (top) or immunoblot with anti-RTA antibody (bottom). (C) 293T cells were transfected with reporter plasmids and plasmids containing RTA, TRAF6, IKKβ, and IKKβΔKD. Luciferase activity normalized against β-galactosidase activity was shown. (D) Phosphorylation of GST fusion proteins, containing the C-terminal 127 amino acids of wild-type RTA, STS/A, or TTS/A variants, by IKKβ was analyzed similarly as in (A). (E) 293T transfection and luciferase reporter assays were carried out as in (C). (F) Whole cell lysates of 293T transfected with plasmids containing wild-type RTA, STS/A, or TTS/A variants were analyzed by immunoblot (top) and used to normalize the basal transcriptional activity of wild-type RTA, the STS/A and TTS/A variants (bottom). Data in (C), (E), and (F) represent the mean ± SEM with indicated P values (*, P<0.05; **, P<0.02) of at least three independent experiments.

Figure 6. Generation and characterization of γHV68 BAC carrying the TTS/A mutation.

(A) Diagram of the strategy to generate recombinant γHV68. Briefly, wild-type RTA or the STS/A and TTS/A alleles were PCR amplified with overlapping PCR primers. Purified PCR products were transfected into NIH3T3 cells, together with the BAC clone containing a transposon within the transactivation domain of RTA. Recombinant viruses in the supernatant were used to infect NIH3T3 cells. Circular BAC DNA was purified and electroporated into DH10B cells. Chl, chloramphenicol; Kan, kanamycin. (B and C) BACs containing γHV68 genome were analyzed by KpnI (B) or EcoRI (C) digestion, and resolved on 0.8% agarose gels stained with ethidium bromide. The white arrows indicate the specific fragment shift caused by homologous recombination within the RTA locus. NR, RTA-null rescued. (D) 293T cells were transfected with γHV68 M3 reporter plasmid and BAC.NR or BAC.TTS/A. At 28 h post-transfection, luciferase activity and β-galactosidase activity were determined and M3 transcriptional activation by RTA was shown. (E) Transfection of 293T cells with the BAC.TTS/A DNA and a plasmid containing TRAF6, and qRT-PCR were carried out as in Figure 4F.

Figure 7. Impaired lytic replication of recombinant γHV68 carrying the TTS/A mutation.

(A) Wild-type MEFs were infected with equal number of genomes, measured by qRT-PCR, of recombinant γHV68.NR (MOI = 0.1) or γHV68.TTS/A. At 30 h post-infection, the RTA mRNA levels were determined by reverse transcription and qRT-PCR. (B) Wild-type, MAVS^−/−, and IKKβ^−/− MEFs were infected with recombinant γHV68.NR (left) and γHV68.TTS/A (right) (MOI = 0.01) and the multi-step growth curves were determined by a plaque assay. Data represent three independent experiments. (C) The lytic replication of recombinant γHV68.NR and γHV68.TTS/A was examined similarly as in (B) with an MOI of 0.1 (γHV68.NR).