Global mRNA degradation during lytic gammaherpesvirus infection contributes to establishment of viral latency

Abstract

During a lytic gammaherpesvirus infection, host gene expression is severely restricted by the global degradation and altered 3' end processing of mRNA. This host shutoff phenotype is orchestrated by the viral SOX protein, yet its functional significance to the viral lifecycle has not been elucidated, in part due to the multifunctional nature of SOX. Using an unbiased mutagenesis screen of the murine gammaherpesvirus 68 (MHV68) SOX homolog, we isolated a single amino acid point mutant that is selectively defective in host shutoff activity. Incorporation of this mutation into MHV68 yielded a virus with significantly reduced capacity for mRNA turnover. Unexpectedly, the MHV68 mutant showed little defect during the acute replication phase in the mouse lung. Instead, the virus exhibited attenuation at later stages of in vivo infections suggestive of defects in both trafficking and latency establishment. Specifically, mice intranasally infected with the host shutoff mutant accumulated to lower levels at 10 days post infection in the lymph nodes, failed to develop splenomegaly, and exhibited reduced viral DNA levels and a lower frequency of latently infected splenocytes. Decreased latency establishment was also observed upon infection via the intraperitoneal route. These results highlight for the first time the importance of global mRNA degradation during a gammaherpesvirus infection and link an exclusively lytic phenomenon with downstream latency establishment.

Figure 1. Isolation of the single-function muSOX mutant R443I.

(A) Diagram of the random PCR mutagenesis screening strategy. (B) HEK293T cells were transfected with a plasmid expressing GFP alone or together with a plasmid expressing HA-tagged wild-type (wt) muSOX or muSOX R443I. GFP fluorescence was monitored 24 h post transfection, along with HA-muSOX, which was detected by immunofluorescence with anti-HA antibodies. Samples were co-stained with DAPI to visualize nuclei. (C) Cells were transfected for 24 h as described above, whereupon GFP mRNA and 18S rRNA levels were analyzed by Northern blot. Shown is a representative northern blot for GFP mRNA and 18S rRNA, and below is the normalized data from five independent experiments with means and standard deviations shown. (D) Total protein was harvested 24 h post transfection with the indicated plasmids, and Western blotted with anti-HA antibodies. The bottom band present in all samples is due to non-specific binding of the antibody. (E) In vitro translated GFP, wild-type muSOX, or muSOX-R443I was incubated with linear DNA for the indicated times. DNA was then extracted and separated by agarose gel electrophoresis. A representative figure from three independent experiments is shown.

Figure 2. Generation of the MHV68.ΔHS virus.

(A) Outline of the strategy for generating MHV68-muSOX-R443I (MHV68.ΔHS), in which the muSOX amino acid arginine at position 443 was mutated to isoleucine, creating an additional PsiI site. The number is derived from the sequenced MHV68 genome (Refseq: NC_001826). (B) MHV68.WT, MHV68.ΔHS, and MHV68.MR BAC DNA were digested with the enzyme PsiI and subsequently resolved on an agarose gel to confirm successful generation of the desired mutants. Arrows indicate the new 1200 bp and 450 bp bands generated after introduction of the PsiI restriction site in muSOX R443I. (C) MHV68.WT, MHV68.ΔHS, and MHV68.MR BAC DNA were digested with the enzyme EcoRI and subsequently resolved on an agarose gel to confirm no unexpected recombination had occurred.

Figure 3. MHV68.ΔHS is defective for host shutoff.

RNA was isolated from NIH 3T3 cells infected with MHV68.WT, MHV68.ΔHS, or MHV68.MR at an MOI of 10 at 20 h post infection. ActB (A), Rplp2 (B), Tubb5 (C), GAPDH (D) and 18S RNA levels were quantified via RT-qPCR. The mean and standard deviation of the normalized mRNA/18S ratio is plotted for at least four independent experiments. The p-values comparing MHV68.ΔHS and MHV68.MR are indicated. (E) COS7 cells were infected with MHV68.WT or MHV68.ΔHS for 24 h. Both viruses express GFP from the BAC vector sequence, which serves as a marker for infection. PABPC localization was monitored by immunofluorescence with anti-PABPC antibodies, and cells were stained with DAPI to visualize nuclei.

Figure 4. MHV68.ΔHS replicates with wild-type kinetics.

Viral replication kinetics were determined by multi-step growth curves in mouse fibroblasts cells following an infection at a MOI of 0.1 with MHV68.WT, MHV68.ΔHS, or MHV68.MR. At the indicated times post infection virus was harvested, and the titer was determined by plaque assay. The mean and standard deviation from at least three independent experiments is graphed.

Figure 5. MHV68.ΔHS generates larger plaques and increases the percentage of lytic antigen-expressing cells.

(A) Shown are representative images of plaques generated 4 dpi in 3T3 cells infected with MHV68.ΔHS or MHV68.MR. The scale bar in the lower right corner of the images represents 1 mm. (B) Plaque diameters of 75 plaques from 5 independent experiments were measured at 4 dpi from cells infected with MHV68.ΔHS or MHV68.MR, and the frequency distribution was graphed. (C) 3T3 cells infected at an MOI of 1 with either MHV68-YFP or MHV68-YFP.ΔHS were analyzed at 18 hpi for YFP and muSOX expression by immunofluorescence using anti-muSOX antibodies. Samples were co-stained with DAPI to visualize nuclei. Shown are representative images from four independent experiments. The numbers of lytically infected cells (YFP- and muSOX-positive) were determined by counting multiple fields of view from four independent experiments, and the percentage of infected cells expressing muSOX is shown.

Figure 6. MHV68.ΔHS replicates to near MHV68.MR levels during the acute phase of infection in the mouse lung.

C57BL/6 mice were infected intranasally with 5×10⁴ pfu MHV68.WT, MHV68.ΔHS, or MHV68.MR. At 3, 5, or 7 dpi lungs were harvested and homogenized, and viral titers were determined by plaque assay. Each point on the graph represents the viral titer from a single lung, and the bar indicates the mean titer for each virus. The dotted line represents the limit of detection at 20 pfu/lung. The p-values comparing MHV68.ΔHS and MHV68.MR are indicated. “n.s.” indicates a p-value greater than 0.05.

Figure 7. MHV68.ΔHS is attenuated for latency establishment.

In two independent experiments, C57BL/6 mice were infected intranasally with 5×10⁴ pfu MHV68.WT, MHV68.ΔHS, or MHV68.MR, and spleens were harvested at 17 dpi. (A) The weight of each spleen is plotted with the mean weight for each virus variant indicated by the bar. (B) Spleen cells were harvested and analyzed for lytic reactivation frequency via infectious center assay. The number of reactivating splenocytes per 10⁷ cells is plotted with the mean for each virus indicated by the bar. The dashed line denotes the limit of detection. (C) DNA was isolated from spleen cells and analyzed by qPCR for the viral glycoprotein B (gB, ORF8) gene and endogenous GAPDH. The normalized gB/GAPDH ratio is plotted along with the mean and standard deviation for each viral infection. The p-values comparing MHV68.ΔHS to MHV68.MR are indicated. (D) The frequency of latently infected cells was determined by limiting dilution-PCR. The percent of wells positive for PCR product is plotted for each dilution. Each point represents the average and standard deviation of three or four mice. The dotted line at 63.2% is used to calculate the frequency of genome harboring cells according to the Poisson distribution.

Figure 8. MHV68.ΔHS fails to traffic to the lymph system and establish latency.

(A) C57BL/6 mice were infected intranasally with 5×10⁴ pfu MHV68.WT, MHV68.ΔHS, or MHV68.MR, and cervical lymph nodes were harvested at 10 dpi. DNA was isolated from the cells and analyzed by qPCR for the viral glycoprotein B (gB, ORF8) gene and endogenous GAPDH. The normalized gB/GAPDH ratio is plotted along with the mean for each viral infection. (B) Mice were infected intraperitoneally with 1×10³ pfu of each of the indicated viruses. DNA was isolated and quantified as above and the normalized gB/GAPDH ratio is plotted along with the mean for each viral infection. Each point on the graph represents the ratio from a single mouse. Data for each figure are compilations of two independent experiments and the p-values are indicated.