Latent gammaherpesvirus 68 infection induces distinct transcriptional changes in different organs

Abstract

Previous studies identified a role for latent herpesvirus infection in cross-protection against infection and exacerbation of chronic inflammatory diseases. Here, we identified more than 500 genes differentially expressed in spleens, livers, or brains of mice latently infected with gammaherpesvirus 68 and found that distinct sets of genes linked to different pathways were altered in the spleen compared to those in the liver. Several of the most differentially expressed latency-specific genes (e.g., the gamma interferon [IFN-γ], Cxcl9, and Ccl5 genes) are associated with known latency-specific phenotypes. Chronic herpesvirus infection, therefore, significantly alters the transcriptional status of host organs. We speculate that such changes may influence host physiology, the status of the immune system, and disease susceptibility.

FIG 1

Latent MHV68 alters host gene expression. Male C57BL/6 mice were infected intranasally with MHV68, ORF73.stop mutant virus (ORF73), or mockulum (mock), and spleens, livers, and brains (3 experiments) were harvested at 28 dpi. RNA was extracted using the ToTALLY RNA kit (Ambion) from tissues incubated in RNAlater-ICE (Ambion). Equal amounts of RNA were pooled from three to four mice per condition for each microarray chip. RNA was labeled and hybridized to Affymetrix M430 2.0 microarrays at the Multiplexed Gene Analysis Core Facility at Washington University in St. Louis, MO. Data were normalized using the Robust Multi-array Average normalization routine in Affymetrix's Expression Console software. Microarray data were analyzed by both factorial analysis and two-way ANOVA. Differentially expressed genes identified by both analyses are shown in the heat map after hierarchical clustering of the genes. The heat map represents the fold change relative to results for mock-infected mice, with red representing the row maximum and blue the row minimum value. Statistical significance was performed by using a standard Student t test in the case of factorial analysis. P values were corrected for multiple hypothesis testing. Gene expression differences were considered significant if the adjusted P value was <0.05 and a >1.5-fold change was observed in expression levels.

FIG 2

Validation of microarray analyses by qRT-PCR. RNA pooled from three to four mice per condition was treated with Turbo DNase (Life Technologies) before cDNA synthesis using 1 μg RNA, oligo(dT)12-18, and Superscript II reverse transcriptase (Life Technologies). qPCR was performed with cDNA from livers (A) or spleens (B) using Power SYBR green master mix (Applied Biosystems) with the indicated gene-specific primers (Table 1). Data were collected from 4 independent experiments. Transcript levels were normalized to the level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) within each sample and compared to the level in mock-infected mice using the ΔΔC_T method (where C_T is the threshold cycle). Data are presented as means ± SEM. Statistical analyses were performed by using the Mann-Whitney U test. ∗, P < 0.05; ns, not significant; nd, could not reliably detect target.

FIG 3

Pathway analysis of differentially expressed genes in latently infected mice. Gene overlap analyses of differentially expressed genes in spleens (A), livers (B), or brains (C) of MHV68-infected mice. A list of latent infection-specific differentially expressed genes was identified by comparing MHV68-infected mice to mock-infected mice and then removing genes that overlapped with those differentially expressed for ORF73.stop-infected mice compared to expression for mock-infected mice. Pathways were identified using MSigDB's Canonical Pathway gene set. The 50 most significant pathways for spleens (A) or livers (B) are displayed. The x axis represents the q value, indicating the significance of enrichment for a given gene set. The values are plotted on a negative log₁₀ scale. Gene sets with values of >1.3 (q < 0.05) were significantly enriched.

FIG 4

Overlap of genes regulated by IFN-γ and latent viral infection. Heat map of differentially expressed genes in livers, spleens, or brains from MHV68-infected mice (compared to mock-infected mice) that significantly overlapped with genes altered by IFN-γ in bone marrow-derived macrophages. The heat map displays the fold change relative to results for mock-infected mice, with red representing the row maximum and blue the row minimum value. Statistical significance was assessed by a hypergeometric test. Nominal P values were used.