Molecular investigation of the 7.2 kb RNA of murine cytomegalovirus

Abstract

Background: HCMV encodes a stable 5 kb RNA of unknown function that is conserved across cytomegalovirus species. In vivo studies of the MCMV orthologue, a 7.2 kb RNA, demonstrated that viruses that do not express the RNA fail to establish efficient persistent replication in the salivary glands of mice. To gain further insight into the function and properties of this conserved locus, we characterized the MCMV intron in finer detail.

Methods: We performed multiple analyses to evaluate transcript expression kinetics, identify transcript termini and promoter elements. The half-lives of intron locus RNAs were quantified by measuring RNA levels following actinomycin D treatment in a qRT-PCR-based assay. We also constructed a series of recombinant viruses to evaluate protein coding potential in the locus and test the role of putative promoter elements. These recombinant viruses were tested in both in vitro and in vivo assays.

Results: We show that the 7.2 kb RNA is expressed with late kinetics during productive infection of mouse fibroblasts. The termini of the precursor RNA that is processed to produce the intron were identified and we demonstrate that the m106 open reading frame, which resides on the spliced mRNA derived from precursor processing, can be translated during infection. Mapping the 5' end of the primary transcript revealed minimal promoter elements located upstream that contribute to transcript expression. Analysis of recombinant viruses with deletions in the putative promoter elements, however, revealed these elements exert only minor effects on intron expression and viral persistence in vivo. Low transcriptional output by the putative promoter element(s) is compensated by the long half-life of the 7.2 kb RNA of approximately 28.8 hours. Detailed analysis of viral spread prior to the establishment of persistence also showed that the intron is not likely required for efficient spread to the salivary gland, but rather enhances persistent replication in this tissue site.

Conclusions: This data provides a comprehensive transcriptional analysis of the MCMV 7.2 kb intron locus. Our studies indicate that the 7.2 kb RNA is an extremely long-lived RNA, a feature which is likely to be important in its role promoting viral persistence in the salivary gland.

Figure 1

Expression kinetics of intron locus transcripts. Total RNA was harvested from infected mouse fibroblasts (MOI = 1.0) at the indicated times and analyzed by northern blot analysis using radio-labeled, antisense RNA probes specific for (A) the intron or (B) exon 2 of the spliced mRNA. To determine the kinetics of intron locus expression, cells were pre-treated with either PAA or CHX for 1 hour prior to infection. Total RNA was prepared at the indicated times.

Figure 2

Mapping of the intron locus. (A) Diagram of the genomic region encompassing the primary transcript of the 7.2 kb intron, illustrating the transcriptional start sites (TSS) and the splice donor (SD) and splice acceptor sites (SA). The location of primers used for primer extension or RLM-RACE are indicated. (B) Diagram of the spliced mRNA and the 5′ and 3′ ends that were identified by RLM-RACE using total RNA harvested from infected mouse fibroblasts at 48 hours post infection (hpi). Putative TATA box directly upstream of the transcriptional start sites is indicated. Also shown are the start and stop codons of the m106 ORF encoded on the second exon of the mRNA as well as the poly-adenylation and cleavage sites used in processing this mRNA. (C) Primer extension analysis was performed on total RNA harvested from either mock-infected (M) or MCMV (WT) infected mouse fibroblasts at 48 hpi. Radiolabeled primers were used to validate the 5′RLM-RACE products (primer 497) and confirm the known 7.2 kb splice donor site (primer 50) as a control. (D) Northern blot analysis of total RNA (T) harvested from infected mouse fibroblasts and fractionated for either polyadenylated (A+) or non-polyadenylated (A-) RNA. The blot was hybridized with a radiolabeled, antisense RNA probe specific for exon 2.

Figure 3

The m106 open reading frame is translated during MCMV infection. (A) Diagram of the GFP cassette insertion within the MCMV Smith BAC clone. (B) Analysis of recombinant virus replication in mouse fibroblasts. Cells were infected at a multiplicity of 0.05 PFU/cell and viral supernatants were collected daily and titrated. Graph represents two biological replicates. (C) Western blot analysis of protein lysates prepared from mock-, MCMV:m106GFP (WT-GFP), or MCMVdelSD:m106GFP (SDM-GFP) infected cells at 48 h p.i. m106-GFP expression was detected using a polyclonal anti-GFP antibody. Asterisk denotes m106-GFP. (D) Mouse fibroblasts were infected with MCMV:m106-GFP (MOI = 0.05). Cells were fixed at 72 h p.i. and m106-GFP protein expression and 7.2 kb intron production was detected by by combined immunofluorescence assay and FISH.

Figure 4

Half-life analysis of the intron locus transcripts. Mouse fibroblasts were infected with MCMV (MOI = 1.0). At 30 hours post infection, cells were treated with 4 ug/ml of Actinomycin D. Total RNA was harvested over the indicated time course. Intron, mRNA, and 18 srRNA transcript levels were quantified by qRT-PCR. Intron and spliced mRNA transcripts were normalized to 18 srRNA. The relative quantitative values at time zero hours were adjusted to 100% and transcript remaining was compared relative to time zero. The fitted curve was modeled by one phase decay using a non-linear regression analysis on four biological replicates for each time point. The half-life (t1/2) shown for each transcript is the best-fit value. Bars represent the mean and error bars represent the standard error of the mean (SEM). (A) Schematic representation of primer probe sets used for qRT-PCR analysis. SD = splice donor sequence, SA = splice acceptor sequence. Half-life decay curves for the (B) 7.2 kb intron using primer probe set a/a’, (C) the second exon of the mRNA (b/b’), (D) the spliced mRNA using primer probe set c/c’, and (E) GAPDH.

Figure 5

Analysis of transcriptional activity of putative intron locus promoter elements. (A) Diagram of genomic location of the viral sequences cloned into pGL3-reporter plasmids. (B) pGL3-reporter constructs were co-transfected into mouse fibroblasts and luciferase induction was assayed 24 h p.i. following either mock co-infection, (C) UV inactivated co-infection, or MCMV co-infection (MOI = 0.05).

Figure 6

Deletion mutations in putative viral promoter elements reveal reduction in transcriptional output in cell culture and decreased recovery of infectious virus in vivo. (A) Diagram of the genomic location of intron locus promoter region deletions. (B) Multi-step growth analysis of replication of all three recombinant viruses in comparison to WT MCMV. Mouse fibroblasts were infected (MOI = 0.05) and culture supernatants were collected every 24 hours and titrated by plaque assay. Graph represents three biological replicates. (C, D, and E) Quantification of intron locus RNAs in cells infected with promoter mutant viruses. Mouse fibroblasts were infected (MOI = 1.0) and total RNA was harvested 48 hours post infection. Intron locus transcript levels are quantified relative to WT MCMV transcript levels by qRT-PCR. The primer probes sets used are the same as shown in Figure  4A. Graphs represent three biological replicates. (F and G) Three-month-old female BALB/c mice were infected with an i.p. dose of 5 x 10⁵ PFU with the indicated viruses. At 14 days post infection, animals were euthanized and tissues collected for analysis of infectious virus yield (F) and viral genome number (G). (F) Salivary gland homogenates were analyzed by plaque assay on mouse fibroblasts. p values represent the Student’s T Test result between WT MCMV infected cells or mice and cells or mice infected with the given recombinant viral mutant for each transcript analyzed (*p < 0.05 **p < 0.01 ***p < 0.001 ****p < 0.0001). WT MCMV = WT; MCMVdel20 = 20DEL; MCMVdel100 = 100DEL; MCMVdel135 = 135DEL; MCMVdelHP = HPM; MCMVdelSD = SDM.

Figure 7

Intron locus RNAs are dispensable for virus dissemination to the salivary glands during acute infection. Three-month-old female BALB/c mice were inoculated i.p. with 5x10⁵ PFU of WT MCMV or MCMVdelSD. At the indicated days post infection, organs were harvested from three mice per infection group to quantify infectious virus by plaque assay. Bars represent the mean and error bars represent the standard error of the mean (SEM). The dashed line indicates the limit of detection.

Figure 8

Detection of viral genomes is significantly reduced in salivary glands during persistence. Three-month-old female BALB/c mice were inoculated i.p. with 5x10⁵ PFU of WT MCMV (WT) or MCMVdelSD (SDM). DNA was harvested from the indicated organs of three mice per infection group at 14 days post infection and viral genomes were quantified by qPCR using a primer probe set specific for the M54 MCMV gene and normalized to the beta-actin cellular gene. Bars represent the mean and error bars represent the standard error of the mean (SEM). p values represent the Student’s T Test result between WT MCMV and MCMVdelSD infected mice (*p < 0.05 **p < 0.01 ***p < 0.001 ****p < 0.0001).