Repair of an Attenuated Low-Passage Murine Cytomegalovirus Bacterial Artificial Chromosome Identifies a Novel Spliced Gene Essential for Salivary Gland Tropism

Abstract

The cloning of herpesviruses as bacterial artificial chromosomes (BACs) has revolutionized the study of herpesvirus biology, allowing rapid and precise manipulation of viral genomes. Several clinical strains of human cytomegalovirus (HCMV) have been cloned as BACs; however, no low-passage strains of murine CMV (MCMV), which provide a model mimicking these isolates, have been cloned. Here, the low-passage G4 strain of was BAC cloned. G4 carries an m157 gene that does not ligate the natural killer (NK) cell-activating receptor, Ly49H, meaning that unlike laboratory strains of MCMV, this virus replicates well in C57BL/6 mice. This BAC clone exhibited normal replication during acute infection in the spleen and liver but was attenuated for salivary gland tropism. Next-generation sequencing revealed a C-to-A mutation at nucleotide position 188422, located in the 3' untranslated region of sgg1, a spliced gene critical for salivary gland tropism. Repair of this mutation restored tropism for the salivary glands. Transcriptional analysis revealed a novel spliced gene within the sgg1 locus. This small open reading frame (ORF), sgg1.1, starts at the 3' end of the first exon of sgg1 and extends exon 2 of sgg1. This shorter spliced gene is prematurely terminated by the nonsense mutation at nt 188422. Sequence analysis of tissue culture-passaged virus demonstrated that sgg1.1 was stable, although other mutational hot spots were identified. The G4 BAC will allow in vivo studies in a broader range of mice, avoiding the strong NK cell responses seen in B6 mice with other MCMV BAC-derived MCMVs.IMPORTANCE Murine cytomegalovirus (MCMV) is widely used as a model of human CMV (HCMV) infection. However, this model relies on strains of MCMV that have been serially passaged in the laboratory for over four decades. These laboratory strains have been cloned as bacterial artificial chromosomes (BACs), which permits rapid and precise manipulation. Low-passage strains of MCMV add to the utility of the mouse model of HCMV infection but do not exist as cloned BACs. This study describes the first such low-passage MCMV BAC. This BAC-derived G4 was initially attenuated in vivo, with subsequent full genomic sequencing revealing a novel spliced transcript required for salivary gland tropism. These data suggest that MCMV, like HCMV, undergoes tissue culture adaptation that can limit in vivo growth and supports the use of BAC clones as a way of standardizing viral strains and minimizing interlaboratory strain variation.

Keywords: BAC; G4; MCMV; sgg1; sgg1.1; tissue tropism.

FIG 1

G4 BAC construction and in vitro characterization. (A) Schematic of BAC construction showing EcoRI restriction enzyme sites (E) and expected fragment sizes (double arrows). (B) RFLP of BAC constructs. The intermediate BAC (pG420) and repaired BAC (pG4209J) were digested with EcoRI and compared to DNA from the parental G4 virus. Solid arrows show DNA fragments unique to the parental virus; dashed arrows show DNA fragments unique to the BAC constructs. (C) Rescued virus (G4209) replicates like the wild type in vivo. Shown are the results of a multistep growth assay performed on MEFs; data are means ± standard deviations (SD) of duplicate data.

FIG 2

In vivo replication of rescued G4 BAC-derived virus, ARG4. Female BALB/c or C57BL/6 mice were infected with 5 × 10³ PFU of SGV stocks of either G4 or ARG4. Titers of virus were assessed by plaque assay on day 3 postinfection in the spleen (A) and liver (B) and on day 17 postinfection in the salivary glands (C). No significant differences between titers of G4 versus ARG4 were noted in any tissue and at any time point. There were 5 mice per group. The dashed line indicates the limit of detection.

FIG 3

The sgg1 locus encodes a novel spliced transcript. (A) Annotation of the sgg1 locus showing the locations of the ORFs from the previous MCMV genome annotation, i.e., m132.1 and sgg1, as well as the extension of the m133 ORF due to a correction to the original sequence of MCMV strain Smith. The red vertical bar denotes the mutation present in G4209 that introduces a premature stop codon into novel sgg1.1, defined by the results of Northern blotting, RNA-Seq, and RACE analysis, as described below in this legend. (B) Northern blot analysis of the sgg1 locus. Total RNA isolated at 6 hpi or 48 hpi from MEFs infected with the indicated viruses was probed with a strand-specific Northern blot probe (designated by a black bar in panel A and described in Materials and Methods). (C) RNA-Seq profile showing read coverage and splice junctions within the sgg1 region at a late stage of MCMV infection. The raw number of spliced reads spanning each splice junction is shown in the arc connecting the corresponding donor and acceptor splice sites. Total read coverage (Cov.) is shown on the y axis. (D) Sequences of the 5′ RACE and 3′ RACE PCR products were aligned to the G4 genome; the 5′ RACE sequences are shown in light blue, 3′ RACE sequences are in tan, and disagreements over the G4 genomic sequence are designated with a thin black rectangle within each RACE amplicon. The putative coding sequence in the newly identified sgg1.1 transcript is shown in light red, and the presumed poly(A) sequence is shown in yellow. An overview of the genomic region containing the sgg1.1 transcript, together with the genomic coordinates of the relevant features in the G4 genome, is shown at the bottom.

FIG 4

The sgg1.1 truncated virus replicates in vivo with similar kinetics to that seen for other sgg1 mutants. To determine if the in vivo phenotype of G4209.M32 (sgg1.1 mutant virus) was similar to that seen for other sgg1 mutant viruses, female BALB/c mice were infected with 5 × 10³ PFU of SGV stocks of G4209.M32 or fully repaired ARG4. Titers of virus were assessed by plaque assay on day 3 postinfection in the spleen (A) and liver (B) and on day 17 postinfection in the salivary glands (C). There were 5 mice per group. The asterisk indicates a significant difference between ARG4 and G4209.M32 in the salivary glands (P = 0.0119 by Mann-Whitney U test). The dashed line indicates the limit of detection.