Identification of a novel betaherpesvirus in Mus musculus

Abstract

Rodent betaherpesviruses vary considerably in genomic content, and these variations can result in a distinct pathogenicity. Therefore, the identification of unknown betaherpesviruses in house mice (Mus musculus), the most important rodent host species in basic research, is of importance. During a search for novel herpesviruses in house mice using herpesvirus consensus PCR and attempts to isolate viruses in tissue culture, we identified a previously unknown betaherpesvirus. The primary PCR search in mouse organs revealed the presence of known strains of murine cytomegalovirus (Murid herpesvirus 1) and of Mus musculus rhadinovirus 1 only. However, the novel virus was detected after incubation of organ pieces in fibroblast tissue culture and subsequent PCR analysis of the supernatants. Long-distance PCR amplification including the DNA polymerase and glycoprotein B genes revealed a 3.4 kb sequence that was similar to sequences of rodent cytomegaloviruses. Pairwise sequence comparisons and phylogenetic analyses showed that this newly identified murine virus is most similar to the English isolate of rat cytomegalovirus, thereby raising the possibility that two distinct CMV lineages have evolved in both Mus musculus and Rattus norvegicus.

Figure 1

Amplification strategy. At the top of the figure above the ruler, the ORFs of UL55 and UL54 are represented by black arrows. Below the ruler, the PCRs are depicted. The degenerate and specific primers are represented by black and open triangles, respectively. Amplified sequences are shown as black lines.

Figure 2

The novel cytomegalovirus of Mus musculus: Multiple sequence alignment and phylogenetic analysis. (a) The MmusCMV-2 DPOL sequence amplified by panherpes consensus PCR (178 bp) was translated into a 59 aa sequence, and a multiple sequence alignment with the corresponding sequences of MCMV, RCMV-M, RCMV-E, HCMV and HHV-6b was generated using the ClustalW module of MacVector™ 10.6. Identical and similar amino acids are boxed, the last in inverse type. Mismatches are given in lower type. (b) A phylogenetic tree was constructed using the nucleic acid sequences encoded by the gB-DPOL segments of MmusCMV-2 and those of known human, mouse and rat cytomegaloviruses (GenBank accession numbers in the Findings section). Abbreviations of common names are used, and those of species names according to the ICTV (International Committee on the Taxonomy of Viruses) are given in parentheses. A multiple alignment of 3.4 kb was analysed with the neighbour-joining method. A rooted phylogram is shown, with HHV-6a as outgroup. The branch length is proportional to evolutionary distance (scale bar). Results of bootstrap analysis (1000 replicates) are indicated at the nodes of the tree. The novel MmusCMV-2 is highlighted in bold type.