The murine cytomegalovirus chemokine homolog, m131/129, is a determinant of viral pathogenicity

Abstract

Chemokines are important mediators of the early inflammatory response to infection and modify a wide range of host immune responses. Functional homologs of cellular chemokines have been identified in a number of herpesviruses, suggesting that the subversion of the host chemokine response contributes to the pathogenesis of these viruses. Transcriptional and reverse transcription-PCR analyses demonstrated that the murine cytomegalovirus (MCMV) chemokine homolog, m131, was spliced at the 3' end to the adjacent downstream open reading frame, m129, resulting in a predicted product of 31 kDa, which is significantly larger than most known chemokines. The in vivo impact of m131/129 was investigated by comparing the replication of MCMV mutants having m131/129 deleted (Deltam131/129) with that of wild-type (wt) MCMV. Our studies demonstrate that both wt and Deltam131/129 viruses replicated to equivalent levels during the first 2 to 3 days following in vivo infection. However, histological studies demonstrated that the early inflammatory response elicited by Deltam131/129 was reduced compared with that of wt MCMV. Furthermore, the Deltam131/129 mutants failed to establish a high-titer infection in the salivary glands. These results suggest that m131/129 possesses proinflammatory properties in vivo and is important for the dissemination of MCMV to or infection of the salivary gland. Notably, the Deltam131/129 mutants were cleared more rapidly from the spleen and liver during acute infection compared with wt MCMV. The accelerated clearance of the mutants was dependent on NK cells and cells of the CD4(+) CD8(+) phenotype. These data suggest that m131/129 may also contribute to virus mechanisms of immune system evasion during early infection, possibly through the interference of NK cells and T cells.

FIG. 1

(A) Genetic organization of the MCMV m131 region. Nucleotide positions refer to the sequenced Smith strain of MCMV. Solid ellipses indicate the position of poly(A) signals on each strand. The specificities of subgenomic fragments used as probes in Northern hybridizations are depicted as solid bars below the ORFs. The probe for the m131 ORF was amplified by PCR from viral DNA. (B) Construction diagrams (not drawn to scale) for plasmids used in this study. Additional details are provided in the text.

FIG. 2

Southern blot analysis of MCMV recombinants. DNA samples were prepared from cells infected with either wt, Δm131Z, Δm131ns, or m131r virus. Samples were digested with SacI or EcoRV, separated on a 1.0% agarose gel, and blotted onto nylon membranes before being hybridized to radiolabelled probes. The 3.8-kb lacZ cassette from pMV10 (19) was used as the lacZ probe; the m131 probe was derived from genomic MCMV DNA by PCR. The predicted positions of SacI and EcoRV sites for each of the viruses are shown schematically below the blot. The open boxes represent the m131 ORF. The solid box represents the lacZ ORF. Predicted fragment sizes are indicated in kilobases.

FIG. 3

Northern blot analysis of RNA isolated from MCMV-infected MEF at immediate-early (lanes 1), early (lanes 2), and late (lanes 3) times p.i. RNA from uninfected MEF was included as a control. Total RNA (5 μg) was separated in a denaturing agarose gel, blotted onto nylon, and hybridized with double-stranded DNA probes. The specificity of the DNA probes is indicated in Fig. 1A. Solid arrows denote the 1-kb m131/129 transcript in wt MCMV and in Δm131ns.

FIG. 4

Nucleotide sequence of the splice junction determined for MCMV m131/129. The intron sequences are displayed in lowercase type. Changes in the nucleotide and amino acid sequence from the published Smith strain are underlined; the cysteine residues conserved with cellular CC chemokines and the predicted initiating methionine residues for m131 and m129 are doubly underlined. The NotI site used for the insertion of the lacZ cassette and oligonucleotide pair is underscored by a dashed line. The predicted signal sequence and N-linked glycosylation sites (N-X-T/S, where X is any amino acid) are shown in italics.

FIG. 5

Multistep growth curves of wt MCMV and Δm131Z in MEF. Confluent monolayers were infected with 0.01 PFU of either wt (■) or Δm131Z (◊) virus per cell. Following virus adsorption (1 h at 37°C), monolayers were washed, fresh medium was added, and the mixture was incubated for various time points before the cells and supernatant were harvested. Samples were stored at −80°C before being used for titer determination on MEF. Each point represents the log₁₀ of the mean PFU per milliliter, with one standard deviation, of triplicate samples. The in vitro growth rates of m131r and Δm131ns were also shown to be similar to those of wt virus and are not presented in the figure.

FIG. 6

Growth of wt (black bars) and Δm131Z (grey bars) virus in the spleens (A), livers (B), and salivary glands (C). Groups of four BALB/c mice were infected i.p. with 10⁴ PFU of the relevant virus, and spleens, livers, and salivary glands were harvested on the days indicated. Organs were homogenized and stored at −80°C before being used for titer determination on MEF. Virus titers are expressed as mean log₁₀ per organ, with standard error. The y axis is cropped to indicate the lower limit of virus detection. N.D, not determined.

FIG. 7

Growth of m131r (shaded bars), Δm131ns (stippled bars), and Δm131Z (solid bars) in vivo. Groups of four BALB/c mice were infected i.p. with 10⁴ PFU of the relevant virus. Organs were harvested and homogenized on the days indicated and stored at −80°C before being used for titer determination on MEF. Virus titers are expressed as the mean log₁₀ per organ, with standard error. The y axis is cropped to indicate the lower limit of virus detection. S.G., salivary glands.

FIG. 8

Effect of NK-cell and CD4⁺ CD8⁺-T-cell depletion on the replication of wt and Δm131Z in the spleen (A), liver (B) and salivary gland (C). The methods used for in vivo depletion of NK cells (stippled bars) and CD4⁺ CD8⁺ T cells (open bars) is described in the text. Undepleted mice (solid bars) were used as controls. All mice received 10^4.3 PFU of either wt or Δm131Z. Virus titers in the spleen and liver were determined on day 5 p.i.; virus titers in the salivary gland were determined in a separate study on day 11 p.i. Each group represents the log₁₀ of the mean PFU per organ, with one standard deviation, from five mice. The y axis is cropped to indicate the lower limit of virus detection.

FIG. 9

Δm131Z elicits a reduced inflammatory infiltrate in the livers of infected mice. Samples were prepared from livers of wt- and Δm131Z-infected BALB/c mice harvested 2 days p.i. (a and b) The number of foci of infection observed for the wt (a) and recombinant (b) viruses. Magnification, ×22.75. Foci of infection are marked by arrows. (c and d) The number of inflammatory cells surrounding each focus of infection for the wt (c) and Δm131Z (d) viruses. Magnification, ×36.4.