Inability of interferon to protect virus-infected cells against lysis by natural killer (NK) cells correlates with NK cell-mediated antiviral effects in vivo

Murine cytomegalovirus (MCMV) is a natural killer (NK) cell-sensitive virus, whereas lymphocytic choriomeningitis virus (LCMV) is an NK cell-resistant virus. Selective depletion of NK cell activity by injection of mice with anti-asialo GM1 antibody enhanced synthesis of MCMV but not that of LCMV when mice were simultaneously infected with the two viruses. This suggests that the NK cell-mediated antiviral effects may depend on target cell susceptibility to NK cell-mediated lysis rather than the ability of a virus to induce a specialized antiviral NK cell. In support of this concept, activated NK cells isolated from either MCMV- or LCMV-infected mice had similar patterns of killing against all targets tested. Mouse embryonic fibroblasts (MEF) infected with MCMV were less sensitive to lysis by activated NK cells than either uninfected or LCMV-infected MEF. However, when MEF were pretreated with IFN, activated NK cell-mediated lysis against MCMV-infected MEF was undiminished and was much higher (up to fourfold) than that against uninfected MEF, whose sensitivity to lysis was almost totally abolished by IFN pretreatment. LCMV-infected MEF were also protected by IFN against activated NK cell-mediated lysis. During infection, the virus-induced IFN may protect uninfected and LCMV-infected cells from IFN-activated, NK cell-mediated lysis, but MCMV-infected cells may remain sensitive to lysis. This could explain how NK cells play a role in resistance to MCMV but not LCMV.     

Enhanced immune response late in primary cytomegalovirus infection of mice.

Murine cytomegalovirus (MCMV) infection has been previously shown to depress humoral and cell-mediated functions to non-MCMV antigens. In this report we show that in C3D2 mice undergoing nonlethal primary infection the depressed anti-sheep RBC plaque-forming cell (PFC) response is followed by an enhanced PFC response. Infected mice often generated twice the number of PFC per spleen than that of control mice. Total numbers of spleen cells as well as the recovery of virus from spleens of infected mice did not distinguish the depressed from the enhanced phase of the response. Investigation of the kinetics of the response revealed a defect in shutdown regulation. This enhanced PFC response during primary MCMV infection was not reflected in measurements of serum hemagglutinin. These findings suggest that MCMV induces an impairment of immunoregulation.     

Murine cytomegalovirus with a deletion of genes spanning HindIII-J and -I displays altered cell and tissue tropism.

Murine cytomegalovirus (MCMV) gene products dispensable for growth in cell culture are likely to have important functions within the infected host, influencing tissue tropism, dissemination, or immunological responses against the virus. To identify such genes, our strategy was to delete large regions of the MCMV genome likely to contain genes nonessential for virus replication in NIH 3T3 cells. Mutant virus RV7 contained a deletion of 7.7 kb spanning portions of MCMV HindIII-J and -I. This virus grew comparably to wild-type (WT) virus in NIH 3T3 fibroblasts, primary embryo fibroblasts, and bone marrow macrophages. However, RV7 failed to replicate in target organs of immunocompetent BALB/c mice and severe combined immunodeficient mice, which are exquisitely sensitive to MCMV infection. This defect in vivo growth may be related to the observation that RV7 grew poorly in the peritoneal macrophage cell line IC-21, which is highly permissive for growth of WT MCMV. Two other mutant viruses with an insertion or smaller deletion in the region common to the RV7 deletion grew comparably to WT virus in the macrophage cell line and replicated in salivary gland tissue. The poor growth of RV7 in IC-21 cells was due to a block in immediate-early gene expression, as levels of RNA from immediate-early gene IE1 were reduced eightfold compared with levels for WT virus in macrophages infected with RV7. Consequently, levels of RNA from early and late genes were also reduced. The lower expression of IE1 in RV7-infected IC-21 macrophages was not due to defective entry of virus into the cells, as equal amounts of viral DNA were present in cells 3 h after infection with RV7 or WT MCMV. These studies demonstrate that deletion of sequences in HindIII-J and -I confer altered cell and tissue tropism.     

Virus progeny of murine cytomegalovirus bacterial artificial chromosome pSM3fr show reduced growth in salivary Glands due to a fixed mutation of MCK-2

Murine cytomegalovirus (MCMV) Smith strain has been cloned as a bacterial artificial chromosome (BAC) named pSM3fr and used for analysis of virus gene functions in vitro and in vivo. When sequencing the complete BAC genome, we identified a frameshift mutation within the open reading frame (ORF) encoding MCMV chemokine homologue MCK-2. This mutation would result in a truncated MCK-2 protein. When mice were infected with pSM3fr-derived virus, we observed reduced virus production in salivary glands, which could be reverted by repair of the frameshift mutation. When looking for the source of the mutation, we consistently found that virus stocks of cell culture-passaged MCMV Smith strain are mixtures of viruses with or without the MCK-2 mutation. We conclude that the MCK-2 mutation in the pSM3fr BAC is the result of clonal selection during the BAC cloning procedure.     

A dominant role for the immunoproteasome in CD8+ T cell responses to murine cytomegalovirus

Murine cytomegalovirus (MCMV) is an important animal model of human cytomegalovirus (HCMV), a β-Herpesvirus that infects the majority of the world's population and causes disease in neonates and immunocompromised adults. CD8(+) T cells are a major part of the immune response to MCMV and HCMV. Processing of peptides for presentation to CD8(+) T cells may be critically dependent on the immunoproteasome, expression of which is affected by MCMV. However, the overall importance of the immunoproteasome in the generation of immunodominant peptides from MCMV is not known. We therefore examined the role of the immunoproteasome in stimulation of CD8(+) T cell responses to MCMV - both conventional memory responses and those undergoing long-term expansion or "inflation". We infected LMP7(-/-) and C57BL/6 mice with MCMV or with newly-generated recombinant vaccinia viruses (rVVs) encoding the immunodominant MCMV protein M45 in either full-length or epitope-only minigene form. We analysed CD8(+) T cell responses using intracellular cytokine stain (ICS) and MHC Class I tetramer staining for a panel of MCMV-derived epitopes. We showed a critical role for immunoproteasome in MCMV affecting all epitopes studied. Interestingly we found that memory "inflating" epitopes demonstrate reduced immunoproteasome dependence compared to non-inflating epitopes. M45-specific responses induced by rVVs remain immunoproteasome-dependent. These results help to define a critical restriction point for CD8(+) T cell epitopes in natural cytomegalovirus (CMV) infection and potentially in vaccine strategies against this and other viruses.     

DNA immunization confers protection against murine cytomegalovirus infection.

The murine cytomegalovirus (MCMV) immediate-early gene 1 (IE1) encodes an 89-kDa phosphoprotein (pp89) which plays a key role in protecting BALB/c mice against the lethal effects of the MCMV infection. In this report, we have addressed the question of whether "naked DNA" vaccination with a eukaryotic expression vector (pcDNA-89) that contains the MCMV IE1 gene driven by a strong enhancer/promoter can confer protection. BALB/c mice were immunized intradermally with pcDNA-89 or with the plasmid backbone pcDNAI/Amp (pcDNA) and then challenged 2 weeks later with either a lethal or a sublethal intraperitoneal dose of the K181 strain of MCMV. Variable results were obtained for the individual experiments in which mice received a lethal challenge. In four separate trials, an average of 63% of the mice immunized with pcDNA-89 survived, compared with 18% of the mice immunized with pcDNA. However, in two other trials there was no specific protection. The results of experiments in which mice were injected with a sublethal dose of MCMV were more consistent, and significant decreases in viral titer in the spleen and salivary glands of pcDNA-89-immunized mice were observed, relative to controls. At the time of peak viral replication, titers in the spleens of immunized mice were reduced 18- to >63-fold, while those in the salivary gland were reduced approximately 24- to 48-fold. Although DNA immunization elicited only a low level of seroconversion in these mice, by 7 weeks postimmunization the mice had generated a cytotoxic T-lymphocyte response against pp89. These results suggest that DNA vaccination with selected CMV genes may provide a safe and efficient means of immunizing against CMV disease.     

Natural Killer Cell Dependent Within-Host Competition Arises during Multiple MCMV Infection: Consequences for Viral Transmission and Evolution

It is becoming increasingly clear that many diseases are the result of infection from multiple genetically distinct strains of a pathogen. Such multi-strain infections have the capacity to alter both disease and pathogen dynamics. Infection with multiple strains of human cytomegalovirus (HCMV) is common and has been linked to enhanced disease. Suggestions that disease enhancement in multi-strain infected patients is due to complementation have been supported by trans-complementation studies in mice during co-infection of wild type and gene knockout strains of murine CMV (MCMV). Complementation between naturally circulating strains of CMV has, however, not been assessed. In addition, many models of multi-strain infection predict that co-infecting strains will compete with each other and that this competition may contribute to selective transmission of more virulent pathogen strains. To assess the outcome of multi-strain infection, C57BL/6 mice were infected with up to four naturally circulating strains of MCMV. In this study, profound within-host competition was observed between co-infecting strains of MCMV. This competition was MCMV strain specific and resulted in the complete exclusion of certain strains of MCMV from the salivary glands of multi-strain infected mice. Competition was dependent on Ly49H⁺ natural killer (NK) cells as well as the expression of the ligand for Ly49H, the MCMV encoded product, m157. Strains of MCMV which expressed an m157 gene product capable of ligating Ly49H were outcompeted by strains of MCMV expressing variant m157 genes. Importantly, within-host competition prevented the shedding of the less virulent strains of MCMV, those recognized by Ly49H, into the saliva of multi-strain infected mice. These data demonstrate that NK cells have the strain specific recognition capacity required to meditate within-host competition between strains of MCMV. Furthermore, this within-host competition has the capacity to shape the dynamics of viral shedding and potentially select for the transmission of more virulent virus strains.     

In vitro and in vivo characterization of a murine cytomegalovirus with a transposon insertional mutation at open reading frame M43

We have recently generated a pool of murine cytomegalovirus (MCMV) mutants by using a Tn3-based transposon mutagenesis approach. In this study, one of the MCMV mutants, RvM43, which contained the transposon inserted in open reading frame M43, was characterized. Our results provide the first direct evidence to suggest that M43 is not essential for viral replication in vitro in NIH 3T3 cells. Moreover, RvM43 exhibited a titer similar to that of the wild-type virus in the lungs, livers, spleens, and kidneys of both BALB/c and SCID mice and was as virulent as the wild-type virus in killing SCID mice that had been intraperitoneally infected with the viruses. In contrast, titers of the mutant virus in the salivary glands of the infected animals at 21 days postinfection were significantly (100 to 1,000-fold) lower than those of the wild-type virus and a rescued virus that restored the M43 region and its expression. Thus, M43 appears to be not essential for viral growth in vivo in the lungs, livers, spleens, and kidneys of infected animals and is also dispensable for virulence in killing SCID mice. Moreover, our results suggest that M43 is an MCMV determinant for growth in the salivary glands. Studies of viral genes required for replication in the salivary glands are important in understanding the mechanism of viral tropism for the salivary glands and shedding in saliva, which is believed to be one of the major routes of CMV transmission among healthy human populations.     

Latency, without persistence, of murine cytomegalovirus in the spleen and kidney.

It is not known if murine cytomegalovirus (MCMV) establishes a state of molecular latency independent of low-level persistent infection. The presence of low levels of infectious MCMV distinguishes persistence from molecular latency. Thus, the distinction between persistence and latency has depended on the sensitivity of plaque assays for detecting low levels of infectious virus in tissue of previously infected mice. To determine whether MCMV establishes molecular latency or remains persistent, we developed two assays for detecting low levels of MCMV in tissue. Using prolonged in vitro culture of virus with either mouse embryonic fibroblasts or the murine 3T12 fibroblast cell line, we reproducibly detected a single PFU of MCMV. Inclusion of undiluted sonicated tissue in this assay decreased sensitivity by up to 100-fold. However, sensitivity was improved to 1 PFU of MCMV when sonicated tissue was appropriately diluted. Severe combined immunodeficient (SCID) mice were also used to detect MCMV in sonicated tissue. Infection of SCID mice with a single PFU of MCMV killed two of eight SCID mice, and the 50% lethal dose of MCMV in SCID mice was 2 to 3 PFU. Applying these two methods, we detected infectious virus in 0 of 34 spleens, 1 of 34 kidneys, and 0 of 37 salivary glands from latently infected mice. Spleens and kidneys assessed for persistent virus contained MCMV DNA by PCR and reactivated after 10 to 50 days in explant cultures. Latently infected kidney cells reactivated after adoptive transfer to SCID mice. Quantitation of the MCMV genome by PCR showed that latently infected spleens without detectable infectious MCMV contained about 3,000,000 copies of the MCMV genome. These results demonstrate that MCMV latency in spleen and kidney exists in the absence of low-level persistent infection. Use of assays with defined sensitivity for detection of MCMV in tissue provides a basis for evaluation of cytomegalovirus gene expression in the spleen and kidney during molecular latency.     

Knockout of the Host Resistance Gene Pkr Fully Restores Replication of Murine Cytomegalovirus m142 and m143 Mutants In Vivo

Murine cytomegalovirus (MCMV) proteins m142 and m143 are essential for viral replication. They bind double-stranded RNA and prevent protein kinase R-induced protein synthesis shutoff. Whether the two viral proteins have additional functions such as their homologs in human cytomegalovirus do remained unknown. We show that MCMV m142 and m143 knockout mutants attain organ titers equivalent to those attained by wild-type MCMV in Pkr knockout mice, suggesting that these viral proteins do not encode additional PKR-independent functions relevant for pathogenesis in vivo.     

Natural killer cells regulate murine cytomegalovirus-induced sialadenitis and salivary gland disease

The transmission of herpesviruses depends on viral shedding at mucosal surfaces. The salivary gland represents a major site of persistent viral replication for many viruses, including cytomegalovirus. We established a mouse model of salivary gland dysfunction after acute viral infection and investigated the cellular requirements for the loss of secretion. Murine cytomegalovirus (MCMV) infection severely impaired saliva secretion independently of salivary gland virus levels. Lymphocytes or circulating monocytes/macrophages were not required for secretory dysfunction. Dysfunction occurred before glandular inflammation, suggesting that a soluble mediator initiated the disruption of acinar cell function. Despite genetic differences in innate resistance to MCMV, NK cells protected the host against acinar atrophy and the loss of secretions under conditions of an exceedingly low virus inoculum. NK cells also modulated the type of glandular inflammation after infection, as they prevented an influx of Siglec-F(+) polymorphonuclear leukocytes (PMNs). Therefore, beyond their recognized role in controlling MCMV replication, NK cells preserve organ integrity and function and regulate the innate inflammatory response within the gland.     

Detection of murine cytomegalovirus DNA in circulating leukocytes harvested during acute infection of mice.

We used virus assay and in situ hybridization with a cloned fragment of the murine cytomegalovirus (MCMV) genome to study MCMV infection of circulating leukocytes harvested from 3-week-old BALB/c, C57BL/6, and C3H mice infected with MCMV intraperitoneally. Infectious virus or MCMV DNA was detected in leukocytes on days 1 through 21 of infection in BALB/c mice and on days 3 through 7 in C57BL/6 mice. On days 5 and 7, MCMV DNA or infectious virus was detected in the leukocytes of 17 (94%) of 18 BALB/c mice and 10 (59%) of 17 C57BL/6 mice. In both strains infection peaked on days 5 and 7, when as many as 0.01 to 0.1% of the circulating leukocytes contained MCMV DNA. In C3H mice, however, infectious virus was rarely recovered from leukocyte fractions and MCMV DNA was detected in the circulating leukocytes of only one animal. Circulating leukocytes may have an important role in the dissemination of CMV infections in susceptible hosts.     

Murine cytomegalovirus-Pseudomonas synergistic infections: comparison of virulent and attenuated virus

The synergistic interactions between Pseudomonas aeruginosa and virulent or attenuated murine cytomegalovirus (MCMV) were compared in vivo. Virulent MCMV challenge at a dose of 5 X 10(5) pfu/mouse intraperitoneally, followed by intranasal superinfection with 5 X 10(6) cfu/mouse of Pseudomonas aeruginosa after 48 h resulted in greater than 80% mortality, apparently owing to a failure of pulmonary clearance mechanisms. Single infections, or the use of attenuated MCMV in synergistic infections, did not result in significant morbidity or mortality. Infection with virulent MCMV in vivo resulted in the rapid spread of virus to the lung, liver, and spleen, followed later by spread to the salivary glands. Attenuated virus was detected in salivary glands only. Virulent MCMV was more effective in adsorbing to, or infecting, spleen cells in vitro than attenuated virus. Viral neutralization experiments using anti-viral serum, rabbit complement, and anti-mouse IgG confirmed the presence of a nonneutralizing antibody on the surface of the virulent virus. Our results suggest that the presence of the nonneutralizing antibody on virulent MCMV allows the virus to preferentially infect, or adsorb to, Fc+ cells in the peritoneum. These cells may then carry the virus, via the lymphatic circulation, to other areas of the body, resulting in the replication of virus in multiple organs. Virus replication in the lung may, in part, be the cause of the observed suppression of pulmonary clearance.     

Development of a vaccine against murine cytomegalovirus (MCMV), consisting of plasmid DNA and formalin-inactivated MCMV, that provides long-term, complete protection against viral replication

We previously demonstrated that immunization of mice with plasmid DNAs (pDNAs) expressing the murine cytomegalovirus (MCMV) genes IE1-pp89 and M84 provided synergistic protection against sublethal viral challenge, while immunization with plasmids expressing putative virion proteins provided no or inconsistent protection. In this report, we sought to augment protection by increasing the breadth of the immune response. We identified another MCMV gene (m04 encoding gp34) that provided strong and consistent protection against viral replication in the spleen. We also found that immunization with a DNA pool containing 10 MCMV genes that individually were nonprotective elicited reproducible protection against low to intermediate doses of challenge virus. Moreover, inclusion of these plasmids into a mixture with gp34, pp89, and M84 DNAs provided even greater protection than did coimmunization with pp89 and M84. The highest level of protection was achieved by immunization of mice with the pool of 13 pDNAs, followed by formalin-inactivated MCMV (FI-MCMV). Immunization with FI-MCMV elicited neutralizing antibodies against salivary gland-derived MCMV, and of greatest importance, mice immunized with both the combined pDNA pool and FI-MCMV had undetectable levels of virus in the spleen and salivary glands after challenge. Intracellular cytokine staining of splenocytes from pDNA- and FI-MCMV-immunized mice showed that pDNA immunization elicited high levels of pp89- and M83-specific CD8(+) T cells, whereas both pDNA and FI-MCMV immunizations generated strong CD8(+)-T-cell responses against virion-associated antigens. Taken together, these results show that immunization with pDNA and inactivated virus provides strong antibody and cell-mediated immunity against CMV infection.     

Pathogenesis of murine cytomegalovirus infection

Infection of mice with murine cytomegalovirus (MCMV) is an established model for studying human cytomegalovirus (HCMV) infection. Similarly to HCMV infection, pathological changes and disease manifestations during MCMV infection are mainly dependent on the immune status of the mouse host. This review focuses mainly on the pathogenesis of MCMV infection in immunocompetent and immunodeficient and/or immature mice and discusses the principles of immunosurveillance of infection and the mechanisms by which this virus evades immune control.     

Laboratory strains of murine cytomegalovirus are genetically similar to but phenotypically distinct from wild strains of virus

Murine cytomegalovirus (MCMV) is widely used to model human cytomegalovirus (HCMV) infection. However, it is known that serially passaged laboratory strains of HCMV differ significantly from recently isolated clinical strains of HCMV. It is therefore axiomatic that clinical models of HCMV using serially passaged strains of MCMV may not be able to fully represent the complexities of the system they are attempting to model and may not fully represent the complex biology of MCMV. To determine whether genotypic and phenotypic differences also exist between laboratory strains of MCMV and wild derived strains of MCMV, we sequenced the genomes of three low-passage strains of MCMV, plus the laboratory strain, K181. We coupled this genetic characterization to their phenotypic characteristics. In contrast to what is seen with HCMV (and rhesus CMV), there were no major genomic rearrangements in the MCMV genomes. In addition, the genome size was remarkably conserved between MCMV strains with no major insertions or deletions. There was, however, significant sequence variation between strains of MCMV, particularly at the genomic termini. These more subtle genetic differences led to considerable differences in in vivo replication with some strains of MCMV, such as WP15B, replicating preferentially in otherwise-MCMV-resistant C57BL/6 mice. CBA mice were no more resistant to MCMV than C57BL/6 mice and for some MCMV strains appeared to control infection less well than C57BL/6 mice. It is apparent that the previously described host resistance patterns of inbred mice and MCMV are not consistently applicable for all MCMV strains.     

Molecular cloning and physical mapping of murine cytomegalovirus DNA.

Murine cytomegalovirus (MCMV) Smith strain DNA is cleaved by restriction endonuclease HindIII into 16 fragments, ranging in size from 0.64 to 22.25 megadaltons. Of the 16 HindIII fragments, 15 were cloned in plasmid pACYC177 in Escherichia coli HB101 (recA). The recombinant plasmid clones were characterized by cleavage with the enzymes XbaI and EcoRI. In addition, fragments generated by double digestion of cloned fragments with HindIII and XbaI were inserted into the plasmid vector pACYC184. The results obtained after hybridization of 32P-labeled cloned fragments to Southern blots of MCMV DNA cleaved with HindIII, XbaI, EcoRI, BamHI, ApaI, ClaI, EcoRV, or KpnI allowed us to construct complete physical maps of the viral DNA for the restriction endonucleases HindIII, XbaI, and EcoRI. On the basis of the cloning and mapping experiments, it was calculated that the MCMV genome spans about 235 kilobase pairs, corresponding to a molecular weight of 155,000,000. All fragments were found to be present in equimolar concentrations, and no cross-hybridization between any of the fragments was seen. We conclude that the MCMV DNA molecule consists of a long unique sequence without large terminal or internal repeat regions. Thus, the structural organization of the MCMV genome is fundamentally different from that of the human cytomegalovirus or herpes simplex virus genome.     

Nasal peptide vaccination elicits CD8 responses and reduces viral burden after challenge with virulent murine cytomegalovirus

Infection of BALB/c mice with murine cytomegalovirus (MCMV) leads to CD8 cell responses to an immunodominant epitope YPHFMPTNL. We presented this epitope as a nasal peptide vaccine in combination with cholera toxin adjuvant, and evaluated immune responses and protection from MCMV challenge. Vaccination of naïve mice generated elevated numbers of peptide‐specific interferon‐7‐secreting splenocytes (median 80/million, range 60 to 490), compared to control mice (median 2/million, range —4.5 to 8; P=0.008, Mann‐Whitney test). Twelve days after challenge with virulent MCMV, vaccinated mice had a 1.1 log₁₀ reduction in salivary gland viral titer compared to unvaccinated controls (5.36±0.24 vs. 6.42±0.12, mean±SD log₁₀ plaque‐forming‐units; P<0.001, t‐test). Mice with chronic MCMV infection had consistent responses to the peptide (183±24/million interferon‐γ‐secreting splenocytes). Nasal peptide vaccination during chronic infection boosted peptide‐specific responses in two of four mice to >900/million interferon‐γ‐secreting splenocytes. Nasal peptide vaccination was immunogenic in naïve and MCMV‐infected mice, and reduced viral burden in naïve mice after virulent MCMV challenge. The nasal route may be useful for peptide presentation by novel human vaccines.     

Pathogenesis of murine cytomegalovirus infection in natural killer cell-depleted mice.

The effect of natural killer (NK) cells on the course of acute and persistent murine cytomegalovirus (MCMV) infection was examined by selectively depleting NK cell activity by inoculation of mice with antibody to asialo GM1, a neutral glycosphingolipid present at high concentrations on NK cells. The dose of MCMV required to cause 50% mortality or morbidity in control C57BL/6 mice dropped 4- and greater than 11-fold, respectively, in mice first treated with anti-asialo GM1. NK cell-depleted mice had higher (up to 1,000-fold) virus titers in their lungs, spleens, and livers at days 3, 5, 7, and 9 postinfection. Spleens and livers of control mice were virus-free by day 7 postinfection, and their lungs showed no signs of active infection at any time. In contrast, MCMV had disseminated to the lungs of NK cell-depleted mice by day 5, and these mice still had moderate levels of virus in their lungs, spleens, and livers at day 9. Markedly severe pathological changes were noted in the livers and spleens of NK cell-depleted, MCMV-infected mice. These included ballooning degeneration of hepatocytes and spleen necrosis. MCMV-infected, NK cell-depleted mice had severe spleen leukopenia, and their spleen leukocytes exhibited a significantly lower (up to 13-fold) response to the T cell mitogen concanavalin A when compared with those of uninfected and MCMV-infected controls. It appeared that NK cells exerted their most potent antiviral effect early in the infection, in a pattern correlating with interferon production and NK cell activation; treatment with anti-asialo GM1 later in infection had no effect on virus titers. The relative effect of NK cell depletion on MCMV pathogenesis depended on the injection route of the virus. NK cell depletion greatly augmented MCMV synthesis and pathogenesis in mice inoculated either intravenously or intraperitoneally but had no effect on the course of disease after intranasal inoculation, at any time point examined. One month after intraperitoneal inoculation of virus, NK cell depletion resulted in a six- to eightfold increase in salivary gland virus titers in persistently infected mice, suggesting that NK cells may be important in controlling virus synthesis in the salivary gland during persistent infection. This treatment did not, however, induce dissemination of virus to other organs. These data support the hypothesis that NK cells limit the severity, extent, and duration of acute MCMV infection and that they may also be involved in regulating the persistent infection.