|Home | About | Journals | Submit | Contact Us | Français|
Murine norovirus (MNV) is a highly infectious but generally nonpathogenic agent that is commonly found in research mouse colonies in both North America and Europe. In the present study, the effects of acute and chronic infections with MNV on immune responses and recovery from concurrent Friend virus (FV) infections were investigated. No significant differences in T-cell or NK-cell responses, FV-neutralizing antibody responses, or long-term recovery from FV infection were observed. We conclude that concurrent MNV infections had no major impacts on FV infections.
Murine noroviruses (MNV) are small single-stranded RNA viruses related to human noroviruses, which are the major cause of nonbacterial gastroenteritis in humans. Noroviruses are highly contagious and spread through the fecal-oral route. Although MNV generally does not cause overt clinical signs in immunocompetent mice, its high prevalence of 32% in North American and European research colonies (15, 23) raises the concern that coinfection with MNV might significantly impact studies being done on other pathogens. In this regard, MNV was recently shown to affect disease progression in bacterium-induced inflammatory bowel disease (19). However, another recent study showed no major impact of MNV on adaptive immunity to coinfection with either vaccinia virus or influenza A virus (14).
Several MNV strains have been described, some of which persist indefinitely and could potentially cause long-term consequences. It has previously been shown that either acute or chronic infection of mice with lactate dehydrogenase-elevating virus suppresses immune responses to coinfection with Friend virus (FV), a mouse retrovirus used by several research groups to study host-virus interactions (20, 25). Thus, there was a precedent for concern over the presence of intercurrent infections in FV studies. MNV replicates predominantly in gut enterocytes but also in antigen-presenting cells (APCs) such as macrophages and dendritic cells (29, 31). MNV infection of APCs could produce downstream effects on FV-specific immune responses that could significantly alter recovery.
FV is a retroviral complex consisting of replication-competent Friend murine leukemia virus and a pathogenic but replication-defective virus known as spleen focus-forming virus. Spleen focus-forming virus encodes a defective env protein (gp55) that binds to erythropoietin receptors on erythroid progenitors, causing them to proliferate, resulting in high numbers of erythroid blasts in hematopoietic tissues such as the spleen. Unless controlled by immune responses, FV infection leads to lethal erythroleukemia in most strains of mice and produces long-term, low-level chronic infections in mice that recover from acute infection (13). Recovery from acute FV infection is highly dependent on both T-cell (11, 24) and virus-neutralizing antibody responses (3, 12, 21), and the quality of these responses is dependent on host genes such as major histocompatibility complex genes (2). To determine whether MNV coinfection affects FV-specific immune responses and ultimately recovery from FV infection, we chose medium-recovery (B10.A × A.BY)F1 mice bearing one susceptible major histocompatibility complex haplotype (H-2a) and one resistant haplotype (H-2b) (2). These mice have detectable, but not optimal, immune responses, thus providing a sensitive model to detect either increased or decreased immune responses due to MNV coinfection.
Acute FV infection induces splenomegaly accompanied by significant alterations in cell subset proportions in the spleen (25), so it was important to determine whether MNV infection alone causes disruptions that might confound data from MNV-FV coinfections. Mice were infected orally with 30 μl of MNV-CR6 p5 containing 3 × 107 PFU. This strain of MNV persists long term (14, 28). MNV shedding was detected by reverse transcription-PCR as previously described (5), except that the starting material was a fresh pellet of mouse feces dissolved by shaking in 0.5 ml phosphate-buffered saline, and RNA was purified with the MagMAX-96 kit from Applied Biosystems. The open reading frame encoding helicase/protease/polymerase was targeted with the forward primer sequence (5′ to 3′) AATCTATGCGCCTGGTTTGC, the reverse primer sequence CTCATCACCCGGGCTGTT, and the probe sequence CACGCCACTCCGC.
Of 20 mice tested for chronic virus shedding, 19 tested positive at 4 to 6 weeks postinfection (wpi). There were no false positives in any of the 13 uninfected controls (data not shown). All mice in the experiment were housed in microisolator cages in the same room, and aseptic husbandry techniques were used to prevent cross contamination. No spread of MNV to mice in uninfected cages was observed over the course of the experiments.
Spleen cells were examined by flow cytometry during both the acute (1 wpi) and chronic (8 wpi) phases of MNV infection as previously described (25). At 1 wpi, there was a slight but statistically significant increase in the percentage of CD11b+ cells (predominantly monocytes and neutrophils) in the spleen, but the difference was no longer significant at 8 wpi (Fig. (Fig.1A).1A). No other significant changes in splenic cell subset proportions were observed in response to MNV infection (Fig. 1A and B).
Splenic CD4+ and CD8+ T cells were analyzed for expression of the activation-induced isoform of CD43. CD43 is a cell surface molecule expressed on effector T cells but not on naïve or memory T cells (10). Interestingly, T cells examined at 1, 2, and 8 wpi showed no activation by MNV alone (Fig. 2A and D). For coinfection studies, the mice were also infected with approximately 2,000 spleen focus-forming units of B-tropic FV complex as described previously (25). FV infections were done either coincidently with MNV (acute MNV) or at 4 wpi with MNV (chronic MNV). The presence of either acute or chronic MNV infection did not significantly alter FV-induced activation of CD4+ T cells (Fig. (Fig.2A)2A) or CD8+ T cells (Fig. (Fig.2D).2D). Two functions important for the control of FV infections were examined, gamma interferon (IFN-γ) production by CD4+ T cells (18) and granzyme B production by CD8+ T cells (32, 34). There was no significant effect of MNV coinfection on the number of CD4+ T cells producing IFN-γ (Fig. (Fig.2B)2B) or the number of CD8+ T cells producing granzyme B (Fig. (Fig.2E).2E). CD8+ T cells were also analyzed with tetramers specific for an immunodominant FV epitope (1, 27). By 1 wpi, tetramer-positive CD8+ T cells expanded slightly in all groups infected with FV (P < 0.05) and reached levels of around five million per spleen by 2 wpi (Fig. (Fig.2F).2F). The levels of expansion were variable from mouse to mouse, but no significant differences between FV-infected groups at any time point were observed.
A hallmark of FV infection is the induction of CD4+ regulatory T cells (Tregs) (6, 17) with significant expansion at 2 wpi (22, 33). CD4+ FoxP3+ Tregs were analyzed at 1, 2, and 8 wpi, with expansion most evident at 2 wpi (Fig. (Fig.2C).2C). Unlike FV infection, MNV infection by itself did not induce expansion of Tregs in the spleen. Again, MNV coinfections did not significantly alter the numbers of Tregs between groups of mice infected with FV. Natural killer (NK) cells also play a protective role in FV infections (16), and a significant proportion of splenic NK cells (NK1.1+) were activated in response to FV infection at all three time points (Fig. (Fig.2G).2G). In contrast, there was no evidence of NK-cell activation in response to MNV only, and concurrent MNV infection did not significantly affect the NK response to FV infection at any of the time points.
Acute FV infection of (B10.A × A.BY)F1 mice produces high levels of plasma viremia that are controlled by virus-neutralizing antibody responses (7, 26). At 1 wpi, all mice had very high viremia levels (Fig. (Fig.3A)3A) with no detectable virus-neutralizing antibody responses (Fig. (Fig.3B).3B). By 2 wpi, virus-neutralizing antibody responses were detectable and viremia was cleared in almost all of the mice. There were no significant differences between the groups, indicating that MNV coinfection did not alter either the antibody responses or viremia levels.
Levels of cellular infection in the spleen were analyzed by flow cytometry to detect expression of the FV-encoded glycosylated gag antigen (8, 9) and to measure the proliferation of erythroid precursors in the spleen. At 1 wpi, there was a high level of both viral antigen expression (monoclonal antibody 34+) and erythroid cell (Ter119+) proliferation in the spleen (Fig. 4A and B). By 2 wpi, FV infection was in decline. Although mice concurrently infected with MNV and FV appeared to have lower levels of FV antigen expression at 2 wpi, the difference was not statistically significant due to the high degree of mouse-to-mouse variability at this time point, when the antiviral immune response is becoming effective (Fig. (Fig.4A).4A). There were also no significant differences associated with MNV coinfection in levels of erythroid cell proliferation or in levels of spleen cells producing live virus, as measured by infectious-center assays (Fig. (Fig.4C).4C). All of the mice in all of the groups that recovered from acute FV infection survived the infection for 8 weeks and had equivalent levels of chronic FV infection at 8 wpi (Fig. (Fig.4C).4C). Thus, there was no indication that MNV affected FV infection levels or recovery from infection.
It has been shown that prior infections with a number of viruses can skew the T-cell repertoire and thereby alter responses to subsequent infections in either beneficial or harmful ways (30). No evidence of such changes was observed in this study, but detailed analyses of the repertoire are required to rule out such effects. However, we can conclude that neither acute nor chronic MNV infection impacted FV-induced immune responses or recovery from FV infection in any major way. This conclusion is basically the same as that reached in studies on the effects of MNV on influenza and vaccinia virus infections (14) but different from that reached in studies on bacterium-induced inflammatory bowel disease, where significant effects were observed (19). A very preliminary conclusion from these results is that the effects of concurrent norovirus infection might be more likely to affect infections of the gastrointestinal tract than of other locations. In the present experiments, little evidence of a systemic immune response to MNV was observed in the spleen, suggesting that the response might be localized to gut-associated lymphatic tissues. It is also possible that the difference between bacterial and viral infections is an important one. It remains important to investigate each type of infection being studied and also to keep in mind that there are numerous strains of MNV (28), some of which might cause different effects. Thus, care must be taken not to generalize to other coinfecting viruses or to other norovirus strains.
We are very grateful to Herbert W. Virgin for sharing MNV stocks with us and providing important advice and criticism.
This work was supported by the Division of Intramural Research of the NIAID, NIH.
Published ahead of print on 7 October 2009.