Suppression of Acute Anti-Friend Virus CD8⁺ T-Cell Responses by Coinfection with Lactate Dehydrogenase-Elevating Virus

Abstract

Friend virus (FV) and lactate dehydrogenase-elevating virus (LDV) are endemic mouse viruses that can cause long-term chronic infections in mice. We found that numerous mouse-passaged FV isolates also contained LDV and that coinfection with LDV delayed FV-specific CD8⁺ T-cell responses during acute infection. While LDV did not alter the type of acute pathology induced by FV, which was severe splenomegaly caused by erythroproliferation, the immunosuppression mediated by LDV increased both the severity and the duration of FV infection. Compared to mice infected with FV alone, those coinfected with both FV and LDV had delayed CD8⁺ T-cell responses, as measured by FV-specific tetramers. This delayed response accounted for the prolonged and exacerbated acute phase of FV infection. Suppression of FV-specific CD8⁺ T-cell responses occurred not only in mice infected concomitantly with LDV but also in mice chronically infected with LDV 8 weeks prior to infection with FV. The LDV-induced suppression was not mediated by T regulatory cells, and no inhibition of the CD4⁺ T-cell or antibody responses was observed. Considering that most human adults are carriers of chronically infectious viruses at the time of new virus insults and that coinfections with viruses such as human immunodeficiency virus and hepatitis C virus are currently epidemic, it is of great interest to determine how infection with one virus may impact host responses to a second infection. Coinfection of mice with LDV and FV provides a well-defined, natural host model for such studies.

In 1957, Charlotte Friend described a filterable agent isolated from leukemic mice that transmitted the induction of malignant tumors following passage to healthy mice (22). The filterable agent was later defined as a viral complex containing a replication-competent retrovirus named Friend murine leukemia virus (F-MuLV) and a replication-defective retrovirus named spleen focus-forming virus (SFFV) (reviewed in reference 31). The structural genes of SFFV are defective, and SFFV cannot spread in the absence of a helper virus such as F-MuLV. However, SFFV is required to induce leukemia in adult mice due to two important factors. First, SFFV encodes a truncated Env glycoprotein (gp55) that binds to erythropoietin receptors (EpoR) on nucleated erythroid precursors and induces a proliferative signal (36). Second, the full transformation of cells to the leukemic state in Friend virus (FV)-infected mice is associated with SFFV provirus integration into both the Spi-1 proto-oncogene (44) and the p53 tumor suppressor gene (46). For type C retroviruses such as FV, integration into the host genome requires actively dividing cells. Thus, gp55 binding to EpoR not only increases the number of target cells for virus entry but also renders them susceptible to provirus integration.

Mouse-passaged retrovirus stocks are typically swarms containing numerous variants, with a range of pathogenic capabilities. Virus stocks of F-MuLV and SFFV clones have been obtained by in vitro culture and have been used to successfully infect mice (32, 39, 40). However, in our experience, such tissue culture-derived virus stocks have been less pathogenic than mouse-passaged virus stocks (unpublished results). Cloned viruses may contain some but not all of the properties of the swarm. For example, a Friend virus clone variant, FIS-2, induces immunosuppression but has weak leukemogenicity (14). Low pathogenicity can also be due to low titers of the SFFV component, which is positively selected in vivo but not in vitro. Consequently, many studies requiring highly pathogenic virus complexes have been done with mouse-passaged stocks containing not only high titers of the pathogenic SFFV component but also naturally arising virus variants. The use of such natural virus swarms has been important in vaccine studies because the viruses present a much stronger challenge to the immune system than the cloned virus stocks, not only in terms of pathogenicity but also in their antigenic complexity. Likewise, genetic studies of host resistance have been greatly facilitated by the use of mouse-passaged virus stocks with sufficient pathogenicity to display phenotypic variability in different mouse strains (8).

Although in vivo passage of FV stocks offers distinct advantages for certain types of studies, there are also inherent disadvantages. For example, an unintended consequence of in vivo passage can be the introduction or propagation of the viruses present in the mice used for passage. Current experiments have shown that some FV stocks passaged in mice for more than 3 decades contain lactate dehydrogenase-elevating virus (LDV). Evidence suggested that LDV was present in FV stocks as early as 1963 (56), and this was confirmed in 1969 (61). Thus, LDV is a long-standing component of the FV complex. This information was overlooked in recent decades, and FV studies have not addressed the effects of LDV on FV replication and pathogenicity in mice.

LDV is an interesting and unusual virus that is endemic in wild-mouse populations. It is an enveloped, positive-stranded RNA virus classified in the order Nidovirales, which also contains coronaviruses, the cause of a recent outbreak of severe acute respiratory syndrome infection in humans in 2003 (17, 34, 52). Although LDV has been eliminated from most experimental mouse colonies, it is still found in some transplantable tumor cell lines and virus stocks (7, 47). Its name derives from its capacity to rapidly infect and cytolyze a small subset of macrophages responsible for scavenging lactate dehydrogenase from the circulation (53). LDV is highly restricted to this macrophage subset. LDV has been described as an ideal persistent virus because of its ability to establish life-long viremic infections in mice regardless of strain, age, sex, or immune status, with no overt clinical signs (54). Virus replication is extremely rapid, and plasma titers reach 10⁹ to 10¹⁰ 50% infective doses (ID₅₀) per milliliter within 12 to 16 h of infection. Due to the loss of permissive target cells for infection, LDV titers gradually decrease until they reach a steady-state level of 10⁴ to 10⁶ ID₅₀ per milliliter, depending on the mouse strain and the rapidity of target cell replenishment.

Interestingly, the immune response has little detectable effect on LDV infections. Antibodies against LDV are generated during the course of infection but appear to be relatively ineffective at neutralizing virus in vivo. LDV circulates as an infectious virus-immunoglobulin G (IgG) complex during chronic infection (4). Neutralizing epitopes appear to be masked by sugars (65), similar to that seen with human immunodeficiency virus (HIV) infection (3). Furthermore, the passive transfer of polyclonal or monoclonal antibodies that neutralize LDV in vitro has little effect on viremia in vivo (24). CD8⁺ T-cell responses are elicited by LDV infection but are ineffective at decreasing virus loads, likely because of the rapid cytolytic infection caused by this virus (63). Thus, specific immune responses are raised against LDV but are unable to control viremia. There is currently no effective vaccine against LDV.

Although immune responses have little apparent effect on LDV titers, LDV induces dramatic effects on the immune system. Among the reported effects are impaired antigen presentation by macrophages (30), polyclonal B-cell activation (12, 13), NK cell activation (42), impaired delayed-type hypersensitivity responses (29), and inhibited cellular immune responses (28). LDV also induces a potent alpha interferon (IFN-α) response within the first day after infection (53). Thus, the presence of LDV in FV stocks could have major impacts on anti-FV immune responses and pathogenesis. In the current study, mice were infected with LDV, FV, and a combination of both viruses (FV/LDV), to study the acute effects of infection on the immune system and on FV replication. Results indicated that the presence of LDV delayed FV-specific T-cell responses by approximately 1 week, thereby extending the period of peak FV replication and increasing pathogenicity.

MATERIALS AND METHODS

Mice.

Experiments were conducted using 12- to 24-week-old female (C57BL/10 × A.BY) F₁ mice bred at the Rocky Mountain Laboratories (RML). The relevant FV resistance genotypes of these mice are H-2^b/b, FV1^b/b, FV2^r/s, and Rfv3^r/s. FV-specific T-cell receptor (TCR) transgenic (Tg) mice (48), which carry a TCR transgene that recognizes the gag leader peptide of FV (6), were also bred at the RML. BALB/c mice were purchased from Harlan Labs (Indianapolis, IN). Mice were treated in accordance with the regulations and guidelines of the Animal Care and Use Committee of the RML and the National Institutes of Health.

Detection of LDV in mouse-passaged FV stocks.

The presence of LDV in mouse-passaged FV stocks maintained at the RML was detected both by lactate dehydrogenase (LDH) assays with serum from infected mice and by TaqMan reverse transcription-PCR with RNA purified directly from FV stocks (see below). All RML mouse-passaged FV stocks, including B-tropic, N-tropic, and the original 1973 NB-tropic stock obtained from the Lilly laboratory (38), tested positive for LDV. All virus stocks maintained in tissue culture were negative for LDV, consistent with the inability of LDV to grow in tissue culture cell lines (reviewed in reference 54).

Generation of LDV-free FV stocks.

The inability of LDV to grow in tissue culture cell lines was employed to obtain a new FV stock free of LDV. Fischer rat embryo cells were infected with the LDV/FV stocks, and the cells were cultured at 37°C in 5% CO₂ for 9 days, with five changes of medium (RPMI medium containing 100 U/ml penicillin, 100 μg/ml streptomycin, 5 × 10⁻⁵ M 2-mercaptoethanol [Sigma-Aldrich, St. Louis, MO], and 10% fetal calf serum [Nova-Tech, Inc., Grand Island, NY]). Culture supernatants were collected on day 9 and used to infect BALB/c mice. Four days after infection, the mice were bled, and sera were shown to be negative for elevated LDH levels. When the spleens were removed on day 14, all were seen to be enlarged with typical-appearing FV-induced splenomegaly, and all tested positive for F-MuLV by infectious center assays. An undiluted sample of the new FV stock was amplified by reverse transcription-PCR (RT-PCR) using LDV-specific primers and probes and yielded no detectable amplification product. In contrast, a 10⁻³ dilution of the same stock tested positive for both the F-MuLV env-specific and the gag-specific primers and probes. The new FV stock also tested positive for F-MuLV by infectious center assays and positive for SFFV by spleen focus assay as described previously (38). To increase the titer of SFFV, the new FV stock was passaged again in BALB/c mice.

Generation of FV-free LDV stock.

To obtain an LDV stock, plasma was taken from mice 14 h after they were coinfected with FV/LDV. At that time point, high levels of LDV virions were released from infected cells, but F-MuLV replication had not been completed. Infection of BALB/c mice with LDV-containing plasma did not induce splenomegaly, a highly sensitive bioassay for the presence of F-MuLV, but the mice tested positive for LDV both by reverse transcription-PCR and by the elevation of serum LDH. The undiluted plasma was amplified by RT-PCR using F-MuLV-specific primers and probes and yielded no detectable amplification product. In contrast, a 10⁻⁵ dilution of the same plasma tested positive for LDV, using LDV-specific primers and probe.

Infections.

Mice were infected by intravenous injection of 0.5 ml of phosphate-buffered balanced salt solution containing 2% fetal bovine serum and 20,000 spleen focus-forming units of FV, alone or in combination with 0.2 ml of a 1-to-100 dilution of sera containing LDV.

Infectious center assay.

An infectious center assay (57) was used to determine the number of spleen cells from infected animals that were producing infectious virus. Serial 10-fold dilutions of spleen cells were plated onto susceptible Mus dunni cells and cocultured for 3 days. Cells were then fixed with ethanol and incubated sequentially with F-MuLV envelope-specific monoclonal antibody (MAb) 720 (57) and peroxidase-conjugated goat anti-mouse IgG (Cappel, West Chester, PA). Foci of infections were identified following development with aminoethylcarbazole substrate.

Virus-neutralizing antibody assay.

To test for virus-neutralizing antibodies, heat-inactivated (56°C; 10 min) plasma samples at titrated dilutions were incubated with an aliquot of F-MuLV stock in the presence of guinea pig complement at 37°C, as previously described (45). The samples were then analyzed by focal infectivity assays, as described above. The titer was defined as the plasma dilution at which greater than 75% of the input virus was neutralized.

LDH assay.

Blood samples were collected by retroorbital bleed and centrifuged (6,000 × g; 10 min at 4°C) to separate the plasma from cellular components. The freshly collected plasma samples were tested for LDH activity, as previously described (7), with some modifications. Briefly, sodium pyruvate and NADH were prepared fresh by dissolving each at 2.5 mg/ml in LDH buffer (0.1 M sodium phosphate, pH 7.4). The freshly prepared sodium pyruvate (0.8 ml) and NADH (0.8 ml) were mixed well with 19.2 ml of LDH buffer. This solution (200 μl per well) was dispensed into a 96-well UV-transparent microtiter plate, and 2 μl of plasma was added and mixed well. The optical densities of the reaction mixture at 340 nm were determined immediately (time zero) and after 1 min. A decrease in the absorbance from time zero to 1 min indicated the oxidation of NADH by LDH. Mice were considered positive for LDV if the LDH activity was at least fivefold higher than that of the uninfected controls.

TaqMan RT-PCR for detection of LDV and FV.

Viral RNA was extracted from 100 μl of plasma or 200 μl of spleen homogenate, using a MagMAX viral RNA isolation kit (Ambion, Foster City, CA). Approximately 200 ng of RNA was reverse transcribed and amplified using an RNA UltraSense one-step quantitative RT-PCR system (Invitrogen, Carlsbad, CA) with LDV 5′ UTR/ORF-1a-specific primers and probe (sense, CGTGCGGTAACCGTCTATTTC; antisense, ATCCCGACTGCATGGTTATAGGT; probe, CTCCTACTATACCTCCCTCTCTAACATTTCCGGG) and two sets of FV-specific primers and probe (FV envelope sense, AAGTCTCCCCCCGCCTCTA; antisense, AGTGCCTGGTAAGCTCCCTGT; probe, ACTCCCACATTGATTTCCCCGTCC; FV gag sense, GCCACGAGACGGCACTTT; antisense, TCCATGTGGGCCAGATGAG; probe, ACCCAGACATTATTACACAGGTTAAGATCAAG). The amplification primers and probes for LDV were designed from the LDV U15146 sequence from the NCBI database. Amplification was detected using an Applied Biosystems 7900HT sequence detection system (Foster City, CA).

Surface and intracellular staining and flow cytometry.

The expression of cell surface markers on splenocytes was analyzed by using fluorochrome-conjugated antibodies to CD4 (clone RM4-5), CD8 (clone 53-6.7), CD11b (M1/70), CD11c (clone HL3), CD19 (clone 1D3), Ter119 (clone Ter119), and CD43 (clone 1B11) (all from BD Pharmingen, San Jose, CA). Cell surface FV glycosylated Gag (glycogag) was detected with MAb 34 (10) stained with allophycocyanin (APC)-labeled goat anti-mouse IgG2b antiserum (Invitrogen, Carlsbad, CA). Tetramers for FV-specific CD4⁺ and CD8⁺ T cells have been described previously (60).

For the detection of intracellular IFN-γ production, spleen cells were stimulated with either plate-bound CD3 antibody (see Fig. 3) (27) or D^bGagL major histocompatibility class I (MHC-I) tetramer (NIAID Tetramer Core Facility) (see Fig. 4) for 5 h in the presence of 10 μg/ml brefeldin A. The cells were then stained for the surface expression of CD4 and CD8, fixed with 2% formaldehyde, permeabilized with 0.1% saponin in phosphate-buffered saline containing 0.1% sodium azide and 1% fetal calf serum, and incubated with allophycocyanin anti-IFN-γ (clone XMG1.2) (BD Pharmingen). Data were acquired using either a FACSCalibur or an LSRII flow cytometer (BD Biosciences, San Jose, CA) and were analyzed using FlowJo version 8.3 (Tree Star, Inc., Ashland, OR) software.

FIG. 3.

No suppression of antibody or general T-cell responses by LDV at 1 week postinfection. Plasma samples were assayed for FV-neutralizing antibody, and titers show the reciprocal of the dilution that produced 75% neutralization of input virus (A). Activation of CD4⁺ (B) and CD8⁺ T (C) cells at 7 days postinfection was measured by the upregulation of CD43 with spleen cells taken directly ex vivo from mice infected with FV, LDV, or FV/LDV. Compared to that of naïve mice, there was significant activation of both subsets of cells in all types of infections (P < 0.008 for all groups). As measured by intracellular cytokine staining following anti-CD3-stimulation, there was no significant increase in the percentage of CD4⁺ T cells producing IFN-γ in any of the groups (D). Increases in percentages of CD8⁺ T cells producing IFN-γ were significant in both of the LDV-infected groups (P < 0.001) but not the FV-infected group (E). Statistical analyses were carried out using an unpaired t test with Bonferroni correction, where n = 8 mice/group.

FIG. 4.

Abrogated FV-specific CD8⁺ T-cell responses in mice coinfected with LDV. Mice were infected either with FV alone or with LDV. Spleens were removed from naive mice (day 0) and from mice at 7 and 14 days after infection. (A) Spleen cells were analyzed by flow cytometry for the expression of cell surface CD4 and FV-specific MHC-II tetramer. The graph shows the mean percentages ± standard errors of the means (SEM) of tetramer-positive CD4⁺ T cells (n = 4 mice per group). Spleen cells were also stained directly ex vivo for cell surface CD8 and FV-specific MHC-I tetramer. The graphs show the mean percentages (± SEM) (B) and numbers (C) of tetramer-positive CD8⁺ T cells (n = 5 to 8 mice per group). (D) Spleen cells were stimulated with MHC-I tetramers in the presence of brefeldin A for 5 h and stained for the surface expression of CD8 and the intracellular expression of IFN-γ (n = 4 to 6 mice per group). The graph shows the mean percentages ± SEM of CD8⁺ T cells that expressed intracellular IFN-γ. Asterisks (*) indicate that the difference between FV infection and FV/LDV coinfection is statistically significant (P < 0.05) as determined by the Mann-Whitney U test.

In vivo T-cell depletions.

FV-infected mice were depleted of CD8⁺ T cells by intraperitoneal (i.p.) injection of 100 μg of anti-CD8 (clone 169.4) at 0, 2, and 5 days postinfection (11). Control mice received injections of 100 μg of rat immunoglobulin (Sigma). Anti-CD8 treatment achieved greater than 98% depletion of CD8⁺ T cells, as determined by flow cytometry. FV/LDV-infected mice were depleted of CD4⁺ T cells by intraperitoneal injection of 100 μg of anti-CD4 (clone 191.1) at 0 and 3 days postinfection or at day −1 and 0 and at 4 days postinfection. Both anti-CD4 treatments resulted in greater than 99% depletion of CD4⁺ T cells and greater than 96% depletion of regulatory T cells (defined as CD25⁺Foxp3⁺).

In vitro suppression assay.

FV-specific TCR Tg CD8⁺ T cells were purified using anti-CD8α magnetic microbeads and the MidiMACS system as recommended by the manufacturer (Miltenyi Biotech). The CD8-depleted spleen cells from TCR Tg mice served as the antigen-presenting cells (APCs), which were incubated with the D^bGagL peptide (5 μg/ml) (6) in Iscove's modified Dulbecco's medium (IMDM; Cambrex, Wilkerville, MD) containing 10% normal mouse serum for 60 min at 37°C, irradiated (3,000 rad), and washed twice with IMDM. TCR CD8⁺ T cells and peptide-pulsed APCs (2 × 10⁵ each) were cultured per well of a flat-bottomed 96-well plate in IMDM containing 10% fetal bovine serum, 2 mM l-glutamine, 50 μM 2-mercaptoethanol, and 100 U/ml each of penicillin and streptomycin at 37°C, 5% CO₂. Naïve CD4⁺ CD25⁻ (2 × 10⁵) T cells were also included in cocultures for optimal IFN-γ expression by stimulated CD8⁺ T cells. CD4⁺ CD25⁺ (6 × 10⁵) T cells from either naïve or chronically infected mice were added to each well for a 2-to-1 ratio of T regulatory (Treg)-to-CD8⁺ T cells. Coculture supernatants were collected after 48 h and assayed for IFN-γ content by enzyme-linked immunosorbent assay, as described previously (59). The percentages of suppression were determined by comparison of the amounts of IFN-γ produced in stimulated CD8⁺ T-cell cultures containing CD4⁺ CD25⁺ T cells from naive mice.

RESULTS

Exacerbation of FV infection by coinfection with LDV.

To examine whether the presence of LDV affected the course of FV infection, we first determined the kinetics of FV infection in mice inoculated with an LDV-free FV stock compared with that in mice infected with our standard RML stock of mouse-passaged FV that contained LDV. FV infection was determined by flow-cytometric detection of viral glycogag protein on splenocytes. At 7 days postinfection, FV-infected mice had a significantly greater percentage of FV antigen-positive spleen cells than mice coinfected with LDV (Fig. 1A). However, by day 14, the percentage of cells expressing FV antigen was reduced to nearly undetectable levels in FV-infected mice. In contrast, FV levels continued to climb after 7 days postinfection in mice coinfected with LDV. By day 28, FV infections were largely resolved in both groups. FV-induced splenomegaly and infectious center data were consistent with those of the above-described results (data not shown). These results suggested that the presence of LDV exacerbated and prolonged FV infection.

FIG. 1.

Exacerbation of FV infection by coinfection with LDV. (A) Mice were infected with LDV-free FV or with a standard RML virus stock, and spleens were removed from naive mice (day 0) or from mice at day 7, day 14, and day 28 after infection. Spleen cells were analyzed by flow cytometry for cell surface expression of the FV glycogag protein, using MAb 34 (10). The graph indicates the percentage of MAb 34-positive cells in nucleated splenocytes. The data were from four mice at each time point. (B) Mice were infected with FV alone or with FV mixed with LDV, and spleens were analyzed at 7 and 14 days after infection. Spleen cells were stained for cell surface expression of the FV glycogag protein, using MAb 34. The graph shows the mean percentages ± standard errors of the means (SEM) of MAb 34-positive cells in nucleated splenocytes, as determined by flow cytometry. (C) Spleen cells were also assayed for virus-producing cells by an infectious center assay, and the graph shows the mean numbers ± SEMs of virus-producing cells per spleen from mice infected with FV, or with mixed FV and LDV. Data were compiled from six to eight mice per group at each time point. Error bars indicate standard errors. Asterisks indicate that the differences between FV infection and FV/LDV coinfection are statistically significant (P < 0.05) as determined by the Mann-Whitney U test. FV-induced splenomegaly results (data not shown) were consistent with the viral antigen and infectious center results.

Because the spleen focus assay used to determine the titers of the above-mentioned virus stocks does not accurately measure twofold differences, a possible explanation for the higher virus levels observed for FV-infected mice at day 7 is that the FV titer was greater in the LDV-free FV stock than in the RML stock containing both viruses. To ensure that each group received the same FV dose, the kinetic studies were repeated with mice inoculated with LDV-free FV, either alone or with FV/LDV. Levels of FV infection were again determined by flow cytometric analysis of viral antigen on splenocytes and also by infectious center assays. At day 7, the levels of FV infection between the two groups were not significantly different (Fig. 1B and C). Interestingly, despite similar FV titers at day 7, the virus levels in mice infected with LDV-free FV again declined dramatically over the next week, while the FV levels in FV/LDV-coinfected mice continued to increase (Fig. 1B and C). Thus, the ability of mice to more quickly resolve FV infections in the absence of LDV was not due to differences in the FV dose or to the relative infection levels at day 7. The resolution of LDV-free FV infection between 1 and 2 weeks postinfection suggested a role for adaptive immunity. Moreover, the inability of coinfected mice to resolve FV infection during this time suggested that LDV infection caused some inhibition of FV-specific immune responses.

The presence of LDV does not alter the cellular distribution of FV infection at 1 week postinfection.

The major organ infected by FV is the spleen, and previous studies using LDV-containing FV stocks showed that the infected cells were primarily proliferating erythroid progenitor cells. Lymphoid and monocytic cells were also infected by FV, albeit to a much lesser extent (16, 26). Because FV requires cell replication for productive infection, polyclonal activation of immune cells by LDV infection (12, 13, 37) could potentially increase the cells’ susceptibility to FV infection (49) and thereby alter their function (50). FV infection with or without LDV induced strong proliferation of erythroid precursor cells (Ter119⁺), and there were no significant differences in the overall cellular composition of the spleens between the two groups of infected mice (Fig. 2A). Furthermore, there was no evidence that the presence of LDV significantly altered the pattern of FV infection in the cellular subsets analyzed, as determined by the percentage (Fig. 2B) or the number (data not shown) of infected cells within each cell subset at 1 week postinfection. We next sought to determine if LDV altered antiviral immune responses.

FIG. 2.

Cellular distribution of FV infection at 1 week postinfection. Mice were infected with FV or FV/LDV, and spleens were removed at day 7 after infection, and single-cell suspensions were made for immediate analysis. Spleen cells were labeled with MAb 34 in conjunction with antibodies to lineage-specific surface molecules and analyzed by flow cytometry. The graphs show the mean percentages ± standard errors of the means of cells in each lineage (A) of Ter119 (erythroid cells), CD11c (dendritic cells), CD11b (monocytes), CD4 (helper T cells), CD8 (cytotoxic T cells), and CD19 (B cells), and the percentage of cells in each lineage expressing FV glycogag (B). Data were compiled from four mice per group.

FV-specific antibody responses.

Since FV-specific antibodies have previously been shown to be essential for recovery from FV infection (9, 25), antibody titers were measured at 1 and 2 weeks postinfection in FV- and FV/LDV-infected mice. FV-neutralizing antibodies were not detectable at 1 week postinfection in either group (Fig. 3A). By 2 weeks postinfection, all mice in both groups had detectable antibody responses, and there were no significant differences in antibody titers between the two groups. Thus, there was no evidence that the antibody response was suppressed by LDV infection.

General T-cell responses.

Previous studies showed that T-cell responses were important in the recovery from FV infection (58). CD8⁺ T-cell responses were especially important in the early phase of recovery, while CD4⁺ T cells were most important after 2 to 3 weeks postinfection (58). General levels of T-cell activation were assessed by measuring the expression of the activation-induced isoform of CD43 on T cells taken directly ex vivo (23). Also, the ability of T cells to produce IFN-γ was measured by intracellular cytokine staining. Compared to the expression level of CD43 in naive mice, the percentages of both the CD4⁺ T cells (Fig. 3B) and the CD8⁺ T cells (Fig. 3C) expressing CD43 were significantly increased in all infected groups at 1 week postinfection. Interestingly, the combination of FV plus LDV did not produce an additive effect compared to that of either virus alone. Although there was no detectable production of IFN-γ by CD4⁺ T cells from any of the groups (Fig. 3D), there was a significant increase in the percentage of CD8⁺ T cells that produced IFN-γ in mice infected either with LDV alone or with FV/LDV (Fig. 3E). Thus, the overall CD8⁺ T-cell IFN-γ response to LDV was significantly more robust than the response to FV (Fig. 3E). There was no indication of a generalized suppression of T-cell responses in LDV-infected mice.

Suppression of anti-FV immune responses by coincident LDV infection.

Although general T-cell responses in FV/LDV-infected mice were not suppressed, we next examined the effect of LDV coinfection with FV-specific CD4⁺ T and CD8⁺ T-cell responses in mice infected with FV alone or with LDV. Splenocytes were isolated at days 7 and 14 following infection, stained for CD4 and CD8 in conjunction with FV-specific MHC-II and -I tetramers, respectively, and analyzed by flow cytometry. At day 7, there were detectable FV-specific CD4⁺ T-cell responses from both groups, with no significant differences between the groups (Fig. 4A). However, there was a striking difference in the FV-specific CD8⁺ T-cell responses. FV-infected mice had vigorous proliferative responses to the immunodominant FV glycogag epitope at day 7, as measured both by percentages (Fig. 4B) and by absolute numbers (Fig. 4C). Furthermore, the CD8⁺ T cells were functional in terms of their ability to produce IFN-γ (Fig. 4D). In comparison, the FV/LDV-coinfected mice had significantly less accumulation of tetramer-positive CD8⁺ T cells by day 7, although vigorous responses were detectable by day 14 (Fig. 4B and C). Thus, coinfection with LDV caused a delay in detectable FV-specific CD8⁺ T-cell responses.

CD8⁺ T-cell responses are required for early recovery.

It was of interest to determine if the failure of FV/LDV-coinfected mice to develop CD8⁺ T-cell responses at day 7 postinfection was sufficient to account for their inability to control FV infection between 1 and 2 weeks postinfection. If that were the case, then mice infected with FV in the absence of a CD8⁺ T-cell response should also fail to display early virus control. We found that FV-infected mice depleted of CD8⁺ T cells prior to infection followed a course very similar to that of FV/LDV-coinfected mice (Fig. 5). Viral loads increased between the first and second week rather than beginning to resolve as they did in the presence of CD8⁺ T cells. These data indicated that LDV-mediated suppression of anti-FV CD8⁺ T-cell responses during the first week of coinfection likely accounted for the delayed control of FV during the acute phase.

FIG. 5.

Resolution of acute FV infection depends on CD8⁺ T-cell responses. FV-infected mice were depleted of CD8⁺ T cells during the first week of infection, and spleen cells were prepared at day 7 and day 14 postinfection to assay for infectious centers. Depletions were greater than 98% complete. The graphs show the mean numbers ± standard errors of the means of infectious centers per spleen from four mice per group at each time point. The difference between the depleted and nondepleted FV-infected mice was statistically significant (P < 0.0001 by Student's t test).

Virus-induced regulatory T cells.

Previous studies showed that acute (66) and chronic (15) infection of mice with the FV complex containing LDV induced Treg cells that affected CD8⁺ T-cell functions. To determine if LDV induced Treg cells that might correlate with the decreased accumulation of FV-specific CD8⁺ T cells, an in vitro suppression assay was performed with CD4⁺ Treg cells obtained from mice infected with either LDV or FV. Since differences in CD8⁺ T-cell accumulation were already evident at day 7, Treg cell activity was tested at days 3 and 7 postinfection. Indeed, mice infected with LDV had significant Treg-mediated suppression of IFN-γ production by FV-specific CD8⁺ T cells at day 3 (Fig. 6A). These findings are consistent with previous results showing that virus-induced Treg cells inhibited CD8⁺ T cells in vitro regardless of the TCR specificity of the CD8⁺ T cell (59). By day 7, the LDV-associated Treg suppression decreased and FV-associated suppression increased. Thus, the timing of the LDV induction of Treg cells fit with a possible role in slowing the CD8⁺ T-cell response.

FIG. 6.

Regulatory T-cell induction. (A) CD4⁺ CD25⁺ T cells were isolated from the spleens of naive mice and from those of mice infected with LDV or FV, at days 3 and 7 following infection, and were assayed in vitro for their ability to suppress IFN-γ production by CD8⁺ T cells. TCR Tg CD8⁺ T cells were stimulated in vitro with peptide-pulsed APCs and cocultured in the presence of enriched CD4⁺ CD25⁺ T cells. After 48 h, the culture supernatants were collected and analyzed for IFN-γ by enzyme-linked immunosorbent assay. The graphs show the mean percentages ± standard errors of the means (SEM) of suppression of IFN-γ production relative to that of cocultures containing stimulated CD8⁺ T cells with CD4⁺ CD25⁺ T cells from naive mice. The data are compiled from four mice per group. (B) Mice infected with FV alone or with mixed FV/LDV were treated with CD4-depleting antibody during the first week of infection. Spleen cells were isolated at 7 days postinfection, stained for cell surface CD8 and FV-specific MHC-I tetramer, and analyzed by flow cytometry. The graph shows the mean percentages ± SEM of CD8⁺ T cells that are tetramer positive. Data were compiled from four to nine mice per group.

If LDV-induced Treg cells were responsible for delayed CD8⁺ T-cell responses, then the depletion of CD4⁺ T cells would be expected to reverse the delay as long as the CD8⁺ T-cell response was CD4⁺ T-cell independent. To determine first if this premise was true, FV-specific CD8⁺ T-cell responses were measured in mice depleted of CD4⁺ T cells by injection of MAb (11). The depletions were highly effective in reducing levels of CD4⁺ T cells, including levels of the Foxp-3⁺ subset, by greater than 99% by day 7 following antibody injection. At day 7, the CD4-depleted mice had FV-specific CD8⁺ T-cell (tetramer-positive) responses as high as those in the control, nondepleted mice (Fig. 6B), indicating that the FV-specific CD8⁺ T-cell response in FV-infected mice was CD4⁺ T-cell independent. In FV/LDV-coinfected mice, depletion of CD4⁺ T cells did not restore a normal CD8⁺ T-cell response at 7 days postinfection (Fig. 6B). Similar results were obtained from mice treated with PC61 anti-CD25 antibody, reported to block Treg cell-mediated suppression (33, 43) (data not shown). Thus, the failure of coinfected mice to develop CD8⁺ T-cell responses by 1 week postinfection could not be attributed solely to LDV-induced Treg cells.

Suppression of anti-FV responses in mice chronically infected with LDV.

Infection with LDV causes numerous acute effects such as extremely high viral load (51), induction of IFN-α (18), and polyclonal immune cell activation (12, 13, 37) that could be responsible for delaying anti-FV CD8⁺ T-cell responses. To address the question of whether any of these factors were responsible for the inhibition of FV-specific T-cell responses, we utilized mice chronically infected with LDV. These mice have 10,000-fold lower titers of LDV compared to that of LDV-infected mice at 16 h postinfection, as determined by quantitative real-time PCR (data not shown), undetectable levels of IFN-α, and no polyclonal lymphocyte activation. Mice chronically infected with LDV (>8 weeks after LDV infection) were infected with FV. At 16 h postinfection with FV, there was no significant induction of IFN-α (Fig. 7A) and no significant polyclonal lymphocyte activation as measured by CD69 upregulation (Fig. 7B). There was a slight increase in the percentage of tetramer-positive CD8⁺ T cells in mice chronically infected with LDV compared to that of naïve mice, but the response was not statistically significant. Regardless, compared to mice with no LDV infection, the LDV chronically infected mice had significantly suppressed FV-specific CD8⁺ T-cell responses (Fig. 7C). As with the mice infected with both FV and LDV simultaneously (Fig. 4), the FV-specific CD8⁺ T-cell response recovered by 14 days postinfection (Fig. 7C). Thus, even in the absence of high titers of LDV, type I IFN responses, and polyclonal T-cell activation, the presence of LDV suppressed FV-specific CD8⁺ T-cell responses.

FIG. 7.

Suppression of FV infection-specific CD8⁺ T-cell responses during chronic LDV infection. Plasma IFN-α responses (A) and the upregulation of CD69 early activation marker on total spleen lymphocytes (B) (representative data) were measured at 16 h postinfection of naive mice or of mice infected with FV alone or with LDV (left panel) and in mice chronically infected with LDV prior to FV infection (chLDV/FV) (right panel). (C) FV-specific tetramers were used to label CD8⁺ T cells at 7 and 14 days following FV infection of naïve mice or of mice that were already chronically infected with LDV. The naïve mice, n = 7; chronic LDV mice, n = 2; FV-infected mice, n = 9; chLDV/FV mice, n = 4. Asterisks (*) indicate that the difference between the FV infection and chLDV/FV infection groups was statistically significant by two-tailed Student's t test (P = 0.0118).

DISCUSSION

The findings of this study demonstrate that the presence of LDV in FV stocks causes suppression of FV-specific CD8⁺ T-cell responses resulting in prolonged acute infection. LDV coinfection did not affect the kinetics of the early FV replication, the cellular tropism of FV, the generation of FV neutralizing antibodies, or the CD4⁺ T-cell responses. However, the FV-specific CD8⁺ T-cell responses were delayed by approximately 1 week, which was sufficient to account for delayed virus clearance. Interestingly, there was no evidence that the LDV-specific CD8⁺ T-cell response was suppressed. Although we did not have LDV-specific tetramers to directly analyze the CD8⁺ T-cell responses, CD8⁺ T-cell expansion, upregulation of activation markers, and production of IFN-γ in LDV-infected and FV/LDV-coinfected mice all suggested that the response to LDV was intact. This interpretation is consistent with previous findings that acute LDV infection induces a cytotoxic T-lymphocyte response detectable by 1 week postinfection (21, 54). Thus, the suppression appeared to be specific for the CD8⁺ T-cell response to FV.

Since LDV titers reach such high levels so quickly (10⁹ to 10¹⁰ ID₅₀/ml by 16 h postinfection), it is possible that the APCs necessary to initiate FV-specific CD8⁺ T-cell responses were overwhelmed with LDV antigens, with no remaining capacity to process and/or present FV antigens present at comparatively minute concentrations. In contrast, APCs responding to FV infections in the absence of LDV would be highly focused and able to prime CD8⁺ T cells. However, we showed that mice with chronic LDV infections, where LDV titers were reduced by 4 log₁₀ (our unpublished data and see reference 51) compared to acutely infected mice also had suppressed anti-FV responses. This result demonstrates that extremely high LDV titers were not necessary to suppress the FV-specific CD8⁺ T-cell response, but it remains possible that chronic LDV levels were still sufficient to swamp the antigen-presenting machinery. It is also possible that LDV peptides have higher affinity for MHC-I molecules and are able to outcompete FV peptides for antigen presentation, thereby slowing the FV-specific CD8⁺ T-cell response. Given that CD4⁺ T cells also need APC function and that the CD4⁺ T-cell response was not suppressed, any impairment of the APC machinery must be predominantly in the MHC-I pathway.

Our experiments rule out the possibility that LDV-induced Treg cells were solely responsible for delayed FV-specific CD8⁺ T-cell responses. Although LDV induced suppressive T cells very early during infection, depletion of those cells did not restore the FV-specific CD8⁺ T-cell response at 1 week postinfection in FV/LDV-coinfected mice. It remains possible that LDV-induced Treg cells play a contributory role in delaying the FV-specific CD8⁺ T-cell response, but at least one other mechanism must be involved. Interestingly, the LDV-induced suppression was transient and waned by 1 week postinfection. In contrast, the FV-induced Treg cell response did not become apparent until the 1-week time point, which is consistent with previous in vivo studies using an FV stock containing LDV (66). Thus, both LDV and FV induced Treg cells but with different kinetics.

The current experiments with mice chronically infected with LDV also addressed the possible roles of acute IFN-α responses and polyclonal lymphocyte activation in suppressing FV-specific CD8⁺ T-cell responses. Since LDV chronic mice lacked those acute responses and were still suppressed, it is unlikely that either IFN-α responses or polyclonal lymphocyte activation was responsible for suppression. In further support of this conclusion, we recently found that mice depleted of plasmacytoid dendritic cells prior to FV/LDV coinfection, which ablated their ability to mount IFN-α responses or the consequent polyclonal lymphocyte activation (5 and data not shown), also had no FV-specific CD8⁺ T-cell responses at 7 days postinfection.

Assuming that the suppressions of FV-specific CD8⁺ T-cell responses during acute and chronic infection are related, the mechanism by which LDV inhibits FV-specific T-cell responses must be due to a long-term effect of LDV on the host. In addition to sustained production of LDV peptides that could affect the presentation of FV antigens, another long-term effect is the lytic depletion of the subset of macrophages that scavenge lactate dehydrogenase (54). It has previously been shown that LDV inhibits antigen presentation by macrophages (30), and it is possible that macrophage loss or dysfunction plays a direct role in delaying the initiation of FV-specific CD8⁺ T-cell responses. It may also be that the effect is indirect via dendritic cells (DCs), which are more commonly involved in priming naïve T cells during viral infections. Although we did not see a difference in levels of FV-infected DCs at 1 week postinfection, it remains possible that there were differences in levels of processed antigens or infection levels at an earlier time point. A negative effect on CD8⁺ T-cell activation could involve the secretion of anti-inflammatory cytokines such as interleukin-10 (2, 20) by DCs responding to cytokines or by breakdown products released from infected or killed macrophages (41). Alternatively, suppression may occur directly with the CD8⁺ T cells rather than via APCs. Whatever mechanism is at play must account for apparently intact LDV-specific CD8⁺ T-cell responses as well as FV- and LDV-specific CD4⁺ T-cell responses. Furthermore, the suppression must be either incomplete or transient because the FV-specific response was only delayed rather than absent.

Given the evidence for the presence of LDV in FV stocks as far back as 1963 (56), and possibly from its original isolation in 1957 (22), it is likely that most current stocks maintained solely by mouse passage still contain LDV. At one time, it appears that LDV was considered an integral component of the FV complex (61) and was also found in stocks of Rauscher leukemia virus (55) and murine sarcoma virus (62). LDV is also common in transplantable tumor and cell lines (7, 47). Although the presence of LDV can go unnoticed because of the lack of overt clinical signs or pathology, it can generate significant impacts on host immune responses and thereby increase the degree and duration of pathology induced by a second viral challenge. Hence, some of the effects previously attributed solely to FV may have been influenced by LDV. Experiments are currently under way to determine whether the presence of LDV plays a role in our current fields of interest: the establishment of FV persistence, the induction of regulatory T cells, and vaccine-induced protection. It is not feasible to retrospectively determine how much of the work described in the published literature on FV also contained LDV because the literature is too vast, and very few of the authors are still actively involved in FV research. However, all studies published henceforth should designate whether or not the experimental FV stocks contain LDV.

Although most of our current understanding of virus-host interactions has come from models of infection with single viral pathogens, there is growing interest in studying the pathogenesis and immune responses elicited by multiple infections. Studies of mixed or sequential infections have shown that the immune response to a particular pathogen can be greatly influenced by the presence of other pathogens (reviewed in reference 64). In nature, multiple acute and chronic infections are frequently observed. Just one example in humans is coinfection with HIV type 1 (HIV-1) and hepatitis C virus (HCV), which is becoming epidemic and is of significant clinical importance (1, 35). Compared to HIV-specific CD8⁺ T-cell responses, those of HCV-specific responses are profoundly impaired in coinfections with HIV (35) and of less breadth than those in HCV monoinfections (19). Thus, the use of characterized animal models of coinfection may provide insights into this important aspect of infectious diseases. In this light, it will be interesting to further characterize the ways in which LDV coinfection influences various aspects of anti-FV immunity.