Theiler's Virus Infection of Primary Cultures of Bone Marrow-Derived Monocytes/Macrophages

Abstract

Theiler's virus, a murine picornavirus, causes a persistent infection of macrophage/microglial cells in the central nervous systems of SJL/J mice. Viral replication is restricted in the majority of infected cells, whereas a minority of them contain large amounts of viral RNA and antigens. For the present work, we infected primary cultures of bone marrow monocytes/macrophages from SJL/J mice with Theiler's virus. During the first 10 h postinfection (p.i.), infected monocytes/macrophages were round and covered with filopodia and contained large amounts of viral antigens throughout their cytoplasm. Later on, they were large, flat, and devoid of filopodia and they contained only small amounts of viral antigens distributed in discrete inclusions. These two types of infected cells were very reminiscent of the two types of infected macrophages found in the spinal cords of SJL/J mice. At the peak of virus production, the viral yield per cell was approximately 200 times lower than that for BHK-21 cells. Cell death occurred in the culture during the first 24 h p.i. but not thereafter. No infected cells could be detected after 4 days p.i., and the infection never spread to 100% of the cells. This restriction was unchanged by treating the medium at pH 2 but was abolished by treating it with a neutralizing alpha/beta interferon antiserum, indicating a role for this cytokine in limiting virus expression in monocyte/macrophage cultures. The role of alpha/beta interferon was confirmed by the observation that monocytes/macrophages from IFNA/BR^−/− mice were fully permissive.

Theiler's murine encephalomyelitis virus (TMEV), a member of the picornavirus family, is a naturally occurring enteric pathogen of mice that causes a persistent infection of the central nervous system (CNS) and a chronic demyelinating disease resembling multiple sclerosis (19). Intracranial inoculation of TMEV in genetically susceptible mice results in a biphasic disease. The first phase is a mild encephalomyelitis during which the virus is found in the gray matter of the brain and spinal cord. During the second phase, the virus disappears from the gray matter and infects the white matter of the spinal cord where it persists for the lifetime of the animal. Viral persistence results in chronic inflammation and primary demyelination. Susceptibility to viral persistence and to the accompanying pathologic changes varies greatly between inbred strains of mice. Some are highly susceptible (e.g., the SJL/J strain), and others are completely resistant (e.g., the C57BL/6 strain). Strains which are resistant to persistent infection clear the infection after the early encephalomyelitis. Susceptibility to TMEV's persistence and to the virus-induced demyelinating disease is multigenic and complex (for a review, see reference 3).

TMEV-induced demyelination appears to be, at least in part, immune mediated. Virus-specific CD4⁺ T cells, responsible for a delayed-type hypersensitivity reaction, have been implicated (6, 10, 11, 12). Myelin epitope-specific CD4⁺ T cells, primed via epitope spreading, could also play a role in the late stages of the disease (17, 24). According to the prevalent view, macrophages recruited and activated by virus- and myelin-specific CD4⁺ T cells are the effectors of demyelination. Indeed, macrophages laden with myelin debris are common in white matter lesions, and the depletion of infiltrating macrophages with mannosylated liposomes reduces demyelinating lesions drastically (30).

Macrophages may play another important role in pathogenesis, as reservoirs of viral persistence. Lipton et al. have shown, by double immunostaining, that the main burden of virus antigen resides in macrophages (21). This has been confirmed by Pena-Rossi et al., using combined immunocytochemistry-in situ hybridization (30). The origin of these infected macrophages, whether they are hematogenous infiltrating monocytes or microglial cells, remained an open question until Pena-Rossi et al. showed that depleting monocytes/macrophages with liposomes cleared the CNS of persistent infection almost completely (30). More recently, Drescher et al. showed that treating mice with transforming growth factor TGF-β2, a potent immunoregulatory mediator, decreased the virus load in the white matter of the spinal cord by reducing the number of infiltrating macrophages (8). Taken together, these results strongly indicate that blood-borne macrophages are the main viral targets during persistent infection.

Two types of infected cells have been observed in the CNS during viral persistence. Some of them contain large amounts of viral RNA and antigens, but the majority contain limited amounts of viral products, indicating restricted viral replication (5, 21). These two types of cells could play different roles in viral persistence. The former might be important to propagate the infection to new target cells, whereas the latter might serve as a reservoir of persistent infection. Studies with macrophage-derived cell lines indicate that permissiveness to TMEV depends on the differentiation/activation status of the cell (14-16, 32). Therefore, it has been hypothesized that the two types of infected cells observed in vivo could correspond to monocytes/macrophages at different stages of differentiation/activation.

Although they offer obvious practical advantages, tissue culture-adapted macrophage cell lines are significantly different from the original cells. Furthermore, as already mentioned, susceptibility to persistence of TMEV depends critically on the genetic background of the mouse, a parameter not taken into account when using macrophage cell lines. Therefore, we decided to study the infection of primary cultures of monocytes/macrophages derived from the bone marrow of SJL/J mice, a strain susceptible to persistent infection. We report that, 10 h postinfection (p.i.), the majority of infected cells were round and covered with filopodia and that viral replication was active in these cells. The morphology of infected cells changed with time. They became flat, with a large cytoplasm and no filopodia. At that stage, viral replication was restricted due to the secretion of alpha/beta interferon (IFN-α/β).

MATERIALS AND METHODS

Viruses.

The DA strain of TMEV was plaque purified on BHK-21 cells, and the virus titers were determined by standard plaque assay on BHK-21 cells (20).

Vesicular stomatitis virus (VSV), strain Indiana, was kindly provided by Eliane Meurs (Pasteur Institute), and the virus titers were determined by conventional plaque assay on NIH 3T3 cells.

Primary cultures of bone marrow-derived macrophages.

SJL/J mice were purchased from Janvier (Le Genest-St-Isle, France). IFNA/BR^−/− mice were bred in the Institut Pasteur animal facility. Bone marrow cells were flushed from femurs and tibias of 8- to 9-week-old female mice. After depletion of red blood cells by using Gey solution and low-speed centrifugation, the cell pellet was resuspended in complete medium consisting of RPMI 1640 medium (Prolabo) supplemented with 1% penicillin-streptomycin (Gibco), 10% heat-inactivated fetal bovine serum (FBS) (Gibco), and 10% L929 cell-conditioned medium as a source of macrophage colony-stimulating factor (M-CSF). The latter was prepared by plating 8 × 10⁶ L929 cells in a T75 flask containing 40 ml of RPMI 1640 medium supplemented with 10% FBS. After 1 week at 37°C in a 5% CO₂ humidified atmosphere, the medium was harvested, filtered, and stored at −20°C until use. Bone marrow cells were seeded at a density of 1.5 × 10⁷ cells onto 100-mm-diameter tissue culture dishes (Falcon). Every other day, one-third volume of complete medium was added to the culture. After 6 days at 37°C in a 5% CO₂ humidified atmosphere, nonadherent cell clusters were gently flushed and centrifuged at low speed. The cell pellet was resuspended in a mixture containing RPMI 1640 medium, 1% penicillin-streptomycin, 10% FBS, and 2.5% L929 cell-conditioned medium (incomplete medium). Cells do not divide in incomplete medium. Cells were dissociated mechanically through needles, then seeded onto 12-mm-diameter coverslips, in 24-well plates, at a density of 2 × 10⁵ cells per well. Cells were cultured overnight at 37°C before being infected. The cells were kept in incomplete medium throughout the experiment.

Infection of primary cultures of bone marrow-derived macrophages.

Cells were washed twice in RPMI 1640 medium to remove serum components, and 0.35 ml of appropriately diluted virus stock solution was added to each well. Cell cultures were infected at a multiplicity of infection (MOI) of 5 PFU per cell. After adsorption of the virus for 2 h at 37°C, cells were washed twice with RPMI 1640 medium and cultured with 0.5 ml of incomplete medium containing only 2% FBS.

Infection of BHK-21 cells.

BHK-21 cells were grown in six-well plates, washed twice in PBS, and infected at an MOI of 5 PFU/cell in 0.5 ml. After adsorption of the virus for 45 min at 25°C, 2.5 ml of medium containing 1% FBS was added to each well.

Cell viability assays.

The Alamar Blue assay (Interchim) consists of an oxidoreduction indicator that changes color with cell growth. At different times postinfection, one-tenth volume of Alamar Blue reagent was added to the medium of TMEV- and mock-infected cultures. After 3 h of incubation at 37°C, the absorbance of the medium was measured at wavelengths of 570 and 600 nm (13).

To determine the number of macrophages in TMEV- and mock-infected cultures, cells were lysed with 0.1 M citric acid containing 0.05% naphtol blue black (Aldrich) and 1% cetrimide (Sigma). Under these conditions, only macrophage nuclei remain intact. Macrophage nuclei were counted in a hemacytometer (1, 28).

For both assays, the results for infected cultures were expressed as the percentage of the number of viable cells in noninfected cultures treated in parallel.

Immunofluorescence labeling.

Immunostaining was done directly on cells seeded on glass coverslips. The cells were fixed with 4% (wt/vol) paraformaldehyde (Electron Microscopy Sciences) for 15 min at room temperature and then quenched with 50 mM NH₄Cl in phosphate-buffered saline (PBS).

Cells were made permeable to reagents by incubating the coverslips for 20 min in PBS-2% (wt/vol) bovine serum albumin-0.1% saponin (Sigma) or for 5 min in PBS-0.1% Triton X-100 for the detection of, respectively, the intracellular MOMA-2 and TMEV antigens.

Nonspecific antibody-binding sites were blocked by incubating the coverslips for 30 min in PBS-2% normal goat serum (Sigma).

The following primary antibodies were used to characterize macrophages: rat monoclonal MOMA-2 (Serotec), rat monoclonal FA/11 (kind gift of Genevieve Milon, Pasteur Institute), rat monoclonal anti-F4/80 (Serotec), biotinylated rat monoclonal anti-Mac-1 (Pharmigen) and fluorescein isothiocyanate (FITC)-conjugated mouse monoclonal anti-I-A^{k, f, r, s} (Pharmingen; clone 10-3.6). Cells were reacted with the primary antibody, diluted as indicated by the manufacturer for 2 h at room temperature. After repeated washing steps in PBS-0.1 mg/ml saponin-2% (wt/vol) bovine serum albumin, coverslips were incubated with Cy3-conjugated goat antirat immunoglobulin G (IgG) (Jackson Immuno Research Laboratories) (for MOMA-2, FA/11, and F4/80) or with Cy3-conjugated Streptavidin (Jackson Immuno Research Laboratories) (for Mac-1). To amplify the signal given by the anti-I-A antibody, cells were incubated with peroxidase-conjugated anti-FITC (Boerhinger) and then with FITC-conjugated tyramide (NEN Life Science).

TMEV antigens were detected with an anticapsid rabbit hyperimmune serum (4) followed by biotinylated antirabbit IgG (Diagnostic Pasteur) and Cy3-conjugated streptavidin (Jackson Immuno Research Laboratories) or with an FITC-conjugated antirabbit IgG (Jackson Immuno Research Laboratories). In some experiments, nuclei were counter-stained with 0.2 μM ethidium homodimer 1 (Sigma).

RESULTS

Bone marrow-derived monocyte/macrophage cultures.

Single-cell suspensions were obtained from the bone marrow of SJL/J mice and grown in vitro with L929 cell-conditioned medium as a source of M-CSF. M-CSF is a growth factor that acts primarily on bone marrow precursors committed to the monocytic lineage (23, 33). It induces their proliferation and differentiation in vitro into clusters of macrophages (25). After 6 days, the cultures contained adherent cells composed mainly of mature macrophages and fibroblasts and numerous clusters of nonadherent cells. These clusters were gently harvested and dissociated. A sample was stained with May-Grunwald Giemsa. More than 95% of the cells were phagocytic mononuclear cells and a small percentage were polymorphonuclear leukocytes (data not shown). The cells were plated on glass coverslips and analyzed by indirect immunofluorescence for the expression of markers of macrophage differentiation. All the cells expressed MOMA-2, FA-11 (Fig. 1A and B), F4/80, and Mac-1 (not shown). Only a few cells (approximately 3%) expressed major histocompatibility complex (MHC) class II molecules (Fig. 1C and D). Thus, these cultured cells had the morphology and the phenotype of phagocytic mononuclear cells differentiated into monocytes/resident macrophages (22). The few MHC class II-positive cells present in the culture could have been dendritic cell precursors.

FIG. 1.

Characterization of bone marrow-derived monocytes/macrophages by immunofluorescence using macrophage-specific antigens. (A) Rat monoclonal MOMA-2. (B) Rat monoclonal FA/11. (C and D) Mouse monoclonal anti-I-A^{k, f, r, s}. For each field, phase contrast is shown on the right. For panels A, B, and C, bars = 100 μm. For panel D, bar = 26.5 μm.

Infection of bone marrow-derived monocytes/macrophages with TMEV.

Monocytes/macrophages grown in vitro were infected with TMEV at an MOI of 5 PFU/cell, as described in Materials and Methods, and observed for 13 days. Some cytopathic effect was observed 24 h after infection but not thereafter. To confirm and quantify this observation, cell viability was measured by using both the Alamar Blue and the naphtol blue black assays. The results are presented in Fig. 2 as percentages of the number of viable cells in mock-infected cultures examined at the same time. As shown in the figure, viability was 100% at 10 h p.i. and then it decreased to about 77% by 27 h p.i. and remained at that level until 13 days p.i.

FIG. 2.

Characterization of bone marrow-derived monocytes/macrophages by immunofluorescence using macrophage-specific antigens. (A) Rat monoclonal MOMA-2. (B) Rat monoclonal FA/11. (C and D) Mouse monoclonal anti-I-A^{k, f, r, s}. For each field, phase contrast is shown on the right. For panels A, B, and C, bars = 100 μm. For panel D, bar = 26.5 μm.

The extent of viral infection in these cultures was examined by detecting TMEV capsid antigens with immunofluorescence. The percentage of cells that contained TMEV antigen was measured at different times p.i. by counting the total number of cells and the number of antigen-positive cells in approximately 10 randomly chosen fields per coverslip (between 400 and 1,000 cells were examined for each coverslip) (Table 1). Approximately 15% of the cells expressed viral antigens at 10 h p.i. This percentage reached a maximum of about 25% between 27 and 54 h p.i. After 27 h, the percentage of infected cells decreased progressively and by 13 days p.i. no viral antigens could be detected in the cultures. To confirm the absence of infection at 13 days p.i., the cells were hybridized in situ with a radioactive probe specific for TMEV RNA (18). No viral RNA could be detected (data not shown). It should be noted that the disappearance of the infection was not due to a decrease of the number of viable cells in the culture (Fig. 2).

TABLE 1.

Fraction of infected cells in primary cultures of bone marrow-derived monocytes/macrophages, at different times p.i.

Time after inoculation	Expt 1		Expt 2		Expt 3
	Total no. of cells counted	No. (%) of antigen-positive cells	Total no. of cells counted	No. (%) of antigen-positive cells	Total no. of cells counted	No. (%) of antigen-positive cells
10 h	702	109 (15.5)	1,430	194 (13.6)	677	114 (16.8)
27 h	1,024	256 (25)	1,022	326 (31.9)	537	114 (21.2)
54 h	1,069	224 (21)	976	307 (31.5)	449	108 (24.1)
4 days	1,043	153 (14.7)	843	167 (19.8)	1,017	86 (8.5)
13 days	NDa	ND	ND	ND	ND	0 (0)

^aND, not done.

Interestingly, two types of virus antigen-containing cells were observed (Fig. 3). At 10 h p.i., approximately 60% of infected cells were round and covered with filopodia (Fig. 3A and 3B). Viral antigens were uniformly distributed in the cytoplasm of these cells, and fluorescent staining was intense, suggesting active virus replication. The other infected cells were larger, flat, and devoid of filopodia. Viral antigens formed inclusions in them, often with a punctate appearance. Overall, fluorescence was weaker in these cells, suggesting restricted viral replication. This second type predominated at 27 h p.i. (Fig. 3C) and was the only type present at 4 days p.i. (Fig. 3D).

FIG. 3.

Morphology of TMEV antigen-containing cells at different times p.i. (A and B) At 10 h p.i., the majority of infected cells were round and covered with filopodia. For panel A, the same field is shown under phase contrast. Panel B shows two different fields. (C) At 27 h p.i., the majority of infected cells were spread out and had a large cytoplasm and no filopodia. (D) This kind of infected cell was the only one detected at 4 days p.i. Bars = 26.5 μm.

Viral titers in the culture medium were measured at different times p.i. by using a plaque assay on BHK-21 cells. The average titers obtained in four independent experiments are shown in Fig. 4. A peak of infectivity was observed at 27 h p.i. Interestingly, this is also the time at which cell death and the number of infected cells were at maximum levels (Fig. 2). For comparison, the figure shows the titers in the medium of BHK-21 cells infected at the same MOI of 5 PFU/cell.

FIG. 4.

Virus titers in the supernatants of infected cells, at different times p.i. □, primary cultures of bone marrow-derived monocytes/macrophages; each point is the average of four independent experiments. ◊, BHK-21 cells; each point is the average of two independent experiments. Standard errors are shown.

Characteristics of infected cells.

We examined whether the infection by TMEV altered the phagocytic activity of monocytes/macrophages and induced the expression of class II antigens.

FITC-labeled latex particles (2 μm diameter, Polysciences) were added to infected and mock-infected cultures at different times p.i. at a particle-to-cell ratio of 50:1. After incubation at 37°C for 1.5 h, the cells were washed several times with PBS, fixed with 4% paraformaldehyde, and stained for TMEV antigens as described in Materials and Methods. The number of particles per infected and uninfected cell was determined. Cells were also incubated with particles at 4°C, a temperature which blocks phagocytosis, to determine the background of adsorbed, as opposed to ingested, particles per cell (26, 29). No statistically significant difference in the number of ingested particles per cell was observed between infected and mock-infected cells at all times p.i. examined. Moreover, phagocytosis was not statistically different in the two types of infected cells described above (data not shown).

The expression of MHC class II molecules on infected cells was examined using double immunostaining for TMEV-antigens and MHC class II molecules. As described above, uninfected cultures contained an average of 3.4% class II-positive cells. The number of class II-positive cells decreased progressively after infection. No class II-positive cells were detected in infected cultures after day 4 p.i. However, class II-positive cells disappeared with time also in mock-infected cultures, indicating that the decrease of expression of these molecules was not due to the virus.

In summary, our data show that infection with TMEV did not modify the capacity of bone marrow-derived monocytes/macrophages to phagocytose latex beads and did not induce the expression of MHC class II molecules.

Monocytes/macrophages infected with TMEV secrete a soluble antiviral factor.

The fact that the infection did not spread to 100% of the cells suggested that either there were two subpopulations of monocytes/macrophages, and only one was permissive to the virus, or that infected cells released a soluble factor which made the other cells in the culture resistant. The experiment outlined in Fig. 5A was designed to test the second hypothesis. The medium of infected cultures (“conditioned medium”) was harvested either at 10, 27, or 96 h p.i. and spun at 100,000 × g to remove viral particles as well as cell debris. Fresh primary cultures were incubated with conditioned medium for 24 h. This conditioned medium was then removed and kept at 37°C, and the cultures were infected with TMEV (5 PFU/cell) in a small volume. After adsorption of the virus, the conditioned medium was returned to the cultures. The cells were examined for the presence of viral antigens by immunofluorescence at various times p.i. The experiment was also performed by using medium from mock-infected cultures as a control.

FIG. 5.

Detection of antiviral factors in supernatants of infected primary cultures of bone marrow-derived monocytes/macrophages. (A) Flow chart of the experiment. (B) Flow chart for neutralization assays with the anti-IFN-α/β polyclonal antibody. (C) Results obtained, in two independent experiments, for primary cultures treated with TMEV-infected (□), or mock-infected (▪) supernatants harvested at 27 h p.i. Standard errors are shown. ∗, P < 0,0001 as determined by a χ² test.

As shown in Fig. 5C, there was a statistically significant decrease in the number of infected cells in cultures treated with conditioned medium. Moreover, the number of infected cells did not increase between 10 and 27 h p.i. in cultures treated with conditioned medium as opposed to cultures treated with control medium. In cultures treated with conditioned medium, all infected cells had large cytoplasms with punctate inclusions of viral antigens. Figure 5C shows the results obtained with conditioned medium harvested at 27 h p.i. Similar results were obtained with conditioned medium collected at 10 or 96 h p.i. (data not shown).

In summary, these data indicate that infected monocytes/macrophages release one or several antiviral soluble factors in the medium.

Infected monocytes/macrophages secrete IFN-α/β.

To determine if the antiviral effect was specific for TMEV, we tested the effect of conditioned medium on the replication of VSV in NIH 3T3 cells. Conditioned medium was centrifuged as described above and diluted serially in Dulbecco's modified Eagle medium (DMEM)-10% FBS. Monolayers of NIH 3T3 cells were incubated for 24 h with dilutions of conditioned medium, rinsed with DMEM, and infected with 70 PFU of VSV. After adsorption, the cells were overlaid with agarose to perform a standard plaque assay. As shown in Fig. 6A, the multiplication of VSV was inhibited in a dose-dependent manner by conditioned medium. The drop in titer was no longer statistically significant for a 1:50 dilution.

FIG. 6.

(A) Inhibition of VSV multiplication on NIH 3T3 cells. VSV was plaque assayed on NIH 3T3 cells in the presence of various dilutions of supernatants of TMEV-infected (□) or mock-infected (▪) primary cultures. The supernatants were harvested at 27 h p.i. (B) Same experiment performed with supernatants treated at pH 2. Two independent experiments were performed. All samples were assayed in duplicate. Standard errors are shown. A statistical analysis was performed using a parametric t test. ∗, P < 0.001; ∗∗, P < 0.002; ∗∗∗, P < 0.05.

IFN-α/β is the only cytokine that is stable after treatment at pH 2 (34). To determine if the inhibitory effect of conditioned medium was due to IFN-α/β, the experiment was repeated after treating the medium at pH 2 for 48 h at 4°C followed by neutralization to pH 7.0. As shown in Fig. 6B, treatment at pH 2 did not impair the inhibitory effect, strongly suggesting that inhibition was due to IFN-α/β.

Neutralization assays with a specific hyperimmune serum were performed to confirm that the inhibition was due to IFN-α/β (Fig. 5B). Conditioned medium, centrifuged at high speed as described above, was incubated for 1 h at 37°C followed by 1 h at 4°C with an amount of hyperimmune sheep antiserum able to neutralize 5,000 U of IFN-α/β (a kind gift of I. Gresser, Institut Curie, Paris) (2, 31). Conditioned medium was also incubated with PBS as control. Fresh monocyte/macrophage cultures were incubated with these media for 24 h at 37°C. The media were removed, and the cells were washed with RPMI and infected with TMEV at an MOI of 5 PFU/cell. After adsorption, the cells were washed again with RPMI and cultured with fresh RPMI-2% FBS-2.5% M-CSF. After 27 h, viral production was measured with a plaque assay using BHK-21 cells.

As shown in Fig. 7, no infectivity was detected when the cells had been treated with nonneutralized conditioned medium, confirming the presence of an inhibitory factor in the medium. Approximately 5 × 10⁴ PFU/ml was detected when the conditioned medium had been treated with the neutralizing anti-interferon serum or when the conditioned medium was from mock-infected cultures. This result confirmed that the antiviral factor released by TMEV-infected monocytes/macrophages was IFN-α/β.

FIG. 7.

IFN-α/β neutralization assays. Fresh primary cultures of bone marrow-derived monocytes/macrophages were treated with the supernatant of an infected primary culture and then infected with TMEV. The supernatant had been incubated with a neutralizing sheep antimouse IFN-α/β serum or with PBS. The same experiment was performed with the supernatant of a mock-infected culture. The supernatants of TMEV- or mock-infected primary cultures were harvested at 27 h p.i. ∗, P < 0.05 in a t test.

Last, we tested the role of IFN-α/β directly by using primary cultures of bone marrow monocytes/macrophages derived from mice with an inactivated gene for the IFN-α/β receptor (IFNA/BR^−/−) (27). The procedure for culturing and infecting these cells was identical to that used for SJL/J mice. Staining by immunofluorescence for viral antigens showed that, in contrast to SJL/J macrophages, 100% of IFNA/BR^−/− macrophages became infected between 10 and 24 h p.i. Infected cells were round and contained large amounts of viral antigens (Fig. 8A). This infection resulted in 100% cell death. As shown in Fig. 8B, viral titers released in the medium were more than 10 times higher than for SJL/J macrophages, although they were still approximately 100 times lower than for BHK-21 cells.

FIG. 8.

Comparison of TMEV infection of bone marrow-derived monocytes/macrophages from SJL/J and IFNA/BR−/− mice. (A) For each mouse strain, the upper panels show the detection of viral capsid antigens by immunofluorescence and the lower panels show the staining of cells with ethidium homodimer 1 for the same field. (B) Viral titers in the medium of infected monocyte/macrophage cultures from SJL/J and IFNA/BR−/− mice. Each point represents the mean of titers obtained from six samples from two independent experiments.

DISCUSSION

Results from several laboratories indicate that infiltrating CNS macrophages are the main viral reservoir in SJL/J mice persistently infected with TMEV (21, 30). A dichotomy has been observed among these cells (5, 21). Some contain large amounts of viral RNA and antigens, whereas viral expression is restricted in others. The former could be important in spreading the infection within the CNS, and indeed it has been shown that macrophages isolated from persistently infected spinal cords can release infectious virus (7).

In the present study, we observed two kinds of TMEV-infected cells in primary cultures of bone marrow-derived monocytes/macrophages from SJL/J mice. At early times p.i., the majority of infected cells were round, covered with filopodia, and contained large amounts of viral antigens diffusely distributed in the cytoplasm (Fig. 3A). At the same time, there was a cytopathic effect in the culture and the production of infectious particles in the medium, suggestive of active viral replication. Later on, infected cells were flat and without filopodia and they contained punctate inclusions of viral antigens. At that time there was no cytopathic effect and little or no infectivity released in the culture medium, although a small amount of infectivity (approximately 10² PFU/10⁴ cells) was found associated with the cells (not shown). These observations suggest that replication becomes restricted at this stage and that there is sequestration of infectious particles in intracellular sites. Cell-associated infectivity in the absence of release of infectious particles could contribute to the persistence of the infection in vivo. Clearly, these two types of infected macrophages are very reminiscent of the two types of infected cells described in vivo in SJL/J mice, using immunofluorescence (21). Experiments are under way, using a newly developed quantitative assay in which immunofluorescence is coupled to fluorescent in situ hybridization, to compare the ratio of expression of viral RNA to that of viral capsid proteins in cultured macrophages and in cells in vivo during persistent infection.

The cells that we infected in vitro were differentiated in the presence of M-CSF (L929 conditioned medium). Importantly, they had the same phenotype (FA-11+, MOMA2+, F4/80+, and Mac-1+) as the macrophages, including the infected ones that are present in vivo in chronic CNS lesions of SJL/J mice (21). It has been shown that growth of TMEV in macrophage cell lines depends on their state of differentiation, suggesting the existence in vivo of a window of susceptibility to TMEV for monocytes entering chronic inflammatory infiltrates (14). Interestingly, we observed in preliminary experiments that increasing the treatment of bone marrow-derived macrophages with M-CSF (increasing the time of treatment or the concentration of M-CSF) diminishes their permissiveness to TMEV (not shown), an observation which is consistent with the above-mentioned hypothesis.

Most of the late restriction observed in cultured macrophages could be attributed to the secretion of IFN-α/β. This raises the possibility that IFN-α/β plays an important role in vivo in restricting viral replication during persistent infection. Unfortunately, this cannot be tested using IFNA/BR^−/− mice since these animals die early of overwhelming meningoencephalomyelitis (9).

At a maximum, 32% of cultured macrophages were infected in our experiments, although the cells were inoculated at an MOI of 5 PFU/cell. Increasing the MOI 10-fold raised the level of infection to only 50% (data not shown). Since the macrophage cultures were not clonal, this result could be due to the presence of a subpopulation of resistant cells. However, the MOI used to infect macrophages corresponds to titers obtained by plaque assay on highly susceptible BHK-21 cells. It is more likely that limited infection of macrophages simply reflects the lower permissiveness of macrophages to TMEV infection. Furthermore, the secretion of IFN-α/β prevented the spread of the infection in the culture, as shown by the fact that the virus did spread to the entire culture when the macrophages were from IFNA/BR^−/− mice. Viral replication was also slower in macrophages than in BHK-21 cells, and viral yield was lower (Fig. 4). Taking into account that approximately 30% of macrophages were viral antigen positive at the peak of viral production, the yield from infected macrophages was approximately 0.5 PFU/cell, whereas it was on the order of 500 PFU/cell for BHK-21 cells (Fig. 4). Interestingly, these yields are very similar to those reported by Trottier et al. for macrophages infected in vivo and for BHK-21 cells (35). In comparison, the yield for IFNA/BR^−/− macrophages was approximately 40 PFU/cell.

Genetically resistant inbred mouse strains clear TMEV infection after 2 to 3 weeks (3). If the infection of macrophages is central to TMEV's persistence, one expects some of the resistance genes to be expressed in these cells. Interestingly, we observed that bone marrow monocytes/macrophages from B10.S mice, a strain that is resistant although it bears the same H-2^s haplotype as the susceptible SJL/J strain, cannot be infected in vitro with the DA strain of Theiler's virus (data not shown). It will be important to determine the origin of this block.

Infiltrating monocytes/macrophages are continuously recruited into white matter perivascular cuffs during persistent infection. It is conceivable that some of these cells, which are permissive to viral replication, although less so than BHK-21 cells, become infected. The secretion of IFN-α/β will restrict replication in the cells and limit diffusion of the infection in the tissues, but recruitment from the periphery will continuously replenish the population of permissive cells. According to this reconstruction of pathogenesis, a combination of replenishment of target cells and of interferon-mediated restriction of viral replication could form the basis of TMEV persistence in the CNS of SJL/J mice. Obviously this view might be an oversimplification. Further work may well reveal a more complex reality.

In summary, we report that primary cultures of mouse monocytes/macrophages could be infected by the DA strain of TMEV. Our results, in particular the detection of two kinds of infected cells, are very reminiscent of observations made in vivo. To our knowledge, we described here the first tissue culture system which mimics in vivo observations. This system makes it possible to study viral persistence at the molecular level in relevant cells.