Theiler’s virus causes a persistent infection with demyelination that is studied as a model for multiple sclerosis. Inbred strains of mice differ in their susceptibility to viral persistence due to both H-2 and non-H-2 genes. A locus with a major effect on persistence has been mapped on chromosome 10, close to the Ifng locus, using a cross between susceptible SJL/J and resistant B10.S mice. We now confirm the existence of this locus using two lines of congenic mice bearing the B10.S Ifng locus on an SJL/J background, and we describe a deletion in the promoter of the Ifng gene of the SJL/J mouse. We studied the expression of IFN-γ, IL-2, IL-10, and IL-12 in the brains of SJL/J mice, B10.S mice, and the two lines of congenic mice during the first 2 wk following inoculation. We found a greater expression of IFN-γ and IL-2 mRNA in the brains of B10.S mice compared with those of SJL/J mice. Also, the ratio of IL-12 to IL-10 mRNA levels was higher in B10.S mice. However, the cytokine profiles were the same for the two lines of resistant congenic mice and for susceptible SJL/J mice. Therefore, the difference of Th1/Th2 balance between the B10.S and SJL/J mice is not due to the Ifng locus and does not account for the difference of susceptibility of these mice to persistent infection.
The DA strain of Theiler’s virus, a murine picornavirus, causes a biphasic neurological disease after intracranial inoculation of genetically susceptible mice. The disease is characterized by an early gray matter encephalomyelitis followed by a persistent infection of the white matter of the spinal cord accompanied by chronic inflammation and primary demyelination. This late disease is studied as a model for multiple sclerosis (1, 2). Genetically resistant strains of mice clear the infection after the first 2 wk following inoculation and do not present with demyelination. Resistance/susceptibility to the persistent infection is a complex phenotype that is under the control of several host genes, including the H-2D region of the MHC (3, 4). Non-H-2 genes are also implicated, because the SJL/J mouse is more susceptible than the B10.S mouse, although both bear the H-2s haplotype. A gene with a major effect on persistence was mapped close to the Ifng gene on chromosome 10 by screening the genome of an (SJL/J × B10.S)F1 × B10.S backcross (5). IFN-γ has a major role in limiting viral persistence and the demyelinating disease, as shown by studying mice that lack the IFN-γR (IFN-γRo/o) (6) or by treating mice with a neutralizing anti-IFN-γ Ab (7, 8). These results and the fact that, according to the current view, IFN-γ plays a central role in Th1/Th2 balance and in the outcome of murine infections in general (9, 10, 11), made the Ifng gene a good candidate gene for the control of Theiler’s virus persistence. In the present work, we studied the Ifng gene and the cytokine profiles of SJL/J and B10.S mice following inoculation with Theiler’s virus. We also used congenic mice to confirm the presence of a susceptibility gene in the Ifng region and to test the role of the Ifng gene and Th1/Th2 balance in the difference of susceptibility between the SJL/J and B10.S strains.
Materials and Methods
SJL/J mice were purchased from Janvier (Saint-Berthevin, France); and B10.S-H2S/Sg McdJ mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Two lines of congenic mice, named SJL.10(70/14) and SJL.10(233/237), were obtained after 10 successive backcrosses of an (SJL/J × B10.S)F1 toward the SJL/J parent followed by brother/sister mating. These mice were selected because they have the B10.S allelic form of the D10 Mit70 and D10 Mit14 markers (line SJL.10(70/14)) or the D10 Mit233 and D10 Mit237 markers (line SJL.10(233/237)) (Fig. 4). The sequence of these markers is available at http://www.genome.wi.mit.edu/cgi-bin/mouse/index. A detailed description of these lines will be given elsewhere (F.B., unpublished observations). Mice of 3–4 wk of age were inoculated intracranially with 104 PFU of the DA strain of Theiler’s virus in 40 μl of PBS buffer.
A 3-kb fragment containing the promoter of the Ifng gene (GenBank accession number M28381) was amplified by PCR from the genomic DNA of SJL/J and B10.S mice. The DNA fragments were subcloned in the Bluescript KS vector (Stratagene, Cambridge, U.K.) and sequenced using the Sequenase kit (Amersham, Arlington Heights, IL). Two polymorphisms were detected between the two mouse strains: a mutation in an SphI restriction enzyme site of the SJL/J genome (GCATGC→ACATGC) and a 16-bp deletion, which we named D10Pas4, just upstream of the TATA box (see Results). The following PCR primers were designed to study D10Pas4: 5′-GAATCCCACAAGAATGGCACAG-3′ and 5′-CGAAGGCTCCTCGGGATTA-3′. PCR reactions with these primers were conducted under standard conditions with buffer (1.5 mM Mg2+) and Taq DNA polymerase from Life Technologies (Eragny, France) (at 94°C for 2 min, followed by 40 cycles of 94°C for 40 s, 55°C for 40 s, 72°C for 15 s). Amplified DNA was analyzed in 5% agarose gels. Microsatellite allele sizes in inbred strains and subspecies were as follows: ∼206 bp for strains 129/Sv, A/J, BALB/cByJIco, B10.S, C3H/HeOuJIco, C57BL/6JIco, C57BR, CB20, CBA/JIco, DBA/2JIco, DDK, DW, FVB/NPas, MAI, MBT, PL/J, PWK/Pas, and SEG; ∼190 bp for the MOLD, SJL/JIco, SPR, and STF strains.
The published sequence of the mouse IFN-γ cDNA (12) was used to prepare a specific RT-PCR probe. BAC 232G19 from Research Genetics (Huntsville, AL) was digested with EcoRI restriction enzyme and hybridized with the probe. The fragments that hybridized were cloned in the Bluescript vector and used to sequence the exon/intron junctions using primers designed from the published cDNA sequence. The exons of the SJL/J and B10.S strains were amplified from genomic DNA using primers designed from the exon/intron junction sequences. They were cloned with the TOPO TA cloning kit (Invitrogen, San Diego, CA) and sequenced with the Sequenase kit. The following primers were used for genomic DNA amplification: For exons 1 and 2, 5′-AAGTTCTGGGCTTCTCCTCC-3′ and 5′-CATGTCACCATCCTTGGGAA-3′; for exon 3, 5′-TCCCTGTTGTTTCTAATGGG-3′ and 5′-CACCTCCTAGCTTTATCAGC-3′; for exon 4, 5′-GATTTCCATCTTCACTGACC-3′ and 5′-TGGGACAATCTCTTCCCCAC-3′.
Quantification of viral RNA in CNS
The assay has been described in detail previously (13). Briefly, total RNA was extracted from the spinal cord. Five-fold dilutions of the RNA solutions were dotted on Hybond-C extra filters (Amersham). The filters were hybridized with a virus-specific 32P-labeled cDNA probe, washed, and exposed overnight against x-ray-sensitive films. In some experiments, the amount of viral RNA was measured by RT-PCR as described for cytokines (see below).
Detection of cytokine mRNA by RT-PCR
The method is similar to that described by Monteyne et al. (14). Brain tissue and cultured cells were lysed in TRIzol reagent (Life Technologies). The mixture was extracted with chloroform, and the RNA was precipitated with isopropanol, washed in ethanol, and finally resuspended in 50 μl of water. Randomly primed cDNA was prepared from ∼5 μg of RNA using 200 U of Moloney murine leukemia virus reverse transcriptase (Life Technologies) according to the manufacturer’s instructions. cDNA was amplified by PCR with a Gene Amp kit (Life Technologies) in a 9600 reactor (Perkin-Elmer Cetus, Norwalk, CT), with 40 cycles for cytokines and 25 cycles for actin and Theiler’s virus mRNA. Using these numbers of cycles avoided reaching saturation. The primers used for the PCR reaction were as follows: IL-2, 5′-GTCCTGCAGGCATGTACAGC-3′ and 5′-AGGGCTTGTTGAGATGATGC-3′; IL-10, 5′-CCCAGAAATCAAGGAGCATTTG-3′ and 5′-CATGTATGCTTCTATGCAGTTG-3′; IL-12, 5′-CAACGTTGCATCCTAGGATCG-3′ and 5′-GCACATCAGACCAGG-3′; IFN-γ, 5′-GACAATCAGGCCATCAGCAAC-3′ and 5′-CGCAATCACAGTCTTGGCTAA-3′; actin, 5′-ATGGATGACGATATCGCTGC-3′ and 5′-GCTGGAAGGTGGACAGTGAG-3′; Theiler’s virus, 5′-TCTAGATCTAGACTATTCAAGTTCGAGAATGGGGA-3′ and 5′-GAATTCGAATTCGAATTCGCCACCATGGGAACGGACAACGCCGAA-3′. The size of the amplified fragment was 500 bp, 250 bp, 1060 bp, 500 bp, 1060 bp, and 800 bp for, respectively, IL-2, IL-10, IL-12, IFN-γ, actin, and Theiler’s virus. The integrity of the RNA and the efficacy of reverse transcription were assessed by amplifying the actin cDNA with 25 PCR cycles.
The PCR products were analyzed in 1% agarose gels containing ethidium bromide, transferred to Hybond-N+ filters (Amersham), and hybridized overnight at 65°C with internal probes labeled with 32P. The intensity of each hybrid band was quantitated with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) under conditions for which intensity is proportional to the amount of radioactivity. The ratios between cytokine (or virus) and actin RNA levels were calculated after subtraction of nonspecific background. The sequences of the internal probes were as follows: IL-2, 5′-CCTGAGCAGGATGGAGAATT-3′; IL-10, 5′-GCTGGAAGACCAAGGTGTCTACAAGGC-3′; IL-12, 5′-TGGAATGGCGTCTCTGTCTGC-3′; IFN-γ, 5′-TCGCCTTGCTGTTGCTGA-3′; actin, 5′-CGTACCACAGGCATTGTGATGG-3′; Theiler’s virus, 5′-CCTATTACAAGTGTGACCTTGA-3′. Each experimental point shown in Results corresponds to the mean value obtained with at least three mice. The cytokine mRNA/actin mRNA ratio varies from experiment to experiment due to small variations of efficiency of PCR amplification, membrane blotting, and hybridization as well as to variations of the specific activity of the probe. Therefore, only ratios obtained in the same experiment can be compared.
In vitro stimulation of spleen cells and lymphocyte proliferation assay
Fresh spleen cells from SJL/J, B10.S, or congenic mice, cultured in RPMI 1640 medium (Life Technologies) supplemented with 10% FCS, were stimulated with 5% of a cell supernatant containing the 145-2C11 anti-CD3 Ab (a gift of O. Leo, Université Libre de Bruxelles, Rhode Saint-Genèse, Belgium). For cytokine mRNA detection, RNA was extracted from 107 cells after 24 h of culture. For the proliferation assay, [3H]thymidine (0.5 mCi/well) was added to the cultures (2.5 105 cells/well) for the last 6 h of a 72-h incubation period.
Results and Discussion
Studies of the Ifng gene and of the cytokine profiles of susceptible and resistant mice
Because the Ifng gene was a good candidate to explain the difference of susceptibility of the SJL/J and B10.S strains, we cloned and sequenced the gene from both strains. There were no differences in the coding part of the gene (data not shown). However, we found a 16-bp deletion, which we called D10Pas4, in the Ifng promoter of the SJL/J mouse (Fig. 1). We screened a large number of inbred mouse strains and found the deletion in strains MOLD, SPR, and STF but not in strains 129/Sv, A/J, B10.S, BALB/cByJIco, C3H/HeOuJIco, C57BL/6JIco, C57BR, CB20, CBA/JIco, DBA/2JIco, DDK, DW, FVB/NPas, MAI, MBT, PL/J, PWK/Pas, and SEG.
The D10Pas4 deletion could have been responsible for the susceptibility of the SJL/J strain for two reasons. First, it is present in the only strain for which the H-2D haplotype does not correlate with susceptibility to persistent infection. Second, it is located 44 nt upstream of the TATA box, between two sequence elements that are conserved among mice, rats, and humans (15). This could be a critical position, because both elements contain an AP-1 · cAMP response binding element-activating transcription factor DNA binding domain. These domains are functional because they are involved in the negative regulation of IFN-γ by glucocorticoids (16). The deletion decreases the distance between the two elements by one and and half helix turns and changes their spatial orientation. Because this could affect the activity of the promoter (17, 18), we measured the level of IFN-γ mRNA in the brains of SJL/J and B10.S mice during the first week after inoculation, a time period during which both strains are infected at similar levels. Because the level of viral RNA starts to decline in resistant mice around day 7 postinoculation, we assume that biochemical events leading to clearance would be operating during the first week. We observed a peak of IFN-γ mRNA expression at day 6 postinoculation in the case of the B10.S mouse. In contrast, the level of IFN-γ mRNA reached a plateau at day 5 postinoculation in the SJL/J mouse (Fig. 2,A). This result was reproduced in three independent experiments. The difference at day 6 postinoculation was highly significant when comparing 12 mice of each strain (Mann-Whitney U test, p = 0.003). As shown in Fig. 2,B, a significantly higher level of IL-2 mRNA was also observed in the brains of B10.S mice (Mann-Whitney U test, p = 0.002 when comparing 12 mice of each strain at day 6 postinoculation). During the same time period, no significant difference in the amount of viral RNA was observed between the two strains (Fig. 2 C).
These results suggested that the response of the B10.S mouse to Theiler’s virus CNS infection was more of the Th1 type than that of the SJL/J mouse. Therefore, we examined the levels of IL-12 and IL-10 mRNA in the brains of B10.S and SJL/J mice inoculated with Theiler’s virus. The expression of IL-12 was higher in B10.S mice than in SJL/J mice, with a peak of expression on day 6 postinoculation (Fig. 3,A). No significant difference was found for IL-10, although its expression tended to be higher in SJL/J mice. Despite some variation of cytokine mRNA levels from experiment to experiment, the ratio of IL-12 to IL-10 mRNA levels was significantly higher in B10.S mice than in SJL/J mice on day 6 postinoculation (Mann-Whitney U test, p = 0.005) (Fig. 3 C). Interestingly, we could not detect IL-4 mRNA by RT-PCR in the brains of either B10.S or SJL/J mice during the first 2 wk following inoculation, although it was found in the spinal cord of SJL/J mice at 45 days postinoculation (data not shown).
Studies with congenic mice
Two lines of congenic mice, named SJL.10(70/14) and SJL.10(233/237), were obtained as described in Materials and Methods. They possess the Ifng gene and varying amounts of the surrounding region of the B10.S parent on an SJL/J background (Fig. 4) (F.B., unpublished observations). Both lines were significantly more resistant to persistent infection than the SJL/J parent (Fig. 4), which demonstrates the existence of a susceptibility locus in the Ifng region. Fig. 4 also shows that this region accounts for 80% of the difference of susceptibility between the SJL/J and B10.S strains. Because Ifng is the only locus for which there is a significant linkage with susceptibility in a (SJL/J × B10.S)F1 × B10.S backcross (5), it is unlikely that the rest of the susceptibility is due to a single locus. Rather, it probably involves several genes and/or interactions between genes.
To determine whether a locus in the Ifng region is responsible for the difference of Th1/Th2 balance described above for the SJL/J and B10.S strains, we examined the levels of IFN-γ, IL-2, IL-10, and IL-12 mRNA in the brains of SJL/J mice and of the two lines of resistant congenic mice during the first week postinoculation. As shown in Fig. 5, the levels of IFN-γ and IL-2 mRNA and the ratio of IL-12 to IL-10 mRNA were the same for the SJL/J and SJL.10(70/14) mice. The same result was obtained with SJL.10(233/237) congenic mice (data not shown). Therefore, the Ifng region is not responsible for the difference of Th1/Th2 balance observed between the SJL/J and B10.S strains, although it accounts for most of the difference of susceptibility between these strains (Fig. 4). These results demonstrate that the Th1/Th2 balance is not the major factor in resistance/susceptibility to persistent infection in this system.
In the past, the fact that IFN-γ limits an infection and that there is a correlation among mouse strains between Th1/Th2 responses and resistance/susceptibility has been taken as evidence for the role of Th1 responses in resistance. Our results illustrate how such an assumption can be improper in the absence of direct genetic proof, such as that provided by congenic animals.
IFN-γRo/o animals (6) and treatment of resistant mice with an anti-IFN-γ Ab (7) showed that IFN-γ restricts Theiler’s virus expression during persistent infection. This finding is not in contradiction with the results reported here. In the present case, the effect of IFN-γ on late disease is the same for the SJL/J and the B10.S strains. Knockout mice and immunological manipulations such as treatments with neutralizing mAbs uncover genes and biochemical pathways with essential roles in pathogenesis but which are not necessarily responsible for differences of susceptibility between strains.
T cell proliferation and IL-2 production after in vitro stimulation
The proliferative response of T lymphocytes after stimulation with anti-CD3 mAbs is controlled by a gene that has been mapped to the region containing the D10Mit14 marker (19). To examine whether this gene was involved in the control of the persistence of Theiler’s virus, we compared the proliferative response of T cells from the SJL/J, B10.S, SJL.10(70/14), and SJL.10(233/237) strains. As shown in Fig. 6, T cells from B10.S mice proliferated more and secreted more IL-2 than T cells from the SJL/J or SJL.10(233/237) strains; there were no differences when comparing T cell proliferation and IL-2 mRNA expression for the SJL/J and SJL.10(233/237) mice. The same result was obtained with the SJL.10(70/14) congenic strain (data not shown). Therefore, the gene that controls the T cell-proliferative response is not in the Ifng region and must be different from that which controls the susceptibility to Theiler’s virus persistent infection.
In summary, the Ifng gene was a good candidate to explain the difference of susceptibility of the SJL/J and B10.S mice to persistent infection by Theiler’s virus. We found a strategically placed deletion in the promoter of the gene of the susceptible SJL/J strain and observed that the response of the resistant B10.S strain to the infection was more of a Th1 type than that of the susceptible SJL/J strain. However, the study of mice congenic for the region ruled out a role for the Ifng gene and the Th1/Th2 balance in the difference of susceptibility between these strains. These studies emphasize how the existence of a correlation between Th1/Th2 responses and resistance/susceptibility to an infectious agent does not demonstrate a causal relationship.
This study was supported by grants from the Institut Pasteur Fondation, the Centre National de la Recherche Scientifique, the National Multiple Sclerosis Society of the United States, the European Community Human Capital and Mobility Program (Contract CHRX-CT94-0670), and the Association pour la Recherche sur la Sclérose en Plaques. P.M. is a Scientific Collaborator of the Fonds National de la Recherche Scientifique, Belgium.