Background
Theiler’s murine encephalomyelitis virus (TMEV)-induced demyelinating disease has been extensively studied as a relevant animal model for human multiple sclerosis (MS). Intracerebrally infected susceptible SJL/J mice develop neuropathological symptoms that exhibit many immunological and pathological similarities to MS and thus serve as an infectious model for this immune-mediated disease [
1,
2]. Immune responses have been implicated in the development of TMEV-induced demyelinating disease [
3‐
5], although the relative importance of virus-specific over autoimmune responses to disease development remains unclear. The majority of immune responses to TMEV are against the capsid proteins encoded by the P1 region in the TMEV-induced demyelinating disease [
6‐
8]. We previously showed that SJL/J mice expressing the TMEV P1 region mounted lower levels of viral epitope-specific CD4
+ and CD8
+ T cell responses throughout the course of viral infection and developed less severe disease [
8]. Lower levels of virus-specific T cell responses in P1-transgenic SJL/J mice led to higher viral persistence compared to their littermates. However, the level of disease development was also dampened due to the reduced number of pathogenic T cells in the transgenic mice. In contrast to the abundant immunological studies on the P1 proteins, very few studies have investigated the effects of the P2 and P3 regions (non-structural proteins).
We previously demonstrated that the majority CD4
+ T cell responses following TMEV infection in susceptible SJL mice but not in resistant C57BL/6 mice targeted the 3D polymerase encoded by P3 [
9]. This result was consistent with an early study that indicated that transgenic expression of the TMEV P2/P3 region did not alter the resistance of C57BL/10 mice (H-2
b) to TMEV-induced disease [
10]. However, their subsequent studies with transgenic mice expressing TMEV 3D in the susceptible FVB background demonstrated resistance to the pathogenesis of demyelinating disease [
11,
12]. Therefore, it is conceivable that 3D-transgenic FVB mice may be self-tolerant to the induction of potentially pathogenic T cell responses. Investigations into the potential contributions of P2/P3-specific immune responses to the development of TMEV-IDD will be valuable for elucidating the potential roles of the TMEV P2 and P3 regions in the pathogenesis of demyelinating disease.
To investigate the effects of the P2/P3 genes in the development of TMEV-IDD, we generated transgenic mice expressing the P2 and P3 regions of TMEV under the control of the hCMV promoter. P2/P3-Tg mice in the resistant B6 background cleared the viral load similar to the control B6 mice. In contrast, P2/P3-Tg mice in the susceptible SJL/J background displayed higher viral loads in the central nervous system (CNS) but less pathogenesis. T cell responses to the P1 (structural capsid) and P2/P3 (non-structural protein) epitopes were significantly reduced in the P2/P3-Tg mice regardless of their backgrounds. Thus, the reduction in both protective and pathogenic T cell responses may cause such increased viral loads but decreased pathogenesis. Interestingly, macrophages and astrocytes from P2/P3-Tg mice infected with TMEV in vitro produced reduced levels of IL-6 and viral messages compared to the cells from naïve control mice. Additionally, NIH-3T3 cells transfected with P2, P3, or P2/P3 showed marked inhibition of viral particle production accompanied by an increase in the type I IFN response compared to untransfected or control plasmid-transfected cells. The inhibition of viral replication in cells from P2/P3-Tg mice appeared to be somewhat virus-specific because the replication of TMEV and closely related encephalomyocarditis virus (EMCV) was inhibited, whereas the inhibition of the replication of unrelated vascular stomatitis virus (VSV) was transient. These results strongly suggest that transgenic expression of non-structural viral genes may be able to induce innate immunity, which inhibits the initial replication of related viruses. Thus, these results open the possibility that gene expression of a part of the viral genome could be used to prevent the virally induced chronic disease caused by related viruses.
Methods
Animals
P2/P3-expressing transgenic (B6 X SJL)F1 founders were generated at the transgenic core facility of Northwestern University. A transgenic founder F1 mouse was bred onto the C57BL/6 and SJL/J backgrounds for more than eight generations in a specific pathogen-free facility prior to use. The breeder C57BL/6 and SJL mice were purchased from Harlen Laboratories (Indianapolis, IN, USA). All experiments were conducted with 6- to 10-week-old females. Experiments using animals were conducted according to the permission (#2011-1316 for Byung Kim) of the Animal Care and Use Committee at Northwestern University.
Synthetic peptides and antibodies
All peptides were purchased from GeneMed (GeneMed Synthesis Inc, South San Francisco, CA, USA) and used as described previously [
13]. All antibodies were purchased from BD Pharmingen (San Diego, CA, USA).
Viruses and cell lines
The BeAn strain of TMEV used in this study was propagated and titered in BHK-21 cells grown in Dulbecco’s modified Eagle medium supplemented with 7.5 % donor calf serum. For intracerebral (i.c.) infection, 30 μl of virus solution containing 3 × 106 PFU was injected into the right cerebral hemisphere of 6–8-week-old mice anesthetized with isoflurane. Clinical symptoms of disease were assessed weekly using the following grading scale: grade 0 = no clinical signs; grade 1 = mild waddling gait or flaccid tail; grade 2 = severe waddling gait; grade 3 = moderate hind limb paresis; and grade 4 = severe hind limb paralysis.
Reverse-transcription PCR
Total cellular RNA was isolated from various tissues of P2/3 transgenic mice, including brain, spinal cord, spleen, thymus, liver, and kidney, using the Trizol® reagent (Invitrogen, CA, USA). First-strand complementary DNA (cDNA) was synthesized from 1 μg of total RNA by utilizing the SuperScript® III First-Strand Synthesis Supermix (Invitrogen, CA, USA) at 55 °C. The relative concentrations of cDNA were equalized among the groups based on the level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) amplification (25 cycles) by polymerase chain reaction (PCR). Primers for the control GAPDH gene and P2/P3 transgene were purchased from Integrated DNA Technologies: GAPDH, 5′-AACTTTGGCATTGTGGAAGG-3′ and 5′-ACACATTGGGGGTAGGAACA-3′ and the P2/P3 transgene (3D), 5′-CTGCAATTGGAACTGACCCAGATG-3′ and 5′-ATGTCGTGACACAG-CCAGAGAT-3′.
Plaque assay
Virus titers in the infected CNS tissues were enumerated using a standard plaque assay on BHK-21 cell monolayers [
14]. To visualize the plaques, the monolayer was stained using 0.1 % crystal violet after fixation with methanol.
Isolation of CNS-infiltrating MNC
Mice were perfused with sterile Hank’s balanced salt solution (HBSS), and the excised brains and spinal cords were homogenized. CNS-infiltrating mononuclear cells (MNCs) were enriched in the 1/3 bottom fraction of a continuous 100 % Percoll (Pharmacia, Piscataway, NJ, USA) gradient after centrifugation for 30 min at 27,000×
g, as described previously [
15].
Intracellular staining of cytokine production
Freshly isolated CNS-infiltrating MNCs were cultured in 96-well round bottom plates in the presence of relevant or control peptides as described previously [
16]. Allophycocyanin-conjugated anti-CD8 (clone Ly2) or anti-CD4 (clone L3T4) antibodies and a PE-labeled rat monoclonal anti-IFN-γ (XMG1.2) antibody were used for intracellular cytokine staining. Cells were analyzed on a Becton Dickinson FACS Calibur or FACS Sort flow cytometer. Live cells were gated based on light scatter properties.
IFN-γ and IL-17 ELISA
Mouse IFN-γ and IL-17 ELISA kits were purchased from BD Biosciences (San Diego, CA, USA) and R&D Systems, Inc. (Minneapolis, MN, USA), respectively. Cytokine levels in splenic culture supernatants were assessed according to the manufacturer’s instructions. Briefly, diluted samples were incubated for 2 h with plate-bound capture antibodies after blocking for 1 h. Cytokine expression levels were visualized using HRP-conjugated detection antibodies in the presence of the HRP substrate TMB (BioFX Laboratories, Owings Mills, MD, USA). The absorbance was measured at 450 nm.
T cell proliferation assay
The proliferative activity of epitope-specific CD4+ T cells was measured based on [3H]thymidine incorporation levels. Spleen cells from P2/P3 transgenic and control mice 7 dpi were stimulated with PBS, the CD4 peptide mix, or plate-bound anti-CD3/CD28 antibodies for 3 days, then pulsed with 1 μCi [3H]thymidine-deoxyribose for 18 h. Cells were harvested and 3H incorporation was measured using TopCount. Data are expressed as CPM ± SD of triplicates.
Generation of H-2Ks tetramers
H-2K
s tetramers were generated as previously described [
17]. Briefly, H-2K
s and the human β2-microglobulin gene were subcloned into the pET28 bacterial expression vector. BL21/DE3 competent cells were transformed and protein expression was induced with IPTG for 4–5 h. Inclusion bodies were purified and refolded in the presence of peptides. The soluble monomeric form of the peptide-MHC complex was biotinylated with BirA at room temperature and tetramerized with streptavidin-PE (Invitrogen, Carlsbad, CA, USA).
Virus replication assay
Plasmids (pMIG) containing the non-structural genes P2, P3, or P2/P3 were generated after inserting the corresponding genes cloned from pSBW. Phoenix cells were transfected with Lipofectamine-2000 (Invitrogen, Carlsbad, CA, USA) to generate retroviruses. NIH-3T3 cells were infected with the recombinant retroviruses at room temperature for 90 min. After 48 h, eGFP-positive cells in the defined gate were sorted using the FACSAria flow cytometer (BD Biosciences, San Diego, CA, USA). After 2 weeks of rest, the cells were infected with TMEV at a multiplicity of infection (MOI) of 10 for 6, 12, 18, and 24 h. Supernatants were collected at each time point for plaque assays. For some experiments, real-time PCR was used to determine gene expression levels. The lowest expression level was set as onefold expression.
Statistical analysis
Data are shown as the mean ± SD of two to three independent experiments or triplicates of one representative experiment from at least three independent experiments. The significance of differences in the mean values was determined by Student’s t test. P values < 0.05 were considered statistically significant.
Discussion
The expression of transgenes in mice has been utilized as an important tool to investigate various immune mechanisms, such as unresponsiveness, autoimmunity, and T cell development [
18‐
20]. We previously demonstrated that transgenic mice carrying the TMEV P1 gene encoding a viral structural protein displayed reduced virus-specific immune responses, resulting in reduced pathogenesis in the transgenic mice despite the increased viral loads in the CNS [
8]. Therefore, viral persistence does not always lead to TMEV-IDD, although viral persistence has long been believed to have a major impact on the development of TMEV-IDD [
21].
In the present study, we established P2/P3-Tg mice to dissect the role of non-structural proteins in the development of TMEV-induced demyelinating disease. We demonstrated that P2/P3-Tg mice (particularly in the susceptible SJL background) displayed elevated viral persistence in the brain and later in the spinal cord (Fig.
2). However, P2/P3-Tg mice developed significantly less severe and less frequent disease (Fig.
2). These data suggest that viral persistence may not directly correlate with disease development. This finding is consistent with our previous data demonstrating that P1-Tg SJL/J mice with higher viral loads at all stages of viral infection compared to their littermates developed less severe disease [
8]. The lack of correlation between viral persistence and disease development implies that viral persistence alone may not be able to drive the development of TMEV-IDD. Indeed, TMEV-infected P2/P3-Tg mice mounted lower levels of anti-TMEV T cell responses (Figs.
3,
4, and
5). The lower levels of protective antiviral T cell responses in the P2/P3-Tg mice under persistent inflammatory environment may have contributed to the elevated viral persistence in the CNS.
The P2/P3-Tg mice were expected to mount unaffected immune responses specific to the epitopes encoded by the P1 genes. However, the P2/P3-Tg SJL mice exhibited severely compromised virus-specific T cell responses to the structural epitopes encoded by the P1 region (Fig.
3b,
c). The presence of excessive levels of type I IFNs is known to exert severe suppression of adaptive immune responses in SJL mice via the inhibition of development and function of antigen-presenting cells [
22]. Therefore, the excessive production of type I IFNs in the cells expressing P2 and/or P3 genes, as well as P2/P3-Tg SJL mice (Figs.
6 and
8) most likely results in the compromised adaptive immune responses to viral antigens, which are responsible for the pathogenesis and the protection. Furthermore, the cells expressing the P2 and/or P3 genes produce lower levels of IL-6 upon TMEV infection (Fig.
6), which may curtail the generation of pathogenic T cells [
23,
24]. Collectively, the lower initial viral replication leading to the low level of anti-viral T cell responses in P2/P3-Tg mice combined with the low level of pathogenic T cell responses may result in elevated viral persistence in the absence of the pathogenesis of demyelinating disease. In contrast to the viral persistence in the infected P2/P3-Tg SJL mice (Fig.
2), cells expressing the P2 and/or P3 genes from the P2/P3-Tg mice or cells transfected with P2 or P3 displayed reduced TMEV replication in vitro (Figs.
6 and
8). Because the viral replication experiments in vitro do not involve adaptive anti-viral immunity, the viral replication level in vitro does not correspond to viral persistence in the infected mice where strong adaptive immune responses are necessary to control viral persistence. Therefore, the protective type I IFNs that are elevated in the P2/P3-expressing cells may be sufficient to hinder viral replication in vitro.
Interestingly, the inhibition of viral replication is partially preferential for the same family of viruses because TMEV and EMCV, which belong to the same cardiovirus genus of the picornavirus family, are selectively inhibited compared to VSV, which is not a member of picornaviridae (Fig.
7). The underlying mechanism of viral replication inhibition is not clear. The elevated production of type I IFNs upon viral infection in cells from P2/P3-Tg mice (Fig.
8) suggests that these cytokine levels may affect viral replication. Recently, the expression of TMEV 3D in susceptible FVB mice showed an antiviral effect through elevated type I interferon expression [
12]. Although the immune responses were not assessed in the TMEV DA strain 3D-Tg FVB mice, they concluded that adaptive immune responses are not an important contributing factor because 3D-expressing Rag-deficient FVB mice, which lack immune responses, provided protection. However, our results obtained with P2/P3-Tg mice (Figs.
3,
4, and
5) indicate that virus-specific adaptive immune responses also contribute to the pathogenesis of TMEV-induced demyelinating disease. Our attempts to individually express the 2A, 2B, 2C, 3A, 3C, and 3D genes to assess their inhibitory functions in the mouse cell line were not successful, suggesting that uncleaved non-structural proteins or contiguous viral genome sequences may be important for the inhibitory function (data not shown).
The protein encoded by the 3A gene in the P3 of poliovirus inhibits protein secretion by blocking endoplasmic reticulum-to-Golgi transport [
25,
26]. This inhibition of protein secretion may play a role in the immune-evading mechanism of poliovirus through the down-regulation of MHC class I-dependent viral antigen presentation [
27] and the reduction of innate cytokine secretion [
28]. However, the inhibitory ability of the TMEV 3A protein was absent [
29]. Interestingly, the uncleaved 2BC protein encoded by the P2 genes (rather than 3A) of foot-and-mouth disease virus (FMDV) exhibited a similar inhibitory function [
30]. Therefore, 2BC of TMEV, which is similar to FMDV with respect to its resistance to bredfeldin A [
30], may play an important role in immune evasion by dampening virus-specific T cell responses. In addition to P2, the expression of the P3 and P2/P3 genes also inhibited viral replication over twofold compared to the controls (Figs.
6 and
8). Therefore, a protein encoded by the genes or a viral genome segment of P3 appears to inhibit viral replication independently of P2. It is conceivable that the 3D region of TMEV in the P3 segment may inhibit viral replication as previously reported [
12]. A very recent study using the TMEV DA 3D-Tg model has suggested that the catalytic site of 3D protein is associated with elevated production of type I IFNs [
31]. Interestingly, cells expressing the P2 or P3 genes displayed elevated production of type I IFNs upon viral infection (Fig.
8). This result is consistent with the previous observation with 3D-Tg mice. However, our results with TMEV BeAn P2/P3-Tg mice suggest that the P2 region, in addition to the P3 (including 3D) is also capable of inducing the production of type I IFNs. Although we failed to detect any measurable levels of viral proteins encoded by the P2 and P3 genes based on the polyclonal rabbit antibodies to the primary amino acid sequences of the N-termini of 2A, 2B, and 3D, the potential involvement of the unfolded protein response (UPR) to the P2/P3 products, which is capable of inducing type I IFNs [
32,
33], cannot be excluded. Nevertheless, distinct mechanisms must be operational between the P2 and P3 gene products in the induction of innate immune responses.
The underlying mechanism responsible for the elevated type I IFN production in cells expressing the P2 and/or P3 genes is not clear. It is conceivable that some host pattern-recognizing molecules (e.g., TLRs such as TLR7/8 and TLR9) recognize either the viral proteins or nucleic acids encoded by the P2/P3 region and consequently trigger innate cytokine production [
34]. We previously showed that TMEV infection triggers TLR-mediated innate cytokine production in mice via interaction with viral double-stranded RNA (dsRNA) replication intermediates [
35‐
37]. In addition, we demonstrated that TMEV infection triggers innate immunity in the host via melanoma differentiation-associated 5 (MDA-5) protein [
38], which involves a group of intracellular responders, the retinoic acid-inducible gene I-like receptors (RLR) [
39]. Signaling pathways of dsRNAs, non-self RNAs, and certain DNAs share the involvement of IPS-1/MAVS, which is associated with RLR-mediated signaling [
40‐
42]. The protection against viral infection in the TMEV 3D-Tg model, which is dependent on the presence of MDA-5 or MAVS signaling [
31], supports this possibility. Because viral DNA is transcribed in transgenic cells to produce viral RNAs encoding the P2 and P3 regions (Fig.
2), structural features of viral RNA branches in the P2 and P3 region of TMEV may be recognized by MDA-5 as previously proposed [
43‐
45]. This possibility is supported by the fact that similar inhibition of viral replication is observed in cells containing the P2 and/or P3 regions as the RNA form after infection with the recombinant retroviruses (Fig.
8). This possibility is also consistent with the results indicating that cells expressing TMEV genes inhibit not only TMEV replication but also replication of closely related EMCV compared to that of unrelated VSV (Fig.
7). Nevertheless, viral DNA and/or RNA transcripts from the viral transgenes may trigger innate cytokine production via TLR-dependent and/or independent pathways [
34,
46‐
48]. Further studies on the mechanism underlying the non-structural gene-mediated antiviral effect might lead to the development of powerful new antiviral drugs or treatments.