IRF7-KO mice on C57BL/6 (B6) background were purchased from Riken BioResource Center (Tsukuba, Japan) and maintained as a breeding colony. Control wild-type B6 mice, which have been shown to be appropriate controls for EAE studies 17, were obtained from Taconic (RY, Denmark). Mice were provided with food and water ad libitum. All experiments were approved by the Danish Justice Ministry Committee on Animal Research (Approval Number 2009/561-1724).

To induce EAE, adult female IRF7-KO and control B6 mice were subcutaneously immunized with 35-55 myelin oligodendrocyte glycoprotein (35-33 MOG) peptide in complete Freund's adjuvant containing 2 mg/ml M. tubercolosis in the flanks. In addition, mice received intraperitoneal injections with 200 μl pertussis toxin (1,5 μg/ml) (Sigma-Aldrich, Brøndby, Denmark) at the time of immunization and two days later. Mice were then caged in groups of 8 (4 WT mice with 4 IRF7-KO mice). The mice were weighed and monitored daily for clinical signs of EAE, which were scored as follows: 0, no symptoms; 1, Weak or hooked tail; 2, Floppy tail; 3, 2 + hind limb paresis (weak hind limbs-assessed by the animal's slowness or splaying limbs when walking or unsteady walk in the cage or on the lid of the cage), Grade 4: 3 + very weak hind limbs or one hind limb paralysed- hind limb paresis-assessed by the animal dragging one or both hind limbs (not complete loss of tonus in one or both hind limbs); 5, 4 + unilateral hind limb paralysis (both hind limbs paralysed); 6, 5 + paresis in one forelimb. Because of ethical reasons, mice were euthanized when they reached a clinical score of 5. In the first experiment the clinical score in first euthanized 4 mice was 3-4, and all other euthanized mice had clinical score 5. At the end of experiment all remaining mice were euthanized. In the second and third experiment, half of the mice were sacrificed at day 15 and the rest either when they scored 4-5 or at the end of experiment. Mice were weighed and scored in a blinded manner.

Mice were deeply anaesthetized and perfused intracardially with ice-cold Phosphate Buffered Saline (PBS). Spinal cords were dissected out and processed as followed:

For Histology: the tissues were placed in 4% paraformaldehyde (PFA) (Sigma-Aldrich) for 60 minutes and overnight in 1% PFA at 4°C. The tissues were then placed in 20% sucrose solution overnight, freeze-embedded in cryo-embedding (Ax-lab, Vedbæk, Denmark), cut in 16-μm cryostat sections, mounted on glass slides and stored at -80°C.

For Flow Cytometry: the tissues were placed in a plate with Hanks Balanced Salt Solution (HBSS) (Invitrogen A/S, Taastrup, Denmark) for further processing.

For Quantitative real-time reverse transcriptase- PCR assay: the tissues were placed in eppendorf tubes containing TRIzol (Invitrogen Life Technologies, Paisley, Scotland, UK), which were then stored in -80°C until further processing.

To investigate the extent and distribution of histopathology Hematoxylin and Eosin staining was performed. Double Immunostaining was used to detect astrocytes and T-cells. In brief, sections were washed in PBS, followed by rinsing in PBS-0.5% Triton (Triton- X-100) (Sigma-Aldrich) (PBST) and blocked in a solution containing PBST and 3% BSA (Sigma-Aldrich). Thereafter, sections were incubated with Cy-3 conjugated mouse anti GFAP antibody (C9205, Sigma-Aldrich), and Rat anti-mouse CD3 (MCA500G, Serotec) antibodies, in order to detect astrocytes and T cells respectively. After several washes in PBST, sections were incubated with donkey anti-rat Alexa Fluor-488 antibody (Invitrogen- Molecular Probes, Taastrup, Denmark), to visualize anti-CD3 antibody. Nuclei were then stained with DAPI (Invitrogen-Molecular Probes). To test the specificity of staining, control sections were treated without primary antibody or with isotype-matched primary antibodies. Control sections displayed no staining comparable with that seen without primary antibodies (not shown). Images were acquired using an Olympus BX51 microscope (Olympus, Denmark) connected to an Olympus DP71 digital camera, and combined using Adobe Photoshop CS version 8.0 to visualize double-labeled cells.

Single cell suspensions of spinal cords and lymph nodes (LN) were prepared by dissociation using a 70 μm cell strainer (BD Biosciences, Brøndby, Denmark). Spinal cord samples were resuspended in 37% Percoll (GE Healthcare Bio-sciences AB, Uppsala, Sweden) and centrifuged to remove myelin. Blocking was performed using Mouse Fc Block (BD Biosciences). Cells were stained with biotinylated anti-mouse CD8, FITC anti-mouse CD4 or PerCP/Cy5.5 anti-mouse CD11b and phycoerythrin (PE) anti-mouse CD45 (BD Biosciences). Data was collected on a FACS Calibur (BD Biosciences), and analyzed using Flowjo software (Tree Star, Ashland, OR).

Single cell suspensions prepared as described above were plated in 96 well plates coated with anti-mouse CD3ε (145-2C11) and cultured for 9 hours to stimulate cytokine production in T cells. GolgiPlug (BD Biosciences) was added two hours after plating. After incubation, cells were washed and stained with V500-rat anti mouse CD4 (BD), PerCP/Cy5.5 anti-mouse CD8α (Biolegend) and either allophycocyanin (APC)-anti-mouse CD196 (CCR6) (Biolegend) or biotin anti-mouse CD183 (CXCR3) (Biolegend) and APC-Streptavidin (BD Biosciences). Intracellular cytokine staining was performed using a Cytofix/Cytoperm kit (BD). PE-rat anti-mouse IL17A (BD Biosciences), PE/Cy7 anti-mouse IFNγ (Biolegend) were used to detect the cytokines. Data was collected on an LSR II (BD Biosciences), and analyzed using FACS DIVA software (BD).

Samples were prepared as described above and stained with V500-rat anti mouse CD4 (BD Biosciences), PerCP/Cy5.5 CD11b (Biolegend), PerCP/Cy5.5 anti-mouse CD8α (Biolegend), and either APC anti-mouse CD196 (CCR6) (Biolegend) or Biotin anti-mouse CD183 (CXCR3) (Biolegend) and APC-Streptavidin (BD Biosciences). Cells were sorted on a FACSVantage/Diva cell sorter (BD Biosciences).

Total RNA was purified using TRIzol RNA isolation reagent (Invitrogen Life Technologies) according to the manufacturer's protocol for whole tissue RNA extraction. One μg of RNA from each spinal cord sample was incubated with Moloney murine leukemia virus RT (Invitrogen Life Technologies) according to the manufacturer's protocol, using random hexamer primers. Quantitative Real-Time Reverse Transcriptase- PCR assay (Quantitative RT-PCR) were performed using ABI Prism 7300 Sequence Detection Systems (Applied Biosystems, Foster City, CA). Quantitative RT-PCR was performed for IRF7, CCL2, CXCl10, TNF-α, IL-1β, IFNγ, IL17 and CD3, using primers and probes as described previously 818. 18s rRNA primers and probes (Applied Biosystems) were used as an endogenous control to account for differences in the extraction and RT of total RNA 8. Each reaction was performed in 25 μl with TaqMan 2× Universal PCR Master Mix (Applied Biosystems), undiluted cDNA, primers, TaqMan probe, and 2× filtered sterile milliQ water. For all genes, PCR conditions were 2 minute at 50°C, 10 minutes at 95°C followed by 40 cycles each consisting of 15 seconds at 95°C and 1 minute at 60°C. To determine the relative RNA levels within the samples, standard curves for the PCR were prepared using cDNA from a reference sample and making fourfold serial dilutions. Relative expression values were then calculated by dividing the expression level of the target gene by the expression level of 18s rRNA.

Data were analyzed by nonparametric, Mann-Whitney t-test using GraphPad Prism software (GraphPad Software Inc., San Diego, California, USA). A p value < 0.05 was considered to be statistically significant. Data are presented as Mean ± SEM.

IRF7 gene expression was measured in spinal cords from WT mice that had been immunized with MOGp35-55+CFA. The results from three experiments are combined in Figure 1 and 1 show that induction of EAE leads to increased IRF7 gene expression. In addition the up-regulation of IRF7 mRNA correlated with the clinical score (Figure 1A). Consistent with the well-known widespread expression of Type I IFN and its response elements, as well as with previous studies 7816, we found expression of IRF7 mRNA by Th1 and Th17 CD4+ T cells, and by macrophages and microglia (additional file 1 and Figure 1B). IRF7 gene expression increased in CD45dimCD11b+ microglia during the course of EAE nearly reaching the levels seen in CD45highCD11b+ myeloid cells infiltrating the CNS (Figure 1B).

To assess the role of IRF7 in EAE, we immunized IRF7-deficient mice with MOG in CFA with pertussis toxin, a standard protocol for the induction of EAE. In four independent experiments, the mean time of disease onset was not significantly different between WT (12.99 ± 1.0 day) and IRF7-KO (12.07 ± 0.7 day) (Table 1). However, the incidence of EAE differed, being 29/39 (74%) versus 28/30 (93%) in WT and IRF7-KO mice, respectively (Table 1). Nearly half of the animals with EAE were euthanized at grade 5 in the IRF7-KO group, compared to only 4/29 of those in the WT group (Table 1). Whereas the number of mice with EAE that did not achieve grade 3 was almost 50% in WT groups, less than a quarter of IRF7-KO mice failed to reach this level of severity (Table 1). Results from one experiment are shown as mean clinical scores in Figure 2A. IRF7 deficient mice developed significantly more severe EAE symptoms (Figure 2A). The need for ethical reasons to euthanize mice with severe disease disallowed study of disease progression in severely affected animals and may have obscured a statistically significant difference in severity. Increased disease severity was paralleled by significantly greater loss of whole body weight (Figure 2B).

We then investigated the effect of IRF7 gene deletion on the infiltration of cells into the CNS. Flow cytometric analysis showed the clinical score of both WT and IRF7-deficient mice was correlated to the number of infiltrating blood-derived cells, as expected. Blood-derived CD45^highCD11b⁺ macrophages were discriminated by their higher level of expression of CD45 from CNS-resident microglia (Figure 3A). The total number of infiltrating CD45^highCD11b⁺ macrophages (p < 0.022, Figure 3A, D), CD4+ (p < 0.022, Figure 3B, E) and CD8+ T cells (p < 0.014, Figure 3C, F) was higher in IRF7-KO compared with WT CNS with more severe EAE.

Fluorescence microscopy was used to localize infiltrating cells in the CNS. CD3+ T cells were increased in number and more diffusely dispersed in white matter in the spinal cord of IRF7-deficient mice with EAE, compared to the more focal and constrained infiltration pattern in WT spinal cord (Figure 4). In contrast to T cells, there was no apparent effect of IRF7-deficiency on numbers or distribution of GFAP+ astrocytes (Figure 4A, B). We further examined CD3ε, IFNγ and IL17 gene expression by quantitative RT-PCR. The content of CD3ε and IFNγ mRNA was increased in spinal cords from both IRF7-deficient and WT mice with EAE, but no significant differences could be measured when CD3ε (p < 0.2086) and IFN-γ mRNA (p < 0.0649) were compared between IRF-7 deficient and WT mice (Figure 4C, D). In contrast, IL-17 gene expression was higher (p < 0.0087) in IRF7-KO spinal cord than in spinal cords from WT mice with EAE (Figure 4E).

To further investigate the role of IRF7 on Th1 and Th17 cells during EAE, we measured IFNγ and IL17 production by CD4+ T cells from spinal cords and LN. IRF7 deficiency did not affect percentages of T cells producing these cytokines in spinal cord (not shown). However, lack of IRF7 resulted in an increase (p < 0.006) in the percentage of CD4+IFNγ+ cells (Figure 5), but not in CD4+IL17+ T cells (not shown) in LN.

We next examined whether lack of IRF7 affected the expression of inflammatory mediators that are known to be involved in induction and regulation of EAE. Levels of the chemokines CCL2, CXCL10, and cytokines IL-1β and TNF-α increased in EAE and correlated to clinical severity (not shown). Quantitative RT-PCR analysis showed that the increases in expression of CCL2 (p < 0.0176, Figure 6A) and CXCL10 (p < 0.0111, Figure 6B), IL-1β (p < 0.0260, Figure 6C) and TNF-α (p < 0.0530, Figure 6D) were higher in IRF7-deficient spinal cord than in spinal cord from WT mice with EAE.