A Defined Diet Combined with Sonicated Inoculum Provides a High Incidence, Moderate Severity Form of Experimental Autoimmune Encephalomyelitis (EAE)

Abstract

Background: Myelin oligodendrocyte glycoprotein 35-55 (MOG_35-55)-peptide induced experimental autoimmune encephalomyelitis (EAE) is a model for inflammation of the brain and spinal cord. However, its severity and incidence vary within and between laboratories. Severe scores can lead to premature termination and are both unnecessary for readouts and detrimental to animal welfare. Ideally, the model would have high incidence, moderate severity, and low interindividual variability to fulfill the "Refine" aspect of the 3R concept. Nevertheless, most efforts to increase incidence also increase the severity. When the effects of potential therapies are tested, moderate severity is sufficient to detect useful drug effects as long as variation is low. Low variation can also reduce group sizes, which supports the "Reduce" aspect of 3R approaches in disease modeling. We set out to reduce variation and control severity by assessing the effects of mouse age, dietary fiber, antigen emulsion, and the dose of MOG and pertussis toxin on incidence, variability, and severity in the MOG-EAE model.

Methods: We compared 14- and 33-week-old female C57BL/6 mice and varied the diet and inoculum in two studies. We measured disease signs in vivo as well as gene expression in the brain and spinal cord and histology by immunofluorescence. Ordinary one-way ANOVA was used for multiple comparisons.

Results: The most reliable induction conditions were with a low-fermentative/fiber diet (AIN 93M) combined with a sonicated emulsion of the MOG_35-55-peptide. High-dose pertussis toxin increased EAE severity and incidence in 14-week-old mice (25% survival) while being more moderate in mature mice (100% survival). Varying all parameters suggests that duration of prefeeding defined diet, emulsion quality, and mouse maturity were factors that increase uniformity of response allowing incidence to reach 100% without excess severity. Microglia and astrocyte-associated markers were upregulated proportionally to score consistent with known EAE pathology.

Conclusions: A defined fiber/high-sugar diet with sonicated inoculum provides for a moderate severity, high incidence, and less variable EAE. The resulting uniformity in animal response and associated cytokine patterns, and the strong link to a defined diet, suggest that this may be a more clinically translatable protocol for the induction of EAE. This is consistent with reported effects of low-fermentable diets on immune modulation in human patients with autoimmune diseases.

Figure 1

Layout of variables assessed in these studies: age, diet, emulsion method, and MOG and PTX dose. Animals were terminated either at the end of the study (weeks 15 and 33) or as the satellite group at week 32 in study 2.

Figure 2

Effect of diet on in vivo results. (a) Summary of the variables tested (b, c) average body weight of study 1 and study 2 mice, respectively; (d, e) average score of study 1 and study 2 mice, respectively; (f, g) incidence (EAE score > 0) in study 1 and study 2; n = 8. Testing the effect of diet on body weight in study 1 yields a p value of 0.016 (*). Testing the effect of diet on score in study 1 yields a p value of 0.0002 (***).

Figure 3

Effect of emulsion preparation (inversion open circles, sonication filled circles) on the time course of EAE development in mice prefed with AIN 93 M for 6 (study 1) or 24 weeks (study 2). (a) Summary of the variables tested. (b, c) Average body weight of study 1 and study 2 mice, respectively; (d, e) average score of study 1 and study 2 mice, respectively; (f, g) incidence (EAE score >0) in study 1 and study 2; n = 8.

Figure 4

Effect of PTX dose on the time course of EAE development in 33-week-old mice prefed with AIN 93 M in study 2 (120 ng of PTX filled diamonds, 240 ng, filled squares). (a) Summary of the variables tested. (b) Average body weight; (c) average score; (d) incidence (EAE score >0); n = 8.

Figure 5

Gene expression analysis of EAE relevant markers in the brain of 33-week-old mice (study 2); (a) overview of study parameters evaluated by qPCR in study 2: diet VRF1 (cyan data points), AIN 93 M (orange data points), and normal diet/control diet (black data points) and MOG_35–55 preparation (300 μg MOG_35–55, 240 ng PTX; sonicated filled data points, inverted open data points); (b–f) quantitative analysis of gene expression levels of immune cell markers (TBX21, FoxP3, GFAP, GM-CSF and H2-Ab1); (f–k) quantitative analysis of gene expression of myelin markers (MBP and MOG) and of CSPG4, a marker for oligodendrocyte progenitor cells. Plotted are the normalized expression levels as fold change relative to the normal diet-fed control group. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001.

Figure 6

Gene expression analysis of EAE relevant markers in the spinal cord of 33-week-old mice (study 2); (a) overview of study parameters evaluated by qPCR in study 2: diet VRF1 (cyan data points), AIN 93 M (orange data points), and normal diet/control diet (black data points) and MOG_35–55 preparation (300 μg MOG_35–55, 240 ng PTX; sonicated filled data points, inverted open data points); (b–f) quantitative analysis of gene expression levels of immune cell markers (TBX21, FoxP3, GFAP, GM-CSF, and H2-Ab1); (f–k) quantitative analysis of gene expression of myelin markers (MBP and MOG) and of CSPG4, a marker for oligodendrocyte progenitor cells. Plotted are the normalized expression levels as fold change relative to the normal diet-fed control group. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001.

Figure 7

Histopathological examination of brains and spinal cords of female C57BL/6J mice (300 μg of MOG_35–55, 240 ng of PTX, sonicated, AIN 93 M and VRF1, study 2). (a) MBP and GFAP positive area on coronal brain sections of VRF1 and AIN 93 M-induced mice vs normal diet/control diet not-induced mice; (b) MBP and GFAP positive area in spinal cord sections of VRF1 and AIN 93 M-induced mice vs normal diet/control diet not-induced mice; (c) MOG_35–55 immunization resulted in inflammation and demyelination in brain and spinal cord. Sonicated-induced VRF1 fed mice displayed moderate inflammation (astrocytes—yellow) and mild demyelination (MBP—green) in brain and spinal cord vs severe inflammation and demyelination in AIN 93 M fed mice. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001. VRF1 results displayed in cyan; AIN 93 M displayed in orange.

Figure 8

Gene expression analysis of EAE relevant markers in the brain of 33-week-old female C57BL/6J mice (study 2) at 7 and 15 days after induction; (a) satellite animals (terminated 7 days after induction) compared to sonicated-induced C57BL/6J mice 300 μg MOG_35–55, 240 ng PTX (terminated 15 days postinduction), and not-induced healthy animals (normal diet/control diet fed) on expression levels in the brain. (b–f) Quantitative analysis of gene expression levels of immune cell markers (TBX21, FoxP3, GFAP, GM-CSF, and H2-Ab1); (f–k) quantitative analysis of gene expression of myelin markers (MBP and MOG) and of CSPG4, a marker for oligodendrocyte progenitor cells. Plotted are the normalized expression levels as fold change relative to the normal diet-fed control group. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001. VRF1 results displayed in cyan; AIN 93 M displayed in orange. Lighter shades of full circles represent termination after 7 days, and darker shades of circles represent termination after 15 days.

Figure 9

Gene expression analysis of EAE relevant markers in the spinal cord of female C57BL/6J mice (study 2) at 7 and 15 days after induction; (a) satellite animals (terminated 7 days after induction) compared to sonicated-induced C57BL/6J mice 300 μg MOG_35–55, 240 ng PTX (terminated 15 days postinduction), and not-induced healthy animals (normal diet/control diet fed) on expression levels in the spinal cord. (b–f) Quantitative analysis of gene expression levels of immune cell markers (TBX21, FoxP3, GFAP, GM-CSF, and H2-Ab1); (f–k) quantitative analysis of gene expression of myelin markers (MBP and MOG) and of CSPG4, a marker for oligodendrocyte progenitor cells. Plotted are the normalized expression levels as fold change relative to the normal diet-fed control group. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001. VRF1 results displayed in cyan; AIN 93 M displayed in orange. Lighter shades of full circles represent termination after 7 days, and darker shades of circles represent termination after 15 days.

Figure 10

Gene expression analysis of EAE relevant markers in the brain of 6-week-old female C57BL/6J mice; (a) effect of diet VRF1 vs AIN 93 M vs normal diet/control diet) on not-induced, naïve female C57BL/6J mice in the brain. (b–e) Quantitative analysis of oligodendrocytes markers (CSPG4, Olig2) and myelin (MBP and MOG). (f–k) Quantitative analysis of gene expression of immune cell markers (H2-Ab1, GFAP, TBX21, FoxP3, GM-CSF, and IL10). Plotted are the normalized expression levels as fold change relative to the normal diet-fed control group. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001. VRF1 results displayed in cyan; AIN 93 M displayed in orange.

Figure 11

Gene expression analysis of EAE relevant markers in the spinal cord of 6-week-old female C57BL/6J mice. (a) Effect of diet VRF1 vs AIN 93 M vs normal diet/control diet on not-induced female C57BL/6J mice in the spinal cord. (b–e) Quantitative analysis of oligodendrocytes markers (CSPG4, Olig2) and myelin (MBP and MOG). (f–k) Quantitative analysis of gene expression of immune cell markers (H2-Ab1, GFAP, TBX21, FoxP3, and GM-CSF). Plotted are the normalized expression levels as fold change relative to the normal diet-fed control group. VRF1 results displayed in cyan; AIN 93 M displayed in orange.