C57BL/6N mice were bred and housed at the Franz-Penzoldt-Zentrum (FPZ), the animal care facility of the University Erlangen-Nuremberg (Germany), under a 12 h day-night-cycle and standardized environmental conditions receiving normal chow (ssniff V1534-300) and tap water ad libitum. All experiments were in accordance with the German laws for animal protection and were approved by the local ethic committees (AZ 55.2 DMS-2532-2-27). For a sugar-rich diet, mice received normal chow (ssniff V1534-300) and tap water containing 10% of D-galactose ad libitum 14 days prior to EAE induction and throughout the observation period. Control mice received normal tap water throughout the adaption and observation period. Otherwise, the galactose-rich and control diet were completely similar.

10–12 weeks-old mice were anesthetized and subcutaneously injected with 200 μg myelin oligodendrocyte glycoprotein (MOG_35–55; Charité, Berlin) and 200 μg Complete Freund’s Adjuvant (Difco) containing 4 mg/ml Mycobacterium tuberculosis (H37RA). Mice received 200 ng Pertussis toxin (List, Germany) intraperitoneally on the day of immunization and two days later. Clinical symptoms were assessed daily according to a 5-point scoring system (4). Body weights of mice receiving either a control or a galactose-rich diet were monitored starting 14 days prior to active immunization and during the course of MOG-EAE.

After perfusion with 4% paraformaldehyde (PFA) on days 16 and 34 post immunization (p.i.), spinal cord tissue was embedded in paraffin after 3 h post-fixation in 4% PFA. 4 µm-thick cross sections were deparaffinized in xylene and rehydrated in decreasing concentrations of ethanol before antigen retrieval in 1mM ethylenediaminetetraacetic acid (EDTA) or citrate buffer. Antigens were blocked with 10% BSA/PBS (bovine serum albumin in phosphate-buffered saline) before incubation with the primary antibody overnight at 4°C. Biotinylated secondary antibodies were applied for 60 min at room temperature followed by peroxidase-coupled avidin-biotin complex (ABC Kit, Vector Laboratories, Inc. Burlingame, CA). Reactivity was visualized with diamino-3,3’benzidine (DAB, Vector Laboratories, Inc. Burlingame, CA). Slides were counterstained using Mayer´s hemalaun solution (Merck, Darmstadt, Germany). The following antibodies were used: anti-CD3 1:200 (CD3-12, BIO-RAD Laboratories), anti-Mac-3 1:200 (M3/84, BD Pharmingen), anti-neurite outgrowth inhibitor A (NogoA) 1:200 (polyclonal; Santa Cruz Biotechnology), anti-oligodendrocyte transcription factor 2 (Olig2) 1:500 (rabbit polyclonal, Merck Millipore), anti-2’,3’-cyclic nucleotide 3’ phosphodiesterase (CNPase) 1:1000 (SMI-91; 1:7500, Sternberger Monoclonals via Covance, Princeton, USA) and anti-NeuN 1:100 (MAB2300, Merck Millipore). For analysis of axonal densities, Bielschowsky silver impregnation was used as described recently (20). Spinal cord cross sections were analyzed by a blinded observer using a BX-51 light microscope. Cellular infiltrates were counted in three lesions within each spinal cord segment (cervical, thoracic and lumbar) at 200x magnification within the margin of a stereological grid and quantified per square millimeter of white matter. Analysis of demyelinated areas in the white matter was performed semi-automatically with the help of CellP software. Axonal density was quantified in silver impregnated sections by counting on a 100 µm diameter grid. Data on axons are presented as relative axonal densities per grid (28).

EAE diseased mice were perfused with 4% PFA for 30 min followed by removal of the spinal cord. Segments of the lumbar spinal cord were immediately fixed overnight at 4°C in 4% glutaraldehyde/4% PFA in 0.1 M phosphate buffer/cacodylate buffer, pH 7.2. After washing with phosphate buffer, the tissue was post-fixed in phosphate buffer containing 1% osmium tetroxide and KaFECN(III) followed by washing with distilled water. Specimens were dehydrated in ascending concentrations of ethanol, embedded in epon and polymerized at 65°C for 72 h. Embedded spinal cord samples were cut at 80 nm with an ultra-microtome and analyzed using a Zeiss EM 906 transmission electron microscope (Carl Zeiss, Jena, Germany).

Spleens of mice receiving a galactose-rich diet or control diet were removed on day 10 or day 16 of EAE and disrupted with a 5 ml glass homogenizer. Cells were filtered through a 100 µM cell strainer followed by erythrocyte lysis. Cells were washed with cold Dulbecco’s Phosphate Buffered Saline (DPBS; Gibco via Life Technologies, Darmstadt, Germany) and used for flow cytometry analysis or the MOG_35-55 restimulation assay.

Spinal cord tissue was removed on day 16 of EAE after perfusion with cold DPBS and disrupted with a 5 ml glas homogenizer. Cells were transferred to a Percoll™ (GE Healthcare, Solingen, Germany) density gradient and centrifuged at 800g for 20 min without break. Cells at the interphases were collected, washed with cold DPBS and analyzed by flow cytometry.

Splenocytes were seeded at a density of 3x10⁶ cells/cm² in Re-medium. MOG_35-55 (20 µg/ml) was added for stimulation and the cells were cultured for 48 or 72 hours at 37°C. Supernatants were harvested after 48h and analyzed by enzyme-linked immunosorbent assay (ELISA) for the secretion of IL-17A and IFN-γ (ELISA DuoSet, R&D) according to the manufacturer’s protocol. To assess the proliferation capacity, splenocytes were labelled with e450 proliferation dye (eBioscience) according to the manufacturer’s protocol and cultured in the presence or absence (control) of MOG_35-55. Proliferation was analyzed 72h later by flow cytometry.

Ex vivo obtained CNS leukocytes and splenocytes were analyzed by extra- and intracellular staining. Dead cells were excluded by a fixable viability dye eFluor^®780 (0.2 μl/test, eBioscience). Nonspecific Fc-mediated interactions were blocked by addition of 0.5 μl anti-CD16/32 (93, eBioscience) for 10 min. For surface staining, cells were stained with the respective fluorochrome- conjugated antibodies for 30 min in PBS. For intracellular cytokine staining, cells were stimulated for 4 h with ionomycin (1 µM; Sigma-Aldrich) and phorbol-12-myristate-13-acetate (PMA; 50 ng/ml; Sigma-Aldrich) in the presence of monensin (2 μM; eBioscience). Cells were stained for surface marker and made permeable by saponin buffer or Fix/Perm (eBioscience) according to the manufacturer’s protocol. Intracellular cytokines were stained with the respective fluorochrome conjugated antibodies for 30-45 min. The following antibodies were used: αCD4-FITC (RM4-5, eBioscience), αCD11b-APC/PE (M1/70, Biolegend), αCD11c-FITC/Pe-Cy7 (HL3, BD Pharmingen), αCD25-APC/BV421 (PC61, BioLegend), αCD44-PE (IM-7, Biolegend), αCD69-BV421 (H1.2F3, Biolegend), αFoxP3-PE/Pe-Cy7 (FJK-16s, eBiosciences), αIL-17A-PE/Pe-Cy7 (eBio17B7, eBioscience), αIFN-γ-APC (XMG1.2, BD Biosciences). Cells were measured with FACSCanto II from BD Biosciences and data were analyzed using FlowJo software (BD Biosciences).

Glycation was performed as described previously (29). Shortly, 10 mg/ml BSA, 500 mM galactose, 1 U/ml penicillin/streptomycin and 1 mM EDTA were dissolved in PBS and filtered through a 0.22 µm pore filter before incubated under sterile conditions at 37°C in the dark over a period of 12 weeks. To check for changes over time, aliquots were taken weekly. Until further analysis, solutions were stored at -20°C. Finally, all aliquots were dialyzed against PBS at 4°C to stop the reaction by removal of unbound sugars. AGEs then were stored in sterile tubes at 4°C. For the preparation of control-BSA, this procedure was performed without the addition of galactose.

Given the special fluorescence characteristics and absorption spectra of some AGE products, auto-fluorescence of pentosidine, a common AGE variant was determined to evaluate the formation of AGEs. Therefore, total protein content was determined by BCA assay (Thermo Scientific) and sample concentrations were adjusted to 1 mg/ml. Arbitrary fluorescent units were measured with a fluorescence plate reader (Victor Multilabel plate reader, Perkin Elmer) at an excitation wavelength of 355 nm and an emission wave length of 460 nm. Fluorescence of PBS alone was subtracted from each data set and results are presented as fluorescent units.

O4+ cells were isolated via magnetic cell separation (MACS) from brains of C57BL/6N mice on day 20 of EAE either receiving control or a galactose-rich diet. The cells were stained with the fixable viability dye eFluor^®780 (0.2 μl/test, eBioscience) for 20 min. Nonspecific Fc-mediated interactions were blocked by addition of 0.5 μl anti-CD16/32 (93, eBioscience) for 10 min before the addition of αO4-APC (130-109-153, Milteniy Biotec) for 30 min. The cells were then labeled with Annexin-V FITC (Invitrogen) according the manufacturers’ instruction. The frequency of Annexin-V⁺ apoptotic cells in O4+ cells was analyzed by flow cytometry.

After dissection of cortices from day 1-2 postnatal mouse pups, MGC were isolated using the Neural Tissue Dissociation Kit (Miltenyi Biotech) according to the manufacturer’s protocol. Obtained MGC were re-suspended in 500 µl MGC media containing DMEM (Gibco) supplemented with 1% penicillin/streptomycin (PAN-Biotech GmBH) and 10% heat inactivated horse serum per brain. Cells were seeded in 10 cm-culture dishes pre-coated with poly-D-lysine/laminin (Sigma) and incubated at 37°C and 5% CO_2. Media was replaced every 2 days. After 14 days in culture, MGC were processed for isolation of oligodendrocytes.

After 14 days in culture, MGC were harvested and labelled with αO4-antibody (Miltenyi Biotech) for separation via magnetic cell sorting (MACS) according to the manufacturer’s instructions. For immunocytochemical analysis, O4⁺ cells were plated on poly-D-lysine -coated cover slips at a density of 5-8 x 10⁵ cells per 4-well chamber and kept at 37°C and 5% CO₂. Half of the media containing MACS Neuro-Medium (Miltenyi Biotech) supplemented with 2% MACS NeuroBrew-21 (Miltenyi Biotech), 1% L-Glutamine (PAN-Biotech GmBH), 1% penicillin/streptomycin, fibroblast growth factor (FGF, 10 ng/ml, Miltenyi Biotech) and platelet-derived growth factor-aa (PDGF-AA, 10 ng/ml, PeproTech) was changed every other day. Cells were either treated with 1.5 mM galactose for 72 h, or with 50 µM AGE-BSA or control-BSA for 72 h.

Oligodendrocytes were washed in PBS and fixed in 1% PFA for 30 min at room temperature. After blocking with 10% BSA+0.1% Triton-X, cells were incubated with primary antibodies over night at 4°C. Cells were incubated with secondary antibodies for 1 hour in the dark at room temperature followed by counterstaining with DAPI (1:20.000 in PBS) for 5 min. Antibodies used include ms/rb anti-Olig2 (AB9610; 1:500; Millipore), ms anti-Ki67 (550,609; 1:75, BD Biosciences) and rb anti-Casp3 (1:300, Cell Signaling) as well as rb/gt anti-ms-Alexa488, gt anti-ms Alexa555 and gt anti-rb Alexa647 conjugated secondary antibodies (1:1000, Invitrogen, ThermoFisher Scientific). For negative controls, cells were incubated with the secondary antibody only. Pictures were taken with Zeiss Observer at a 20x magnification. Quantitative analysis was performed with the help of ImageJ by counting at least 300 Olig2-positive cells in at least 10 different visual fields.

iPNs were generated as previously described (30). In brief, renal proximal tubule epithelial cells (RPTECs) were isolated from the urine of participants. After cultivation, RPTECs were transfected via electroporation with the episomal plasmids pCXLE-hOCT3/4-shp53-F, pCXLE-hSK and pCXLE-hUL (Addgene). Resulting induced pluripotent stem cell (iPSC) colonies were transferred into a free-floating state, forming embryoid bodies (EBs). Differentiation into meso- and entoderm was inhibited by using 10 µM SB431542 (Biozol) and 5 µM dorsomorphin (Sigma). After inhibition, EBs were transformed into adherent state again by transferring EBs on 0.002% poly-L-ornithine (Sigma) and 10 µg/ml laminin (Sigma) coated dishes. For the induction of the formation of neural rosettes, EBs were cultivated in neural stem cell (NSC) medium (DMEM/F12 GlutaMAX™ (Thermi Fisher Scientific), supplemented with 20 µg/l insulin (Sigma), 1.6 g/l L-glucose (Applichem), 1 µl/ml B27™ supplement (Life Technologies), 1 µl/ml N2 supplement (Life Technologies), 10 ng/ml basic fibroblast growth factor (PAN Biotech), 10 ng/ml epidermal growth factor (PAN Biotech). After the emergence of neuronal rosettes, neuronal progenitor cells (NPCs) were isolated from donut-shaped structures and transferred to another dish. For the differentiation of NPCs to iPNs, medium was changed to neuronal differentiation medium composed of DMEM/F12 GlutaMAX™, supplemented with 50 µg/ml L- ascorbic acid (Carl Roth), 50 µg/ml apo-transferrin (Sigma), 2x B27™-, and 2x N2-supplement, and cells were initially treated with 500 ng/ml sonic hedgehog (SHH; Peprotech) and 2 µM retinoic acid (RA; Sigma) for the induction of differentiation. Afterwards, neuronal medium was supplemented with 10 ng/ml brain-derived neurotrophic factor (BDNF; Peprotech) and 20 ng/ml glial-derived neurotrophic factor (GDNF; Peprotech) for maintenance and maturation. iPNs were used for experiments at the age of 21-23d.

8,000-10,000 cells/well were plated onto a 96-well plate, 14-16 days after induction of neuronal differentiation. Cells were treated with 1.5 mM galactose for 24 h, 48 h, or 72 h or with 50 µM AGE-BSA or control-BSA for 48 h. For cell death analysis, cells were treated, except for the control condition, with 5 µM H₂O₂ for 2 h. For this time neuronal medium was replaced with DPBD+/+.

Staining was performed after treatment incubation by changing the medium to DMEM F12/Glutamax (Gibco) + 4 µg/ml Höchst (Thermo Fisher) for 1.5 h to stain the nuclei. A subsequent staining of dead cells by 5 µl/ml 7AAD (eBiosciece) in DMEM F12/Glutamax was executed for 10 min at 37°C. After incubation, the well plate was analyzed under an inverse fluorescence microscope. Images were acquired at 20,000 x magnification. Randomly, three visual fields were obtained per well. Per condition a minimum of three wells was analyzed. Total cell amount per visual field (Höchst+) and dead cells (Höchst+/7AAD+) were manually counted using the cell counter tool from ImageJ.

Microglia were isolated from brain tissue of EAE diseased mice (d20 p.i.) via magnetic cell separation (MACS, Miltenyi Biotech) using CD11b microbeads. Microglia were identified as CD11b+CD45- cells via flow cytometry and analyzed for CD69 and MHCII expression. Cells were cultured in MACS Neuro-Medium (Miltenyi Biotech) supplemented with 2% MACS NeuroBrew-21 (Miltenyi Biotech) at a density of 2x10^5 cells/well in poly-L-Lysine (0.01%, Sigma) coated 12 well plates for 24 h. Supernatants were analyzed for the secretion of IL-1β, IL-6 and TNFα (ELISA DuoSet, R&D).

Spinal cord tissue was homogenized in RLT buffer with an Ultra-Turrax for 30s. Total RNA was isolated using the RNase Mini Kit #A2791 (Qiagen) following the manufacturer’s instructions. Reverse transcription was performed by using the GoScript Reverse Transcription Mix Oligo(dT) (Promega). PCR reactions were performed at a 5 μl scale on a qTower 2.0 real time PCR System (Analytic Jena, Germany) in triplicates. Relative quantification was performed by the ΔΔCT method, normalizing target gene expression on β-Actin as housekeeping gene. The following primer (TaqMan^®, Thermo Fisher Scientific) were used: Ager Mm01134790_g1, Nos2 Mm00440502_m1, Nrf2 (Nfe2l2) Mm00477784_m1, Actb Mm00607939_s1.

Statistical analysis was performed using GraphPad Prism (GraphPad Software Inc., La Jolla, CA). All in vitro and ex vivo data were analyzed by one-way ANOVA followed by Tukey’s post-test, unpaired t-test or Wilcoxon rank sum test after checking for normal distribution (unless indicated otherwise in the legends). EAE data were analyzed by Mann–Whitney U test. Data are presented as mean ± SEM; *p< 0.05, **p< 0.01, or ***p<0.001 were considered to be statistically significant.

To examine the impact of a galactose-rich diet on the course of EAE, young C57BL/6N mice were administered a diet consisting of a 10% galactose in drinking water. This diet was started 14 days prior to active induction of MOG_35-55-EAE and disease severity was compared to control mice receiving normal drinking water. The galactose-rich diet significantly worsened disease symptoms in the effector phase and impaired the recovery in the chronic phase of EAE compared to controls ( Figure 1A ). No major differences between both groups were observed in terms of disease incidence ( Table 1 ) or mortality (data not shown). As changes in the diet may affect physical and metabolic parameters, mice were also monitored for changes in body weight and blood sugar levels. A galactose-rich diet was neither associated with significant body weight changes during the 14-day adaptation period prior to EAE induction nor in the period after immunization compared to the control ( Table 1 ). Blood glucose levels were similar in both groups prior to EAE induction and at the maximum of the disease ( Figure 1B ).

To analyze potential mechanisms of galactose-action in MOG-EAE, we performed histopathological analyses on spinal cord cross sections from control treated mice and mice receiving a galactose-rich diet during the chronic phase of the disease on day 34 post immunization (p.i.). In accordance with the observed exacerbation of clinical symptoms, CNPase staining revealed a more pronounced demyelination in mice receiving a galactose-rich diet ( Figure 2A ). Moreover, NogoA+ mature oligodendrocytes were significantly reduced by high galactose intake ( Figure 2B ). Further histological analysis showed a significant loss of NeuN⁺ neurons in the anterior horn of these mice ( Figure 2C ). We next studied the effect of a galactose-rich diet on the neuro-axonal integrity during EAE. Bielschowsky’s silver impregnation of spinal cord cross sections revealed a reduced axonal density in mice receiving a galactose-rich diet compared to the control group ( Figure 2D ). Using electron microscopy, we further analyzed the effect of galactose on axonal integrity and detected an increased number of axolytic axons ( Figure 2E ). In order to exclude an effect of a galactose-rich diet on CNS cells in the absence of autoimmune processes, we also histologically analyzed spinal cord tissue of naïve young mice (11.2 weeks ± 5 days) receiving a galactose-rich diet for 7 weeks. No obvious differences in the number of oligodendrocytes, neurons or astrocytes were observed as compared to naïve mice fed a normal diet ( Supplementary Figure 1 ).

In contrast to the evident effects of galactose on the oligodendroglial and neuronal compartment, no changes in numbers of infiltrating CD3+ T cells and Mac-3+ macrophage/microglia were observed at the maximum of EAE ( Figures 3A, B ). Further flow cytometry analysis of spinal cord infiltrating cells at day 16 of EAE revealed no obvious differences between mice on a galactose-rich diet and controls. There were neither effects on the frequencies of Th1, Th17 and Treg cells ( Figure 3C ) nor on CD11b⁺/CD11c⁺ antigen-presenting cells ( Figure 3D ). Additional analysis of T cell activation also revealed no differences in CD44+ and CD25+ expression ( Figure 3E ). CD69 expression in CNS infiltrating macrophages and brain resident microglia also revealed no differences in galactose fed mice compared to the controls ( Figure 3F ). To additionally investigate potential effects of galactose on the immune system, we also performed flow cytometry analysis of splenic cells prior to the onset of EAE symptoms on day 10 p.i. as well as during acute stages of the disease on day 16 p.i. A galactose-rich diet did not affect the frequency of splenic Th1, Th17 or Treg cells ( Figure 3G ), T cell activation ( Figure 3H ) or CD11b⁺/CD11c⁺ antigen-presenting cells ( Figure 3I ) prior to disease onset or at later stages (data not shown). Moreover, upon antigen-specific restimulation of splenocytes in vitro, neither the proliferation of lymphocytes ( Figure 3J ) nor the secretion of the pro-inflammatory cytokines IFNγ and IL-17 showed any difference between cells from mice receiving a control or galactose-rich diet ( Figure 3K ).

Our histopathological observations suggested that galactose might impair the neuro-axonal integrity during MOG-EAE by decreasing oligodendrocyte and neuron numbers. Additional analysis of ex vivo obtained oligodendrocytes from EAE diseased mice revealed an enhanced Annexin-V uptake in cells isolated from galactose-fed mice compared to the controls, suggesting a potential effect of increased galactose concentrations on cell apoptosis ( Figure 4A ). We therefore investigated a potential direct effect of elevated galactose concentrations in in vitro culture assays. The addition of 1.5 mM galactose, mimicking a high yet still physiological blood concentration, the proliferation of isolated O4⁺ oligodendrocytes as assessed by Ki67 staining ( Figure 4B ). Moreover, human iPNs derived from iPSCs were cultured in the presence of 1.5 mM galactose for one to three days. Yet, the addition of galactose did not impact on cell survival in iPNs as compared to H₂O₂ treatment as a positive control ( Figure 4C ). Given our findings on the missing effect of galactose on CNS cells in vitro, we hypothesized that not galactose itself, but some reactive products may be responsible for the exacerbation of clinical symptoms and the neuroglial damage in the EAE model. Galactose is a reducing sugar and during protein glycation, amino groups of proteins may react with these sugars in a non-enzymatic reaction (31). Indeed, incubation of 10 mg/ml BSA with 0.5 M galactose at 37°C for up to 12 weeks resulted in the formation of advanced glycation end product (AGE)-coupled BSA as measured by auto-fluorescence of pentosidine as a common AGE variant ( Figure 4D ). While relative fluorescent units of control-BSA were at about one to two after 12 weeks of incubation, relative fluorescent units of BSA incubated with galactose were 20-fold higher after 5 weeks of incubation and reached almost 50 units after 12 weeks of incubation, demonstrating an increased AGE production due to galactose addition ( Figure 4D ). Interestingly, fluorescence measurements of spinal cord protein isolated from EAE diseased mice on the maximum of EAE also revealed higher fluorescent units in mice receiving a galactose-rich diet compared to the controls ( Figure 4E ). These data imply an increased concentration of fluorescent AGEs in the spinal cord of galactose-treated mice, indicating a potential role of AGEs in the observed neuropathological effects during MOG-EAE. We therefore tested the impact of AGE treatment on cell survival in human iPSC-derived iPNs and murine oligodendroglial cell cultures. iPSC-derived iPNs were treated with H₂O₂ to induce cell death. While the combined addition of H₂O₂ and control-BSA did not further increase neuronal cell death compared to H₂O₂ treatment alone, the addition of H₂O₂ together with galactose-induced AGEs significantly aggravated cell survival compared to H₂O₂ treated neurons ( Figure 4F ). Yet, the difference on neuronal cell death between H₂O₂+BSA and H₂O₂+AGE-BSA treated cells is not statistically significant when comparing all four groups (p=0.0507). However, the frequency of dead cells in H₂O₂+BSA treated cultures is comparable to the frequency of dead cells in H₂O₂ cultures (H₂O₂: 36.8 ± 3.7% vs. H₂O₂+BSA: 37.5 ± 3.8%). In contrast, the treatment with H₂O₂+AGE-BSA increased the cell death to 48.5 ± 3.4%. Hence, H₂O₂+AGE-BSA treatment increases neuronal cell death by around 32% compared to H₂O₂ or 29% compared to the H₂O₂+BSA treatment. Similar to the deleterious effect of galactose-induced AGEs on neuronal cell viability in vitro, oligodendrocytes cultured in the presence of galactose-induced AGEs showed enhanced apoptosis compared to control-BSA treated cells. More precisely, treatment of O4⁺ oligodendrocytes with galactose-induced AGEs for 72 h resulted in a more pronounced cell apoptosis compared to BSA treatment as assessed by counting Olig2⁺ cells with shrunken cell nuclei ( Figure 4G ) or caspase-3 positive Olig2+ cells ( Figure 4H ). Twice as many cells presented features of apoptosis in comparison to control-BSA treated cells, strengthening our concept that galactose-induced AGEs adversely affect neuronal and oligodendroglial cell viability in vitro and during MOG-EAE. We performed additional experiments to identify a potential mechanism of enhanced oligodendrocyte apoptosis due to increased galactose intake. Yet, we observed no differences in the expression of the receptor for AGEs (RAGE) in the CNS or in oligodendroglial cells during MOG-EAE ( Supplementary Figures 2A, B ). The analysis of nitrite levels, nitric oxide synthase 2 (Nos2) and Nuclear factor erythroid 2-related factor 2 (Nrf2) as markers for oxidative stress revealed no differences in galactose-fed mice compared to the controls ( Supplementary Figures 2C-E ). Moreover, the analysis of microglia cells isolated from EAE-diseased mice receiving either control or a galactose-rich diet demonstrated no differences in both groups, arguing for a direct effect of galactose or galactose induced AGEs on oligodendroglial cells during neuroinflammation.