ORIGINAL RESEARCH published: 14 June 2019 doi: 10.3389/fimmu.2019.01374 Manipulation of Gut Microbiota Influences Immune Responses, Axon Preservation, and Motor Disability in Edited by: Marcella Reale, Università degli Studi G. d’Annunzio a Model of Progressive Multiple Chieti e Pescara, Italy Reviewed by: Sclerosis Yoshiro Ohara, Kanazawa Medical University, Japan Leyre Mestre 1,2*† , Francisco Javier Carrillo-Salinas 1†‡ , Miriam Mecha 1,2 , Ana Feliú 1,2 , Javier Ochoa-Reparaz, Carmen Espejo 2,3,4 , José Carlos Álvarez-Cermeño 2,5 , Luisa María Villar 2,5 and Eastern Washington University, Carmen Guaza 1,2* United States 1 Neuroimmunology Group, Functional and Systems Neurobiology Department, Instituto Cajal, CSIC, Madrid, Spain, 2 Red *Correspondence: Española de Esclerosis Múltiple (REEM), Barcelona, Spain, 3 Servei de Neurología-Neuroimmunología, Centre d’Esclerosi Leyre Mestre Múltiple de Catalunya, Vall d’Hebron Institut de Recerca, Hospital Universitari Vall d’Hebron, Barcelona, Spain, 4 Universitat leyre@cajal.csic.es Autònoma de Barcelona, Bellaterra, Spain, 5 Immunology Department, Hospital Universitario Ramón y Cajal de Investigación Carmen Guaza Sanitaria (IRYCIS), Madrid, Spain cgjb@cajal.csic.es † These authors have contributed Gut microbiota dysbiosis has been implicated in MS and other immune diseases, equally to this work ‡ Present Address: although it remains unclear how manipulating the gut microbiota may affect the disease Francisco Javier Carrillo-Salinas, course. Using a well-established model of progressive MS triggered by intracranial Department of Immunology, Tufts infection with Theiler’s murine encephalomyelitis virus (TMEV), we sought to determine University School of Medicine, whether dysbiosis induced by oral antibiotics (ABX) administered on pre-symptomatic Boston, MA, United States and symptomatic phases of the disease influences its course. We also addressed the Specialty section: effects of microbiota recolonization after ABX withdrawn in the presence or absence of This article was submitted to probiotics. Central and peripheral immunity, plasma acetate and butyrate levels, axon Multiple Sclerosis and Neuroimmunology, damage and motor disability were evaluated. The cocktail of ABX prevented motor a section of the journal dysfunction and limited axon damage in mice, which had fewer CD4+ and CD8+ T Frontiers in Immunology cells in the CNS, while gut microbiota recolonization worsened motor function and Received: 02 April 2019 Accepted: 30 May 2019 axonal integrity. The underlying mechanisms of ABX protective effects seem to involve Published: 14 June 2019 CD4+ CD39+ T cells and CD5+ CD1d+ B cells into the CNS. In addition, microglia Citation: adopted a round amoeboid morphology associated to an anti-inflammatory gene profile Mestre L, Carrillo-Salinas FJ, in the spinal cord of TMEV mice administered ABX. The immune changes in the spleen Mecha M, Feliú A, Espejo C, Álvarez-Cermeño JC, Villar LM and and mesenteric lymph nodes were modest, yet ABX treatment of mice limited IL-17 Guaza C (2019) Manipulation of Gut production ex vivo. Collectively, our results provide evidence of the functional relevance Microbiota Influences Immune Responses, Axon Preservation, and of gut microbiota manipulation on the neurodegenerative state and disease severity Motor Disability in a Model of in a model of progressive MS and reinforce the role of gut microbiota as target for Progressive Multiple Sclerosis. MS treatment. Front. Immunol. 10:1374. doi: 10.3389/fimmu.2019.01374 Keywords: Theiler’s virus model, gut microbiota, Treg and Breg cells, neuroinflammation, multiple sclerosis Frontiers in Immunology | www.frontiersin.org 1 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS INTRODUCTION Th17 accumulation in the spinal cord and restored EAE development (20). By contrast, the colonization of the same Multiple sclerosis (MS) is a chronic demyelinating and mice with Bacteroides fragilis and its polysaccharide A dampened inflammatory neurodegenerative disease of the central nervous symptomatology in the EAE model (21). Oral administration system (CNS), and one of the most prevalent neurological of Lactobacillus spp. and Bifidobacterium bifidus reduced the diseases. The precise etiology of MS still remains unclear, yet clinical score, increasing Treg cell number in EAE mice (22). it is largely accepted that both the onset and progression The possibility that components of the gut microbiota may favor of the disease reflects a complex interaction of genetic and CNS autoimmunity was also recently suggested when microbiota environmental factors (1–4). As such, MS could be considered a transplants from twins with MS augmented the incidence of neurological disease that provokes an autoimmune inflammatory spontaneous EAE (23). Complementary studies show that gut reaction or an autoimmune disease that induces an attack microbiota transplants from MS patients to GF mice exacerbated on the myelin sheath. MS could be initiated within the CNS the development of EAE following myelin oligodendrocyte as a neurodegenerative process or it might even be triggered glycoprotein (MOG) immunization (24). by an external agent, for example through viral infection (5, Early researches of RR-MS (Relapsing-Remitting MS) patients 6). Particularly, in susceptible strain of mice the intracranial found reductions in Firmicutes, Proteobacteria, and Bacteroidetes inoculation of TMEV induces a biphasic disease characterized (25), and more recently, a study investigated 60 RRMS patients, by an early acute disease associated with replication of the virus 28 untreated patients and 43 healthy controls showing increased in the CNS gray matter that reach the peak at day 10 post amounts of Methanobrevibacter and Akkermansia and reduced infection. TMEV initially infects neurons during the first 2 weeks amounts of Butyricimonas in untreated patients (14). Because spreading thereafter through axonal transport. The susceptible all the studies reporting potential dysbiosis in MS focus on RR- strains of mice fail to complete clear the virus that mainly persist MS, further studies are necessary to determine whether dysbiosis in monocytes/macrophages and glial cells albeit viral titers are occurs in all forms of MS. Hence, it is relevant to investigate gut scarce at the chronic stages. After a latency period, depending dysbiosis in a well-established model of progressive MS, such as on viral strain, the mice develop a late chronic demyelinating that triggered by intracranial TMEV infection. In an attempt to disease with extensive demyelinating lesions, axonal damage, and explore the effects of altered gut microbiota in the progressive mononuclear infiltrates in the spinal cord that resembles the forms of MS, we studied the effect of gut dysbiosis produced pathological and clinical features of the progressive forms of by oral antibiotic administration (an antibiotic cocktail—ABX) MS (7). Autoimmunity in TMEV-induced demyelinating disease on pre-symptomatic and symptomatic stages of TMEV-IDD. (TMEV-IDD) is first detected as a delayed type hypersensitivity We also assessed the effects of microbiota recolonization after response to the myelin proteolipid (PLP), PLP139−151 peptide ABX withdrawn in the presence or absence of probiotics. CNS around 60 days after the infection (8, 9). A chronic autoimmune and peripheral (spleen and mesenteric lymph nodes, MLNs) disease is subsequently established, as epitope spreading induces immune responses, plasma acetate and butyrate levels, microglial autoimmune responses against other myelin peptides. The events activity, spinal cord cytokine gene expression, axon damage, triggering autoimmunity in MS are still not fully understood, and motor disability were evaluated. We provide evidence of yet it seems that genetic predisposition in conjunction with the functional relevance of gut microbiota manipulation on the environment factors, including viral infection, could drive the neurodegenerative state and disease severity in mice subjected onset and progress of the disease. As such, the Theiler’s to TMEV-IDD. virus model brings together these genetic and environmental components of the disease in the context of virus-induced MATERIALS AND METHODS autoimmunity (10). In recent years, the incidence of MS in developed countries Animals and TMEV Infection has risen hugely. One hypothesis to explain this increase in the TMEV-IDD susceptible female SJL/J mice (Harlan, Barcelona, incidence of MS is the modification of the gut microbiota as Spain) were maintained at the Instituto Cajal (CSIC, Madrid, a result of changes in diet, combined with the widespread use Spain) under conventional controlled conditions: 12 h light/dark of antibiotics (11). Indeed, there is evidence that dysbiosis of cycle, temperature 20◦ C (±2◦ C), 40–50% relative humidity, the gut microbiota is implicated in MS (12–14), as well as in and with ad libitum access to food and water. The right other immune related diseases like rheumatoid arthritis, type brain hemisphere of 4-week-old mice was inoculated with 2 1 diabetes or inflammatory bowel disease (15, 16). However, a × 106 plaque forming units (pfu) of the Daniel (DA) strain functional link between the gut bacteria and MS still remains of TMEV, diluted in 30 µL of DMEM supplemented with to be established. In terms of experimental models of MS, the 10% fetal calf serum (FCS) (26). TMEV was purified in the severity of EAE (experimental autoimmune encephalomyelitis) Dr. Rodríguez’s lab (Department of Neurology, Mayo Clinic) was reduced by oral antibiotic administration (17–19). Moreover, (27). Sham control mice received 30 µL of the vehicle alone. in a spontaneous relapsing–remitting model of EAE, germ- Each experimental group were housed in different cages and free (GF) transgenic SJL/J mice were protected against disease, in different isolated racks in the same room, never mixing while gut colonization by commensal microbiota restored disease mice exposed to different experimental conditions. Each cage susceptibility (12). Likewise, mono-colonization of the gut of contains 5 mice subjected to the same treatment, receiving the C57BL/6 mice with segmented-filamentous bacteria promoted same food and sterile water, and manipulated by the same Frontiers in Immunology | www.frontiersin.org 2 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS person. Therefore, the microbial environment was the same Sample Collection for each mouse, although coprophagy may lead to a more Fresh fecal samples were collected from each mouse on day 70 homogeneous microbiota among animals in the same cage that and 85 pi and stored at −80◦ C. The mice were anesthetized with are subjected to the same experimental conditions, but not in pentobarbital (Dolethal, 50 mg/kg body weight) on day 85 pi, mice housed in different cages. All experiments were performed and the spleen was removed, placed in cold RPMI and processed in strict accordance with EU (Directive 2010/63/EU) and Spanish for flow cytometry. MLNs were also removed and processed for regulations (Royal Decree 53/2013 BOE n◦ 34 and “Comunidad T CD4 studies ex-vivo. Blood samples were collected in EDTA- de Madrid” decree: ES 280790000184). The Ethics Committee treated BD vacutainers to obtain the plasma, which were kept on Animal Experimentation at the Instituto Cajal (CSIC) also cold on ice. Anesthetized mice were perfused transcardially with approved all the procedures employed in this study (protocol saline (0.9% NaCl), and the brain and spinal cord were removed number: 2013/03 CEEA-IC). and processed as we explain in the following sections. Treatments Metabolite Analysis At pre-symptomatic stages, 55 days after virus infection, mice Blood samples were centrifuged for 10 min at 1,500 g in a were provided autoclaved drinking water supplemented with refrigerated centrifuge within 2 h of extraction, and 0.1 ml the ABX cocktail over 15 days: ampicillin, 1 g/L [Sigma- aliquots were stored at −20◦ C. The protein in the samples was Aldrich, Madrid, Spain]; metronidazole, 1 g/L [Sigma-Aldrich, precipitated with an equal volume of acetonitrile (ACN—ACN: Madrid, Spain]; neomycin sulfate, 1 g/L [Gibco-Invitrogen, H2 O, 1:1) and after a 10 min centrifugation at 4◦ C, derivatization Barcelona, Spain]; and vancomycin, 0.5 g/L [Sigma-Aldrich was carried out by mixing 40 µL of samples with 20 µL of 3- Madrid, Spain]. This time course is consistent with standard NPH and 20 µL of 120 mM EDC in 6% pyridine, and incubating antibiotic administration used in other studies to alter the for 30 min and 40◦ C. The samples were then diluted in 920 µL microbiota (28, 29). On day 70 post infection (pi—symptomatic of 10% ACN and finally, the samples were filtered through a phase), the mice treated with ABX were divided into three 0.22 µm PTFE filters and analyzed by mass spectrometry (LC- groups: (i) a group that continued to receive ABX until the ESI-QQQ 8030 Shimadzu). A mixture of standards was prepared end of the experiment (day 85 pi: TAA group, n = 10); (ii) a in ACN: H2 O (1:1), with 500 mg/L of acetic, and butyric acid group that stopped receiving ABX and that received autoclaved (Sigma-Aldrich, Madrid, Spain), and derivatization was carried water (TA, n = 5); and, (iii) a group that stopped receiving ABX out as indicated above. The injection volume was 10 µL and the but that were administered 3 × 108 cfu (colony forming units, gradient mode analytical conditions were: 20% phase B for 2 min, 100 µl) of a commercial probiotic mix, Vivomixx (see below), to 40% phase B to 7 min; from 40 to 100% phase B till 7.5 min; administered by oral gavage three times a week (TAVx, n = 5). return to the initial conditions from 8 to 9 min. Phase A H2 O + Vivomixx is a multispecies probiotic composed of: Lactobacillus 0.01% FA; Phase B ACN + 0.01% FA; flow rate 0.6 mL/min; and paracasei DSM 24734, Lactobacillus plantarum DSM 24730, run time 10 min. Lactobacillus acidophilus DSM 24735, Lactobacillus delbrueckii subspecies bulgaricus DSM 24734, Bifidobacterium longum DSM CD4 T Cells ex vivo Experiments 24736, Bifidobacterium infantis DSM 24737, Bifidobacterium CD4+ T cells were isolated from the mesenteric lymph nodes breve 24732, and Streptococcus thermophilus DSM 24731. The (MLN) using a commercial cell isolation kit and following the corresponding control groups used in these experiments were: manufacturer’s instructions (Miltenyi, Biotec Inc.; San Diego, Sham or TMEV mice that received the vehicle alone throughout CA). Cells from each animal were plated (106 cells/well), cultured the experiment (S group or T group, n = 10); and Sham mice and polarized to Th17 cells in RPMI supplemented with 2- administered ABX (SA group, n = 5). The number of mice β Mercaptoethanol (50 µM), immobilized hamster anti-mouse described above corresponds to two independent experiments. CD3ε (500 µg/ml: BD biosciences), hamster anti-mouse CD28 (2 µg/ml: BD biosciences, San Diego, CA, USA), rat anti-mouse Evaluation of Motor Function IFN-γ (10 µg/ml: BD biosciences, San Diego, CA, USA), rat anti- The development and progression of the demyelinating disease mouse IL-4 (10 µg/ml: BD biosciences, San Diego, CA, USA), was followed in the mice by evaluating spontaneous motor hTGF-β (5 ng/ml: Peprotech, Inc, UK), recombinant murine function. Locomotor activity was screened on day 70 and 85 pi in IL-6 (20 ng/ml: Peprotech, Inc, UK) recombinant murine IL- an activity cage (Activity Monitor System Omnitech Electronics 23 (BioLegend, CA, USA, 5 ng/ml). Three days later the cells Inc., Columbus, OH, USA) coupled to a Digiscan Analyser, were stimulated for 5 h with PMA (100 ng/ml: Tocris Bioscience, evaluating spontaneous motor activity. Horizontal (HACTV) Bristol, UK) and ionomycin (free acid, 500 ng/ml: Tocris and vertical (VACTV) activity was evaluated through the total Bioscience, Bristol, UK), and the supernatants were collected and number of beam interruptions at the horizontal and vertical stored at −20◦ C for cytokine measurement. sensor over a 10 min session. Motor coordination was assessed in a rotarod test, the apparatus (Ugo Basile, Milan, Italy) consisting ELISA of a suspended rod rotating at a constant or accelerating speed. IL-17 production by Th17 polarized cells purified from MLN was Three days after training for the mice to habituate to the test, measured by solid phase sandwich ELISA using Quantikine kits animals were subjected to a trial test for a maximum of 5 min. (R&D systems Inc, MN, USA), according to the manufacturer’s Frontiers in Immunology | www.frontiersin.org 3 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS instructions: reference number M1700 for IL-17 detection. The analyzed in triplicate on an Applied Biosystems PRISM 7500 assay’s sensitivity was 5 pg/ml for IL-17. Sequence detection system, normalizing the mRNA expression to the Rps29 gene in each sample and quantifying gene expression Tissue Processing by the 2−11Ct method. The results are expressed relative to the sham mice for each time point. and Immunohistochemistry The spinal cord was fixed overnight in 4% paraformaldehyde (PFA) prepared in 0.1 M phosphate buffer (PB), cryoprotected Flow Cytometry in a 30% solution of sucrose in 0.1 M PB and frozen at −80◦ C. A CNS or spleen leukocyte suspension was obtained as described Free-floating transversal cryostat sections (30 µm thick) of the previously (30). Freshly isolated cells (106 ) were incubated spinal cord were rinsed in PB, permeabilized in 0.1 M PB with a Fc block (anti-mouse CD16/CD32: eBioscience, San containing 0.1% Triton X-100 (PBT) and blocked for 1 h at Diego, CA, USA) and labeled with anti-mouse antibodies: PE room temperature in blocking buffer (PBT plus 5% normal goat anti-CD25; PE anti-CD3; PerCP-Cy5.5-conjugated anti-CD45; serum—NGS). Microglia were labeled with a primary antibody PerCP-Cy5.5-conjugated anti-B220; PE-Cy7-conjugated anti- against Iba-1 (ionized calcium binding adaptor molecule 1, CD39; PE-Cy7-conjugated anti-CD8a; PE-Cy7-conjugated anti- 1:1,000: Wako Chemical Pure Industry, Osaka, Japan), while CD19; APC anti-CD3; APC-Cy7-conjugated anti-CD4; APC- axon integrity was assessed by staining with Neurofilament-H Cy7-conjugated anti-CD11b (all from BD Pharmingen, San (NF-H; 1:1,000: Millipore, Temecula, CA, USA). After washing, Diego, CA, USA); APC anti-CD3; and PE-Cy7-conjugated anti- the sections were incubated with an Alexa Fluor-594 conjugated CD39 (both from eBioscience, San Diego, CA, USA). For Foxp3 goat anti-rabbit antibody (1:1,000: Molecular Probes Inc, Eugene, detection the cells were suspended in fixation/permeabilization OR, USA), washed and mounted with Mowiol. In all cases, the buffer for 30 min and stained with a PE anti-Foxp3 antibody (BD specificity of staining was confirmed by omitting the primary Pharmingen, San Diego, CA, USA). Cell viability was assessed antibody. Five spinal cord sections were obtained per animal with the LIVE/DEAD Fixable Green Dead Cell Stain Kit or and at least three mice per group were used to quantify the area LIVE/DEAD Fixable Red Dead Cell Stain Kit (Life Technologies immunostained by Iba-1 or NF-H using ImageJ software (NIH, Corporation, OR, USA). At least 10,000 events were acquired in Bethesda, MD, USA). White matter area from ventral and lateral each experiment on a FACSaria flow cytometer (BD Biosciences, horns were analyzed by immunofluorescence on a Leica TC SP5 San Diego, CA, USA) and the data were analyzed using FlowJo confocal microscope. Representative images of the surface of software (v.10; Tree Star, Ashland, OR, USA). single microglial cell were obtained with Imaris software. 16S rRNA Microbial Community Analysis RNA Extraction, Reverse Transcription, Fecal DNA was isolated using QIAamp DNA Stool Mini Kit and RT-PCR (Qiagen; cat. no. 51504) following the stool pathogen detection After removal, the cervical spinal cord was frozen immediately protocol. DNA was extracted following the manufacturer’s and kept at −80◦ C. Given the high content of lipids found instructions with the subsequent modification: heating step in in the spinal cord, total RNA was extracted using the RNeasy ASL buffer at 95◦ C and eluting the DNA in 50 µl AE buffer. Lipid RNA extraction kit (Qiagen, Manchester, UK). Genomic Regions V3 and V4 of the bacterial 16S rRNA gene were amplified DNA contamination was avoided by DNAse I (Qiagen) treatment using as primer forward CCTACGGGNGGCWGCAG and and the RNA yield was determined with a Nanodrop R reverse GACTACHVGGGTATCTAATCC. PCRs were initiated spectrophotometer (Nanodrop Technologies, Wilmington, DE, with the incubation at 98◦ C for 30 s, and the PCR amplification USA). Total RNA (1 µg) was reverse transcribed into cDNA was performed over 20 cycles of 98◦ C for 10 s and 50◦ C for using poly-dT primers and a reverse transcription kit (Promega 20 s and 72◦ C for 20 s. Additional elongation step at 72◦ C for Biotech Ibérica S.L., Madrid, Spain). Real-time PCR (RT-PCR) 20 s were performed. PCR conditions were changed to 28 cycles was performed with SYBR R and the oligonucleotide primer of amplification and 57◦ C of TM for the following groups sequences were as follows: mIL-1β, forward 5′ -TGG TGT GTG (Sham+ABX, 70 dpi; TMEV+ABX, 70 dpi; SA, 85 dpi; TAA, ACG TTC CCA TT-3′ , reverse 5′ -TCC ATT GAG GTG GAG 85 dpi. AGC TTT C-3′ ; mTNF-α, forward 5′ -AGA GGC ACT CCC CCA Quality analyses of the reads were performed using FastQC AAA GA-3′ , reverse 5′ -CGA TCA CCC CGA AGT TCA GT-3′ ; software (http://www.bioinformatics.babraham.ac.uk/projects/ mIL-4, forward, 5′ -CCACGGATGCGACAAAAATC-3′ , reverse fastqc/). The microbiome analysis of the paired-end reads was 5′ -GACGTTTGGCACATCCATCTC-3′ ; mIL-10, forward 5′ - performed using the QIIME2 v2018.8 (Qualitative Insights into TGA ATT CCC TGG GTG AGA AGC TGA-3′ , reverse 5′ -TGG Microbial Ecology) software (31). The sequences were grouped CCT TGT AGA CAC CTT GGT CTT-3′ ; mIL-6, forward 5’-TCC in “sequence variants” (SVs), equivalent of 100% Operational AGA AAC CGC TAT GAA GTT C-3′ , reverse 5′ -CAC CAG CAT Taxonomic Units (OTUs). Representative sequences from each CAG TCC CAA GA-3′ ; mRps29 forward 5’-GCC GCG TCT GCT cluster were queried against the Greengenes database (version CCA A-3’, reverse 5′ -ACA TGT TCA GCC CGT ATT TGC-3′ . 13_8) to obtain the taxonomic assignments at a 97% sequence PCRs were initiated by incubation at 50◦ C for 2 min and 95◦ C identity using BLAST. These representative sequences were for 10 min, and PCR amplification was performed over 40 cycles further used for phylogenetic analysis and diversity metric at 95◦ C for 15 s and 60◦ C for 1 min. The cDNA samples were calculations. Sequences classified within the Streptophyta clade, Frontiers in Immunology | www.frontiersin.org 4 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS most likely coming from chloroplast DNA in murine feed (32), received ABX the abundance of several KEGG pathways was were extracted for further metagenomic analysis. enhanced, specifically those related to Infectious Diseases, Cell motility and secretion, Neurodegenerative Diseases, Signal Metagenome Functional Prediction transduction, Cell Growth and Death, or Biosynthesis and Metagenomic predictions were made using PICRUSt biodegradation of secondary metabolites in treatment. By (Phylogenetic Investigation of Communities by Reconstruction contrast, the KEGG pathways Endocrine System, Carbohydrate of Unobserved States) software (33). PICRUSt uses the metabolism or Energy metabolism pathways were significantly normalized OTU abundance table and a predicted functional dampened relative to the TMEV mice (Figure 1G). trait abundance to produce a table of functions by sample assuming, by default, that KEGG (Kyoto Encyclopedia of Genes Oral Administration of Probiotics Modifies and Genomes) Orthologs predictions are sought. Microbiota Recolonization After Antibiotic Statistical Analysis Cessation, Influencing Deambulatory A principal component analysis (PCA) and an extended error Activity, and Plasma Microbiota bar analysis with a 95% confidence interval (CI) were performed Metabolites in TMEV-IDD using Statistical Analysis of Metagenomic Profiles (STAMP) In order to examine the effects of microbiota recolonization in software (34). The rest of the data are expressed as the mean ± the TMEV-IDD model, mice that received ABX until day 70 pi SEM and they were analyzed using GraphPath Prism5 (GrapPad were divided into three groups: one that continued to receive Software; RRID: SCR_002798). One-Way ANOVA followed by ABX until day 85 pi (TAA); another that received vehicle alone the Bonferroni post-hoc test were used to determine the statistical (TA); and the third group that received the probiotic Vivomixx significance. In the case of non-parametric analyses, a Kruskal- (TAVx: see Figure 2A). Motor deficits were more clearly evident Wallis Mann-Whitney U-test was applied. in the group of TMEV mice that received the vehicle alone (Figure 2B), which displayed reduced deambulatory capacity (p RESULTS < 0.001), less vertical activity (p < 0.01) and a shorter latency to fall in the rotarod test (p < 0.01). By contrast, oral ABX Dysbiosis by Oral Antibiotics Improve treatment dampened the motor symptomatology (as also seen Motor Function of TMEV Mice at 70 dpi), evident through the horizontal (p < 0.05) and Intracranial infection of a susceptible strain of mice with vertical activity (p < 0.05) in the activity cage, and while rotarod Theiler’s virus produces motor disability that arises ∼60– performance appeared to improve relative to the TMEV mice that 70 dpi (35–37) and moreover, such infection is associated received the vehicle alone, it did not reach statistical significance with changes in the gut microbiota (30). Here we assessed (p = 0.06). When ABX treatment was disrupted between 70 whether the ABX cocktail administered orally during the and 85 dpi, the motor behavior of the mice assessed in the pre-symptomatic phase affected the onset of the motor activity cage and rotarod worsened (Figure 2B). Interestingly, effects (Figure 1A). Behavior in the activity cage and rotarod the deambulation of mice that received the probiotic Vivomixx performance indicated that oral ABX administration for 15 days during the recolonization from day 70 to 85 pi improved (p < before the appearance of symptoms apparently prevented the 0.05) although the probiotic failed to restore vertical activity that motor dysfunction that develops in TMEV mice (Figure 1B). is the best indicator of hind limb weakness. Collectively these ABX treatment induced dysbiosis, since the bacterial 16S data show that microbiota removal by ABX affects the motor gene was not amplified from fecal pellets of TMEV-ABX symptomatology being protective against TMEV-IDD, while mice under standard PCR conditions (Figure 1C). However, recolonization disrupts this beneficial effect. Our results also in forced PCR conditions to amplify bacterial DNA from suggest that the probiotic administration may help recolonization TMEV samples, the bacterial composition of the stools analyzed in a positive manner. by Illumina MiSeq sequencing revealed that the microbiota To corroborate the recolonization of the microbiota, fecal clustered according to the treatment received (ABX or vehicle: samples were collected at 85 dpi and analyzed in a similar Figure 1D). ABX treatment induced a decrease in the relative manner to the samples from mice at day 70 pi. Amplification abundance of Bacteroidetes and Firmicutes, and an increase of 16S bacterial genes from fecal pellets showed that cessation in Actinobacteria and Proteobacteria phyla (Figure 1E). At the of ABX followed by probiotic administration induced a rapid genus level, ABX induced a decrease in the relative abundance of recolonization (Figure 2C). In addition, the analysis of the Oscillospira, Ruminococcus (phylum Firmicutes), Anaeroplasma microbiota confirmed that samples from mice that no longer (phylum Tenericutes) Mucispirillum (phylum Deferribacteres) received ABX or that were administered probiotics cluster and Bilophila, and Sutterella (phylum Proteobacteria), while there together with those that never received ABX (Figure 2D). As we was an increase in Anaerococcus, Streptococcus, Bacillus (phylum could expect, the oral administration of Vivomixx significantly Firmicutes), Flavobacterium, Prevotella (phylum Bacteroidetes), increased members of both the Bacteroidetes and Firmicutes Acinetobacter, Ochrobactrum (phylum Proteobacteria), and phyla. However, it was noteworthy that ABX cessation mainly Corynebacterium (phylum Actinobacteria: Figure 1F). induced an increase in members of the phylum Bacteroidetes We used PICRUSTs to assess the functional content of (Figure 2E). Despite this initiation of recolonization, the the microbiota based on the 16S data. In TMEV mice that Bacteroides (phylum Bacteroidetes) Mucispirillum (phylum Frontiers in Immunology | www.frontiersin.org 5 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 1 | ABX-treatment prevented motor dysfunction in TMEV mice. (A) Diagram of the experimental groups. Mice were subjected to oral ABX treatment from day 55 to 70 pi or they received the vehicle alone. (B) Horizontal (HACTV) and vertical motor activity (VACTV) was evaluated for 10 min in the Activity cage, while motor (Continued) Frontiers in Immunology | www.frontiersin.org 6 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 1 | coordination was measured over 5 min on a Rotarod apparatus. The data are presented as the means ± SEM: ***p < 0.001 vs. Sham + vehicle; # p < 0.05 vs. TMEV + vehicle; ## p < 0.01 vs. TMEV + vehicle. (C) Agarose gel (1.5%) to visualize the amplicon amplified by 16S (V3–V4) PCR. (D) A principal components analysis of the faecal microbiota from TMEV-vehicle mice (black dots; n = 10) and TMEV mice treated with ABX (red dots; n = 12). Extended error bar plots, at phylum (E) and genus level (F), shows the significant differences in the mean proportions of bacterial taxa in TMEV (black) and TMEV-ABX (red) mice. Corrected p-values are shown on the right. (G) Extended error bars show significantly different KEGG pathway maps in the TMEV-vehicle and TMEV-ABX mice. Deferribacteres), and Coprococcus (phylum Firmicutes) were TMEV mice. However, there was a significant reduction in CD8+ relatively less abundant in the TA mice than in the TMEV T cells in the TAVX mice (p < 0.01) relative to the TMEV mice. mice (T), while there was significantly less Ruminococcus, CD39 expressing CD4+ T cells are important modulators Dehalobacterium, and Oscillospora (phylum Firmicutes) and of autoimmune diseases like MS (39–42). Intriguingly, FACS Mucispirillum (phylum Deferribacteres) in the ABX-TMEV analysis showed fewer Foxp3+ CD39+ CD4+ Treg cells in mice that received Vivomixx (TAVx) than in the TMEV mice. the CNS of TMEV mice and while there was a significant The continued treatment with ABX until day 85 maintained increase after oral administration of ABX, this was reversed a decrease in the relative abundance of Oscillospora and upon cessation of ABX treatment (Figures 3D,E). ABX treatment Ruminococcus (phylum Firmicutes), and an increase in that of also augmented the subpopulation of Foxp3− CD39+ T cells Flavobacterium (phylum Bacteroidetes) and Corynebacterium in the CNS while no significant variations were found in this (phylum Actinobacteria) in the fecal pellets of TAA mice subpopulation under recolonization conditions (Figures 3D,F). (Figure 2F). The experimental evidence obtained in the EAE model There is evidence of the importance of gut microbiota derived suggests that distinct subpopulations of B cells with regulatory fermentative metabolites like short chain fatty acids (SCFAs), phenotypes are involved in protection from the disease (43). such as acetate and butyrate, in autoimmunity and MS (38). Interestingly, there was an increase in the percentage of B Altered SCFA levels in blood could be associated to changes in cells (B220+ CD19+ ) in the CNS of TMEV mice during the gut microbiota so, we analyze the concentrations of acetate, and chronic phase when compared to the sham conditions (p < butyrate in plasma of mice subjected to the different treatments 0.01, Figures 4A,B). CNS of mice treated with ABX show similar (Figure 2G). We observed a significant increase in the levels of percentage of B than TMEV mice, whereas cessation of ABX acetate during recolonization in the mice that received Vivomixx (p < 0.001) as well as ABX withdrawn plus the administration relative to the TMEV mice (p < 0.05). The elevated levels of Vivomixx (p < 0.05) provoked a significant reduction in B of butyrate evident during the recolonization period in the cells. One interesting observation (Figures 4A,C) was that the absence of probiotics (p < 0.05 vs. TMEV mice) likely indicate percentage of regulatory B cells (CD5+ CD1high) increased in that the adjustments to the microbiota populations after ABX TMEV mice (p < 0.01) and that the ABX cocktail enhanced this cessation promote butyrate production, and the elevated plasma population of CNS regulatory B cells (p < 0.05) relative to that levels of this metabolite suggest that the oral administration found in TMEV mice. In the rest of the conditions there were of Vivomixx reinforces butyrate producing bacteria very few or none of these specific Breg cells. Collectively, our (p < 0.01 vs. TMEV mice). results reflect the importance of changes in the microbiota for CNS immunity in the context of the chronic neurodegenerative Gut Microbiota Recolonization Modifies stage of TMEV-IDD. the CNS Immunological Changes Provoked Microglial Changes and Axonal Damage by Oral Antibiotic Treatment A flow cytometry study of the immune response revealed a Associated With Alterations in the significant increase in CD4+ T cells in the CNS of TMEV Microbiota During the Chronic Phase infected mice compared to the sham mice (p < 0.001), which is of TMEV-IDD indicative of CNS infiltration during the chronic phase of TMEV- As innate immune cells, microglia are critical to maintain IDD (Figures 3A,B). In accordance with the data from EAE CNS homeostasis and interestingly, the microbiota may shape mice, oral ABX administration reduced the presence of CD4+ microglial activity (44–46). Here we found a greater area of T cells relative to the TMEV mice (p < 0.05, Figure 3B), and Iba-1 microglial staining in the spinal cord of TMEV mice this reduction in the number of CD4+ T cells persisted during during the chronic phase relative to that in sham mice (p the recolonization period (p < 0.05) and it was potentiated by < 0.05, Figure 5A), consistent with our previous data (30). the presence of Vivomixx (p < 0.01). A similar response was Nevertheless, microglia undergo morphological changes when observed in the population of CD8+ T cells (Figures 3A,C), confronted by different microbiota challenges (Figure 5A). For whereby an increase in the CD8+ T cells that infiltrated the CNS example, cells adopt a transitional morphology in TMEV mice of TMEV mice (p < 0.001) was limited in the TMEV mice that that received ABX, shifting from cells with thin cell bodies were treated with ABX throughout the experiment (TAA group, with numerous branched extensions to more round amoeboid- Figures 3A,C). While there were fewer CD8+ T cells in the CNS like cells with fewer branches and more thickened then, the of mice whose ABX treatment was withdrawn at 70 dpi (TA area of Iba-1 staining increased relative to that in TMEV mice group), this was not significantly different to that observed in (p < 0.01). During recolonization, the microglia maintained Frontiers in Immunology | www.frontiersin.org 7 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 2 | Partial recolonization after cessation of ABX aggravates motor disability in TMEV-IDD. (A) Scheme of the experimental groups: (S) black open squares, sham mice orally administered autoclaved water; (SA) red open square, Sham mice subjected to ABX treatment from day 55 to 85 pi; (T) black square, TMEV mice (Continued) Frontiers in Immunology | www.frontiersin.org 8 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 2 | administered autoclaved water; (TAA) red square, TMEV mice administered ABX from day 55 to 85 pi; (TA) pink square, TMEV mice administered ABX from day 55 to 70 pi followed by autoclaved water; (TAVx) green square, TMEV mice administered ABX from day 55 to 70 pi followed by oral Vivomixx administration from 70 to 85 dpi. (B) Horizontal (HACTV) and vertical activity (VACTV) was evaluated for 10 min in the Activity cage, while motor coordination was measured for 5 min on the Rotarod apparatus. The data are presented as the means ± SEM: **p < 0.01 vs. S; ***p < 0.001 vs. S; # p < 0.05 vs. T. (C) Agarose gel (1.5%) to visualize the amplicon amplified by 16S (V3–V4) PCR. (D) Principal components analysis of the faecal microbiota from the T (black; n = 4); TAA (red; n = 5); TA (blue; n = 3); TAVx (pink; n = 5) experimental groups. (E) Extended error bar plot identifying significant differences between the mean proportions of bacterial taxa upon recolonization following termination of the ABX treatment or recolonization upon Vivomixx administration. (F) Extended error bar plot to visualize significant differences in the bacterial taxa between the T and TAA group, or the TA or TAVx experimental groups. (G) Quantification of the plasma levels of acetate and butyrate. The data are presented as the means ± SEM: # p < 0.05 vs. T; ## p < 0.01 vs. T. their amoeboid shape but the area of Iba-1 staining reduced in the TMEV mice (p < 0.001 vs. Sham), and these cells significantly relative to the ABX-TMEV mice (p < 0.001). By were less prevalent in the spleen of mice that received oral contrast, the mice that received oral Vivomixx had microglia with ABX (p < 0.001 vs. TMEV). Both, recolonization alone or more branches than those observed during recolonization in the in conjunction with Vivomixx did not modify the reduction absence of probiotics. 3D reconstruction videos of representative in the Foxp3+ CD39+ CD4+ and Foxp3− CD39+ CD4+ T cell microglial cells to visualize the morphological changes of subpopulations provoked by ABX (Figure 6C). Again, in contrast microglia from TMEV and ABX-TMEV mice are provided to the CNS, we did not observe changes in the frequency of B as Supplementary Material (T, Supplementary Video 1; TAA, cells under the distinct conditions used (Figures 7A,B). However, Supplementary Video 2). when we gated for B220low CD19+ CD5+ there was a small, yet The expression of the genes encoding the pro-inflammatory significant, increase in CD1dhigh+ cells in ABX TMEV mice cytokines mIL-1β and mTNF-α was assessed in the spinal cord of relative to the TMEV mice (p < 0.05, Figure 7C). Therefore, the mice subjected to the different conditions studied, together with manipulation of the intestinal microbiota affected differentially those encoding the anti-inflammatory cytokines mIL-4 and mIL- the central and peripheral (spleen) immune responses in the 10 (Figure 5B). Treatment with ABX diminished the levels of context of neurodegenerative inflammation. mIL-1β and mTNF-α induced by TMEV, and the recolonization TMEV-infected mice show a significant increase (p < 0.05 after cessation of ABX and probiotic Vivomixx administration vs. sham mice) in the proportion of CD4+ T cells in MLNs, reversed this effect in the case of mTNF-α. Regarding the but no significant changes were observed in this population anti-inflammatory cytokines, oral ABX treatment significantly under the different conditions of gut microbiota manipulation increased mIL-4 and mIL-10 mRNA expression relative to TMEV (Figure 8A). The cytokine IL-17 was quantified by ELISA in mice (p < 0.05), these genes being expressed more strongly than CD4+ lymphocytes harvested from the MLNs of mice after their in sham mice (p < 0.001). After cessation of ABX administration polarization to the Th-17 phenotype. CD4+ T cells obtained from there was a tendency for mIL-4 mRNA expression to decrease, TMEV mice produced significantly more IL-17 (p < 0.001) than whereas Vivomixx treatment reversed IL-4 gene expression to a those from sham mice, whereas oral treatment with ABX reduced similar level as in sham mice. However, the changes in mIL-10 the production of IL-17 in TMEV mice (p < 0.05) (Figure 8B). gene expression were more subtle during such recolonization. A Strikingly, there was significantly less IL-17 after ABX cessation histological analysis of spinal cord sections stained for NF-H was relative to TMEV mice and the presence of Vivomixx potentiated performed to quantify the axonal integrity. The axonal damage this reduction (p < 0.01; Figure 8B). in TMEV mice was prevented by ABX treatment (p < 0.01) and recolonization tended to limit this effect, inasmuch as the damage was not significantly different to that in TMEV mice (Figure 5C). DISCUSSION Peripheral Immunological Changes in In the last 10 years, evidence has emerged of a relationship between gut microbiota and MS not only in animal models but TMEV-IDD Mice Associated With Oral also in patients (13, 14, 47–49). However, the influence of the gut Antibiotic Treatment and Recolonization microbiota on progressive forms of MS is an issue little explored. With or Without Probiotics Hence, in a progressive model of MS we investigated how Minimal changes were detected in the populations of CD4+ microbiota manipulation influences the central and peripheral and CD8+ T cells in the spleen of the TMEV-mice (Figure 6A). immune responses, axonal preservation integrity and motor However, it was noteworthy that CD4 fluorescence intensity function as an index of disease progression. was higher in TMEV mice than in sham, fact that was During the chronic phase of TMEV-IDD, once the prevented by ABX (Figure 6B). In addition, the proportion of symptomatology had been established, changes in the FoxP3− CD39+ CD4+ cells were more prevalent in the TMEV composition of the microbiota were evident between sham mice than in the sham mice (p < 0.001), whereas oral ABX and TMEV mice (30). In the present study the oral ABX cocktail administration significantly reduced the proportion of these cells administered during the pre-symptomatic transitional phase relative to the TMEV mice (p < 0.01), contrasting with the prevented motor dysfunction of TMEV mice. The analysis of data from the CNS. A similar profile was observed for the the bacterial composition in feces revealed a reduction of the subpopulation FoxP3+ CD39+ CD4+ cells, which was enhanced relative abundance of bacteria belonging to Firmicutes and Frontiers in Immunology | www.frontiersin.org 9 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 3 | Changes in T cells in the CNS associated with both dysbiosis and recolonization of the gut microbiota. A single cell leukocyte suspension was obtained from the brain and spinal cord. (A) Representative flow cytometry plots of CD45+ CD3+ CD4+ T cells, CD45+ CD3+ CD8+ T cells. (B,C) Quantification of the change in number of CD45+ CD3+ CD4+ and CD45+ CD3+ CD8+ T cells relative to S mice. (D) Representative flow cytometry plots of CD39+ Foxp3 cells gated on CD45+ CD3+ CD4+ T cells. (E,F) Percentage of Foxp3+ CD39+ cells and Foxp3− CD39+ T cells gated on CD45+ CD3+ CD4+ T cells, respectively. The data represents the change relative to S mice. Groups S, T, and TAA (n = 10), and groups SA, TA, TAVx (n = 5). Statistics: ***p < 0.001 vs. S; # p < 0.05 vs. T; ## p < 0.01 vs. T; + p < 0.05 vs. TAA; ++ p < 0.01 vs. TAA. Frontiers in Immunology | www.frontiersin.org 10 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 4 | Changes in B cells in the CNS associated with both dysbiosis and recolonization of the gut microbiota. A single cell leukocyte suspension was obtained from the brain and spinal cord. (A) Representative flow cytometry plots of B220+ CD19+ B cells and CD5+ CD1d cells, gated on B220+ CD19+ B cells. (B) Quantification of the change in the percentage of B220+ CD19+ B cells relative to the S mice. (C) Percentage of CD5+ CD1dhigh cells in the CNS gated on B220+ CD19+ B cells. Groups S, T and TAA (n = 10), and groups SA, TA, TAVx (n = 5). Statistics: **p < 0.01 vs. S; # p < 0.05 vs. T; ### p < 0.001 vs. T; + p < 0.05 vs. TAA; +++ p < 0.001 vs. TAA. Bacteroidetes phyla, while increasing bacteria belonging to leukocyte antigen (HLA) class II transgenic mouse model of EAE, Actinobacteria and Proteobacteria. Accordingly, previous studies increasing human gut-derived commensal Prevotella was able to on EAE-mice treated with the same ABX cocktail indicated suppress the disease (54). In our study the relative abundance of a decrease of Firmicutes abundance (50) or a macroscopic Prevotella was enhanced in TMEV-mice treated with ABX that GF-like phenotype (51). Particularly, vancomycin reduced the displayed a good performance in the motor tests, supporting the abundance of Firmicutes and Bacteroidetes phyla as well as possible therapeutic role of Prevotella in MS. increased Proteobacteria abundance (52). Although an increase Recolonization following the cessation of ABX had a negative of relative abundance of Proteobacteria has been related to both, effect on TMEV-IDD progression, as reflected by the worsened pro and anti-inflammatory roles in MS (53), our observations motor function and in accordance to previous studies on and the data published in the EAE model suggest a beneficial role GF mice after directed intestinal colonization with segmented of this commensal population. It must be note that variations on filamentous bacteria (20). Indeed, there is significant support for microbiota composition depend on the antibiotic cocktail, doses, the gut microbiota playing a critical role in shaping the immune timing of treatment, or even mice strain. system (55–58). Here, we provide evidence of marked differences The association of the different bacterial communities to between central and peripheral immune responses in mice in the gut-brain-immune system interaction is currently under which the microbiota had been manipulated. Interestingly, oral study. It is therefore significant that Prevotella abundance is less ABX limits the increase in the number of CD4+ and CD8+ T prominent in MS patients (13, 14) and its presence is enhanced cells in the CNS of TMEV mice, which unexpectedly persisted by disease-modifying therapy (14). Moreover, using a human in TMEV mice when the ABX treatment stopped. However, Frontiers in Immunology | www.frontiersin.org 11 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 5 | Dysbiosis and the recolonization of the gut microbiota are associated with changes to microglia and axonal damage in the spinal cord. (A) Representative images of transversal spinal cord sections immunostained with Iba-1 to label microglia. Scale bar 50 µm. In the bottom it is represented higher magnification reconstruction images (63x) of Iba1 stained cells to show the changes in morphology. Scale bar 10 µm. These images are accompanied by a quantification of the relative area occupied by Iba-1 staining. (B) Expression of the inflammatory markers (mIL-1b, mTNF-α, mIL-4, or mIL-10) in the spinal cord was assessed by qRT-PCR analysis and using the 2−11Ct method. (C) Representative images of axonal damage in the spinal cord visualized by Neurofilament-H staining (Scale bar 50 µm). Higher magnification pictures are also presented (Scale bar 25 µm) along with quantification of the area stained with Neurofilament-H. Statistics: *p < 0.05 vs. S; ***p < 0.001 vs. S; # p < 0.05 vs. T; ## p < 0.01 vs. T; + p < 0.05 vs. TAA; +++ p < 0.01 vs. TAA. analyzing the subpopulations of CD4+ T cells in more depth, pathways by which the commensal microbiome affects host recolonization reversed the increase in both Foxp3+ CD39+ T immunity at the distal site is by regulating the activity of cells and Foxp3− CD39+ T cells induced by ABX treatment in Treg cells through the ectonucleotidase CD39 (59). Particularly, TMEV infected mice, supporting the beneficial role of CD39+ Foxp3+ CD39+ T cells isolated from MS patients supress IL-17 Treg cells in preventing MS progression. In fact, one of the production and there was a significant reduction in the CD39+ Frontiers in Immunology | www.frontiersin.org 12 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 6 | Gut microbiota manipulation effect on the spleen T cells in mice. A spleen cell suspension was stained to analyse CD4+ , CD8+ T cell, and CD39+ T cell (A) Representative flow cytometry plots of CD45+ CD3+ CD4+ T cells and CD45+ CD3+ CD8+ T cells are presented along with the corresponding quantification (Continued) Frontiers in Immunology | www.frontiersin.org 13 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 6 | below. (B) Histogram of CD3 and CD4 fluorescence intensity emitted by CD45+ CD3+ CD4+ splenocytes from representative mice of S, T, and TAA experimental groups. The arrow marks the increase of CD4 fluorescence intensity by splenocytes from TMEV mice (C) Flow cytometry plots of Foxp3CD39+ T cells gated on CD45+ CD3+ CD4+ T cells shown together with the quantification of the Foxp3− CD39+ and Foxp3+ CD39+ T cells. Statistics: ***p < 0.001 vs. S; # p < 0.05 vs. T; ## p < 0.01 vs. T; ### p < 0.001 vs. T. FIGURE 7 | Gut microbiota manipulation effect on the spleen B cells in TMEV mice. (A) Representative flow cytometry plots of percentage of B220+ CD19+ B cells are presented along with the corresponding quantification below (B). (C) Normalized histogram of CD1d expression in B220low CD19+ CD5+ cells is presented along with the quantification of the percentage of B220low CD19+ CD5+ CD1dhigh Breg cells. Statistics: # p < 0.05 vs. T. Treg cells in MS patients (39). CD39+ cells have been associated autoimmune disease, potentially involving both the IL-10 and with the suppression of neuroinflammation by Bacteroides fragilis TGF-β produced by Foxp3+ CD39+ and Foxp3− CD39+ cells, in the EAE model (60) in accordance with studies showing respectively (62). that polysaccharide A (PSA) was not able to protect against Although MS has classically been considered a predominantly EAE in CD39− deficent mice (59). Moreover, CD39+ Treg T cell-dependent autoimmune disease, B cells also regulate cells are thought to be involved in the beneficial effects of immune responses and contribute to disease pathogenesis Teriflunomide in the EAE model (40) and an increase in CD39 (63, 64). In particular, regulatory CD5+ CD1high B cells expressing cells occurs in MS patients that receive Fingolimod that produce IL-10 were recently shown to impede EAE (61). It is well-known that CD39 Treg cells play an essential initiation (65), and broad spectrum antibiotic administration role in constraining pathogenic Th17 cells and preventing to EAE mice increased the CD5+ CD1high B cell population Frontiers in Immunology | www.frontiersin.org 14 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS FIGURE 8 | Manipulation of the gut microbiota affects the Th17 responses in the MLN. (A) The frequency of CD4+ T cells increased in TMEV mice relative to sham mice. (B) CD4+ T cells isolated using Miltenyi beads were polarized to Th17 in culture, and IL-17 production was measured by ELISA. The data represent the mean ± SEM of the change relative to Sham (data from two experiments performed in triplicate). Statistics: *p < 0.05 vs. S. ***p < 0.001 vs. S; # p < 0.05 vs. T; ## p < 0.01 vs. T. at distal lymphoid sites (43). Here, ABX treatment spinal cord gene expression profile during recolonization was was seen to augment the regulatory CD5+ CD1dhigh B pro-inflammatory. Most importantly in this context, is that cells in the spleen and CNS, which could be partially oral ABX administration preserves the axonal integrity in the responsible for the increase in IL-10 in the spinal cord of spinal cord of TMEV mice while recolonization increased the TMEV-ABX mice. Indeed, other leukocyte populations, axonal damage. including T lymphocytes, monocytes, and intrinsic CNS To shed further light on our data, we measured in microglia might be contributing to the increased gene plasma the SCFAs, acetate and butyrate (68). Remarkably, the expression of IL-10. Hence, the findings presented here levels of acetate and butyrate in the plasma only increased suggest that gut commensal bacteria might mediate CNS during recolonization, particularly after the application of immune homeostasis. Vivomixx, and probably, in association with an increase of The main changes referred to gut microbiota manipulation Bacteroidetes and Firmicutes in the feces. Therefore, plasma in the TMEV-IDD model were in the CNS where microglia are acetate and butyrate changes seem to reflect critical events the baseline sentinels. The alterations in microglial morphology in the recolonization period after long term oral ABX induced by oral ABX are indicative of the dynamic polarization administration, suggesting the interest of using probiotics to states of these cells. For instance, microglia can retract help recolonization. Although SCFAs ameliorate the disease their processes and enlarge their cell body like activated course of EAE (68, 69), few studies have addressed the phagocytic cells, increasing the relative area of Iba-1 staining. role of these metabolites in MS patients. While MS patients As commented before, the anti-inflammatory gene expression might exhibit a reduction in SCFA producing bacteria (70), profile in the spinal cord was associated with reduced IL-1β, in our study only mice that received Vivomixx during and increased IL-4 and IL-10 mRNA expression. Classically, recolonization improved their deambulation. Whether the microglial activity is influenced by cytokines, chemokines and enhanced deambulatory activity is related to an increase in other micro-environmental factors, shaping their homeostatic acetate and butyrate remains unclear, an issue that merits and pathogenic program (66). Recent studies show striking further study. Our findings also indicate that alterations to differences in the microglia derived from GF and specific bacterial populations during recolonization induce changes in pathogen free (SPF) mice at the genetic and morphological level CNS immunity, being the most significant the diminished (44), and signals from maternal gut microbes could influence CD39+ Treg and Breg cells following microbiota depletion by microglia even at birth (67). We suggest that the changes oral ABX. in the microbiota produced by oral ABX maintain microglia Unlike EAE, the main changes provoked by manipulating in a protective state, attenuating the damage associated with the gut microbiota in the TMEV-IDD model were evident chronic neuroinflammation that occurs in progressive MS. at the CNS with modest changes in the periphery. Although Ameboid microglial cells were also present in the spinal cord EAE and TMEV-IDD share pathogenic mechanisms, these two during recolonization. However, the Iba-1 stained area was models have important differences that may reflect mouse strain significantly smaller than that observed in TMEV mice that (C57BL/6 vs. SJL/J) specific responses regarding immunity, received ABX, likely suggesting a reduction in the protective neuroinflammation and clinical outcomes. This would explain microglia recruited to inflammatory sites. More ramified why our main findings after gut microbiota manipulation occur microglia was associated to Vivomixx treatment, although the in the CNS while the alterations in peripheral compartments are Frontiers in Immunology | www.frontiersin.org 15 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS mild in comparison with the changes observed in EAE in the AUTHOR CONTRIBUTIONS same context. At spleen level, ABX inhibited the induction of CD4 expression observed in the CD4+ T cells from TMEV mice; LM and CG designed the research. LM, FC-S, MM, and AF this fact is relevant since these CD4high cells, and not CD4norm performed the research. CE, JÁ-C, and LV were advisors involved cells, cause EAE after adoptive transfer to naïve recipients in the discussion of the results. LM and CG wrote the manuscript. (71). The contribution of the MLN to pro-inflammatory Th- All the authors contributed to and approved the final version of 17-cell generation during inflammation of the small intestine the manuscript submitted for publication. has been reported previously (72) and dendritic cells exposed to PSA migrate to the MLN, where they favor the presence of FUNDING IL-10 producing Treg cells regulating the Th1/Th17 response in Crohn’s disease (73). Our ex vivo experiments with sorted This work was supported by grants from the Ministerio de CD4+ T cells from MLNs cultured in conditions favoring Th- Economía y Competitividad (MINNECO SAF2016-76449-R) 17 polarization generated interesting data. Firstly, CD4+ T and the Red Española de Esclerosis Múltiple (REEM: cells from TMEV mice were capable of further enhancing IL- RD16/0015/0021), sponsored by the Fondo de Investigación 17 production and secondly, CD4+ cells from ABX-TMEV Sanitaria (FIS). mice released less IL-17, consistent with previous studies (18). However, recolonization failed to reverse the reduction in IL- ACKNOWLEDGMENTS 17 and Vivomixx was not sufficient to recover the ABX- induced suppression of IL-17 by CD4+ T cells sorted from We thank Laura Ramos and Nieves López de Andradas for their the MLN. excellent technical support. We are also grateful to Dr. Moses One of the main messages of our study is precisely that Rodríguez (Department of Immunology and Neurology, Mayo at the chronic neurodegenerative phase of TMEV-IDD, the Clinic/Foundation, Rochester, MN, USA) for the generous gift main immunological changes related to the manipulation of the Theiler’s virus DA strain. We also thank the CBMSO- of the gut microbiota occur within the CNS. Our data also Genomics & Massive Sequencing service, CBMSO-Cytometry support a link gut microbiota-CNS axis, since microbiota service, the Fundación Parque Científico de Madrid, and Instituto dysbiosis provoked by oral ABX exerts neuroprotective Cajal-microscopy Unit for their brilliant technical assistance. effects, diminishing motor disability and axonal damage. The authors would also like to thank José L. Luque-García Nevertheless, exactly how the gut microbiota modulates and Héctor Estévez-Sánchez (Grupo de trazas, especialización the development and disease course of TMEV-IDD is still y proteómica, Dept Química Analítica, UCM, Madrid) for unclear. Further research is necessary to better understand the the detection of the plasma metabolites. We also acknowledge mechanisms and signaling pathways involved in coordinating support of the publication fee by the CSIC Open Access the gut microbiota and the CNS in progressive forms Publication Support Initiative through its Unit of Information of MS. Resources for Research (URICI). DATA AVAILABILITY SUPPLEMENTARY MATERIAL The raw data supporting the conclusions of this manuscript will The Supplementary Material for this article can be found be made available by the authors, without undue reservation, to online at: https://www.frontiersin.org/articles/10.3389/fimmu. any qualified researcher. 2019.01374/full#supplementary-material Supplementary Video 1 | 3D reconstruction of microglia cells from TMEV-mice. Digital reconstruction by IMARIS software of microglial cells surface obtained from ETHICS STATEMENT 63x confocal image from the spinal cord of TMEV infected mice. Red cells correspond to Iba-1 positive cells. The morphology of representative aisled cell are All experiments were performed in strict accordance with presented in yellow and DAPI in blue. EU (Directive 2010/63/EU) and Spanish regulations (Royal Supplementary Video 2 | 3D reconstruction of microglia cells from ABX-TMEV Decree 53/2013 BOE n◦ 34 and Comunidad de Madrid mice. 63x confocal image from transversal spinal cord section from TMEV mice decree: ES 280790000184). The Ethics Committee on Animal subjected to ABX treatment were processed by IMARIS software and 3D reconstruction of microglial cells were obtained. Red cells correspond to Iba-1 Experimentation at the Instituto Cajal (CSIC) also approved positive cells. The morphology of representative cells is showed in yellow and an all the procedures employed in this study (protocol number: isolated cell could not be dis distinguished. DAPI staining for nucleus is present 2013/03 CEEA-IC). in blue. REFERENCES and the risk of multiple sclerosis. Arch Neurol. (2010) 67:1034–5. doi: 10.1001/archneurol.2010.182 1. Ramagopalan SV, Hanwell HE, Giovannoni G, Knappskog PM, Nyland 2. Ascherio A. Environmental factors in multiple sclerosis. Expert Rev Neurother. HI, Myhr KM, et al. Vitamin D-dependent rickets, HLA-DRB1, (2013) 13:3–9. doi: 10.1586/14737175.2013.865866 Frontiers in Immunology | www.frontiersin.org 16 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS 3. Belbasis L, Bellou V, Evangelou E, Ioannidis JP, Tzoulaki I. Environmental 21. Mao YK, Kasper DL, Wang B, Forsythe P, Bienenstock J, Kunze WA. risk factors and multiple sclerosis: an umbrella review of systematic Bacteroides fragilis polysaccharide A is necessary and sufficient for acute reviews and meta-analyses. Lancet Neurol. (2015) 14:263–73. activation of intestinal sensory neurons. Nat Commun. (2013) 4:1465. doi: 10.1016/S1474-4422(14)70267-4 doi: 10.1038/ncomms2478 4. Miyake S, Yamamura T. Gut environmental factors and multiple sclerosis. J 22. Lavasani S, Dzhambazov B, Nouri M, Fak F, Buske S, Molin G, et al. A novel Neuroimmunol. (2019) 329:20–3. doi: 10.1016/j.jneuroim.2018.07.015 probiotic mixture exerts a therapeutic effect on experimental autoimmune 5. Trapp BD, Nave KA. Multiple sclerosis: an immune or encephalomyelitis mediated by IL-10 producing regulatory T cells. PLoS ONE. neurodegenerative disorder? Annu Rev Neurosci. (2008) 31:247–69. (2010) 5:e9009. doi: 10.1371/journal.pone.0009009 doi: 10.1146/annurev.neuro.30.051606.094309 23. Berer K, Gerdes LA, Cekanaviciute E, Jia X, Xiao L, Xia Z, et al. Gut 6. Sato F, Martinez NE, Stewart EC, Omura S, Alexander JS, Tsunoda I. microbiota from multiple sclerosis patients enables spontaneous autoimmune “Microglial nodules” and “newly forming lesions” may be a Janus face of early encephalomyelitis in mice. Proc Natl Acad Sci USA. (2017) 114:10719–24. MS lesions; implications from virus-induced demyelination, the Inside-Out doi: 10.1073/pnas.1711233114 model. BMC Neurol. (2015) 15:219. doi: 10.1186/s12883-015-0478-y 24. Cekanaviciute E, Yoo BB, Runia TF, Debelius JW, Singh S, Nelson CA, et al. 7. Oleszak EL, Chang JR, Friedman H, Katsetos CD, Platsoucas CD. Theiler’s Gut bacteria from multiple sclerosis patients modulate human T cells and virus infection: a model for multiple sclerosis. Clin Microbiol Rev. (2004) exacerbate symptoms in mouse models. Proc Natl Acad Sci USA. (2017) 17:174–207. doi: 10.1128/cmr.17.1.174-207.2004 114:10713–18 doi: 10.1073/pnas.1711235114 8. Katz-Levy Y, Neville KL, Padilla J, Rahbe S, Begolka WS, Girvin AM, 25. Bhargava P, Mowry EM. Gut microbiome and multiple sclerosis. Curr Neurol et al. Temporal development of autoreactive Th1 responses and endogenous Neurosci Rep. (2014) 14:492. doi: 10.1007/s11910-014-0492-2 presentation of self myelin epitopes by central nervous system-resident 26. Lledo A, Borrell J, Guaza C. Dexamethasone regulation of interleukin-1- APCs in Theiler’s virus-infected mice. J Immunol. (2000) 165:5304–14. receptors in the hippocampus of Theiler’s virus-infected mice: effects on doi: 10.4049/jimmunol.165.9.5304 virus-mediated demyelination. Eur J Pharmacol. (1999) 372:75–83. 9. Arevalo-Martin A, Molina-Holgado E, Guaza C. A CB(1)/CB(2) 27. Rodriguez M, Leibowitz JL, Lampert PW. Persistent infection of receptor agonist, WIN 55,212-2, exerts its therapeutic effect in oligodendrocytes in Theiler’s virus-induced encephalomyelitis. Ann Neurol. a viral autoimmune model of multiple sclerosis by restoring (1983) 13:426–33. self-tolerance to myelin. Neuropharmacology. (2012) 63:385–93. 28. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, Edberg S, Medzhitov R. doi: 10.1016/j.neuropharm.2012.04.012 Recognition of commensal microflora by toll-like receptors is required 10. Croxford JL, Olson JK, Miller SD. Epitope spreading and molecular for intestinal homeostasis. Cell. (2004) 118:229–41. doi: 10.1016/j.cell.2004. mimicry as triggers of autoimmunity in the Theiler’s virus-induced 07.002 demyelinating disease model of multiple sclerosis. Autoimmun Rev. 29. Morgun A, Dzutsev A, Dong X, Greer RL, Sexton DJ, Ravel J, et al. Uncovering (2002) 1:251–60. doi: 10.1016/S1568-9972(02)00080-0 effects of antibiotics on the host and microbiota using transkingdom gene 11. Kap YS, Bus-Spoor C, van Drie N, Dubbelaar ML, Grit C, Kooistra SM, networks. Gut. (2015) 64:1732–43. doi: 10.1136/gutjnl-2014-308820 et al. Targeted diet modification reduces multiple sclerosis-like disease in adult 30. Carrillo-Salinas FJ, Mestre L, Mecha M, Feliu A, Del Campo R, Villarrubia marmoset monkeys from an outbred colony. J Immunol. (2018) 201:3229–43. N, et al. Gut dysbiosis and neuroimmune responses to brain infection doi: 10.4049/jimmunol.1800822 with Theiler’s murine encephalomyelitis virus. Sci Rep. (2017) 7:44377. 12. Berer K, Mues M, Koutrolos M, Rasbi ZA, Boziki M, Johner C, doi: 10.1038/srep44377 et al. Commensal microbiota and myelin autoantigen cooperate 31. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello to trigger autoimmune demyelination. Nature. (2011) 479:538–41. EK, et al. QIIME allows analysis of high-throughput community sequencing doi: 10.1038/nature10554 data. Nat Methods. (2010) 7:335–6. doi: 10.1038/nmeth.f.303 13. Miyake S, Kim S, Suda W, Oshima K, Nakamura M, Matsuoka T, et al. 32. Ubeda C, Taur Y, Jenq RR, Equinda MJ, Son T, Samstein M, et al. Vancomycin- Dysbiosis in the gut microbiota of patients with multiple sclerosis, with a resistant Enterococcus domination of intestinal microbiota is enabled by striking depletion of species belonging to clostridia XIVa and IV clusters. PLoS antibiotic treatment in mice and precedes bloodstream invasion in humans. ONE. (2015) 10:e0137429. J Clin Invest. (2010) 120:4332–41. doi: 10.1172/JCI43918 14. Jangi S, Gandhi R, Cox LM, Li N, von Glehn F, Yan R, et al. Alterations of the 33. Langille MG, Zaneveld J, Caporaso JG, McDonald D, Knights D, Reyes human gut microbiome in multiple sclerosis. Nat Commun. (2016) 7:12015. JA, et al. Predictive functional profiling of microbial communities using doi: 10.1038/ncomms12015 16S rRNA marker gene sequences. Nat Biotechnol. (2013) 31:814–21. 15. Vatanen T, Franzosa EA, Schwager R, Tripathi S, Arthur TD, Vehik K, et al. doi: 10.1038/nbt.2676 The human gut microbiome in early-onset type 1 diabetes from the TEDDY 34. Parks DH, Tyson GW, Hugenholtz P, Beiko RG. STAMP: statistical analysis study. Nature. (2018) 562:589–94. doi: 10.1038/s41586-018-0620-2 of taxonomic and functional profiles. Bioinformatics. (2014) 30:3123–4. 16. Pianta A, Arvikar SL, Strle K, Drouin EE, Wang Q, Costello CE, et al. doi: 10.1093/bioinformatics/btu494 Two rheumatoid arthritis-specific autoantigens correlate microbial immunity 35. Mestre L, Docagne F, Correa F, Loria F, Hernangomez M, Borrell J, et al. A with autoimmune responses in joints. J Clin Invest. (2017) 127:2946–56. cannabinoid agonist interferes with the progression of a chronic model of doi: 10.1172/JCI93450 multiple sclerosis by downregulating adhesion molecules. Mol Cell Neurosci. 17. Ochoa-Reparaz J, Mielcarz DW, Ditrio LE, Burroughs AR, Foureau DM, (2009) 40:258–66. doi: 10.1016/j.mcn.2008.10.015 Haque-Begum S, et al. Role of gut commensal microflora in the development 36. Loria F, Petrosino S, Hernangomez M, Mestre L, Spagnolo A, Correa of experimental autoimmune encephalomyelitis. J Immunol. (2009) 183:6041– F, et al. An endocannabinoid tone limits excitotoxicity in vitro and 50. doi: 10.4049/jimmunol.0900747 in a model of multiple sclerosis. Neurobiol Dis. (2010) 37:166–76. 18. Ochoa-Reparaz J, Mielcarz DW, Ditrio LE, Burroughs AR, Begum-Haque S, doi: 10.1016/j.nbd.2009.09.020 Dasgupta S, et al. Central nervous system demyelinating disease protection 37. Feliu A, Bonilla Del Rio I, Carrillo-Salinas FJ, Hernandez-Torres G, Mestre L, by the human commensal Bacteroides fragilis depends on polysaccharide A Puente N, et al. 2-Arachidonoylglycerol reduces proteoglycans and enhances expression. J Immunol. (2010) 185:4101–8. doi: 10.4049/jimmunol.1001443 remyelination in a progressive model of demyelination. J Neurosci. (2017) 19. Yokote H, Miyake S, Croxford JL, Oki S, Mizusawa H, Yamamura T. NKT cell- 37:8385–98. doi: 10.1523/JNEUROSCI.2900-16.2017 dependent amelioration of a mouse model of multiple sclerosis by altering gut 38. Rooks MG, Garrett WS. Gut microbiota, metabolites and host immunity. Nat flora. Am J Pathol. (2008) 173:1714–23. doi: 10.2353/ajpath.2008 Rev Immunol. (2016) 16:341–52. doi: 10.1038/nri.2016.42 20. Lee YK, Menezes JS, Umesaki Y, Mazmanian SK. Proinflammatory T- 39. Fletcher JM, Lonergan R, Costelloe L, Kinsella K, Moran B, O’Farrelly C, cell responses to gut microbiota promote experimental autoimmune et al. CD39+Foxp3+ regulatory T Cells suppress pathogenic Th17 cells encephalomyelitis. Proc Natl Acad Sci USA. (2011) 108(Suppl. 1):4615–22. and are impaired in multiple sclerosis. J Immunol. (2009) 183:7602–10. doi: 10.1073/pnas.1000082107 doi: 10.4049/jimmunol.0901881 Frontiers in Immunology | www.frontiersin.org 17 June 2019 | Volume 10 | Article 1374 Mestre et al. Gut Microbiota and Progressive MS 40. Ochoa-Reparaz J, Colpitts SL, Kircher C, Kasper EJ, Telesford KM, Begum- capacity of CD39(+) CD4 T regulatory subsets in the suppression of Haque S, et al. Induction of gut regulatory CD39(+) T cells by teriflunomide neuroinflammation. Gut Microbes. (2014) 5:552–61. doi: 10.4161/gmic. protects against EAE. Neurol Neuroimmunol Neuroinflamm. (2016) 3:e291. 29797 doi: 10.1212/NXI.0000000000000291 60. Telesford KM, Yan W, Ochoa-Reparaz J, Pant A, Kircher C, Christy 41. Ochoa-Reparaz J, Kasper LH. The influence of gut-derived CD39 regulatory MA, et al. A commensal symbiotic factor derived from Bacteroides T cells in CNS demyelinating disease. Transl Res. (2017) 179:126–38. fragilis promotes human CD39(+)Foxp3(+) T cells and Treg function. doi: 10.1016/j.trsl.2016.07.016 Gut Microbes. (2015) 6:234–42. doi: 10.1080/19490976.2015.10 42. Pant AB, Wang Y, Mielcarz DW, Kasper EJ, Telesford KM, Mishra M, 56973 et al. Alteration of CD39+Foxp3+ CD4 T cell and cytokine levels in 61. Muls N, Dang HA, Sindic CJ, van Pesch V. Fingolimod increases CD39- EAE/MS following anti-CD52 treatment. J Neuroimmunol. (2017) 303:22–30. expressing regulatory T cells in multiple sclerosis patients. PLoS ONE. (2014) doi: 10.1016/j.jneuroim.2016.12.010 9:e113025. doi: 10.1371/journal.pone.0113025 43. Ochoa-Reparaz J, Mielcarz DW, Haque-Begum S, Kasper LH. Induction of 62. Kochetkova I, Thornburg T, Callis G, Pascual DW. Segregated a regulatory B cell population in experimental allergic encephalomyelitis by regulatory CD39+CD4+ T cell function: TGF-beta-producing Foxp3- alteration of the gut commensal microflora. Gut Microbes. (2010) 1:103–8. and IL-10-producing Foxp3+ cells are interdependent for protection doi: 10.4161/gmic.1.2.11515 against collagen-induced arthritis. J Immunol. (2011) 187:4654–66. 44. Erny D, Hrabe de Angelis AL, Jaitin D, Wieghofer P, Staszewski O, David E, doi: 10.4049/jimmunol.1100530 et al. Host microbiota constantly control maturation and function of microglia 63. Yanaba K, Bouaziz JD, Matsushita T, Magro CM, St. Clair EW, Tedder in the CNS. Nat Neurosci. (2015) 18:965–77. doi: 10.1038/nn.4030 TF. B-lymphocyte contributions to human autoimmune disease. 45. Wang Y, Wang Z, Wang Y, Li F, Jia J, Song X, et al. The gut-microglia Immunol Rev. (2008) 223:284–99. doi: 10.1111/j.1600-065X.2008.0 connection: implications for central nervous system diseases. Front Immunol. 0646.x (2018) 9:2325. doi: 10.3389/fimmu.2018.02325 64. Sospedra M. B cells in multiple sclerosis. Curr Opin Neurol. (2018) 31:256–62. 46. Abdel-Haq R, Schlachetzki JCM, Glass CK, Mazmanian SK. Microbiome- doi: 10.1097/WCO.000000000000563 microglia connections via the gut-brain axis. J Exp Med. (2019) 216:41–59. 65. Matsushita T, Yanaba K, Bouaziz JD, Fujimoto M, Tedder TF. Regulatory doi: 10.1084/jem.20180794 B cells inhibit EAE initiation in mice while other B cells promote disease 47. Banati M, Csecsei P, Koszegi E, Nielsen HH, Suto G, Bors L, et al. progression. J Clin Invest. (2008) 118:3420–30. doi: 10.1172/JCI36030 Antibody response against gastrointestinal antigens in demyelinating 66. Gertig U, Hanisch UK. Microglial diversity by responses and responders. Front diseases of the central nervous system. Eur J Neurol. (2013) 20:1492–5. Cell Neurosci. (2014) 8:101. doi: 10.3389/fncel.2014.00101 doi: 10.1111/ene.12072 67. Thion MS, Low D, Silvin A, Chen J, Grisel P, Schulte-Schrepping J, et al. 48. Mielcarz DW, Kasper LH. The gut microbiome in multiple sclerosis. Curr Microbiome influences prenatal and adult microglia in a sex-specific manner. Treat Options Neurol. (2015) 17:344. doi: 10.1007/s11940-015-0344-7 Cell. (2018) 172:500–16 e516. doi: 10.1016/j.cell.2017.11.042 49. Chen J, Chia N, Kalari KR, Yao JZ, Novotna M, Paz Soldan MM, et al. Multiple 68. Haghikia A, Jorg S, Duscha A, Berg J, Manzel A, Waschbisch A, et al. Dietary sclerosis patients have a distinct gut microbiota compared to healthy controls. fatty acids directly impact central nervous system autoimmunity via the small Sci Rep. (2016) 6:28484. doi: 10.1038/srep28484 intestine. Immunity. (2015) 43:817–29. doi: 10.1016/j.immuni.2015.09.007 50. Hill DA, Hoffmann C, Abt MC, Du Y, Kobuley D, Kirn TJ, et al. Metagenomic 69. Mizuno M, Noto D, Kaga N, Chiba A, Miyake S. The dual role of short fatty analyses reveal antibiotic-induced temporal and spatial changes in intestinal acid chains in the pathogenesis of autoimmune disease models. PLoS ONE. microbiota with associated alterations in immune cell homeostasis. Mucosal (2017) 12:e0173032. doi: 10.1371/journal.pone.0173032 Immunol. (2010) 3:148–58. doi: 10.1038/mi.2009.132 70. Melbye P, Olsson A, Hansen TH, Sondergaard HB, Bang Oturai A. Short-chain 51. Reikvam DH, Erofeev A, Sandvik A, Grcic V, Jahnsen FL, Gaustad fatty acids and gut microbiota in multiple sclerosis. Acta Neurol Scand. (2018) P, et al. Depletion of murine intestinal microbiota: effects on gut 139:208–19. doi: 10.1111/ane.13045 mucosa and epithelial gene expression. PLoS ONE. (2011) 6:e17996. 71. Li J, Ridgway W, Fathman CG, Tse HY, Shaw MK. High cell surface doi: 10.1371/journal.pone.0017996 expression of CD4 allows distinction of CD4(+)CD25(+) antigen-specific 52. Murphy EF, Cotter PD, Hogan A, O’Sullivan O, Joyce A, Fouhy F, effector T cells from CD4(+)CD25(+) regulatory T cells in murine et al. Divergent metabolic outcomes arising from targeted manipulation experimental autoimmune encephalomyelitis. J Neuroimmunol. (2007) of the gut microbiota in diet-induced obesity. Gut. (2013) 62:220–6. 192:57–67. doi: 10.1016/j.neuroim.2007.09.004 doi: 10.1136/gutjnl-2011-300705 72. Kawabe T, Suzuki N, Yamaki S, Sun SL, Asao A, Okuyama Y, et al. Mesenteric 53. Shahi SK, Freedman SN, Mangalam AK. Gut microbiome in multiple lymph nodes contribute to proinflammatory Th17-cell generation during sclerosis: the players involved and the roles they play. Gut Microbes. (2017) inflammation of the small intestine in mice. Eur J Immunol. (2016) 46:1119– 8:607–15. doi: 10.1080/19490976.2017.1349041 31. doi: 10.1002/eji.201545907 54. Mangalam A, Shahi SK, Luckey D, Karau M, Marietta E, Luo 73. Sakuraba A, Sato T, Kamada N, Kitazume M, Sugita A, Hibi T. N, et al. Human gut-derived commensal bacteria suppress CNS Th1/Th17 immune response is induced by mesenteric lymph node inflammatory and demyelinating disease. Cell Rep. (2017) 20:1269–77. dendritic cells in Crohn’s disease. Gastroenterology. (2009) 137:1736–45. doi: 10.1016/j.celrep.2017.07.031 doi: 10.1053/j.gastro.2009.07.049 55. Lathrop SK, Bloom SM, Rao SM, Nutsch K, Lio CW, Santacruz N, et al. Peripheral education of the immune system by colonic commensal Conflict of Interest Statement: The authors declare that the research was microbiota. Nature. (2011) 478:250–4. doi: 10.1038/nature10434 conducted in the absence of any commercial or financial relationships that could 56. Wu HJ, Wu E. The role of gut microbiota in immune homeostasis and be construed as a potential conflict of interest. autoimmunity. Gut Microbes. (2012) 3:4–14. doi: 10.4161/gmic.19320 57. Honda K, Littman DR. The microbiota in adaptive immune homeostasis and Copyright © 2019 Mestre, Carrillo-Salinas, Mecha, Feliú, Espejo, Álvarez-Cermeño, disease. Nature. (2016) 535:75–84. doi: 10.1038/nature18848 Villar and Guaza. This is an open-access article distributed under the terms of 58. Colpitts SL, Kasper LH. Influence of the gut microbiome on autoimmunity the Creative Commons Attribution License (CC BY). The use, distribution or in the central nervous system. J Immunol. (2017) 198:596–604. reproduction in other forums is permitted, provided the original author(s) and the doi: 10.4049/jimmunol.1601438 copyright owner(s) are credited and that the original publication in this journal 59. Wang Y, Begum-Haque S, Telesford KM, Ochoa-Reparaz J, Christy M, Kasper is cited, in accordance with accepted academic practice. No use, distribution or EJ, et al. A commensal bacterial product elicits and modulatesmigratory reproduction is permitted which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 18 June 2019 | Volume 10 | Article 1374