LP and LB improved EAN
To investigate the therapeutic effect of Lactobacillus on EAN, animals in the EAN group and the treatment groups were administered 300 µg P0180-199 emulsified to a complete Freund adjuvant (10 mg/ml M. tuberculosis H37Ra) with bilateral plantar injection of 150 µl per rat on days 1 and 7 of immunization. The treatment group started gavage LP or LB (5*109 CFU/ml) daily for one week before immunization, and the model group and the control group started gavage saline. Onsets were observed on day 9 postimmunization in the EAN group and on day 11 in the treatment groups. Disease progression was rapid, peaking at day 18 postimmunization, and the score was 1.67 ± 0.235 in the LP group, 2.0 ± 0.471 in the LB group, and 2.833 ± 0.235 in the EAN group on day 18 postimmunization. The scores in the treatment groups were significantly lower than in the EAN group (p < 0.05) (Fig. 1A). Weights were recorded between treatment and model groups during disease progression and there was no statistically significant difference in body weights (Fig. 1B).
LP and LB improved the ratio of CD4 + T/CD8 + T cells in peripheral blood
The CD4 + T/CD8 + T cells ratio was critical in the immune balance of EAN development at 11th (early onset) and 18th (peak) after immunization. Tail vein blood was collected to detect the ratio. At 11th (early onset) after immunization, CD3 + T in the EAN model group was significantly lower than that of normal control (p < 0.05), but the CD4 + T/CD8 + T ratio was normal (Fig. 2C). On day 18 after immunization (peak), CD3+T downregulation was more pronounced, CD4+T/CD8+T ratio was significantly upregulated (p < 0.05) and CD4 + T/CD8 + T were downregulated after LP and LB treatments, and LP improved CD3+T downregulation (Fig. 2D).
LP and LB improved the release of inflammatory factors in peripheral blood
Peripheral blood was collected from the tail vein at the beginning of the disease (11th day after immunization) and at the peak of the disease (18th day after immunization). Supernatant was taken after centrifugation to detect IL-1, IL-6 and TNF-α levels. In the early stages, IL-1, IL-6 and TNF-α in peripheral blood did not change significantly (Fig. 3A). At the peak of the disease, IL-1, IL-6, TNF-α were significantly upregulate in the EAN model group, and LP could downregulate IL-1, IL-6 and TNF-α, while LB had little effect on inflammatory factors secretion (Fig. 3B).
Improvement of pathological changes in sciatic nerve after treatment with LP or LB
At peak (day 18 after immunization), the rat sciatic nerve was taken, and pathological changes of the sciatic nerve were assessed by electron microscopy after sectioning, while HE and LFB staining were performed after section embedding to assess the degree of inflammatory cell infiltration and demyelination of the sciatic nerve. The results showed that the axons of nerve fibers in the control group were completely arranged in order, the morphology of myelin sheath was regular, and the morphology of nucleus was normal (Fig. 4A). The nerve axons of the EAN model group were incomplete, the myelin sheath was shed, and the morphology was clearly irregular and loosely arranged. Compared with the EAN model group, the LP treatment group had a regular axon arrangement and improved myelin morphology. However, the sciatic nerves in the LB treatment group were loosely arranged, and the thickness of the myelin sheath was significantly thinner (Fig. 4A).The HE staining results showed that the inflammatory cell infiltration in EAN group was evident, and inflammatory cell infiltration was significantly improved in the LP treatment group. LB had no obvious effect on inflammatory cell infiltration (Fig. 4B). Figure 4C was the LFB staining of the sciatic nerve, and the results showed that the myelin sheath in the control group was intact and neatly arranged, while it was loosely arranged and had a large area of myelin shedding in the EAN model group, and the LP treatment group improved the myelin shedding situation. There was no significant improvement in EAN demyelination in the LB treatment group. There was a statistical analysis of the neural demyelinating staining score, with 1 being mild demyelinating, 2 moderately demyelinating, and 3 severely demyelinating(Fig. 4C).
Effects of LB and LP on the intestine
Colonic specimens were taken at the peak of the disease for HE staining and Alixinlan staining to observe inflammatory infiltration and mucosal integrity. The results showed that intestinal mucosa was destroyed, structural disorder and inflammatory cell infiltration were evident in the EAN model group, while intestinal inflammation in the LB or LP treatment group was significantly improved, and intestinal mucosal structure was regular (Fig. 5A, B). At the same time, the expression of Occludin, ZO-1 and reg3γ genes was detected at mRNA level, and IL-1, IL-6 and TNF-α gene expression levels were detected by ELISA. The results showed that Occludin and ZO-1 gene expression levels were significantly downregulated in the EAN model group, while the LP or LB treatment group improved this downregulation to be close to that of the NC normal control group (Fig. 5C). Reg3γ in the EAN model group was significantly upregulated, while reg3γ could be downregulated in the LB or LP treatment group. In addition, IL-1β and TNF-α were upregulated in the EAN model group, while LP treatment improved this upregulation (p < 0.05).
Fecal differential metabolite analysis
Rat feces were collected for metabolomic determination, and 1797 metabolites were identified, which were classified as lipids (23.205%), lipid-like molecules (21.48%), organic heterocyclic compounds (15.693%), and benzene compounds (11.964%) according to their chemical taxonomy attribution information, organic acids and derivatives (Fig. 6A). Among NC group and EAN group, 2,3-quinoxalinedione, 1,4-dihydro-6,7-dinitro-, Cis-vaccenic acid, Podocarpic acid, D-galacturonic acid, Sarcosine, Corticosterone, D-2-aminobutyric acid, D-proline, Lecanoric acid were upregulated. Magnolol, Mitoxantrone, Methylprednisolone succinate, Mestranol, 4-hydroxybenzoate, 6-hydroxyhexanoate, Cyclohexylpentanorprostaglandin were downregulated (Fig. 6B). There were 34 metabolic differences between the LP treatment group and the EAN model group, among which the upregulated metabolites were Indole − 3−acetaldehyde, 4'−hydroxychalcone, Glutaric acid, Cochlioquinone acid, Dodecanedioic acid, Palmitic acid, Ethanone, Glabridin, Mitoxantrone. Downregulated are Caproic acid, Isovaleric acid, Picolinic acid, Fosamine, Phenylpyruvate, D − ornithine, D − proline, 2,3 − quinoxalinedione, 1,4 − dihydro − 6,7 − dinitro−, Sarcosine, 2 − ketohexanoic acid, Glycine, Oxypurinol, DL − threonine, D − 2−aminobutyric acid, Coumatetralyl, Pseudouridine, Lecanoric ACID (Fig. 6C). 2,3-quinoxalinedione, D-proline, Lecanoric acid, D-2-aminobutyric acid, Sarcosine, and Caproic acid that were upregulated in the EAN model group were all differentially downregulated after LP treatment. Mitoxantrone, which was down-regulated in the EAN model group, was differentially up-regulated in the LP-treated group. Indole-3-acetaldehyde, which was the most significant upregulation in the LP treatment group, was an NF-KB inhibitor. Magnolol, was also an NF-KB inhibitor.
KEGG differential metabolic pathway analysis
KEGG (http://www.kegg.jp/) is one of the most commonly used database of gene pathways. KEGG database includes pathway information, genetic information processing, environmental information processing, cellular processes, biological systems, human diseases, and drug development. The results of metabolic pathway enrichment analysis were presented in a histogram, and the EAN model group was enriched in three metabolic pathways. These included Arginine and proline metabolism, Regulation of actin cytoskeleton, and prion diseases (Fig. 7A). Differential metabolites in the LP group were enriched for Biosynthesis of amino acids; Glycine, serine and threonine metabolism; Protein digestion and absorption; Pantothenate and CoA biosynthesis; Vitamin digestion and absorption; beta − Alanine metabolism; Mineral absorption; Lysine degradation; Aminoacyl − tRNA biosynthesis; Central carbon metabolism in cancer; Valine, leucine and isoleucine Biosynthesis; Arginine biosynthesis; HIF-1 signaling pathway (Fig. 7B). The Differential Abundance Score of the enriched pathways showed that the metabolic pathways of the differential metabolites in the EAN model group were significantly upregulated, while the metabolic pathways of the differential metabolites in the LP treatment group were significantly downregulated (Fig. 7C-D). In order to facilitate the observation of the expression of each differential metabolite annotated in the KEGG metabolic pathway, the KEGG metabolic pathway with a number of differential metabolites greater than 2 was selected in this study, and co-metabolic differences containing LP and EAN were screened, and the differential metabolites were displayed as heat maps, and the results showed that the differential metabolites DL-proline, D-proline, Sarcosine in Arginine and proline metabolism significantly upregulated. Upregulation of Pyruvate in the LP-treated group and down-regulation of D − proline and Sarcosine enriched the same pathway in Arginine and proline metabolism, which was downregulated in the LP-treated group compared to the EAN model group (Fig. 7E-F).
Analysis of microbial community structure and differences
The Community Structure Component Map showed the community structure of each group at different classification levels. Figure 8A showed the top ten species at the family level, of which the top five were Lactobacillaceae, Ruminococcaceae, Muribaculaceae, Lachnospiraceae, Peptostreptococcaceae. Ruminococcaceae and Lachnospiraceae were upregulated in the EAN model group, while LP treatment downregulated Ruminococcaceae and Lachnospiraceae. The EAN model group downgraded Muribaculaceae. Alpha Diversity was used to analyze the diversity of microbial communities within the sample (Within-Community), and there was no statistically significant difference in the results indicators Shannon index, ACE index, Observed OUT, PD whole tree (P > 0.05) (Fig. 8B). Further Anosim analysis was performed based on Unweighted UnifracBeta distance to analyze the β diversity. Further Anosim analysis was a nonparametric test to test whether the differences between groups were significantly greater than the differences between groups, and the results showed that the group community structure had significant differences (Fig. 8C).
.