doi: 10.1038/s41421-024-00725-5.
Sodium oligomannate disrupts the adherence of Ribhigh bacteria to gut epithelia to block SAA-triggered Th1 inflammation in 5XFAD transgenic mice
Affiliations
- PMID: 39557828
- PMCID: PMC11573985
- DOI: 10.1038/s41421-024-00725-5
Item in Clipboard
Sodium oligomannate disrupts the adherence of Ribhigh bacteria to gut epithelia to block SAA-triggered Th1 inflammation in 5XFAD transgenic mice
Cell Discov.
.
Abstract
Sodium oligomannate (GV-971), an oligosaccharide drug approved in China for treating mild-to-moderate Alzheimer's disease (AD), was previously found to recondition the gut microbiota and limit altered peripheral Th1 immunity in AD transgenic mice. As a follow-up study, we here made advances by pinpointing a Lactobacillus murinus (L.m.) strain that highly expressed a gene encoding a putative adhesin containing Rib repeats (Ribhigh-L.m.) particularly enriched in 5XFAD transgenic mice. Mechanistically, Ribhigh-L.m. adherence to the gut epithelia upregulated fecal metabolites, among which lactate ranked as the top candidate. Excess lactate stimulated the epithelial production of serum amyloid A (SAA) in the gut via the GPR81-NFκB axis, contributing to peripheral Th1 activation. Moreover, GV-971 disrupted the adherence of Ribhigh-L.m. to gut epithelia via direct binding to Rib, which corrected the excess lactate, reduced SAA, and alleviated Th1-skewed inflammation. Together, we gained further insights into the molecular link between gut bacteria and AD progression and the mechanism of GV-971 in treating AD.
© 2024. The Author(s).
Conflict of interest statement
Figures
a Shannon diversity of the gut fecal microbiota of WT and 5XFAD Tg mice at week ages 24, 33, and 41 sequenced by 16S rRNA-seq. Statistical significance was determined using two-way ANOVA. Effects of Time and Genotype were significant (Time: P < 0.001; Genotype: P < 0.001). Post-hoc analysis for single experimental conditions using Tukey’s Honest Significant Difference (HSD) test: w24_WT vs w24_Tg: *, Padj = 0.0201389; w33_WT vs w33_Tg: *, Padj = 0.0319952; w41_WT vs w41_Tg: **, Padj = 0.0028909; w24_Tg vs w41_Tg: **, Padj = 0.0037371. Blue, WT; red, Tg. b PCoA on Operational Taxonomy Units (OTU) level of gut fecal microbiota of WT and Tg mice at week ages 24, 33 and 41 (n = 4–10) sequenced by 16S rRNA-seq. PCo1, principal coordinate 1. Statistical significance was determined using two-way ANOVA. Effects of Time and Genotype were significant (Time: P < 0.001; Genotype: P < 0.001). Post-hoc analysis for single experimental conditions using Tukey’s HSD test: w24_WT vs w24_Tg: ***, Padj = 0.0002200; w33_WT vs w33_Tg: **, Padj = 0.0060697; w41_WT vs w41_Tg: **, Padj = 0.0021552; w24_Tg vs w41_Tg: *, Padj = 0.0195302. Blue, WT; red, Tg. c Linear discriminant analysis effect size (LEfSe) and linear discriminant analysis (LDA) based on OTU were used to differentiate key bacteria between the fecal microbiota of WT and Tg mice at week age 41 (w41) from phylum to genus level sequenced by 16S rRNA-seq. The log10LDA cut-off score was set to 4.0 to indicate representative microbiota of each group that had significant differential power between WT and Tg mice at w41. p, phylum; c, class; o, order; f, family; g, genus. Green, WT; red, Tg. d, e Changes of the relative abundance of the gut fecal microbiota of WT and Tg mice that are significantly higher (d) or largely lower (e) in Tg relative to WT at w41 at species level, revealed by metagenomics. Relative abundance is calculated based on total reads per million (TPM). n = 4. *P < 0.05, two-group comparison by Student’s t-test. Blue, WT; red, Tg. f Changes of the relative abundance of functional pathways enriched in the gut fecal microbiota of WT and Tg mice at w41, revealed by metagenomics. Relative abundance is calculated based on TPM. Data are represented as mean ± standard deviation (SD) (n = 4). *P < 0.05, two-group comparison by Student’s t-test. Green, WT; red, Tg. g Mean relative abundance changes of several key enzymes of microbial pathways related to glycolysis and lactate production of WT and Tg mice feces at w41. The relative abundances of XFP (P < 0.05) and ACK (P < 0.05) and LDH (P < 0.05) were significantly more abundant in the gut microbiota of the w41 Tg mice feces (n = 4 for each group). Yellow-filled brackets: enzymes that up-regulated in w41 Tg gut feces compared to WT feces. ACDH, Acetaldehyde dehydrogenase; ACK, Acetate kinase; ACS, Acetyl-CoA synthetase; ADH, Alcohol dehydrogenase; ALDH, Aldehyde dehydrogenase; LDHC, Lactate dehydrogenase (cytochrome); LDH, Lactate dehydrogenase; MAE, Malate dehydrogenase; PDC, Pyruvate decarboxylase; PDH, Pyruvate dehydrogenase E1; POXL, Pyruvate oxidase; PTA, Phosphate acetyltransferase; PYC, Pyruvate carboxylase; PYK, Pyruvate kinase; XFP, Xylulose-5-p/f-6-p phosphoketolase. TCA cycle, tricarboxylic acid cycle. *P < 0.05, by Student’s t-test. h Volcano plot shows differential detected metabolites in the gut fecal microbiota of WT and Tg mice at w41 using untargeted metabolomics. Red dots: up-regulated in Tg mice feces; blue dots: down-regulated in Tg mice feces; gray dots: no-sig, not significantly changing metabolites. The significance level is defined as a P value less than or equal to 0.05 and the absolute value of Fold Change (FC) of metabolites abundances in Tg feces vs WT feces value greater than 1.2 or smaller than 0.83. i Representative HE staining histology sections from the ileum tissue of WT and Tg mice at w41. Red triangle, epithelial lesions reflecting impaired epithelial integrity status; black arrow, epithelial edema. The scale bar is 100 μm. WT: n = 5, 5 M; Tg: n = 6, 4 M + 2 F. M, male mice; F, female mice. j–l Statistics of pathology evaluation scores on gut integrity (j), edema (k), and infiltration of inflammatory cells (l) of ileum tissue of WT and Tg mice at w41. Higher pathology scores represent less integrity, more severe edema, and more infiltration of inflammatory cells, respectively. Blue, WT; red, Tg. The small red diamond symbol at the center of the violin plot is the mean value of all score points of each specific group. **P < 0.01, two-group comparison by Student’s t-test. Box plots represent the median (the horizontal line within boxes) and the 75th and 25th percentiles (the top and bottom of each box, respectively). The upper and lower whiskers represent 1.5× IQR from the top and bottom of the box, respectively. w24_WT: n = 10, 5 M + 5 F; w24_Tg: n = 10, 5 M + 5 F; w33_WT: n = 8, 4 M + 4 F; w33_Tg: n = 9, 5 M + 4 F; w41_WT: n = 4, 3 M + 1 F; w41_Tg: n = 4, 4 F. M, male mice; F, female mice.
a Volcano plot shows differentially expressed genes (DEGs) of the ileum tissue of 7-month-old Tg recipient mice that received 41-week-age WT (+WT feces) and 5XFAD (Tg) feces (+Tg feces) in FMT assays. Up, down: significantly expressed genes up-regulated and down-regulated respectively in ileum tissue of 7-month-old Tg recipient mice that received 41-week-age Tg feces (+ Tg feces) vs WT feces (+ WT feces). NoDiff, not differentially expressed genes. The significance level is defined as Padj < 0.05 and the absolute value of log2FC of +Tg feces vs +WT feces > 1. b Heatmap shows the differential gene expression of the ileum tissue. The cut-off P value is 0.05, while the cut-off Padj value is 0.3. Red or blue colors indicate up- or down-regulation in the ileum tissue of Tg mice that received Tg fecal samples compared to those receving WT feces, respectively. c Representative IHC images stained with SAA antibody of the ileum tissue slices of 7-month-old Tg recipient mice that received 41-week-age WT (+WT feces) and Tg feces (+Tg feces) in FMT assays. Data are represented as mean ± standard error of the mean (SEM). The yellow staining area is the positive SAA staining signal. The positive rate is calculated as the ratio of positive area × positive staining intensity. *P < 0.05, by Student’s t-test, n = 5–6. d Heatmap shows the changes of cytokines secreted by THP1-Dual cells stimulated by hSAA. THP1-Dual cells were treated with 0.8 µg/mL hSAA for 24 h, and the secretion of cytokines in the culture supernatant was determined by cytokine array. n = 3. e Vehicle-induced or hSAA-induced (10 µg/mL, 24 h) THP1-Dual cells were co-cultured with naïve CD4+ T cells for 3 days, and the frequency of Th1 cells was detected by the flow cytometer. Data are mean ± SEM. *P < 0.05, by Student’s t-test, n = 3. f BMDCs were incubated with 10 µg/mL mSAA for 24 h, and the secretion of cytokines in the culture supernatant was determined by cytokine array, and the significantly changed cytokines in mSAA-treated group vs control group were shown as heatmap. n = 3. g Level of the blood SAA of WT and Tg mice at w41 measured by enzyme-linked immune absorbent (ELISA) assays. Data are represented as mean ± SEM. *P < 0.05 by Student’s t-test, n = 7. WT: n = 7, 3 M + 4 F; Tg: n = 7, 3 M + 4 F. M, male mice; F, female mice. h Changes in the frequency of blood Th1 cells of WT and Tg mice at w 41 measured by flow cytometry. Data are represented as mean ± SEM, **P < 0.01 by Student’s t-test, n = 5–6. WT: n = 5, 5 M; Tg: n = 6, 4 M + 2 F. M, male mice; F, female mice. i Mouse blood SAA levels of the 12-month-old C57 WT recipient mice that received 41-week-age WT or Tg fecal samples in FMT assays. Data are represented as mean ± SEM. n = 6–7. j Cell frequency of the blood monocyte of the recipient 7-month-old Tg mice that received 41-week-age WT or Tg feces in FMT assay. Data are represented as mean ± SEM. *P < 0.05 by Student’s t-test, n = 4–6. k Frequency of blood Th1 cells of the 12-month-old C57 WT recipient mice that received 41-week-age WT (+ WT feces) or Tg fecal samples (+ Tg feces) in FMT assays. Data are represented as mean ± SEM. *P < 0.05, by Student’s t-test, n = 7. All recipient mice in FMT assays are male mice.
a Changes in the relative abundances of the top 15 VFs of WT and Tg mice feces at w41 from the VFDB (n = 4). Data are represented as mean ± SD. *P < 0.05, by Student’s t-test. b Changes in the relative abundances of the Rib of the gut microbiota of 7-month-old Tg and ABX-treated Tg mice. Box plots represent the median (the horizontal line within boxes) and the 75th and 25th percentiles (the top and bottom of each box, respectively). The upper and lower whiskers represent 1.5× IQR from the top and bottom of the box, respectively. **P < 0.01 by Wilcoxon rank-sum test, n = 8. Tg: n = 8, 4 M + 4 F; Tg + ABX: n = 8, 4 M + 4 F. M, male mice; F, female mice. c Changes in the relative abundances of the Rib of the gut microbiota of 7-month-old Tg recipient mice that received w41 WT (+WT feces) and Tg feces (+ Tg feces) in FMT assays. Data are represented as mean ± SEM. *P < 0.05, by Wilcoxon rank-sum test, n = 5–6. d Schematic design of experiments to understand the differences between L.m. species isolated from 41-week-age WT and Tg feces. After collecting the feces from both mice, serial dilution was applied and further cultured on the MRS selection plate under an anaerobic chamber for the enrichment of L.m. species. PCR-guided sequencing was used to validate the bacteria’s identity. After that, we cultured the individual species to the logarithm phases to obtain rapidly propagating species and then performed single-base whole genome sequencing and transcriptomics. Functional assays were further designed to compare the function of these species, including the in vitro aggregation test, and the in vivo FMT assays using either WT mice or GF mice. e Genomic map of Tg-L.m. sequenced by bacteria complete map sequencing and plotted using Circos software. Outer ring to inner ring signifies information of genome size (number in megabase pair, Mbp), genomic information of coding sequence (CDS) area on positive chain and negative chain (color labels showing different clusters of orthologous group (COG) function of CDS), location of Rib (purple line), location of genomic island (GI, in yellow line), location of insertion sequence (IS, in black line), GC% contents (outer red or inner blue bar indicating that the GC% of this area is higher or lower than the average GC content of the whole genome. The higher the bar, the larger the differences), and GC-Skew value (or (G − C)/(G + C) value used to define leading and lagging strands). f Volcano plots of DEGs in WT and Tg-L.m. sequenced by single-bacteria transcriptomics. The x and y axes represent log2FC of genes in Tg-L.m. vs WT-L.m. and the negative log10 of Padj of these genes. Red, up-regulated in Tg-L.m.; green, down-regulated in WT-L.m.; gray, not significant (nosig); Rib, the Rib repeats. g Gene ontology (GO) enrichment analysis of up-regulated genes in Tg-L.m. vs WT-L.m. FDR, false-discovery rate. h In vitro bacteria aggregation ability of L.m. bacteria isolated from WT and Tg mice feces at 15–120 min. **P < 0.01, by Unpaired t-test, n = 3.
a FC and Padj of the levels of differential metabolites of MRS-broth cultured with Tg-L.m. (Ribhigh-L.m.) compared with MRS-broth alone without bacteria detected by untargeted metabolomics. Color legends indicate up-regulate (red) or down-regulate (green) metabolites in MRS-broth cultured with Tg L.m. (MRS + Ribhigh-L.m.); noDiff, no difference between groups. b Lactate level in the 24-h anaerobic MRS culture medium of Riblow-L.m and Ribhigh-L.m. ***P < 0.001 by Student’s t-test. c Schematic design of in vitro and in vivo assays on host–bacteria interaction using Ribhigh-L.m and Riblow-L.m. d Lactate changes in the 48 h L.m. MRS-HT29 condition medium (+Ribhigh-L.m.) vs the MRS-HT29 condition medium without L.m. culture (Con). ***P < 0.001 by Student’s t-test. e The in vivo intestine lactate levels of the recipient GF mice that received FMT of single Riblow-L.m. and Ribhigh-L.m. species. * P < 0.05, by Student’s t-test. f Effect of 10 mM lactate on the protein level of GPR81, phosphorylated NF-κB p65 (p-NF-κB p65), and SAA in HT29 cells. Cells were treated with lactate for 2 h followed by cell lysis and western blotted with the indicated antibodies. s, short exposure; l, long exposure. g Effect of transient knockdown of GPR81 (small interference RNA, or siGPR81) on the protein level of GPR81, p-NF-κB p65, and SAA in HT29 cells that stimulated by 2 h treatment of 10 mM lactate. All siRNAs were added to HT29 cells with RNAimax 6 h before stimulating by lactate. After 2 h stimulation, HT29 cells were subjected to western blotting with the indicated antibodies. h–j Ileum IHC staining results of GPR81 (h), p-NF-κB p65 (i), and SAA (j) of the recipient WT mice transplanted with two-week oral gavage of 1 × 109 CFU/mL single-bacteria suspension of Ribhigh-L.m. Representative images were shown on the left, with statistical results shown on the right. ***P < 0.001, by Student’s t-test. n = 9–10 per group. k Plasma SAA levels from the recipient germ-free (GF) mice transplanted with two-week oral gavage of 1 × 109 CFU/mL single-bacteria suspension of Riblow-and Ribhigh-L.m., detected using ELISA. PBS was given the same volume as bacteria suspension as control. *P < 0.05, by Mann–Whitley test, n = 8 per group. l Blood Th1 cell frequency of the recipient GF mice transplanted with two-week oral gavage of 1 × 109 CFU/mL single-bacteria suspension of Riblow-and Ribhigh-L.m. PBS was given the same volume as bacteria suspension as control. *P < 0.05, by Mann–Whitley test, n = 8 per group. All recipient mice in FMT assays are male mice. Data are represented as mean ± SEM.
a Schematic outline of patient selection and validation tests workflow. Stool and plasma samples from health control (HC) and AD patients were collected after clinical diagnosis including cognition test, neuroimaging using magnetic resonance imaging (MRI), and blood test on a series of AD markers including Aβ 42/40, total-Tau, NFL, and GFAP, etc. Plasma samples were tested for SAA using ELISA, while stool samples were collected for metagenomics sequencing and FMT assays on ABX-treated WT mice. Colon SAA of the recipient mice was detected via antibodies targeting SAA using IHC and was statistically analyzed by an experienced pathologist on positive-staining areas. The fecal lactate of the recipient mice was detected using targeted metabolomics. b Relative abundance of Rib in the gut feces samples from HC and AD patients. CPM, Counts per million mapped reads. HC group: n = 14, AD group: n = 24. c Plasma SAA level of HC and AD patients detected by ELISA. HC group: n = 24, AD group: n = 58. ***P < 0.001, by Mann–Whitney test. d, e Representative IHC staining images (d) and positive area statistics (e) of colon SAA from the recipient ABX-treated WT mice transplanted with feces samples from AD patients with high Rib and HC subjects with low Rib abundances in FMT assays. n = 5–6, **P < 0.01, by Student’s t-test. f Levels of fecal lactate level from the recipient ABX-treated WT mice transplanted with feces samples from AD patients with high Rib and HC subjects with low Rib abundances in FMT assays. n = 5–6, P = 0.0510, by Student’s t-test. The high and low Rib abundances were chosen by ranking, i.e., we chose the top 6 samples with the highest Rib abundances in AD feces and the bottom 5 samples with the lowest Rib abundances in HC feces. All recipient mice in FMT assays are male mice. Data are represented as mean ± SEM.
a Protein structure of the Rib domain from L.m., which we named Rib_gene1738. The black bracket demonstrates the protein structure of one representative Rib repeated domain, of which the repeated GIANLDKL residues were present (also see Supplementary Fig. S6a). b Detailed interaction of 2 to 6 sugar units of GV-971 (solid sticks) with repeated GIANLDKL residues of the Rib_gene1738 (color ribbon) protein from Ribhigh-L.m. The binding sites are shown. Hydrogen bonds and salt bridges are depicted as yellow and purple dashed lines, respectively. Different oligomers of GV-971 are shown as sticks. The repeated GIANLDKL residues are displayed in green, orange, and blue ribbons. c Type of connection between GV-971 and GIANLDKL residues of the Rib_gene1738 protein (color ribbon) from Ribhigh-L.m. d, e Molecular dynamics simulations of 5 sugar units of GV-971 interacting with repeated GIANLDKL residues of the Rib_gene1738 protein (color ribbon) from Ribhigh-L.m. using AMBER at total 100 nanoseconds (ns) simulation. RMSD values through 100 ns (d) and interacting forces including hydrogen bonds and salt bridges are depicted as yellow and purple dashed lines, respectively (e). f SPR sensor grams of the Rib_gene2117-GV-971 interaction. Concentrations of GV-971 containing 2–10 sugar units (from bottom to top) are 0.5, 1, 2, 5, and 10 mM, respectively. KD value is 10−5 M. g The effect of GV-971 of two different doses (200 μg/mL and 400 μg/mL, final concentration) on the adhesion of Ribhigh-L.m. to HT29 cells at 2-h incubation time. The numbers of bacteria per HT29 cells were calculated in 4 to 5 separate areas for each well and replicated 3 times. Data are represented as mean ± SD. ***P < 0.001 by Student’s t-test, n = 15. h The effect of GV-971 of two different doses (200 μg/mL and 400 μg/mL, final concentration) on GPR81-NF-κB signaling and SAA levels of HT29 cells in adhesion assays in g. s, short exposure; l, long exposure. i The effect of 100 mg/kg GV-971 on the relative abundance of the 41-week-age Tg mice gut microbiota species L.m. Data are represented as mean ± SEM. *P < 0.05 by Student’s t-test, n = 5–7. Tg: n = 7, 3 M + 4 F; Tg+GV-971: n = 5, 3 M + 2 F. M, male mice; F, female mice. j The relative abundance of fecal L.m. in the WT recipient mouse that received 40-week-age Tg mouse feces treated or not treated by 100 mg/kg GV-971. Data are represented as mean ± SEM. **P < 0.01 by Student’s t-test, n = 7–8. k Effect of 100 mg/kg GV-971 on fecal lactate levels of the 41-week-age Tg mice feces detected by targeted metabolomics. Data are represented as mean ± SD. *P < 0.05 by Student’s t-test, n = 8–10. l The relative abundance of fecal lactate in the WT recipient mouse that received 40-week-age Tg mouse feces treated or not treated by 200 mg/kg GV-971. The levels of lactate were detected by targeted metabolomics. Data are represented as mean ± SEM, n = 5–8. m Effect of one-month treatment of 100 mg/kg GV-971 on Ileum SAA levels of 9-month-old Tg mice. Data are represented as mean ± SEM, n = 7. Tg: n = 7, 7 M; Tg+GV-971: n = 7, 5 M + 2 F. M, male mice; F, female mice. n The relative abundance of ileum SAA in the WT recipient mouse that received 40-week-age Tg mouse feces treated or not treated by 200 mg/kg GV-971. Data are represented as mean ± SEM. *P < 0.05 by Student’s t test, n = 7. o Effect of 100 mg/kg GV-971 on blood SAA levels of the 41-week-age Tg mice. The levels of SAA were determined by ELISA. Data are represented as mean ± SEM, n = 7–8. Tg: n = 8, 4 M + 4 F; Tg + GV-971: n = 7, 4 M + 3 F. M, male mice; F, female mice. p Effect of 100 mg/kg GV-971 on blood Th1 cells of the 41-week-age Tg mice. Data are represented as mean ± SEM. ***P < 0.001 by Student’s t-test, n = 7. Tg: n = 7, 7 M; Tg+GV-971: n = 7, 5 M + 2 F. M, male mice; F, female mice. All recipient mice in FMT assays are male mice.
a The colonization of L.m. via adhesion-related Rib domain initiated metabolic reprogramming including the altered amino acid and lactate. The release of lactate could activate SAA via GPR81-p-NF-κB p65 signaling pathway. SAA then activates monocytes and DCs which subsequently stimulate Th1 cells, mediating AD-associated neuroinflammation. Other species bearing Rib, including Lactobacillus reuteri, Enterococcus faecium, Streptococcus suis, etc., may have similar adhesive potential and Th1 stimulating function in AD development. During this process, phenylalanine, isoleucine, and other amino acids also increase to stimulate Th1 cells as mentioned in our last report. b Oral administration of GV-971 inhibits Rib-bearing L.m. adhesion and decreases bacteria colonization, suppresses the lactate-GPR81-SAA axis, and inhibits phenylalanine, and isoleucine, which finally reduces Th1 cell activation.
Similar articles
-
Sodium oligomannate alters gut microbiota, reduces cerebral amyloidosis and reactive microglia in a sex-specific manner.Mol Neurodegener. 2024 Feb 17;19(1):18. doi: 10.1186/s13024-023-00700-w. Mol Neurodegener. 2024. PMID: 38365827 Free PMC article.
-
Sodium oligomannate therapeutically remodels gut microbiota and suppresses gut bacterial amino acids-shaped neuroinflammation to inhibit Alzheimer's disease progression.Cell Res. 2019 Oct;29(10):787-803. doi: 10.1038/s41422-019-0216-x. Epub 2019 Sep 6. Cell Res. 2019. PMID: 31488882 Free PMC article.
-
Inhibition of microfold cells ameliorates early pathological phenotypes by modulating microglial functions in Alzheimer's disease mouse model.J Neuroinflammation. 2023 Nov 27;20(1):282. doi: 10.1186/s12974-023-02966-9. J Neuroinflammation. 2023. PMID: 38012646 Free PMC article.
-
Sodium Oligomannate: First Approval.Drugs. 2020 Mar;80(4):441-444. doi: 10.1007/s40265-020-01268-1. Drugs. 2020. PMID: 32020555 Review.
-
Activation of aryl hydrocarbon receptor (AhR) in Alzheimer's disease: role of tryptophan metabolites generated by gut host-microbiota.J Mol Med (Berl). 2023 Mar;101(3):201-222. doi: 10.1007/s00109-023-02289-5. Epub 2023 Feb 9. J Mol Med (Berl). 2023. PMID: 36757399 Free PMC article. Review.
References
Grants and funding
LinkOut - more resources
Full Text Sources
Molecular Biology Databases
Research Materials
