Inhibition of microfold cells ameliorates early pathological phenotypes by modulating microglial functions in Alzheimer's disease mouse model

Abstract

Background: The gut microbiota has recently attracted attention as a pathogenic factor in Alzheimer's disease (AD). Microfold (M) cells, which play a crucial role in the gut immune response against external antigens, are also exploited for the entry of pathogenic bacteria and proteins into the body. However, whether changes in M cells can affect the gut environments and consequently change brain pathologies in AD remains unknown.

Methods: Five familial AD (5xFAD) and 5xFAD-derived fecal microbiota transplanted (5xFAD-FMT) naïve mice were used to investigate the changes of M cells in the AD environment. Next, to establish the effect of M cell depletion on AD environments, 5xFAD mice and Spib knockout mice were bred, and behavioral and histological analyses were performed when M cell-depleted 5xFAD mice were six or nine months of age.

Results: In this study, we found that M cell numbers were increased in the colons of 5xFAD and 5xFAD-FMT mice compared to those of wild-type (WT) and WT-FMT mice. Moreover, the level of total bacteria infiltrating the colons increased in the AD-mimicked mice. The levels of M cell-related genes and that of infiltrating bacteria showed a significant correlation. The genetic inhibition of M cells (Spib knockout) in 5xFAD mice changed the composition of the gut microbiota, along with decreasing proinflammatory cytokine levels in the colons. M cell depletion ameliorated AD symptoms including amyloid-β accumulation, microglial dysfunction, neuroinflammation, and memory impairment. Similarly, 5xFAD-FMT did not induce AD-like pathologies, such as memory impairment and excessive neuroinflammation in Spib^-/- mice.

Conclusion: Therefore, our findings provide evidence that the inhibiting M cells can prevent AD progression, with therapeutic implications.

Keywords: Alzheimer’s disease; Microbiota-Gut-Brain axis; Microfold cells; Microglia.

Fig. 1

Microfold cell counts were increased in the colons of 5xFAD and 5xFAD-FMT mice. A, B Expression of M cells-related genes in the colons of 5xFAD A and 5xFAD-FMT B mice (n = 4–5 per group). C and E Representative images of GP2⁺ cells in the colons of 5xFAD C and 5xFAD-FMT E mice; scale bar, 20 μm. The quantification of GP2⁺ cells in the colons of 5xFAD D and 5xFAD-FMT F mice (n = 3–5 per group). G and I The level of total bacteria was measured in the colons of 5xFAD G and 5xFAD-FMT I using qRT-PCR (n = 5–9 per group). H, J Correlation analysis of total bacteria and Gp2 was analyzed in the colons of WT and 5xFAD mice H and WT-FMT and 5xFAD-FMT mice J. Bars represent as mean ± standard deviation. Statistical analysis included the Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. WT or WT-FMT mice

Fig. 2

Depletion of microfold cells alleviated memory dysfunction and Aβ pathologies in 5xFAD mice. A Spontaneous alternations (%) of WT, WT/Spib^−/−, 5xFAD, and 5xFAD/Spib^−/− mice (WT, n = 17; WT/Spib^−/−, n = 17; 5xFAD, n = 18; 5xFAD/Spib^−/− mice, n = 16). B Escape latencies of WT, WT/Spib^−/−, 5xFAD, and 5xFAD/Spib^−/− mice over 10 days. C, D In the probe trial on day 11, crossing the platform number C and spending time in the target quadrant D were recorded and analyzed. Statistical analysis included the two-way ANOVA and Tukey’s post hoc test. Brain sections from 6-month-old mice were stained with thioflavin S (ThS). E Representative images of ThS staining of the hippocampus and cortex; scale bar, 100 μm. F The quantification of ThS-stained area (%) (n = 6 per group). G, H Analysis of Aβ_1-40 G and Aβ_1-42 H levels in the hippocampus and cortex using ELISA kits (n = 6 per group). Statistical analysis included the Student’s t-test. I and J Images of Iba-1 I and GFAP J immunostaining in the hippocampus and cortex; scale bar, 100 μm. K–N The quantification of Iba-1 in the hippocampus K and cortex L and GFAP in the hippocampus M and cortex N positive cells area (%) (n = 4–6 per group). Statistical analysis included the two-way ANOVA and Tukey’s post hoc test. Bars represent as mean ± standard deviation. Bars of the Fig. 2B represent as mean ± standard error of the mean. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, and ^####P < 0.0001 vs. WT mice. *P < 0.05, **P < 0.01, and ***P < 0.001 vs. 5xFAD mice

Fig. 3

Depletion of microfold cells enhanced microglial phagocytosis in 5xFAD mice. A Images of Aβ plaques (thioflavin S, green) and microglia (Iba-1, red); scale bar, 20 μm. B The quantification of microglia surrounding Aβ plaques (n = 9 per group). C Immunofluorescence images of lysosome (CD68, green), microglia (Iba-1, red), and Aβ (6E10, blue); scale bar, 20 μm. D, E The quantification of phagosome (CD68⁺Iba-1⁺/Iba-1⁺) area (%) D and internalized Aβ (Iba-1⁺6E10⁺/6E10⁺) area (%) E (n = 9 per group). F–I Flow cytometric analysis of the quantification of and CD68-positive microglia (CD45^intCD11b⁺) F and G and MeX04-positive microglia (CD45^intCD11b⁺) H and I in the brain (n = 8 per group). Bars represent as mean ± standard deviation. Statistical analysis included the Student’s t-test. *P < 0.05 vs. 5xFAD mice

Fig. 4

Depletion of microfold cells increased the count of brain infiltrating T_H2 cells in 5xFAD mice. A Gating strategy for detection of brain infiltrating CD45^highCD4⁺ (CD4⁺ T), CD45^highCD4⁺IFN-γ⁺ (T_H1) cells, and CD45^highCD4⁺IL-4⁺ (T_H2) cells. B–D Graph displaying the calculated percentage of infiltrating CD45^highCD4⁺ (CD4⁺ T) cells B, CD45^highCD4⁺IFN-γ⁺ (T_H1) cells C, and CD45^highCD4⁺IL-4⁺ (T_H2) cells D in WT, WT/Spib^−/−, 5xFAD, and 5xFAD/Spib^−/− mouse brains (n = 13–17 per group). E Histograms are representative of B220⁺ B cells at the brain. F Graph displaying the calculated percentage of B220⁺ B cells in WT, WT/Spib^−/−, 5xFAD, and 5xFAD/Spib^−/− brains (n = 4–6 per group). Bars represent as mean ± standard deviation. Statistical analysis included the two-way ANOVA and Tukey’s post hoc test. ^#P < 0.05, ^###P < 0.001, and ^####P < 0.0001 vs. WT mice. **P < 0.01 vs. 5xFAD mice

Fig. 5

Depletion of microfold cells altered the composition of gut microbiota in 5xFAD mice. A and D The distribution of the relative abundance of gut microbiota at the phylum A and class D levels was observed in the four groups (n = 5 per group). B, C Relative abundance of Bacteroidetes B and Firmicutes C at the phylum level. E and F Relative abundance of Actinomycetia E and Clostridia F at the class level. G–J The gene expression of colonic inflammatory mediators was measured with qRT-PCR (n = 6–8 per group). Bars represent as mean ± standard deviation. Statistical analysis included the two-way ANOVA and Tukey’s post hoc test. ^#P < 0.05 and ^##P < 0.01 vs. WT mice. *P < 0.05 and ***P < 0.001 vs. 5xFAD mice

Fig. 6

The effect of 5xFAD-fecal microbiota transplantation (FMT) on behavioral changes and neuroinflammation in Spib^−/− mice. A Experimental design B Spontaneous alternations (%) of WT-FMT, WT-FMT/Spib^−/−, 5xFAD-FMT, and 5xFAD-FMT/Spib^−/− mice (WT-FMT, n = 16; WT-FMT/Spib^−/−, n = 15; 5xFAD-FMT, n = 19; and 5xFAD-FMT/Spib^−/− mice, n = 17). C Escape latencies of WT-FMT, WT-FMT/Spib^−/−, 5xFAD-FMT, and 5xFAD-FMT/Spib^−/− mice over 6 days. D, E In the probe trial on day 7, crossing the platform number D and spending time in the target quadrant E were recorded and analyzed. F, G Immunofluorescence images of Iba-1 F and GFAP G staining in the hippocampus and cortex; scale bar, 100 μm. H–K The quantification of Iba-1 in the hippocampus H and cortex I and GFAP in the hippocampus J and cortex K positive cells area (%) (n = 4–6 per group). L–Q The inflammatory cytokines were measured in the hippocampus L–N and cortex O–Q using qRT-PCR (n = 4–5 per group). Bars represent as mean ± standard deviation. Bars of the Fig. 6C represent as mean ± standard error of the mean. Statistical analysis included the two-way ANOVA and Tukey’s post hoc test. ^#P < 0.05, ^###P < 0.001 and ^####P < 0.0001 vs. WT-FMT mice. *P < 0.05 and **P < 0.01 vs. 5xFAD-FMT mice