Regulatory effects of mangiferin on LPS-induced inflammatory responses and intestinal flora imbalance during sepsis

Abstract

Studies suggest that mangiferin (MAF) has good therapeutic effects on chronic bronchitis and hepatitis. Also, it is one of the antiviral ingredients in Anemarrhena asphodeloides Bunge. However, its effect on the LPS-induced inflammation and intestinal flora during sepsis remains unclear yet. In the present study, LPS-stimulated inflammation RAW264.7 cells and LPS-induced sepsis mice were used to evaluate the efficacy of MAF in vitro and in vivo. 16S rDNA sequencing was performed to analyze the characteristics of intestinal flora of the sepsis mice. It has been demonstrated that MAF (12.5 and 25 μg/mL) significantly inhibited protein expressions of TLR4, MyD88, NF-κB, and TNF-α in the LPS-treated cells and reduced the supernatant TNF-α and IL-6 levels. In vivo, MAF (20 mg/kg) markedly protected the sepsis mice and reduced the serum TNF-α and IL-6 levels. Also, MAF significantly downregulated the protein expressions of TLR4, NF-κB, and MyD88 in the livers. Importantly, MAF significantly attenuated the pathological injuries of the livers and small intestines. Further, MAF significantly increased proportion of Bacteroidota and decreased the proportions of Firmicutes, Desulfobacterota, Actinobacteriota, and Proteobacteria at phylum level, and it markedly reduced the proportions of Escherichia-Shigella, Pseudoalteromonas, Staphylococcus at genus level. Moreover, MAF affects some metabolism-related pathways such as citrate cycle (TCA cycle), lipoic acid metabolism, oxidative phosphorylation, bacterial chemotaxis, fatty acid biosynthesis, and peptidoglycan biosynthesis of the intestinal flora. Thus, it can be concluded that MAF as a treatment reduces the inflammatory responses in vitro and in vivo by inhibiting the TLR4/ MyD88/NF-κB pathway, and corrects intestinal flora imbalance during sepsis to some degree.

Keywords: TLR4/myD88/NF‐κB; inflammation; intestinal flora; lipopolysaccharide; mangiferin.

FIGURE 1

Effect of MAF on cell viability of RAW264.7 cells. (a) Cell viability in the absence of LPS (x̅ ± s, n = 3). *p < .05, **p < .01 vs. Vehicle. The cells were, respectively, cocultured with various concentrations of MAF (1.56–100 μg/mL) for 24 h followed by adding 10‐μL CCK‐8 to culture for another 2 h. The OD value of the solution was measured at 450 nm; (b) Cell viability in the presence of LPS (x̅ ± s, n = 3). *p < .05, **p < .01 vs. Vehicle; ^## p < .01 vs. LPS. The cells were treated with LPS (200 ng/mL) and MAF (6.25–25 μg/mL) for 24 h. Then, 10‐μL CCK‐8 was added to culture for another 2 h. After that, the OD value of the solution was measured at 450 nm.

FIGURE 2

MAF downregulates the protein expressions of TLR4, myD88, NF‐κB, and TNF‐α in LPS‐stimulated cells at regular time points (x̅ ± s, n = 3). *p < .05, **p < .01 vs. Vehicle; ^# p < .05, ^## p < .01 vs. LPS. The cells were treated with MAF in the presence of LPS for 4 h (a), 8 h (b), and 12 h (c), respectively. The cell precipitations were collected and then lysed. The total proteins were then separated by using 10% SDS‐PAGE. After semidry transfer and coculture with the primary and secondary antibodies, the membranes were imaged with ECL.

FIGURE 3

MAF protects LPS‐induced sepsis mice. (a) MAF decreases serum TNF‐α and IL‐6 levels in sepsis mice (x̅ ± s, n = 6). **p < .01 vs. Vehicle, ^# p < .05, ^## p < .01 vs. LPS. Before the LPS injections, the mice were, respectively, treated with MAF (5, 10, and 20 mg/kg) and DEX (4 mg/kg) for 5 days continuously and once a day. LPS (20 mg/kg) was intraperitoneally injected into the mice 4 h post the last treatments except for the vehicle controls. Then, the peripheral blood samples were collected by cutting tails. The serum TNF‐α and IL‐6 levels were assayed by using the ELISA method; (b) MAF increases the survival rate of the sepsis mice. *p < .05, **p < .01 vs. LPS. After the LPS injections, deaths of the animals in groups were recorded and plotted to draw a survival curve. The differences among groups were examined by Kaplan–Meier method; (c) MAF downregulates protein expressions of TLR4, MyD88, and NF‐κB in livers of sepsis mice (x̅ ± s, n = 6). **p < .01 vs. Vehicle, ^# p < .05, ^## p < .01 vs. LPS. The livers were immediately collected after the animals were dead. Parts of the livers were for western blotting analysis and others were for pathological analysis. The liver tissues were homogenized and then lysed. The total proteins were separated and transferred onto PVDF membranes. The membranes were blocked, and then cocultured with anti‐TLR4, myD88, and NF‐κB antibodies, respectively. After coculturing with the corresponding secondary antibodies, the membranes were imaged with ECL. (d) Immunohistochemical analysis of TLR4 and NF‐κB in livers (x̅ ± s, n = 6). **p < .01 vs. Vehicle, ^# p < .05, ^## p < .01 vs. LPS. The liver samples were fixed, dehydrated, embedded, and then cut into sections followed by dewaxing, hydration, and antigen repair. Anti‐TLR4 and NF‐κB antibodies were added and then cocultured with the secondary antibodies. After that, DAB was used to terminate the reaction. The positive granules labeled with brownish yellow were observed under an inverted microscope.

FIGURE 4

Pathological injuries of livers and small intestines of sepsis mice. The collected livers were fixed, dehydrated, and embedded. After that, the tissue samples were cut into sections. Next, the sections were dewaxed and hydrated followed by staining with H&E. The pathological changes especially edema, inflammation, and hemorrhage in the tissues were observed under an inverted microscope.

FIGURE 5

α and β diversity of intestinal flora in sepsis mice (x̅ ± s, n = 6–9). (a) The feces samples of the sepsis mice were collected, and 16S rDNA sequencing was then conducted. α diversity indices including Chao1, Shannon, and Simpson were assayed among groups to compare the differences among the groups; (b) PCoA assay was used to assay the similarity among the clusters.

FIGURE 6

Intestinal flora composition and metabolism pathways. (a) Intestinal flora composition at phylum level (x̅ ± s, n = 6–9). *p < .05, **p < .01 vs. Blank; ^# p < .05, ^## p < .01 vs. LPS. The differences in the proportions of Bacteroidota, Firmicutes, Desulfobacterota, and Actinobacteriota were compared at phylum level among groups. (b) Intestinal flora composition at genus level (x̅ ± s, n = 6–9). *p < .05, **p < .01 vs. Blank; ^# p < .05, ^## p < .01 vs. LPS. The differences in the proportions of Escherichia‐Shigella, Enterorhabdus, Pseudoalteromonas, Vibrio, Staphylococcus, and Prevotellaceae_UCG‐001 were compared at genus level among groups. (c) Pathway analysis of intestinal flora among groups.