Liuwei Dihuang formula ameliorates chronic stress-induced emotional and cognitive impairments in mice by elevating hippocampal O-GlcNAc modification

Abstract

A substantial body of evidence has indicated that intracerebral O-linked N-acetyl-β-D-glucosamine (O-GlcNAc), a generalized post-translational modification, was emerging as an effective regulator of stress-induced emotional and cognitive impairments. Our previous studies showed that the Liuwei Dihuang formula (LW) significantly improved the emotional and cognitive dysfunctions in various types of stress mouse models. In the current study, we sought to determine the effects of LW on intracerebral O-GlcNAc levels in chronic unpredictable mild stress (CUMS) mice. The dynamic behavioral tests showed that anxiety- and depression-like behaviors and object recognition memory of CUMS mice were improved in a dose-dependent manner after LW treatment. Moreover, linear discriminate analysis (LEfSe) of genera abundance revealed a significant difference in microbiome among the study groups. LW showed a great impact on the relative abundance of these gut microbiota in CUMS mice and reinstated them to control mouse levels. We found that LW potentially altered the Uridine diphosphate-N-acetylglucosamine (UDP-GlcNAc) biosynthesis process, and the abundance of O-GlcNAcase (OGA) and O-GlcNAc transferase (OGT) in CUMS mice, which was inferred using PICRUSt analysis. We further verified advantageous changes in hippocampal O-GlcNAc modification of CUMS mice following LW administration, as well as changes in the levels of OGA and OGT. In summary, LW intervention increased the levels of hippocampal O-GlcNAc modification and ameliorated the emotional and cognitive impairments induced by chronic stress in CUMS mice. LW therefore could be considered a potential prophylactic and therapeutic agent for chronic stress.

Keywords: Liuwei Dihuang formula; O-GlcNAc; O-GlcNAc transferase; O-GlcNAcase; gut microbiota.

FIGURE 1

The experimental timeline and the effects of CUMS on mice. The dosage and days of intragastric administration and the specific time arrangement of each behavior after the end of CUMS-14D (A), CUMS-28D (B).

FIGURE 2

Liuwei Dihuang formula (LW) ameliorates the emotional and cognitive impairments of CUMS mice. Intragastric administration was started 7 days in advance and behavioral tests were performed after chronic stress. The distance in the center zone (A1) in the OFT, open arm time (A2) in the EPM, sucrose preference rate (A3) in SPT, preference index (A4) in the NOR test, and emotion and cognition z-score (A5) of mice on day 14 of CUMS. The center distance (B1), open arm time (B2), sucrose preference rate (B3), preference index (B4), and z-score (B5) of mice on day 28 of CUMS. *P < 0.05, ^**P < 0.01, ^***P < 0.001 were one-way ANOVA followed by Dunnett’s post-hoc test. The values are mean ± S.D., n = 10.

FIGURE 3

Liuwei Dihuang formula increases the community diversity of gut microbiota in CUMS mice. Mice to be sampled were placed in a clean cage covered with sterile filter paper after behavioral tests, and fecal samples were collected immediately after the mice defecated. This was used to detect gut microbiota. Different mouse samples should be replaced with new filter paper. The community richness of mice on days 14 (A1,A2) and 28 (B1,B2) of CUMS were analyzed by sobs and chao. The community diversity of mice on days 14 (A3,A4) and 28 (B3,B4) of CUMS were analyzed by Shannon and Simpson. *P < 0.05, ^**P < 0.01 were one-way ANOVA followed by Dunnett’s post-hoc test. The values are mean ± S.D., n = 10.

FIGURE 4

Liuwei Dihuang formula restored the beta diversity of gut microbiota in CUMS mice. After the basic collection of mouse feces, DNA extraction, and PCR amplification were performed to process the constructed Illumina Miseq sequencing data. Principle coordinate analysis (PCoA) plots were calculated by the Bray-Curtis similarity methods based on the microbial genus of CUMS-14D (A1), CUMS-28D (B1) mice. Characterization of gut microbiota composition in CUMS-14D (A2), CUMS-28D (B2) mice based on PCoA scores. *P < 0.05 was one-way ANOVA followed by Dunnett’s post-hoc test. The values are mean ± S.D., n = 10.

FIGURE 5

Liuwei Dihuang formula reversed the relative abundances of the predominant bacterial taxon of gut microbiota in CUMS mice. The average abundance of the predominant bacterial phylum (with relative abundance >1% in at least one sample) in CUMS-14D (A1) and -28D (B1) mice. Gut microbiota variations at the phylum level in CUMS-14D (A2) and -28D (B2) mice. (C) The linear discriminant analysis (LDA) effect size (LEfSe) classifying the most differentially abundant taxa at the genus level in CUMS-28D mice (| LDA| > 2 and P-value < 0.05 were shown). *P < 0.05, ^**P < 0.01, ^***P < 0.001 were one-way ANOVA followed by Dunnett’s post-hoc test. The values are mean ± S.D., n = 10.

FIGURE 6

Liuwei Dihuang formula altered the O-GlcNAc-related biosynthesis pathways and enzymes of gut microbiota in CUMS mice. PICRUSt was used to standardize the OTU abundance table, that is, to remove the influence of the copy number of the 16S marker gene in the genome. Then, COG and KEGG functional annotations of OTU were carried out according to the corresponding green-gene id of each OTU, and annotation information of OTU at COG and KEGG functional levels and abundance information of each function in different samples were obtained. (A) The functional prediction of KEGG pathways using the phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt) based on OTUs of CUMS mice. Distribution of relative abundances of OGT (B1,B2), OGA (C1,C2), and the ratio of OGT/OGA (B3,C3) in CUMS-14D and -28D mice. *P < 0.05, ^**P < 0.01 were one-way ANOVA followed by Dunnett’s post-hoc test. The values are mean ± S.D., n = 10.

FIGURE 7

Liuwei Dihuang formula elevated hippocampal O-GlcNAc modification of CUMS mice. After behavioral tests, the mice were sacrificed, the hippocampus was isolated from the brain tissue and hippocampus protein was extracted. (A1) Representative Western blot of O-GlcNAc from whole hippocampal lysates of CUMS mice. Quantification of hippocampal O-GlcNAc modification levels in CUMS-14D (A2) and -28D (A3) mice, represented as mean optical intensity normalized to GAPDH. The concentrations of OGT (B1,C1), OGA (B2,C2), and the ratio of OGT/OGA (B3,C3) in the hippocampus of CUMS-14D and -28D mice were obtained by ELISA. *P < 0.05, ^**P < 0.01 were one-way ANOVA followed by Dunnett’s post-hoc test. The values are mean ± S.D., n = 10.

FIGURE 8

Sketch map for the mechanisms of LW on CUMS induced emotional disorders and cognitive impairment. LW decreases the ratio of OGT/OGA levels in CUMS-14D mice but increases the hippocampal ratio of OGT/OGA in CUMS-28D mice. LW intervention increased the levels of hippocampal O-GlcNAc modification and ameliorated the emotional and cognitive impairments induced by chronic stress in CUMS mice.