Effective sequence analysis of IDHS diarrhea mice intestinal mucosa flora
As could be seen in Table 1, the average length of the effective sequence after quality filtering was in the range of 1478 bp-1483 bp, and the resulting sequence length was consistent with the target length range, which could effectively reflect the true situation of the microorganisms in the sample.
Table 1
Data preprocessing statistics table
Sample
|
CCS
|
Filtered
|
Average Length
|
ctcm1
|
8564
|
7911
|
1482
|
ctcm2
|
12740
|
11924
|
1482
|
ctcm3
|
13999
|
13122
|
1481
|
ctcm4
|
16566
|
15571
|
1482
|
ctcm5
|
14125
|
13267
|
1482
|
ctmm1
|
11663
|
10888
|
1483
|
ctmm2
|
17319
|
16302
|
1482
|
ctmm3
|
10461
|
9725
|
1482
|
ctmm4
|
11757
|
10594
|
1483
|
ctmm5
|
14361
|
13564
|
1478
|
cttm1
|
20104
|
18679
|
1480
|
cttm2
|
12463
|
11689
|
1482
|
cttm3
|
15491
|
14391
|
1483
|
cttm4
|
5110
|
3976
|
1478
|
cttm5
|
11058
|
10290
|
1483
|
Note: CCS: circular consensus sequence; Filtered: the number of effective CCS after removing primers and length filtering; Average Length: average length of effective sequence. Con1-5: control group sample 1–5; Mol1-5: model group sample 1–5; GD1-5: GD group sample 1–5. |
The effect of GD on the number of OTUs in the intestinal mucosa of IDHS diarrhea mice
The number of OTUs in the ctcm, ctmm and cttm were respectively 274,301 and 287, including 119 OTUs in the intersection of OTUs in the three groups(Fig. 2). Among them, 91 OTUs in the cttm were indicated that the number of intestinal OTUs in model mice decreased after treating with GD.
The effect of GD on the diversity of intestinal mucosa in IDHS diarrhea mice
Combined with the Alpha diversity analysis, Chao1 index in the ctmm was lower than that of the ctcm and cttm. Chao1 index in the cttm was close to the ctcm, but there was no significant difference between the three groups, indicating that GD could recover and to some extent increased the richness of intestinal mucosa flora(Fig. 3A). Compared with the ctmm, Shannon and Simpson index of the cttm increased, which reflected that GD had a restorative effect on the intestinal mucosa flora diversity in mice with diarrhea caused by IDHS(Fig. 3B、3C).
Beta diversity analysis showed that the main coordinate variable 1 was 32.36% and the main coordinate variable 2 was 21.33% (Fig. 3D). From the results, the samples in the ctcm and the ctmm were evenly distributed and relatively concentrated. The distance between the samples of the cttm and the ctmm was close, indicating that they had a higher similarity in bacterial community composition. The samples of the cttm and the ctcm could be clearly separated, suggesting that there was a certain difference in the bacterial community structure.
The effect of GD on the genus level of intestinal mucosa in IDHS diarrhea mice
Comparing the NCBI database, there were totally 187 genera in the ctcm, ctmm, and cttm(Fig. 4 and Table 2). Among these, Lactobacillus was the first dominant genus. The abundance of Lactobacillus in the intestinal mucosa of the ctmm was lower than that of the ctcm, accounting for 82.29%. In the GD, the Lactobacillus (79.88%) was slightly lower than that of the ctmm, showing that the modelling had an inhibitory effect on Lactobacillus, while GD had a regulatory effect on the growth of Lactobacillus. The abundance of Streptococcus in the ctmm(5.03%) and ctcm(6.41%) was higher than that in the ctcm(4.45%), which was suggested that GD could promote the growth of Streptococcus. Compared with the ctcm, the Neisseria in the ctmm(1.23%) was significantly increased, while in the cttm(0.78%) was similar to that in the ctcm(0.80%). Thus, GD regulated the abundance of Neisseria to reach the normal level by inhibiting its growth. The Clostridium and Muribaculum in the ctmm was very low, respectively accounting for 0.09% and 0.01%, while the Clostridium (1.09%) and Muribaculum (1.08%) in the cttm was significantly higher than that of the ctcm, suggesting that GD could promote the growth of Clostridium and Muribaculum.
Table 2
Effect of GD treatment on the abundance of genus level of the intestinal mucosal flora in IDHS diarrhea mice (top 20)
Genus name
|
The relative abundance(%)
|
ctcm
|
ctmm
|
cttm
|
Lactobacillus
|
0.8427 ± 0.0244
|
0.8229 ± 0.7823
|
0.7988 ± 0.0844
|
Streptococcus
|
0.0455 ± 0.0047
|
0.0503 ± 0.0083
|
0.0641 ± 0.0212
|
Neisseria
|
0.0080 ± 0.0015
|
0.0123 ± 0.0044
|
0.0078 ± 0.0021
|
Clostridium
|
0.0090 ± 0.0076
|
0.0090 ± 0.0003
|
0.0109 ± 0.0103
|
Muribaculum
|
0.0079 ± 0.0000
|
0.0001 ± 0.0000
|
0.0108 ± 0.0000
|
Lautropia
|
0.0054 ± 0.0015
|
0.0055 ± 0.0021
|
0.0078 ± 0.0058
|
Curvibacter
|
0.0050 ± 0.0024
|
0.0081 ± 0.0019
|
0.0053 ± 0.0022
|
Prevotella
|
0.0049 ± 0.0012
|
0.0077 ± 0.0040
|
0.0053 ± 0.0016
|
Porphyromonas
|
0.0064 ± 0.0027
|
0.0040 ± 0.0020
|
0.0047 ± 0.0016
|
Ureaplasma
|
0.0042 ± 0.0008
|
0.0039 ± 0.0016
|
0.0036 ± 0.0028
|
Staphylococcus
|
0.0037 ± 0.0026
|
0.0014 ± 0.0009
|
0.0063 ± 0.0040
|
Bacteroides
|
0.0013 ± 0.0007
|
0.0006 ± 0.0003
|
0.0094 ± 0.0089
|
Gemella
|
0.0028 ± 0.0011
|
0.0020 ± 0.0012
|
0.0031 ± 0.0011
|
Rothia
|
0.0033 ± 0.0019
|
0.0023 ± 0.0012
|
0.0021 ± 0.0010
|
Cutibacterium
|
0.0042 ± 0.0036
|
0.0009 ± 0.0006
|
0.0018 ± 0.0015
|
Helicobacter
|
0.0004 ± 0.0001
|
0.0027 ± 0.0000
|
0.0011 ± 0.0000
|
Weissella
|
0.0011 ± 0.0002
|
0.0006 ± 0.0003
|
0.0043 ± 0.0037
|
Pantoea
|
0.0019 ± 0.0016
|
0.0014 ± 0.0006
|
0.0019 ± 0.0016
|
Veillonella
|
0.0016 ± 0.0007
|
0.0020 ± 0.0015
|
0.0015 ± 0.0012
|
Granulicatella
|
0.0007 ± 0.0005
|
0.0005 ± 0.0002
|
0.0036 ± 0.0033
|
The effect of GD on the species level of intestinal mucosa in IDHS diarrhea mice
It could be seen from Fig. 5 and Table 3 that the top 5 dominant species bacteria in the three groups were Lactobacillus crispatus, Streptococcus oralis, Muribaculum intestinale, Lautropia mirabilis, and Curvibacter lanceolatus. Lactobacillus crispatus was the most abundant in the three groups. The abundence of Lactobacillus crispatus(82.20%) in the ctmm was lower than that in the ctcm(83.92%), whereas in the cttm(79.72%) was not significantly different from that in the ctcm, showing that Lactobacillus crispatus had a regulating effect but had little effect after the treatment of GD. The Muribaculum intestinale in the ctmm(0.01%) was lower than that in the ctcm(0.79%), while the abundance in the cttm(1.08%) was significantly higher than that of the ctmm, which suggested that GD had a role in the growth of Muribaculum intestinale enhancement. Compared with the ctmm(0.81%), the Curvibacter lanceolatus in the ctcm(0.50%) and cttm(0.53%) were decreased. It indicated that GD had a certain inhibitory effect on Curvibacter Lanceolatus. The Streptococcus oralis and Lautropia mirabilis in the ctmm(4.96%, 0.55%) and the cttm(6.26%, 0.78%) were higher than the ctcm(4.34%, 0.54%). The two species bacteria in the ctcm and cttm had the similar contents, revealing that GD had little effect on the reproduction of Streptococcus oralis and Lautropia mirabilis.
Table 3
Effect of GD treatment on the abundance of species level of the intestinal mucosal flora in IDHS diarrhea mice (top 20)
Species name
|
The relative abundance(%)
|
ctcm
|
ctmm
|
cttm
|
Lactobacillus crispatus
|
0.8392 ± 0.0208
|
0.8820 ± 0.0405
|
0.07972 ± 0.0887
|
Streptococcus oralis
|
0.0434 ± 0.0045
|
0.0496 ± 0.0081
|
0.0626 ± 0.0200
|
Muribaculum intestinale
|
0.0079 ± 0.0000
|
0.0001 ± 0.0000
|
0.0018 ± 0.0000
|
Lautropia mirabilis
|
0.0054 ± 0.0011
|
0.0055 ± 0.0021
|
0.0078 ± 0.0058
|
Curvibacter lanceolatus
|
0.0050 ± 0.0024
|
0.0081 ± 0.0019
|
0.0053 ± 0.0022
|
Neisseria mucosa
|
0.0054 ± 0.0020
|
0.0074 ± 0.0032
|
0.0051 ± 0.0008
|
Bacteroides fragilis
|
0.0011 ± 0.0005
|
0.0004 ± 0.0001
|
0.0091 ± 0.0088
|
Porphyromonas gingivalis
|
0.0037 ± 0.0018
|
0.0014 ± 0.0011
|
0.0027 ± 0.0022
|
Cutibacterium acnes
|
0.0042 ± 0.0031
|
0.0009 ± 0.0006
|
0.0018 ± 0.0015
|
Prevotella intermedia
|
0.0016 ± 0.0002
|
0.0033 ± 0.0024
|
0.0015 ± 0.0012
|
Helicobacter typhlonius
|
0.0004 ± 0.0000
|
0.0053 ± 0.0000
|
0.0001 ± 0.0000
|
Weissella hellenica
|
0.0007 ± 0.0000
|
0.0005 ± 0.0002
|
0.0043 ± 0.0037
|
Porphyromonas endodontalis
|
0.0016 ± 0.0011
|
0.0023 ± 0.0009
|
0.0014 ± 0.0011
|
Granulicatella_adiacens
|
0.0007 ± 0.0005
|
0.0005 ± 0.0002
|
0.0036 ± 0.0033
|
Fusobacterium nucleatum
|
0.0001 ± 0.0005
|
0.0007 ± 0.0005
|
0.0016 ± 0.0013
|
Aerococcus_viridans
|
0.0048 ± 0.0003
|
0.0026 ± 0.0005
|
0.0017 ± 0.0009
|
Prevotella_veroralis
|
0.0009 ± 0.0006
|
0.0010 ± 0.0007
|
0.0012 ± 0.0007
|
Enterococcus_mundtii
|
0.0003 ± 0.0001
|
0.0026 ± 0.0020
|
0.0005 ± 0.0002
|
Parvimonas_micra
|
0.0012 ± 0.0007
|
0.0020 ± 0.0016
|
0.0001 ± 0.0002
|
Abiotrophia_defectiva
|
0.0014 ± 0.0007
|
0.0011 ± 0.0001
|
0.0009 ± 0.0003
|
Investigation of the mechanism of action of GD against diarrhea by network pharmacology
Active ingredients-targets network analysis
Using the established filter conditions, OB ≥ 30% and DL ≥ 0.18, 146 active ingredients were identified in five TCM in GD from the TCMSP (Fig. 6A). It was entered into the TCMSP database and a total of 269 targets with 146 active ingredients were found in the search. Though the UniProt database, 269 targets were found 252 corresponding gene names. The GD active ingredients-targets network diagram was constructed based on the interactions among the 5 herbs, 146 active ingredients and 252 targets associated with GD (as shown in Fig. 6B). The network contained of 384 nodes (representing active ingredients and targets) and 2694 edges (representing the interaction between the active ingredient and the target). Of the 146 active ingredients, 18 of them have more than 30 targets; for example, quercetin (degree = 284), formononetin (degree = 76), beta-sitosterol(degree = 74), kaempferol (degree = 58), and wogonin (degree = 45) had a high degree values and were located at central positions in the network. From the perspective of the targets, 26 targets worked with more than 30 ingredients. The top five targets were PTGS2, HSP90, CALM, AR and ESR1, which interacted with 121,95, 94, 93 and 92 ingredients. Based on this, we found the phenomenon that some ingredients in GD could act on multiple targets, and different ingredients worked together on the same target, which reflected the mechanism of interaction between multiple ingredients and multiple targets in TCM.
PPI core network analysis
Using the Venny 2.1 online mapping tool platform, the 328 disease targets and the 252 active ingredient targets were used to draw a Venn diagram, getting 31 common targets were obtained for both (Fig. 7). The results showed that GD played a cooperative role in treating diarrhea through multiple potential targets. Using 31 potential targets entered into the String database to obtain the protein interaction data(Fig. 8A), the Cytoscape 3.7.2 software was used to map these targets to the human protein-protein interaction network, a total of 31 targets and 186 edges related to gout were obtained(Fig. 8B). The DC, BC and CC of each node was calculated by the Cytoscape 3.7.2. For further research, 13 potential targets with the median value of DC, BC, and CC greater than 13,0.017 and 0.6 were finally obtained, including TNF, IL-6, EGFR and so on (Fig. 8C). It was suggested that the mechanism of GD for treating diarrhea was closely related to these core targets. In summary, it was very likely that GD exerted its pharmacological effects by acting on these core targets.
GO and KEGG enrichment analysis
In order to further elucidate the possible effects of GD on diarrhea, the biological processes and signaling pathways of 13 key targets were carried out through the gene enrichment analysis plug-in ClueGO. The results showed that the biological processes (P < 0.05) were largely related to the Regulation of reactive oxygen species biosynthetic process, positive regulation of neuroinflammatory response, positive regulation of ATP biosynthetic process and positive regulation of receptor-mediated endocytosis(Fig. 9A), and the signaling pathways (P < 0.05) were mainly involved with the AGE-RAGE signaling pathway in diabetic complications, HIF-1 signaling pathway, Adipocytokine signaling pathway and VEGF signaling pathway (Fig. 9B).