Overall structural changes in microbiota composition
Good’s coverage for the six groups was greater than 98.0%, indicating a considerable sequencing depth for the analysis of the microbiota (Fig. 1a-c). For alpha diversity analysis, Chao and ACE estimators and the Shannon and Simpson indexes were used to assess community richness and diversity, respectively. As shown in Fig. 1a, the richness of the bacterial population (Chao1 and ACE indices) in the DSC group was significantly higher than that in the NC group (p < 0.01), while there was no significant difference between the species richness of intestinal flora in CHD and NCHD group (Fig. 1b), and there was no significant difference between the species richness of the reproductive tract of DSM and that of NM (Fig. 1c). For the Shannon and Simpson analyses, we can see that the Shannon index of the DSC group is significantly lower than that of normal group (p < 0.05), which indicated that the diversity of intestinal flora of children with DS was less than that of healthy children, while the diversity index of children with congenital heart disease was not significantly different from that of children without congenital heart disease, and the reproductive tract microbial species diversity of DS mothers was not significantly different from that of healthy children.
To measure the extent of the similarity between microbial communities, PCoA plots of unweighted and weighted UniFrac distances were generated. From the PCoA analysis, we can see that the composition of intestinal flora in children with DS is significantly different from that in the healthy control group (Fig. 1d). Among these algorithms, the PC1 distribution with the largest contribution rate revealed changes of 16.32%, and the PC2 distribution revealed changes of 6.5%, which showed that there was a significant difference between the two groups. However, the distribution of flora with CHD and NCHD was similar. There was a partially significant difference between the reproductive tract flora of DSM and NM (Fig. 1e). Among these algorithms, the PC1 distribution with the largest contribution rate revealed changes of 57.1%, and the PC2 distribution revealed changes of 18.1%; that is, the total microbial composition was not similar between the two groups. We also used cluster analysis to explore the similarity between samples. From Fig. 1f, we can see that DSC group samples tended to gather together and had similar microbial composition and diversity distribution, and it can be seen that the composition and diversity of intestinal flora in NC were similar as a whole, with Bacteroides as the majority. At the same time, it shows the accuracy of the results of our sample collection. We found that the composition of the diversity of the microbial flora in the group of mothers of children with DS tended to clump together, had a similar microbial composition and had a relatively high abundance of Lactobacillus (Fig. 1g), indicating that there is a significant difference in the overall composition compared with the mothers of healthy children.
Taxonomy-based comparisons at the phylum and genus levels
From the phylum-level analysis (Fig. 2a), we could clearly see that the DSC group was mainly composed of 16.67% Bacteroides, 37.35% Firmicutes, 26.13% Proteobacteria, 50% Verrucomicrobia and 15.64% Actinobacteria. The normal group was mainly composed of 45.30% Bacteroides, 22.85% Firmicutes, 15.64% Proteobacteria, 28% Verrucomicrobia and 3.02% Actinobacteria. Therefore, we found that Bacteroides is the most different intestinal flora between children with DS and healthy children. The DSM group was mainly composed of 9.36% Bacteroides, 28.99% Firmicutes, 6.55% Proteobacteria and 29.96% Actinobacteria. The NM group was mainly composed of 13.39% Bacteroides, 25.42% Firmicutes, 6.51% Proteobacteria and 24.14% Actinobacteria. The most different intestinal flora were Firmicutes and Bacteroides.
We further analyzed the composition of intestinal flora at the genus level (Fig. 2b), and we could clearly see that the DSC group was mainly composed of 18.42% Escherichia, 14.93% Bifidobacterium and 14.66% Bacteroides. The normal group was mainly composed of 44.30% Bacteroides, 9.94% Escherichia and 2.90% Bifidobacterium. We found that the relative abundance of Escherichia in the intestines of children with DS was significantly higher than that of healthy children (p < 0.01) (Fig. 2c), followed by that of Bifidobacterium (Fig. 2d), and the relative abundance of Bacteroides in the intestines of children with DS was significantly lower than that of healthy children (p < 0.01) (Fig. 2e). Therefore, we speculated that the greater abundance of Escherichia and the lower abundance of Bacteroides in the intestines of children with DS might be related to the susceptibility of children with DS to intestinal diseases.
Through the analysis of the intestinal flora of the DSC and NC groups at the level of Metastats (Fig. 2f), we found that the main differences between the two groups were Bacteroides, Bifidobacterium, Escherichia, Clostridioides, Erysipelatoclostridium, Faecilibacterium, Haemophilus, Klebsiella, Parabacteroides and Ruminococcus. The results show that the distribution and differential flora of DS patients in the intestinal flora are different from those of healthy children and provide a theoretical basis for revealing the related disease mechanisms of DS. Through analysis of the intestinal flora of groups CHD and NCHD through Metastats (Fig. 2g), we found that the main differences between the two groups were Enterococcus and Erysipelatoclostridium, indicating that DS patients with CHD and DS patients without CHD have significant distribution of intestinal flora, which indicated that the relative abundance of these two intestinal bacteria was higher in children with CHD. This related discovery provides a certain theoretical basis for exploring the pathogenesis of CHD; then we analyzed the flora of DSM and NM at the genus level of Metastats (Fig. 2h), and we found the main differences in flora between the two groups were Lactobacillus, Atopobium and Corynebacterium. The findings provide a basis for exploring the correlation between children with DS and their mothers.
The differences in the dominant members of the microbiota
From 3.2, we determined the different species in the DSC and DSM groups from the relative abundance level of OTUs. To verify and further determine the more significant microorganisms in different groups, we also conducted LEfSe analysis. LEfSe was used to identify the specific phylotypes related to the DSC and DSM groups. As shown in Fig. 3a-c, the main differential microbial species between the control group and the DSC groups were Bacteroides, Clostridioides, Pseudomonadales, Bifidobacterium, Streptococcus and Escherichia. The main differential microbial species between the CHD group and the NCHD groups were Enterococcus, Pseudomonadales, Bacteroides, Haemophilus, Romboutsia and Paeniclostridium. The main differential microbial species between the DSM group and the NM groups were Lactobacillus, Pseudomonas, Corynebacterium, Peptostreptococcus, Sneathia and Gaiella.
Functional gene prediction
The intestinal flora plays a certain role in the intestine and is closely related to the function of the body. Since we found that the intestinal microbes changed through 16S rRNA sequencing technology, we further analyzed and predicted the functions between different groups.
Through the L1 level KEGG analysis of the gene pathway (Fig. 4a), we found that the expression of the functional pathway of the group NC was mainly concentrated in Metabolism, Cellular Processes and Organic Systems, and the DSC and CHD groups were significantly lower than the normal group in terms of the Genetic Information Processing pathway, indicating that DS patients were weak in the transmission and expression of genetic information. The expression of group DSM functional pathways was mainly concentrated in Genetic Information Processing and Environmental Information Processing, and the functional pathways of group NM were significantly lower in Metabolism, Cellular Processes and Organic Systems. Among other groups, the difference in flora between mothers of children with DS and healthy mothers is also revealed here, which provides a reference basis for exploring the relevant mechanism of mothers' influence on children with DS.
We further analyzed the L2 KEGG of the gene pathway (Fig. 4b) and found that group DSC and CHD significantly downregulated Immune System, Cell Growth and Death, Replication and Repair, Nucleotide Metabolism and Folding, Sorting and Degradation compared with the other groups. The expression of signaling molecules and interacting genes was significantly upregulated, indicating that the immune function of DS patients was weak, and their cell growth and nucleotide metabolism were lower than those of healthy children, which was consistent with the content reported in the literature. The poor immunity of DS children easily causes infection and various complications. Children with DS with poor immunity easily develop infections and various complications. Normal group NC had significantly increased Cell Motility, Lipid Metabolism, Carbohydrate, Metabolism, Amino Acid Metabolism, Metabolism of Terpenoids and Polyketides, Endocrine System, Digestive System, Biosynthesis of Other Secondary Metabolites, Energy Metabolism, Glycan Biosynthesis and Metabolism, which were expressed more in lipid and amino acid metabolism than group DSC. Group NM significantly upregulates Excretory System, Neurodegenerative Diseases and Sensory System and significantly downregulates Signal Transduction, Xenobiotics Biodegradation and Metabolism, Transport and Catabolism, Metabolism of Cofactors and Vitamins, Cell Motility, Lipid Metabolism, Carbohydrate Metabolism, Cellular Community -prokaryotes, Metabolism of Other Amino Acids and Amino Acid Metabolism Pathways. Group DSM significantly downregulated Metabolism of Terpenoids and Polyketides and Endocrine System. The analysis of the L2 level metabolic pathway can provide a basis for exploring the pathogenesis and prevention of DS.