Succession of Gut Microbial Structure in Twin Giant Pandas During the Dietary Change Stage and Its Role in Polysaccharide Metabolism

Abstract

Adaptation to a bamboo diet is an essential process for giant panda growth, and gut microbes play an important role in the digestion of the polysaccharides in bamboo. The dietary transition in giant panda cubs is particularly complex, but it is an ideal period in which to study the effects of gut microbes on polysaccharide use because their main food changes from milk to bamboo (together with some bamboo shoot and coarse pastry). Here, we used 16S rDNA and internal transcribed spacer 1 (ITS1) DNA sequencing and metagenomic sequencing analysis to investigate the succession of the gut microbial structure in feces sampled from twin giant panda cubs during the completely dietary transition and determine the abundances of polysaccharide-metabolizing genes and their corresponding microbes to better understand the degradation of bamboo polysaccharides. Successive changes in the gut microbial diversity and structure were apparent in the growth of pandas during dietary shift process. Microbial diversity increased after the introduction of supplementary foods and then varied in a complex way for 1.5-2 years as bamboo and complex food components were introduced. They then stabilized after 2 years, when the cubs consumed a specialized bamboo diet. The microbes had more potential to metabolize the cellulose in bamboo than the hemicellulose, providing genes encoding cellulase systems corresponding to glycoside hydrolases (GHs; such as GH1, GH3, GH5, GH8, GH9, GH74, and GH94). The cellulose-metabolizing species (or genes) of gut bacteria was more abundant than that of gut fungi. Although cellulose-metabolizing species did not predominate in the gut bacterial community, microbial interactions allowed the giant pandas to achieve the necessary dietary shift and ultimately adapt to a bamboo diet.

Keywords: Clostridium; GH5; cellulose degradation; functional gene prediction; giant panda cub; gut microbes; succession.

FIGURE 1

Fecal bacterial α-diversity in giant panda cubs during the dietary change stage. (A) S_obs index; (B) Chao index; (C) Shannon’s index; (D) Simpson’s index. Colored bars represent the different stages of dietary change. Blue bars represent individual diversity indices in stage 1 and stage 2 (formula and chewing bamboo), purple bars represent those in stage 3 (increasing bamboo consumption), and green bars represent those in stage 4 (bamboo diet). Results are expressed as means ± standard errors. Intestinal bacterial α-diversity is indicated by the height of the column at each sampling time, and the overall fluctuations in the four indices are expressed by the black trend lines.

FIGURE 2

Fecal bacterial and fungal β-diversity in giant panda cubs during the period of dietary change. Colored symbols represent different stages of dietary change. (A–C) describe the results for bacterial β-diversity; (D,E) describe the results for fungal β-diversity. (A) PCoA analysis of genera showed the relative distances between samples in different stages of dietary change; each pattern represents a different individual in a different stage, so the individual differences are also reflected in the relative distances. (B) and (D) are based on an Adonis analysis; “Between” represents differences between sample groups (according to sampling time), and the other bars, named “s (sample) + year (last two digits) + month (two digits),” represent the individual differences at each sampling time. Y-axis scale represents rank of distance. (C,E) Hierarchical clustering tree at the genus level. The lengths between branches represent the distances between samples marked by (Age + name); in (C), sampling time points in S1 and S2 are colored blue, those in S3 are colored purple, and those in S4 are colored green; in (E), sampling time points in S1 and S2 are colored blue and those in S3 and S4 are colored purple.

FIGURE 3

Bar-plot analysis of bacterial communities in panda cubs during their growth and dietary changes. (A,B) Bar-plot analysis of bacterial communities in BanBan and YueYue (other non-dominant taxa, with abundance <1%, were combined), respectively. Different colors represent different taxa at the genus level, and lengths are proportional to the abundance of the genus.

FIGURE 4

Bar-plot analysis of fungal communities in panda cubs during their growth and dietary changes. (A,B) Bar-plot analysis of fungal communities in BanBan and YueYue (others non-dominant taxa, with abundance <5%, were combined), respectively. Different colors represent different taxa at the genus level, and lengths are proportional to the abundance of the genus.

FIGURE 5

Genes encoding polysaccharide (starch, hemicellulose, and cellulose)-metabolizing enzymes were analyzed with functional prediction and metagenomic sequencing. Bacterial genes mainly encoding the polysaccharide-metabolizing enzymes cellulase [EC 3.2.1.4 (A)], hemicellulase [EC3.2.1.37 (B)], and α-amylase [EC3.2.1.1 (C)] were inferred based on 16S rRNA sequences and were predicted with PICRUSt 2, and the fungal genes encoding the polysaccharide-metabolizing enzymes cellulase [EC 3.2.1.4 (D)] and hemicellulase [EC3.2.1.8 (E)] were inferred based on ITS sequences and predicted with PICRUSt 2. Colored points represent samples from different stages of dietary change. Blue points represent samples from stage 1 and stage 2 (formula and bamboo chewing), purple points represent samples from stage 3 (increasing bamboo consumption), and green points represent samples from stage 4 (bamboo diet). Black trend lines indicate the average values for the two cubs and show the changes in the abundances of functional genes. (F) Heatmap of the dominant GHs and CBMs that correspond to cellulase (EC 3.2.1.4) and hemicellulase (EC 3.2.1.8 and 3.2.1.37), based on a metagenomic sequencing analysis and annotated with the CAZy database. Deeper red color indicates higher abundance of the gene; deeper blue color indicates lower abundance of the gene.

FIGURE 6

Microbial interaction grid for cellulose and hemicellulose degradation based on Spearman’s correlation analysis. Almost all the dominant bacterial genera (species) involved in cellulose and hemicellulose degradation are included in this grid; the same species have the same color. The sizes of the nodes represent the abundance of the species: the greater the abundance, the larger the node. The color of the connecting line indicates a positive or negative correlation: red indicates a positive correlation between species and functions, and green indicates a negative correlation between species and functions. The thickness of the line indicates the magnitude of the correlation coefficient: a thicker line indicates a stronger correlation. The more lines, the closer the relationships between this species/function and other species/functions.