Gut microbiota provides an effective strategy for sympatric proximal species to coexist in interspecific competition. In the present study, 16SrDNA was used to investigate the gut microbial of the Cricetulus longicaudatus and Apodemus agrarius, which are two species distributed in the same domain, under natural ambient and varying dietary situations. Our data revealed that there were significant differences in gut microbial structure and diversity between the two species. Specifically, the voles demonstrated high alpha diversity and abundance of Lactobacillus, whereas the mice shown substantial enrichment of Verrucomicrobiota. Wild voles had a more complex co-occurrence network, with a low level of positive correlation rate, however, after being fed various diets, the network structure was simplified, and the positive correlation rate increased. On the contrary, wild mice had a simple co-occurrence network, with a high level of positive correlation rate, after exposure to different diets, the network structure became more complex, accompanied by a decrease in the positive correlation rate. Our results also revealed differences in dietary adaption between the two species. Voles exhibited a greater microbial adaptability under high-fat and high-fiber diet than mice, as indicating by a significant rise in the Firmicutes/Bacteroidetes ratio. While mice demonstrated reduced adaptation to dietary changes, and it had a stronger ability to adapt to high-fat diet than to high-fiber diet. Finally, our data revealed significant alterations in carbohydrate metabolism pathways between two species. This study provides new insights into how the gut microbiota of symmetrically distributed rodents provides effective survival strategies for species in the face of competition.

In this study, we conducted two experiments.

Design of experiment one:

First, we investigates the food adaptation strategies of two syntopic species in the wild by comparing the differences in gut microbiota of two types of mice under natural field conditions. The experimental design is outlined as follows: In the summer of 2024, six samples each of healthy adult male C. longicaudatus and A. agrarius were obtained using cage catches in Houyin Village, Xixian County, Shanxi Province (36.729°N, 110.839°E). Rectal fecal samples were promptly collected in the field and preserved at -80°C until the 16S rDNA sequencing was conducted.

Design of experiment two:

We tested the adaptive alterations in gut microbiota of two species using an indoor experiment including the feeding of animals with food from various nutritional layers under differing dietary circumstances. The experimental design is outlined as follows: Healthy adult male C. longicaudatus and A. agrarius were collected in Huyan Village, Xixian County, Shanxi Province (36.729°N, 110.839°E). Following a two-week adaptation stage in the laboratory, the experiment will commence. All animals are kept in individual cages under circumstances of 23±2℃ temperature, 55±2% relative humidity, a natural light cycle, and a sufficient diet. C. longicaudatus and A. agrarius were respectively divided at random into three groups according to their body mass: Con (Control, standard rodent diet, n=6), HG (High-fiber diet, n=6), and HF (High-fat diet, n=6). Food composition are shown in Table 1. The animals continuous feeding for 28 days, with free access to water and food during the period. After 28 days, rectal feces from all experimental samples were collected and preserved at -80°C until 16S rDNA sequencing was conducted in the indoor experimental group. The experimental procedure was rigorously executed in compliance with the "Regulations on the Administration of Experimental Animals" (2017 Revision) established by the Ministry of Science and Technology of the People's Republic of China.

Bioinformatics analysis

Following sequencing, the two brief Illumina reads were combined using PEAR software (version 0.9.8) based on their overlap, and the fastq files were processed to produce distinct FAST and QUAL files, which could subsequently be evaluated using standard methodologies. The effective tags were grouped into operational taxonomic units (OTUs) of ≥ 97% similarity using Usearch software (version 11.0.667). Chimeric sequences and singleton OTUs (containing only one read) were eliminated, following which the residual sequences were categorized into each sample according to the OTUs. The most abundant tag sequence was chosen as the typical sequence for each cluster. Bacterial and fungal OTU representative sequences were taxonomically identified by using BLAST searches against the RDP Database and the UNITE fungal ITS Database, respectively.

Data analysis

The α-diversity indices (including Chao1, Simpson, and Shannon indices) were quantified in terms of OTU richness. To assess sample adequacy, rarefaction curves of the observed numbers of OTUs were constructed, all α-diversity indices were calculated with Mothur software (version 3.8.31). The OTU rarefaction curve and rank abundance curves were plotted in R (version 3.6.0). To estimate the diversity of the microbial community of the sample, we calculated the within-sample (alpha) diversity by T-test for two groups and multiple group comparisons were made using ANOVA test. Beta diversity evaluates differences in the microbiome among samples and is normally combined with dimensional reduction methods such as principal coordinate analysis (PCoA) to obtain visual representations. These analyses were visualized using Rvegan package (version 2.5-6), and finally the inter-sample distances were presented as scatterplots. Venn diagrams were drawn using R 3.2.6 (Venn Diagram package) to describe common and gap flora between groups. Difference comparison is used to identify features with significantly different abundances between groups using LefSe (version 1.1.0 ). The data were further examined and network analysis was produced using R 3.2.6 and Gephi v.0.9.2 software. Functional prediction analysis of bacteria and archaea using PICRUSt (v1.1.4) software, by comparing existing 16S rDNA gene sequencing data with a microbial reference genome database of known metabolic functions, enables the prediction of bacterial and archaeal metabolic functions. The experimental results were expressed as mean ± standard deviation (mean ± SD). SPSS 26.0 software was used to calculate samples among groups by one-way ANOVA analysis. Statistical differences between treatments were considered significant at P < 0.05, P < 0.01, and P < 0.001.