3. RESULTS

3.1 Sequencing quality and ASV analysis

In this study, the composition of gut microbiota in Catharsius molossus under starvation and refeeding conditions was analyzed. A total of 12 samples were sequenced, yielding a range of 49,222 to 69,752 sequences after removing chimeric sequences. The average number of valid sequences per sample was 66,620. The hungry group exhibited 1,289 ASVs, while the refed group displayed 2,847 ASVs. There were 419 common ASVs between the two groups, with 870 ASVs exclusive to the hungry group and 2,428 ASVs exclusive to the refed group (Fig. 1A). This indicates shared bacteria within the two gut microbiota groups while also highlighting their distinctive microbial communities. The sparse curve analysis showed a plateauing trend with increasing extracted sequence counts, indicating that the sequencing data volume and depth were appropriately set for the samples (Fig. 1B).

3.2 Composition and differential analysis of two gut microbiota

Through ASV annotation, a total of 26 phyla, 45 classes, 68 orders, 177 families, and 399 genera were identified across the two gut microbiota. At the phylum level (abundance > 1%), the predominant phyla in Catharsius molossus under starvation conditions were Proteobacteria (51.93%), Firmicutes (33.11%), Actinobacteria (5.65%), Bacteroidetes (5.63%), Others (2.66%), and Chloroflexi (1.05%), with the highest abundance observed in Proteobacteria (Fig. 2A). In contrast, the predominant phyla in Catharsius molossus under refed conditions were Firmicutes (44.88%), Proteobacteria (19.73%), Bacteroidetes (15.16%), Actinobacteria (4.78%), Synergistetes (4.15%), Candidatus_Saccharibacteria (3.04%), Planctomycetes (2.31%), Others (2.18%), No_Rank (1.49%), Acidobacteria (1.21%), and Chloroflexi (1.08%), with the highest abundance observed in Firmicutes. This reveals distinct predominant phyla between the two conditions, with the gut microbiota of refed beetles displaying higher phylum-level diversity (Fig. 2B).
At the genus level, the predominant genera in Catharsius molossusunder starvation conditions were Unassigned (51.02%), Others (15.42%),Vagococcus (15.11%), No_Rank (8.93%), Dysgonomonas(2.66%), Sphingobacterium (1.71%), Gordonia (1.68%),Acinetobacter (1.26%), and Paracoccus (1.13%), with the highest classified and abundant genus being Vagococcus . In contrast, the predominant genera in Catharsius molossus under refed conditions were Others (25.96%), Romboutsia (13.81%), No_Rank (12.24%), Unassigned (10.09%), Clostridium_XI(7.67%), Proteiniphilum (6.33%), Cloacibacillus(3.83%), Clostridium_sensu_stricto (3.75%), Saccharibacteria_genera_incertae_sedis (3.04%), Turicibacter (2.69%), Lysinibacillus (1.9%), Corynebacterium (1.62%),Luteimonas (1.38%), Hydrogenophaga (1.22%),Ercella (1.22%), Dysgonomonas (1.08%), Serpens(1%), Anaerovorax (1%), with Romboutsia being the highest classified and abundant genus. This indicates higher genus-level diversity in the gut microbiota of refed beetles, with different dominant genera observed between the two conditions (Fig. 2C).
Based on Metastats analysis, comparisons of phylum-level classifications between the starvation and refed gut microbiota of Catharsius molossus revealed significant differences (P < 0.05) in Hydrogenedentes, Proteobacteria, Synergistetes, Planctomycetes, Verrucomicrobia, Fusobacteria, and Bacteroidetes (Table 2). At the genus level, 22 genera exhibited significant differences (P < 0.05) (Table 3).

3.3 Alpha diversity and differential analysis of two gut microbiota

Alpha diversity analysis was conducted based on the Wilcoxon rank-sum test to compare the richness and diversity indices of the gut microbiota between the starvation and refed states of Catharsius molossus . The Chao1 and ACE indices, representing species richness(M. Yang et al., 2022), were found to be significantly higher in the refed group than in the starvation group (P < 0.05), indicating a greater species diversity in the refed microbiota. Additionally, the Shannon index, reflecting microbial diversity(Zhang et al., 2022), was also higher in the refed group, while the Simpson index, indicating lower microbial diversity, was lower in the refed group (P < 0.05). These results collectively suggest that the gut microbiota of the refed group exhibited both higher richness and diversity compared to the starvation group, underscoring significant differences between the two feeding conditions (Table 4).

3.4 Beta diversity and differential analysis of the two gut microbiota

Principal Coordinate Analysis (PCoA) was employed to evaluate the similarity in the structure of gut microbiota between the starved and refed groups(Song et al., 2021). The PCoA plot visualized the primary sources of dissimilarity among the samples along the horizontal (Axis 1) and vertical (Axis 2) axes, which accounted for the most significant variations. Distinct colors on the PCoA plot represented different groups, with shorter distances indicating greater similarity and reduced dissimilarity in microbial structures between paired samples. Employing unweighted UniFrac distances, the PCoA analysis revealed that Axis 1 contributed to 27.89% of the variance, while Axis 2 contributed to 15.47% (Fig. 3). Importantly, the gut microbiota of Catharsius molossus exhibited clustering patterns corresponding to their feeding conditions. Notably, the starvation group displayed more scattered microbial compositions among samples, whereas the refed group demonstrated a higher degree of intra-group clustering. These findings imply that under conditions of starvation, Catharsius molossusmay employ diverse strategies in response to environmental changes, leading to differences in their gut microbiota configurations.

3.5 Functional prediction using PICRUSt

Functional gene annotations predicted by PICRUSt provided a comprehensive insight into the metabolic potential of the gut microbiota in Catharsius molossus. These annotations were mapped to primary, secondary, and tertiary pathways within the KEGG database, encompassing various hierarchical levels of metabolic classification. The primary pathways fell under five fundamental metabolic categories: metabolism, genetic information processing, environmental information processing, cellular processes, and organismal systems. Within these primary pathways, a total of 27 secondary-level pathways and 175 tertiary pathways were identified. Remarkably, metabolic pathways were dominant within the primary metabolic category, accounting for approximately 80% of the total annotations. Among the secondary pathways, amino acid metabolism exhibited the highest gene abundance, constituting around 15%. Notably, utilizing the Wilcoxon rank-sum test, 32 tertiary metabolic pathways displayed significant differences (P< 0.05), highlighting potential alterations in the metabolic landscape between groups at the granularity of these specific pathway levels (Fig. 4).