Background

Insects, as the most diverse and abundant class of animals, thrive in a wide range of habitats and utilize various substrates [1, 2]. In the long-term process of evolution, insects and gut microbiota have established an interdependent symbiotic relationship [3]. Previously, numerous studies have demonstrated that diverse microbes colonizing the insects gut, which played integral roles in their hosts, such as affecting the host metabolism, promoting efficient digestion, aiding the defense and detoxification ability [4,5,6,7,8,9,10,11]. Associations between symbiotic microorganisms and insects are common. For example, the gut bacterial communities found in the camellia weevil Curculio chinensis exhibit consistency with a potential microbial role in the detoxification of the defensive chemicals produced by Camellia trees [11]. The gut microbiota of Cerambycidae species offered a potential function in assisting with degrading lignocellulose [12].

Dorcus is a genus of the Lucanini tribe, belonging to the Lucaninae subfamily and the Lucanidae family, commonly found in China, Japan, Korea, and other East Asian countries. Currently, little research is done on the Dorcus hopei species, including the entire Dorcus genus. D. hopei hopei (Dhh in short, AKA the Giant Chinese Stag Beetle, Fig. 1), one subspecies of D. hopei, is predominantly concentrated in southern mainland China’s broad-leaf forests at an altitude of approximately 200–2000 m, following a wide distribution in East Asia [13]. It is one of the most abundant subspecies in the Dorcus genus and one of the most common stag beetles in mainland China.

Fig. 1
figure 1

Overview of the development stages of Dorcus hopei hopei

The larvae are exceptionally efficient in the decomposition of timber and fungus, while the adults are important consumers of plant exudate. Dhh plays a vital role in the decomposition of dead wood and the carbon cycle and acts as a critical species in the maintenance of healthy forest flora. Although Dhh was still rarely studied, D. hopei binodulosus (another subspecies of D. hopei [14, 15]) was proven to function during nitrogen fixation [16]. Termite gut microbiota has shown to play a vital role in delignification including the Oryctes rhinoceros beetle, known for its consumption in rotten wood., The gut microbiota composition of the latter is similar to the gut microbiota of termites [17,18,19]. The gut microbiota of O. rhinoceros beetle may generate cellobiase to help the host cellulose degradation process [19]. Similarly, the wood-feeding beetle Odontotaenius disjunctus harbors a distinct fiber-associated microbiome, like the termite [20]. These results may suggest that the gut microbiota of wood-feeding insects such as Dhh might play a similar role.

Furthermore, the aim is to outline distinct differences between the gut microbiota composition of wild and domestic Dhh larval individuals. The study compared the gut microbiota disparities between different-sized larval Dhh to establish relationships between larval health, larval size, and specific microbial composition. These results could provide valuable insights toward future studies for the biochemistry field and the microbial conditions of stag beetles in the Dorcus genus. The present study is the first experimental investigation of the gut microbiota functioning and composition of the Dorcus Stag Beetle.

Results

Sequencing information of gut microbiota in wild and domestic Dorcus hopei hopei

A total of 25 samples, including 13 domestic insect gut samples and 12 wild insect gut samples, were sequenced. A total of 1,888,381 raw tages and 1,667,599 valid tages (Table S1) were obtained from 16 S rDNA amplicon sequencing. A total of 4,013 amplicon sequence variants (ASVs) were clustered, in which 33 phyla, 92 classes, 205 orders, 331 families, 732 genera, and 1,072 species were identified. The number of ASVs varied from 74 to 848 per sample, with an average of 316 ASVs per sample observed. Among the 4,013 ASVs, 2,406 ASVs were solely found in the intestinal content of Dhh_W, 1,048 ASVs in the intestinal content of Dhh_D, and 559 ASVs were shared by two Dhh groups (Fig. 2A). Rarefaction curves plotted from observed OTUs have reached a plateau with a high Good’s coverage index (> 99%) that indicated that the sequence depth was able to represent the majority of the ASVs present in all the 25 fecal samples (Fig. 2B, C).

Fig. 2
figure 2

Sequencing Information of gut microbiota in wild and domestic Dorcus hopei hopei. A Venn diagram of ASVs distribution, (B) rarefaction curve of observed_OTUs, (C) rarefaction curve of goods coverage

Intestinal bacterial alpha and beta diversity

The alpha diversity index indicated differences in gut microbiota flora between wild and domestic Dhh groups. According to the observed OTUs and Chao1 index, the wild group is enriched with significantly greater numbers of gut microbiota species compared to the domestic group (Fig. 3A, B). Pielou_e index illustrated that the wild group is slightly more even than the domestic group. (Fig. 3C). The Simpson index showed that the wild group had a higher value (Fig. 3D). The two indexes above showed that the diversity of the microbial communities for the wild group was higher.

Fig. 3
figure 3

Violin plots of alpha diversity and PCoA analysis between the Wild and Domestic Dhh groups. A observed_otus, (B) chao1, (C) pielou_e, (D) Simpson’s index, (E) Principal coordinates analysis (PcoA) and Adonis analysis revealed the differences between the Wild and Domestic groups

Similarity analysis (ANOSIM) at ASV level was based on the Bray-Curtis dissimilarity algorithm to explore the overall difference among wild and domestic Dhh groups (R = 0.2463, p = 0.002). Principal coordinates analysis (PCoA) based on the Bray-Curtis distance algorithm was used to compare the community similarities between wild and domestic Dhh groups. The PCoA scatter plot showed that the samples of Dhh_D were relatively more gathered, compared to the samples of Dhh_W, and the two groups didn’t separate thoroughly (Fig. 3E).

Characteristics comparison of the gut microbiota between wild and domestic dhh

The wild Dhh fecal microbiotas’ majority of abundance was occupied by three phyla: Firmicutes, with a relative abundance (RA) of 24.56–91.36% (mean RA: 62.02%); Proteobacteria, with a RA of 4.28–58.57% (mean RA: 33.71%); and Bacteroidetes, with a RA of 0.47–6.39% (meanRA:1.78%) (Fig. 4A). In the studied wild Dhh, Erysipelotrichaceae was the predominant family within the phylum Firmicutes, with Ruminococcaceae being the second-most dominant family belonging to Firmicutes. These two families are among the top 5 most abundant families in the wild and the overall sample (Fig. 4B). Other phyla observed in low abundance include Fusobacteriota (mean RA: 0.18%) and Actinobacteriota (mean RA: 1.02%), in which Fusobacteriota was almost exclusively seen in wild Dhh fecal samples. In the domestic Dhh samples, the most dominant phylum was Proteobacteria, with a RA of 28.38–89.80% (mean RA: 67.08%), followed by Firmicutes, with a RA of 2.56–67.07% (mean RA: 22.00%); and Actinobacteriota, with a RA of 0.25–10.97% (mean RA: 3.89%). Although at low abundances, multiple trace phyla expressed by domestic Dhh microbiota are almost exclusive to domestic samples, including Methylomirabilota (mean RA: 0.03%) and Desulfobacterota (mean RA: 0.21%) (Fig. 4A).

Fig. 4
figure 4

Difference between wild and domestic Dhh larvae feces. A Bar chart depicting the RA of the major detected phyla in the Wild and Domestic groups. B Circos plot depicting the top 5 families of the overall sample & their RA distribution in the wild and domestic groups. C Top genera in the overall sample & their RA distribution in the Wild and Domestic groups

The RA of Firmicutes was higher in wild Dhh compared to that of domestic Dhh (Wilcoxon test, P < 0.01), whilst the RA of Proteobacteria in wild Dhh was lower compared to that of domestic Dhh (Wilcoxon test, P < 0.05). Bacteroidetes RA showed no significant difference between the two Dhh groups (Wilcoxon test, P = 0.87). Burkholderiaceae, belonging to the phylum Proteobacteria, was the predominant family in both wild and domestic Dhh samples and did express significant differences between its abundance in the two Dhh groups. Its RA for the combined sample reached an impressive 30.71% on average, with a mean RA of 41.02% and 19.54% for the domestic and wild samples, respectively. Another family belonging to the Proteobacteria phylum, Rhodocyclaceae, is among the top 5 most abundant bacterial families for the overall & separate samples (Fig. 4B).

From the genus levels, Ralstonia was the predominant genus in the combined sample and both individual Dhh groups (mean RA: 67.08%). However, there were still stark differences in terms of RA of Ralstonia between the wild and domestic samples, with Ralstonia’s RA being significantly higher in domestic Dhh compared to wild Dhh (40.63% vs. 19.17% respectively). The second overall most abundant genus was Turicibacter (mean RA: 9.99); its mean RA was only 2.75% in domestic Dhh compared to the 17.84% in the wild sample. The abundance of other genera, such as Tyzzerella, Methyloversatilis, and (Candidatus) Soleaferrea, also demonstrated a significant difference in wild Dhh compared to domestic Dhh (Wilcoxon test, P < 0.05). These three genera, combined with Ralstonia and Turicibacter, constitute the five dominant genera in terms of relative abundance in the overall sample (Fig. 4 C).

The Linear discriminant analysis effect size (LEfSe) analysis further identified numerous taxonomic clades that demonstrate statistically significant differences between wild and domestic Dhh (Fig. 5A). A Cladogram of the phylum to species was drawn to fully understand the distribution of different taxa at various taxonomic levels. It was found that the bacteria in two phyla (i.e., Actinobacteria and Proteobacteria), three classes (i.e., Actinobacteria, Gammaproteobacteria, and Alphaproteobacteria), and one order (i.e., Burkholderiales) were more significantly abundant in Dhh_D. Bacteria in one phylum (i.e., Firmicutes), two classes (i.e., Clostridia and Firmicutes_unclassified), six orders (i.e., Erysipelptrichales, Christensenellales, Clostridiales, Lachnospirales, Oscillospirales and Firmicutes_unclassified). The logarithmic discriminant analysis (LDA) results showed that the genus level biomarkers in the gut microbiota of domestic Dhh group were Ralstonia, and Methyloversatilis, while the enriched genera of wild Dhh group were Turicibacter, Gandidatus_Soleaferrea, Tyzzerella, and two unclassified genera of Firmicutes and Ruminococcaceae respectively (Fig. 5B).

Fig. 5
figure 5

Bacterial taxon differences between the wild and domestic Dhh larvae feces microbiota by LEfSe in genus level. A Linear Discriminant Analysis Efect Size (LEfSe) cladogram. B LEfSe histogram. The threshold for the logarithmic discriminant analysis (LDA) score was 4

Network analysis of wild and domestic dhh gut microbiota

Interaction network analysis of the bacterial microbiome in each group was used to select the key species and to unveil the relationship of the microbiome in the community (Fig. 6). The network analysis found 48 nodes and 109 edges in Dhh_W and 51 nodes and 69 edges in Dhh_D. Compared to the Dhh_D, the edges, and average triangles presented much higher, which indicated that a more complex network was observed in Dhh_W. Also, the average weighted degree was much higher in Dhh_W, which probably indicates that the nodes in the network of Dhh_W were more important than those of Dhh_D. More network topological attributes are present in Fig. 6C. Furthermore, different dominant genera were observed in three Dhh groups individually (estimated by weighted degree). In Dhh_D, Tyzzerella, Ralstonia, (Candidatus) Soleaferrea, Turicibacter, and Vibrio greatly contributed to the interaction complexity. In Dhh_W, Tyzzerella, (Candidatus) Soleaferrea, Pseudomonas, Christensenellaceae_R-7_group, and Ralstonia played an important role in the complexity of community.

Fig. 6
figure 6

Interaction network of microbiota at the genus level in (A) wild and (B) domestic Dhh groups. The edge colors represent positive (red) and negative (blue) correlations. The size of each genus is positively correlated with its importance, and the thicker line indicates a stronger correlation. The Only significant interactions are shown (using Sparcc, |r| > 0.2; P < 0.05). C Topological property of the bacterial network of Dhh_W and Dhh_D

Predicted functions of wild and domestic dhh gut microbiota

The microbiota functions were predicted using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2) based on the detected 16 S rRNA gene amplicon sequences. The anticipated pathways and functions showed significant differences in microbiota functionalities between wild and domestic Dhh. Kyoto Encyclopedia of Genes and Genomes (KEGG) level 2 and level 3 pathways (Fig. 7), such as “Nucleotide metabolism”, “replication and repair pathways”, “environmental adaptation”, “starch and sucrose metabolism”, “ribosome biogenesis pathway”, “sporulation”, and “bacterial toxin synthesis” pathways were significantly more enriched (Student’s t-test, P < 0.01) in wild Dhh. On the other hand, “glycan biosynthesis and metabolism pathway”, “lipid metabolism pathway”, “metabolism of other amino acids”, “xenobiotics5 biodegradation and metabolism”, “energy metabolism”, “TCA cycle”, “glyoxylate and dicarboxylate metabolism”, and “oxidative phosphorylation pathways” were enriched significantly greater in domestic Dhh group.

Fig. 7
figure 7

Prediction of altered KEGG pathways using PICRUSt2 analysis of the fecal microbiota for Dhh groups Domestic vs. Wild. The prediction of altered KEGG pathways in (A) level 2 and (B) level 3. Bar plots on the left side display the mean proportion of each KEGG pathway. Dot plots on the right show the differences in mean proportions between the two indicated groups. P-value was calculated using t-test

Microbiota differences of larval dhh of different sizes

All 4 indexes used previously for the alpha diversity of wild vs. domestic Dhh group, demonstrated no significance in terms of gut microbe OTUs between different Dorcus size groups (Figure S1, Table S2). Similarity analysis (ANOSIM) at the ASV level based on Bray-Curtis dissimilarity explored no significant difference among Dhh_S, Dhh_M, and Dhh_L. The PCA (P = 0.416) and PCoA (p = 0.342) analyses were used to visualize the results, which coincides with the ANOSIM analysis result (Figure S2). The bacterial community possessed a similar structure among Dhh_S, Dhh_M, and Dhh_L, which indicated the size of Dhh makes no significant difference in the gut bacterial microbiota.

At the phylum level, the overall top 5 most dominant phyla remain to be Proteobacteria, Firmicutes, Actinobacteriota, Bacteroidetes, and Cyanobacteria. The column chart depicted a significant differences in the relative abundance of the top 5 phyla in larval Dhh of different sizes (Figure S3A). The depicted disparities of relative genera abundance were also less pronounced compared to the wild vs. domestic analysis, with only 9 of the 718 tested genera revealing statistically significant differences for relative abundance between the three Dhh groups. Of the 9 genera, only Turicibacter was among the top 5 most dominant genera in the overall sample (Figure S3B).

Very minimal predicted pathways and functions were shown to have different levels of enrichment between the three Dorcus size groups. The “signaling molecules and interaction” pathway (level 2) was shown to be distinguished enriched among the three groups (Figure S4A). The “ribosome biogenesis” function (level 3) is only enriched at inferior levels in the “Small” group and the “cyanoamino acid metabolism” (level 3) is only greatly enriched in the “Large” group (Figure S4B). The “Wnt signaling pathway” was also enriched at statistically significant levels. Although significant, this pathway’s depicted deficient proportion was deemed unfit for further inspection (Figure S4C).

Discussion

The investigation of the gut bacterial community of Dhh_W and Dhh_D was conducted through high-throughput sequencing, which was the first study of the gut bacteria of Dhh. Firmicutes and Proteobacteria were dominant in the gut bacteria of Dhh_W and Dhh_D, which is consistent with the prior studies in other beetles, including Phalacrognathus muelleri [21], Anoplophora glabripennis [22], Hylobius abietis [23], Nicrophorus vespilloides [24], and Popillia japonica [25]. Previous research has indicated that Firmicutes play a significant role in the degradation of plant carbohydrates, while Proteobacteria contribute to nitrogen fixation and metabolism [21, 26].

As a result, the alpha diversity of Dhh_W was higher than that of Dhh_D. This finding is consistent with a previous study found in species of Bos gaurus [27], Mus musculus [28]. Intestinal bacteria communities are regarded as originating from the environment and diet [29]. Therefore, the high diversity displayed in Dhh_W may be due to the high microbial colonization from a relatively diverse dietary intake in the wild environment. PCoA and ANOSIM analysis revealed a significant difference between Dhh_W and Dhh_D communities, suggesting that the host environment serves a crucial role in shaping intestinal bacterial structure. Previous studies demonstrated that gut bacteria communities can influence the host diet [30,31,32,33]. Although the study tried to provide Dhh_D with the same food source as Dhh_W, there may still be some unrealized differences in food composition, which may be one of the reasons for community differences.

Although differences between gut microbial communities were observed, there are also observable similarities as well. For instance, although Firmicutes were expressed with higher RA and Proteobacteria expressed at a lower RA in wild Dhh compared to the domestic group, the two phyla combined comprised more than 85% of the bacterial species in the domestic group and more than 95% in the wild group.

The statistics reflect the significance of these two phyla in forming larval Dhh gut microbiota; furthermore, the discrepancy in RA of the two major phyla among the two Dhh groups demonstrates the environmental and ecological impact on the Firmicutes-to-Proteobacteria ratio and overall microbiota compositional balance. The top 5 most dominant families in the overall sample are Burkholderiaceae, Erysipelotrichaceae, Ruminococcaceae, Lachnospiraceae, and Rhodocyclaceae. Three belong to Firmicutes and two to Proteobacteria, parallel to the obtained phylum-level statistics. Ralstonia was responsible for the abundance of Burkholderiaceae, Turicibacter for Erysipelotrichaceae, Tyzzerella for Lachnospiraceae, and Methyloversatilis for Rhodocyclaceae.

The family Ruminococcaceae was distinctive as its abundance was the second-most dominant family belonging to Firmicutes in wild Dhh groups, which some taxa in this family presumably share certain beneficial features favorable for polysaccharides biodegradation in Dhh gut. This speculation was supported by some general Ruminococcaceae known capacity for structural polysaccharides breakdown [34, 35] and further testified by the strong positive association between Ruminococcaceae and tetrapod herbivory [34]. The study cannot conclude that the specific function of Ruminococcaceae in Dhh gut microbiota will be identical to that of other tetrapod herbivores due to physiological differences, current knowledge about biochemical potentials of Ruminococcaceae does imply high probabilities of this bacterial family conducting similar roles in Dhh gut microbiota.

Previous studies have utilized microbial correlation network analysis to investigate the differences in microbial community structure [36,37,38,39]. Microbial correlation network analysis allows the identification of the key species in the entire network, by identifying the species with high connectivity throughout the network, and these species may have a crucial role in the structure and function of the microbial community [40]. Our study performed microbial correlation network analysis and observed several topological characteristics of the network. Specifically, the study found an increasing number of edges and higher average triangles, indicating an enhanced complexity of community in Dhh_W. Besides, the proportion of positive association decreased, and that of negative association increased in Dhh_W, which indicated a more balanced community association. The key nodes with a high degree, in the network, probably exert an essential function in maintaining gut microbiota and in exerting specific beneficial functions. Network analysis indicated that a more complex and stable bacterial community was observed in Dhh_W, which coincided with our anticipation. It’s probably the complex environment and relatively diverse food intake that leads to the complex intestinal bacterial community and its potential function.

Overall, although existing at different abundances, both Wild and Domestic Dhh group exhibited significant gut microbiota communities associated with xenobiotic degradation and metabolism. Out of the top 5 genera, the ones responsible for metabolism and plant material breakdown (Turicibacter, Tyzzerella, Soleaferrea) were more significantly enriched in the Wild group; genera with xenobiotic degradation abilities (Ralsonia, Methyloversatilis) were more abundant in the Domestic group, completely aligning with trends identified by PICRUSt2 results. Xenobiotics-degrading microbes’ elevated presence is potentially the direct effect of the artificial exacerbation of Dhh larvae’s habitat ecology.

As the most abundant genus in Dhh_W and Dhh_D, the soil-borne bacterial genus Ralstonia was widely regarded as a plant pathogen, causing bacterial wilt in plants [41]. Ralstonia includes mostly plant pathogenic bacteria capable of growing on plant oils/fatty acids, and expresses genes such as aceA1 and aceA2 that code for enzymes involved in the glyoxylate cycle [42]. This indicates that Ralstonia potentially provides Dhh with metabolites derived from previously undegradable plant lipids and aids in Dhh cells aerobic respiration. Ralstonia could utilize an array of aromatic compounds and heavy metals for energy for detoxification [43]. These compounds are primarily xenobiotic and hazardous to organism health. Ralstonia species are especially valuable gut microbes to form symbiotic relationships with larval Dhh, whose nutrition sources often experience contamination during decay. PICRUSt2 reinforced our hypothesized function of Ralstonia in larval Dhh gut microbiota. The functions of “lipid metabolism,” “xenobiotics biodegradation and metabolism,” “TCA (tricarboxylic acid) cycle,” and “glyoxylate metabolism” were all significantly enriched in the domestic group, the group of higher Ralstonia RA. These results indicate that the microorganisms being regarded as plant pathogens probably play a vital role in insects by improving metabolism and detoxification. Furthermore, since the rearing of domestic larvae occurred in human infrastructure, the domestic group is indeed expected to be exposed to greater levels of xenobiotics, validating PICRUSt2 data and Ralstonia’s predicted functions.

The genera (Candidatus) Soleaferrea and Tyzzerella are denoted to contribute to plant-matter breakdown. Both were identified as core microbiota genera of Macrotermes falciger (a termite species), a species sharing similar diets with Dhh larvae (xylophagous & mycophagous) [44]. Furthermore, Tyzzerella was significantly enriched in the gut microbiota of Bark beetles which served as forest damaging agents [45].

The microbiota differences between the Dorcus size groups (L, M, S) are of less significance, as could be seen from abundance analysis and PICRUSt2 results. One actual difference between the Dhh group’s gut microbiota is the Turicibacter RA, which depicts a descending tendency as larvae size decreases. This suggests that larger Dhh larvae growth was potentially attributed to reinforced metabolic capabilities and internal hormone regulation provided by Turicibacter. The “large” group was the most enriched, the “medium” group came second, and the “small” group was the least enriched. Overall, the patterns demonstrated by Dhh larvae of different sizes contrasted with the former research results, as size and microbiota composition/diversity were demonstrated to be associative in the stag beetle Odontolabis fallaciosa [46].

This present study results could be beneficial for improving the current artificial rearing methods of Dhh by suggesting advantageous microbial communities for larval growth, potentially leading to modified feeding procedures that cultivate more aesthetic and healthier commercial Dhh. More importantly, the obtained data are beneficial for developing future environmental protection and conservation methods. One hypothesized technique to be considered is the release of artificially bred Dhh larvae or adults with modified microbiota profiles imitating wild individuals into designated areas to facilitate material decomposition and nutrient cycling in any required habitat, such as the insect-based agri-food waste valorization [47]. Modifying the gut microbiome reduces the released specimens’ repulsive responses to a feral environment and lowers any risks of introducing foreign gut microbes into the wild population. This could alleviate nutrient cycling issues in the specified environment with minimal contamination of the gut microbiome and overall biochemical diversity of the local species.

As this study is one of the very few research focusing on the gut microbiota of the D. hopei and specifically the Dhh subspecies, these results provided additional insights into the area of Lucanidae gut microbiota and outlined several new research orientations to be conducted for further exploration or validation of information regarding gut microbiota of Dhh.

Conclusions

The present study utilized 16 S rDNA sequencing to compare the fecal bacteria of wild and domestic Dhh larvae and between larvae of different sizes. The relative abundance of fecal microbiota showed a notable difference between wild and domestic Dhh. Overall, the domestic Dhh individuals contained more gut microbial taxa with xenobiotic degrading functions, such as genera Ralstonia and Methyloversatilis, while the wild sample possesses gut microbiota compositions more appropriate for energy metabolism and potential growth, such as Turicibacter and Tyzzerella. On the other hand, Dhh larvae of different Dorcus size groups exhibited significantly fewer disparities in larval gut microbiota composition. This finding suggested that habitat-ecology factors, rather than individual growth potential, are responsible for the assemblage of detected distinguished microbiota patterns. These results provided preliminary insights into the gut microbiota of Dhh and its most common subspecies and increased our current understanding of Lucanidae microbiota composition. Results also indicated future research orientations for exploring insect gut microbiome. Furthermore, the significant enrichment of xenobiotics-degrading gut microbiota functionalities in Dhh larvae, especially in the domestic group, demonstrates the inherent environmental presence of factors pernicious to the survival of Dhh larvae. This calls for conservational actions aimed at alleviating the influence of artificial infrastructure and contaminants on local ecology and biodiversity.

Materials and methods

Sample collection and rearing condition

The individuals from the wild were collected in decayed timber located in Hangzhou, Zhejiang Province, China. Out of the 12 wild larvae, four specimens were from Yangjiapai Village (30° 14’ 57.0768’’ N, 120° 3’ 54.3384’’ E), four from Yuewang Temple (30°09’07.0"N, 120°13’46.9"E), three from Wenbi Mountain (30°12’34.6"N, 120°05’36.6"E), and one from Baoshi Mountain (30°15’40.0"N 120°08’38.9"E). Larvae were excavated out of the initial shelter, rotten logs in broad-leaved forests shaded area, from the range of altitude between 200 and 600 m. with proper instruments, no larvae were harmed in the process, and each health condition was examined before being accepted as a sample individual. Each individual was assigned a number; their physical data (head capsule diameter) and feces samples were obtained.

Domestic larvae were raised in captivity. The larvae were given either fermented wood flakes or bottled Coriolus versicolor mushroom packets, instead of rotten timber presence in the wild group. All the domestic larvae were reared in a thermal controlled environment between strict temperatures of 22–24 °C for the larval life cycle. The fungus bottles and wood flakes’ humidity remained in the good range between 55 ± 5% constantly. The fungus bottle and fermented wood flake (compressed in a bottle) were changed periodically about every 3 months to prevent food decay. Larvae were kept in a dark environment and the container remained a viable ventilation, and any other human disturbance was prohibited to eliminate confounding variables. Likewise, each individual was assigned a number, and their physical data & feces samples were obtained.

Body data collection and grouping

The head capsule diameter was measured for each larva. The larvae were held up, and the horizontal diameter of the head capsule was measured using a caliper. The results were recorded and used for grouping. The utilization of head capsule size data instead of other features as growth-potential indicators has been justified in preceding texts. The body weight of each larva was measured using a balance. Body weight was seen as the standard for fat accumulation, which is directly associated with material processing and nutrient absorption. In the process, the larvae body surface feces and wood dust were cleaned before measuring to prevent deviation.

Two grouping plans were used in this study. The purchased and foraged individuals were isolated into “Domestic” and “Wild” Dhh groups (Dhh_D and Dhh_W), each including 11 and 12 Dhh sample individuals, respectively. The same experimental individuals were assigned into groups “Large,” “Medium,” and “Small” Dhh groups (Dhh_L, Dhh_M, Dhh_S). Each corresponds to a head capsule size of equal/over 11 mm, less than 11 mm & equal/over 9 mm, and less than 9 mm), each including 6, 10, and 6 Dhh sample individuals respectively. Analyses concerning the two grouping systems were separately done.

Fecal sample collection method

The development stages overview of Dhh is shown in Fig. 1. The collected samples were all in the third instar larvae stage. Wild individuals were collected and installed in a container filled with the wood dust processed from the original sheltering rotten timber. Domestic individuals were taken out of their feeding medium and installed in containers. larvae were brought to the laboratory under safety and were left undisturbed for a 3-day acclimation period, in which they were monitored for any potential repulsive behaviors or health condition changes. Throughout the transportation and acclimation process, the living temperature of the larvae was controlled at 24 °C. After three days, the larvae were gathered for feces collection. The larvae’s abdomen and anal region were disinfected by wiping them with 75% alcohol-soaked gauze. The feces were then naturally excreted by the larvae individually into a sterile EP tube. Sterile forceps and scoopulas were used to remove 1 gram of feces sample from each tube and place it on a sterile foil, measuring on an analytical balance to ensure it is the precise amount. After reaffirmation, feces samples were transferred into iLongsee Stool Storage Kit (Longsee Biomedical, China), and kept in -80℃ refrigerator until DNA extraction.

PCR amplification and 16 S rDNA sequencing

DNA from different samples was extracted using the E.Z.N.A. ® Stool DNA Kit (D4015, Omega, Inc., USA) according to the manufacturer’s instructions. Nuclear-free water was used for blank. Total DNA was eluted in 50 µL of Elution buffer and stored at -80 °C until amplification in the PCR by LC-Bio Technology Co., Ltd, Hangzhou, Zhejiang Province, China.

The V3-V4 hypervariable region of the 16 S rRNA gene was amplified using indicated primers 341 F (5’-CCTACGGGNGGCWGCAG-3’) and 805R (5’-GACTACHV GGGTATCTAATC C-3’) (Logue et al., 2015). PCR amplifications were performed in 25 µL of reaction mixture including 25 ng of template DNA, 12.5 µL of PCR Premix, 2.5 µL of each primer, and PCR-grade water. The PCR conditions go as follows: 98 °C for 30 s; 32 cycles of 98 °C for 10 s, 54 °C for 30 s, and 72 °C for 45 s; and then finally 72 °C for 10 min. PCR products were verified with agarose gel electrophoresis. Throughout the DNA extraction process, high-purity water will be used to exclude possible false-positive PCR results. The PCR products were purified by AMPure XT beads (Beckman Coulter Genomics, Danvers, MA, USA) and quantified by Qubit (Invitrogen, USA). Amplicon pools were prepared for sequencing; the amplicon library size and quantity were assessed on Agilent 2100 Bioanalyzer (Agilent, USA) and with the Library Quantification Kit for Illumina (Kapa Biosciences, Woburn, MA, USA), respectively. The libraries were sequenced on the NovaSeq PE250 platform.

Data analysis

Samples were sequenced on an Illumina NovaSeq platform according to the manufacturer’s recommendations provided by LC-Bio. Paired-end reads were assigned to samples based on their individual barcoding and trimmed by cutting off the primer sequence. Paired-end reads were merged using FLASH. Raw reads filtering was performed under specific filtering conditions to obtain precise tags according to fqtrim (v0.94). Chimeric sequences were filtered using Vsearch (v2.3.4). After dereplication, feature tables and sequences were obtained using DADA2. 5 indices of Alpha diversity including Chao1, Observed species, Goods coverage, Shannon, and Simpson, were calculated with QIIME2 according to SILVA (release 132) classifier, and feature abundance was normalized using the relative abundance of each sample. Beta diversity were calculated by QIIME2, and the graphs were drawn by R package. BLAST was used for sequence alignment, and the feature sequences were annotated via SILVA database. Additional statistical analyses were manipulated utilizing R package (v3.5.2). Principal Coordinates Analysis (PCoA) [48] was calculated using ade4 package and Principal Components Analysis (PCA) [49] was calculated using stat package. Analysis of similarities (ANOSIM ), a non-parametric statistical test, used to test the difference between Dhh groups, was calculated using vegan package [50]. Log-linear discriminant analysis effect size algorithm (LEfSe) analysis was performed to identify potential taxonomic groups (LDA scores > 4.0, p < 0.05) that can be regarded as biomarkers for different Dhh groups [51]. A bacterial interaction network was performed using the Fastspar [52] with a Sparcc correlation coefficient (|correlation r)|>0.2, and p < 0.05), and the network map was visualized using Gephi 0.9.4. Prediction of the microbial function and pathway in KEGG database based on the 16 S rRNA sequencing data was undertaken using the PICRUSt2. And PICRUSt2 Analysis was performed using the OmicStudio Analysis at https://www.omicstudio.cn/analysis/. Other diagrams were visualized using the R package (v3.5.2). Wilcoxon rank-sum test, Kruskal-Wallis H test, or student’s t-test were employed to determine statistical significance (P < 0.05 is statistically significant).