The HiPure Soil DNA Kits were purchased from Magen biotechnology (Magen Pvt. Ltd. China). The Kod Plus buffer and DNA polymerase were purchased from TOYOBO (Japan) along with, other reagents required for PCR amplifications. The AxyPrep^™ gel DNA extraction kit and StepOnePlus^™ Real-Time PCR system were procured from Axygen (Axygen Biosciences, CA, United States) and ABI Life Technologies (Life Foster City, United States) respectively.

The wood-feeding C. formosanus termites were collected from Shaoguan City (24°55′8″ N, 113°57′44″ E), Guangdong province, P. R. China. Since C. formosanus is among the 100 worst alien invasive species causing tremendous damage to wooden structures and forests, feeding predominantly on pine trees (Lee et al., 2021). We reared and maintained these termites in the laboratory at 26 ± 2° with R.H. ~60–80% by feeding ad libitum on pine wood, Pinus massoniana (Mikaelyan et al., 2015; Dar et al., 2022). To avoid the contamination from environmental microbes including bacteria, the pine wood blocks were autoclaved at 121°C for 15 min before feeding to the termites (Mitaka and Vargo, 2023). A total of 1,050 worker termites were carefully dissected in a biosafety hood using sterilized instruments (Dar et al., 2022). The termites used for the experimentation comprised three different groups that were reared from the same stock colony. Within each group, approximately 350 ± 20 worker termites were subjected to dissection to reveal the individual gut regions. Prior to dissection, the adult worker caste termites were surface sterilized with 50 and 70% ethanol in distilled water each for 30 s, followed by a brief wash of sterile double distilled water (SDDW). After dissection, the gut systems were divided into foregut, midgut- and hindgut (Supplementary Figure S1) with the aid of 30X portable optical loupe by using sharp and sterilized needles. Then, each of the gut-regions was separately placed in 1.5 mL micro-centrifuge tubes (MCT), frozen immediately in liquid nitrogen, and stored at −80°C till further use. The dissection and processing of the gut-regions were carried out carefully to eliminate the chances of bias caused due to the mixing of the gut contents. The gut-regions were suspended in 500 μL of sodium phosphate buffer solution (50 mM; pH 7.0) and subsequently homogenized with polypropylene micro-pestles. The experiments were replicated three times, with each replication involving the consideration of 350 ± 20 termites to retrieve the individual gut regions.

The microbial community DNA was extracted by using the HiPure Soil DNA extraction Kit (Magen, Guangzhou, China) following the manufacturer’s instructions. The extracted DNA was quantified and quality checked on a nanodrop Biospectrophotometer. The V3 and V4 regions of the 16S rDNA genes of the bacterial community were PCR amplified (Liu et al., 2019) by using the primers listed in Table 1. The PCR amplifications were operated at 94°C for 2 min, 30 cycles at 98°C for 10 s, 55°C for 30 s, and 68°C for 30 s followed by a final extension step at 68°C for 5 min. The PCR amplifications were performed in triplicates using a 50 μL reaction mixture comprising 5 μL of 2 mM dNTPs, 3 μL of MgSO₄ (25 mM), 5 μL of 10× KOD Buffer, 1.5 μL of each primer (10 μM), and 1 μL of KOD Polymerase added to 100 ng of template DNA. The successful PCR amplicons were extracted from agarose gels (2%) and then purified by using an AxyPrep DNA Gel extraction kit (Axygen Biosciences, CA, United States) according to the manufacturer’s instructions followed by quantification with ABI StepOnePlus Real-Time PCR System (Life Technologies Foster City, United States). The quantified amplicons were pooled and used to construct paired-end DNA libraries. The constructed libraries were sequenced (250 paired-end sequencing) on the Illumina MiSeq platform according to the standard protocols.

The obtained raw sequences were processed by using the Quantitative insights into microbial ecology software v1.9.1 (QIIME, Caporaso et al., 2010). Before analysis, the raw sequence reads containing adapters or low-quality reads were quality-filtered using FASTP software v0.18.0 (Chen et al., 2018). The quality clean-up was carried out by removing the sequence reads containing more than 10% unknown nucleotides and base quality (Q-value) > 20. Subsequently, the paired-end clean reads were merged into consensus sequences using FLASH v1.2.11 (Magoč and Salzberg, 2011) based on the overlaps longer than 10 bp and a mismatch error rate of 2%. All chimeric tags were removed by UCHIME algorithm v4.2 (Edgar et al., 2011), resulting in effective tags that were used for further analysis. These effective tags were clustered into operational taxonomic units (OTUs) based on ≥97% sequence similarity using UPARSE pipeline v9.2.64 (Edgar, 2013). The number of OTUs was summarized with USEARCH 7.0 to generate OTU data table for each group. The annotation of the OTUs was performed by the Ribosomal Database Project classifier (RDP), and then the representative sequences of each OTU were selected for taxonomic information with an identity threshold of 0.8 (Wang et al., 2007). The initial OTU matrix often contains several OTUs showing extremely low abundance that reduces the number of OTUs having high abundances thereby increasing the complexity of the data analysis. The elimination of these rare OTUs depicts a negligible effect on bacterial diversity, while significantly increasing the efficiency of data analysis. Therefore, OTUs that showed a relative abundance of less than 0.1% of all OTUs were excluded from the analysis (Bokulich et al., 2013). The rarefaction curves representing the bacterial diversity of each sample were drawn in the Mothur software package (Kemp and Aller, 2004) to determine the sequencing depth.

The sampling depth of the obtained OTUs was estimated from rarefaction curves by using good’s coverage (Good, 1953). An overview of the work flow of methodology used for the present study is provided in Figure 1. The Alpha diversity index such as expected richness (Chao, 1984), diversity (Chao and Shen, 2003), and Evenness in terms of Chao, Shannon, and Simpson indices of the bacterial communities were calculated for each sample in QIIME (Caporaso et al., 2010). The community structure was compared between the gut regions by using the taxonomy-dependent Bray-Curtis metrics, a statistical test used to quantify the compositional dissimilarity among samples (Bray and Curtis, 1957). To visualize the dissimilarity among the gut-regions, the high dimensionality of the pairwise dissimilarity scores were compressed into two dimensions and plotted on non-metric multidimensional scaling (NMDS) using vegan package v2.5.3 (Oksanen et al., 2010) in R program. Similarly, the covariance between the community structure of gut compartments was determined by permutation multivariate analysis of variance (PERMANOVA) and subsequently visualized by principal coordinate analysis (PCoA) using the adonis function implemented in the vegan package (Ramette, 2007). Later, the biomarker features in each group were screened by LEfSe software version 1.0 (Segata et al., 2011), in the R project. Similarly, the ternary plots depicting the species abundance were plotted using R ggtern package v3.1.0 (Hamilton and Ferry, 2018). Further, the tests of significance for both intragroup and intergroup Unifrac distances were carried out by using the Monte Carlo permutation test in the QIIME software. To explore the differences between the intergroup distances, an analysis of similarity (ANOSIM) was performed and effects were visualized through biplots in the R project using the vegan package.

The functional characterization of bacterial communities residing in the gut system of animals is the key element to understanding host-symbiont interactions. To accomplish this, the KEGG pathway analysis of the OTUs was inferred by using Tax4Fun v1.0 (Aßhauer et al., 2015) by predicting the potential functions of observed microbial communities within the termite gut. Further, the microbiome phenotypes were classified using Bug Base (Ward et al., 2017). The functional annotation of the prokaryotic taxa database (FAPROTAX) v1.0 (Louca et al., 2016) was used for generating the ecological functional profiles of the gut bacteria. The analysis of function difference between bacterial groups was calculated by Tukey’s HSD test in the R project using the Vegan package (version 2.5.3, Oksanen et al., 2010).

The sequences obtained in this study are deposited to the NCBI Sequence Read Archive database with Bioproject ID, PRJNA1013132. The accession numbers for the sequence reads are SRR25919998 to SRR25920006.

The results are presented as means and standard deviations derived from three replicates. The data were subjected to statistical analyses in R program v4.3.1 (R Core Team, 2023), and the principal component analysis were performed to generate the biplots of first two components by using ggtern package v3.1.0 (Hamilton and Ferry, 2018). A p ≤ 0.05 was considered statistically significant. Each experiment was replicated at least three times to ensure the reproducibility of the results.

A total of 1.17 million pair-end reads were obtained from 9 metagenomic libraries by sequencing the V3 and V4 regions of the bacterial communities, derived from the foregut, midgut, and hindgut of the wood-feeding termite, C. formosanus. These pair-end reads were processed into 1,172,293 clean-end reads. The high-quality cleaned reads were further binned and filtered into 1,072,995 raw tags (Table 2). These raw tags were further processed for chimera removal resulting in 1,034,446 effective tags. Subsequently, the effective tags were classified into 893,035 taxon tags after removing 141,411 singleton tags. The high-quality reads based on the read length of ~450 bp led to the assignment of 3,505 OTUs after defining 97% sequence similarity (Figure 2). The calculated rarefaction curves inferred from species richness reached a plateau, except for a few more diverse samples from hindgut (Supplementary Figure S2). However, the good’s coverage depicted efficient sequencing depth covering over >99.4% of the bacterial communities harbored by the termite. The presence of the Archaeal sequences was not detected in the microbial metagenomes. Since bacteria comprise an integral component of the termite gut microbiota, all of the divisions found in our metagenomic analysis were bacterial lineages, suggesting their indispensable role in fostering digestive symbiosis among termites.

Among the gut regions, the highest number of taxon tags were achieved with foregut (109101) followed by midgut and hindgut which contributed 101,092 and 87485.3 taxon reads, respectively. Of the 3,505 OTUs observed in the gut system of C. formosanus, the indicator species analysis revealed an abundance of 2,177 OTUs only, among the three gut-regions of the termite. Among them, the maximum number of OTUs was harbored by foregut having 1,287 OTUs. Though lower than the foregut, the midgut and hindgut regions sheltered an average of 1,146 and 1,130 OTUs, respectively (Figure 2). Approximately 500 OTUs assigned to species-level classification were shared between the foregut and hindgut. The highest number of 711 OTUs were shared between the foregut and midgut regions. The total OTUs shared by the midgut with hindgut was found to be 644 while 469 OTUs were shared between all the three gut-regions under consideration. The number of unique OTUs were observed as 545, and 455 for foregut, and hindgut respectively, while being lowest for the midgut which restricted the unique OTUs to 260 only (Figure 3).

The Shannon index of 5.48 together with Simpson and Pielou indices (0.87 and 0.54 respectively) suggested higher bacterial diversity in the foregut than midgut and hindgut. However, the nonparametric estimation of the bacterial species demonstrated highest species richness in hindgut depicting a chao1 value of 1,722 (Figure 4, Supplementary Figure S2) which was further supported by ACE value of 1884.7. After hindgut, the chao1 and ACE indices revealed higher species richness in the midgut (1,620 and 1,710) while least in the foregut (1,454 and 1,469). The phylogenetic diversity (PD) analysis of the hindgut metagenome having PD-tree value of 128.2, was significantly different from foregut metagenome (98) showing statistical significance of p < 0.005. Similarly, the Tukey HSD test based on Sob index revealed significant variation in alpha diversity (p < 0.01) of the bacteria in the gut-regions of C. formosanus. Significant differences in bacterial species richness and diversity were observed among the three gut-regions of the termite, in terms of an evaluation by Kruskal-Wallis test (p < 0.05).

The ANOSIM R was equal to 0.835 (p < 0.01) which suggested differences in the intergroup similarity ratio for unweighted Unifrac distance (Figure 5, Table 3). The unweighted Unifrac distance-based NMDs analysis revealed higher intergroup distances than the intragroup variants (Figure 6). A similar trend (R2 = 0.4274, p < 0.01) was also shown by the unweighted-Unifrac distance based on the Adonis (PERMANOVA) analysis (Table 3). The weighted-Unifrac distance measurements (R = 0.547, p < 0.05) further indicated noticeable differences between intergroup and intragroup gut microbiomes. The PERMANOVA analysis (p < 0.01) based on the Unifrac distance matrices for bacterial community profiles confirmed that each of the gut-regions possessed unique bacterial diversity which might be due to the physiology and pH milieu of the gut environments.

To characterize the phylogenetic affiliations of the bacterial communities associated with different gut-regions of the termite, C. formosanus, we analyzed gene sequencing of the community DNA. Overall, the identified bacteria were represented by 26 different phyla in the gut system of C. formosanus. All of the gut regions represented dominance of the Bacteroidetes with 30.2, 43.5, and 56.9%, respectively, in foregut, midgut and hindgut of the termite. However, after Bacteroidetes, the gut systems exhibited variable dominance of different phyla such as foregut, sheltered 28.9% of the Proteobacteria accounting for second dominant phylum in that region (Figure 7). In contrast, midgut and hindgut were dominated by Firmicutes and Spirochaetes with 21.2 and 23.3% abundance, respectively. In the foregut and midgut regions, the bacteria from 7 different phyla viz., Bacteroidetes, Spirochaetes, Cyanobacteria, Firmicutes, Proteobacteria, Planctomycetes and Actinobacteria comprised over 97.73% and 96.58% of the total prokaryotic communities. However, the hindgut region was dominated by the members of Bacteroidetes, Spirochaetes and Firmicutes together with Verrucomicrobia that accounted for over 94% of the observed bacteria. One astonishing feature was the predominance of Verrucomicrobia in the hindgut where it contributed 4.78%, however its load in the foregut and midgut regions was very low (0.89 and 0.95% respectively). The members of the phylum Planctomycetes showed higher abundance in the foregut among the three gut-regions where it comprised over 2.39% as compared to midgut (0.33%) and hindgut (0.03%). Among the gut regions, the highest number of bacterial phyla were represented in midgut, i.e., 23 phyla followed by foregut that harbored the members of 22 bacterial phylotypes (Figure 7). However, bacteria belonging to only 18 phyla were represented in the hindgut signifying the autochthonous characteristic of some particular phylotypes like Bacteroidetes and Spirochaetes. The other phyla such as Actinobacteria, Verrucomicrobia, Synergistetes, Elusimicrobia and Tenericutes etc. were represented in lower fraction ranging from 0.03 to 4.8% in different regions of the gut of C. formosanus.

A total of bacteria affiliated to 142 families were observed in the gut system of C. formosanus, being highest, i.e., 120 families in the foregut, and then midgut (117 families). Out of the total observed families of bacteria, highest dominance was exhibited by Porphyromonadaceae being 53.5, 40.6 and 28.1% in the hindgut, midgut and foregut, respectively. After Porphyromonadaceae, the second most abundant family was Spirochaetaceae showing, 21.6, 16.5, and 9.0% in hindgut, midgut, and foregut, respectively. Similarly, the abundance of the Lachnospiraceae were higher in the midgut represented by 16.8%. The members of all other families such as Enterobacteriaceae, Moraxellaceae, Streptococcaceae, Caulobacteraceae, Rikenellaceae, etc. were found in lower numbers.

Coptotermes formosanus harbored a considerable diversity of bacteria belonging to 26 different phyla and 352 known genera. The members of the phyla like Deinococcus thermus, Chlamydiae, Candidate phylum_TM6, and Armatimonadetes, etc. were specific to particular gut regions (Supplementary additional file 1). The bacteria belonging to lineages of Candidate_phylum_TM6, and Chlamydiae were uniquely observed in foregut while the members of the bacterial groups such as Deinococcus thermus, and Armatimonadetes were shared by foregut and midgut regions.

Similar to the phyla distribution, the lowest number (80) of bacterial families were observed in the hindgut. At the family level description, the top 10 families including Porphyromonadaceae contributed about 69.3% of the foregut bacteria (Figure 8). The midgut was dominated by Lachnospiraceae (16.8%) and Spirochaetaceae (16.5%) after Porphyromonadaceae (40.6%). Unlike foregut and midgut, the hindgut possessed members of 80 bacterial families being dominated by the Porphyromonadaceae (53.5%) and Spirochaetaceae (21.6%). The most dominant 10 families occupied about 84.9 and 84.7% of the bacterial families in the midgut and hindgut, respectively. The members of the families like Streptococcaceae, Enterobacteriaceae, and Leuconostocaceae were found more in the foregut region while species of Lachnospiraceae were uniquely abundant in the midgut region.

The identified bacteria from all the gut regions were found affiliated with 352 known genera apart from some unclassified species (Figure 9). The top 20 genera and their abundances in the gut-regions are shown in Table 4. At the genus level, Candidatus Azobacteroides was the most dominant throughout the gut system represented by 27.5, 39.8, and 51.7%, respectively, in the foregut, midgut, and hindgut. The Lactobacillaceae represented by Weissella, and Lactobacillus were majorly found in the foregut. Noteworthy, gut cluster 15 was highly abundant (15.8%) in the midgut after Candidatus Azobacteroides. However, the Moraxellaceae represented by Acinetobacter showed slight variations in the abundance between midgut (3.2%) and foregut (3.3%). The members of the genera like Candidatus and Treponema were most dominant in the hindgut showing 1.8 and 1.2-fold abundances when compared with foregut and midgut, respectively (Figure 9).

Among the gut-regions, Ruminicoccaceae was more abundant in hindgut that were classified as Termite group_aaa (OTU_13, OTU_73, OTU_79), Termite cockroach cluster (OTU_134), Gut_cluster_6 (OTU_152), Gut_cluster_7 (OTU_224), and uncultured_12 (OTU_76, OTU_33), Ruminicoccus_2 (OTU_71, OTU_116) (Supplementary additional file 2). The NCBI BLAST analysis for many OTUs from hindgut matched with uncultured Treponema 16S rRNA gene clones isolated from the gut of other termites such as Reticulitermes speratus and Formosan subterranean termite, etc. However, some Treponema represented OTUs showed confidence levels <90%, signifying the presence of possible novel Treponema phylotypes in C. formosanus (Figure 10).

Out of the total OTUs, only 44 OTUs were enriched in individual gut regions. Among these unique OTUs, majority 28 (63.6%) were sheltered in the foregut, while only 2 OTUs (OTU_03, and OTU_94) representing gut cluster_15, and gut cluster_13 from Lachnospiraceae were predominantly enriched in the midgut. The number of the unique OTUs enriched in the hindgut was 14 that were affiliated with Ruminicoccaceae, Termite gut cluster, Rhodocyclaceae, Porphyromonadaceae, Spirochaetaceae, Veillonellaceae besides some uncultured taxa (Figure 10). Among the 26 bacterial families that enriched in the gut system, the LEfSe indicated the enrichment of the Spirochaetaceae, Veillonellaceae, and Rhodocyclaceae in the hindgut while none of the families were enriched in the midgut (Supplementary Figure S3). The highest and most diverse bacteria were enriched in the foregut (23 families) probably due to its immediate contact with the surrounding environment ingested some environmental bacteria with food.

The functional profiling of the metagenomic data revealed shared metabolic signatures of bacterial communities residing in the foregut, midgut, and hindgut of C. formosanus. The bacteria showed a diverse array of symbiotic functions, such as carbohydrate and amino acid metabolisms, detoxification, and degradation of xenobiotics and terpenoids besides many other pivotal roles (Figure 11). The carbohydrate and energy metabolisms were perpetual processes for the gut microbiota system of the lower wood-feeding termite, showing 12–13% and ~7% relative abundances among the gut-regions, respectively (Figure 11). For energy metabolism, the Tukey’s HSD test was significant when the functions of bacteria from hindgut were compared with foregut and midgut regions (p < 0.05). However, the relative abundance of different metabolic modules assigned to the carbohydrate metabolism (Figure 11B) was statistically insignificant among the gut-regions indicating it as a continuous process.

Besides energy and glycan metabolisms, the bacterial communities of the hindgut were chiefly involved in cellular processes like signal transduction and cell motility. Since the foregut acts as the main recipient of the environmental bacteria it showed higher (>4.1%) relative abundance toward detoxification mechanisms like xenobiotic and terpenoid degradation. The relative abundance of lipid metabolism was above 3.8 and 3.7% in the anterior gut (foregut and midgut) when compared to the hindgut which showed a lower amount of sequence reads, i.e., 3.3% (Figure 11). The functional profiles calculated from the KEGG Orthology (KO) database of prokaryotes revealed sequence reads that encode for environmental adaptation were predominant in the hindgut with a relative abundance of 0.0025. The BugBase analysis demonstrated a high prevalence of the gram-positive and biofilm-forming bacteria in the midgut (4.2%), followed by the foregut (3.9%) while the hindgut showed a higher abundance of aerobic, and gram-negative bacteria affiliated with Bacteroidetes (Supplementary Figure S4). The foregut being a major hub for reception of the environmental microorganisms showed predominance of anaerobic/facultative anaerobes and mobile-element-containing bacteria having high stress tolerance potentially oxidative stress imposed by the lignin contents of the diet besides the toxicity caused by some pathogens ingested from the environment. The representation of the gram-positive bacteria was observed least in the hindgut. The major community of the biofilm-forming bacteria was comprised of the members belonging to the phyla of Spirochaetes, Proteobacteria, and Firmicutes from midgut (Supplementary Figure S3E). Likewise, most of the Bacteroidetes members formed anaerobic consortium that were found abundant in the hindgut, however, least in the foregut.

Indeed, our data showed that the composition of symbiotic bacteria in C. formosanus varies significantly concerning the structure as well as gut physiology of the termite. The highest α-diversity was observed in the foregut followed by the hindgut as revealed by Shannon and Simpson indices. This could be attributed primarily to the foregut’s exposure to the environmental bacteria ingested through food (Bignell et al., 2011). The higher diversity of bacterial symbionts in the hindgut is due to its large size and available surface area provided by flagellates, accommodating endo and ecto-symbionts (Smith et al., 2017; Nalepa, 2020). Flagellates also influence the composition of bacterial communities in termite guts (Reid et al., 2014). Several species of Verrucomicrobia, Candidate phylum_SR1, Candidate phylum_BD1-5 besides Bacteroidetes, and Spirochaetes, dominant in the data sets of hindguts are well-known as ecto- and endo- symbionts of flagellates in termites (Hongoh et al., 2007a, 2008a; Sato et al., 2014, 2009). Notably, bacterial lineages like Porphyromonadaceae cluster_V, Spirochaetaceae_Treponema I, and Pseudomonadaceae, identified as endosymbionts of C. formosanus residing within the cells of cellulolytic protist Pseudotrichonympha grassii, have been associated with high redox potential in other termites and Surinam cockroach, Pycnoscelus surinamensis (Hongoh et al., 2008a; Lampert et al., 2019). Additionally, many sequences (OTU_135) from the hindgut represented by bristle-like uncultured bacteria were classified as Candidatus Symbiothrix, a taxon exclusive to termite guts, acting as ectosymbionts of protists coexisting with Spirochaetes (Hongoh et al., 2007a). Although we observed a low abundance of Candidatus Symbiothrix in C. formosanus, it has been reported as highly dominant in other termites, including C. niger (Kohler, 2011; Reid et al., 2014). The observed low abundance of OTUs representing Candidatus Symbiothrix could be attributed to different geographical distributions and evolutionary history, as C. niger is native to neotropics having distinct biogeographical attributes (Chouvenc et al., 2016).

The spatial structure of the bacteria in C. formosanus also showed an abundance of Proteobacteria (28.9%), Actinobacteria (4.5%), and Planctomycetes (2.3%) in the foregut. This could be attributed to their procurement from the environment through wood particles as Proteobacteria is the most diverse and largest bacterial group in the environment (Guimaraes et al., 2020). The predominance of these bacterial phylotypes aligns with previous findings (Utami et al., 2018; Dar et al., 2022), suggesting their symbiosis and functional roles in C. formosanus. These inferences are further evidenced by our previous observations that demonstrated Proteobacteria, as the dominant culturable phylotype in the termite’s gut-regions (Dar et al., 2022). Among the Proteobacteria, members of the Enterobacteriaceae and Moraxellaceae of the class Gamma proteobacteria were well represented across the foregut (7.39 and 6.26%) and midgut (1.9 and 3.46% relative abundance). Similarly, Lachnospiraceae (16.8%), and Caulobacteraceae (2.43%) were more abundant in the midgut as compared to the hindgut among the dominant families of class Clostridia, and Alphaproteobacteria, respectively. The abundances of Enterobacteriaceae and Moraxellaceae have been influenced by factors like low pH and high partial pressure of hydrogen ions in the cockroach gut (Lampert et al., 2019). Majority of the OTUs representing Actinobacteria (OTUs_93, OTU_185, OTU_48, OTU_144, and OTU_195) from Streptomyces lineage, displayed higher abundance in the foregut, reflecting their role in secreting antibiotics and secondary antimicrobial compounds to inhibit incoming pathogens in termite guts (Visser et al., 2012; Arango et al., 2016). The foregut of C. formosanus was also densely colonized by several deep-branching lineages of Planctomycetes (such as Planctomyces_1, Isosphaera, Gemmata), akin to the hindgut of soil feeding higher termite, Cubitermes spp. (Köhler et al., 2008).

The physicochemical microenvironment of the gut has been recognized as a strong selective factor that determines the composition of the gut-region-specific bacterial lineages in termites (Bauer et al., 2015; Mikaelyan et al., 2017). Alterations in intestinal pH induce modifications in microbial composition, reducing or eliminating the pathogenic bacteria sensitive to acidic conditions while favoring acid-resistant microbes (Mroz et al., 2002). The higher presence of the Lactobacillus (OTU_45, OTU_175, and OTU_176) in the foregut corroborates with their role in the putative fermentation of ingested sugars created due to maceration of wood leading to the slightly acidic environment. However, the low amount of Lactobacillus in the midgut and hindgut could be attributed to low concentrations of lactate caused by the high turnover of the compound (Tholen and Brune, 2000). Moreover, the microbial activities in gut compartments give rise to steep radial gradients of oxygen, hydrogen, and potentially other metabolites that in turn determine the community structure of symbiotic bacteria (Kappler and Brune, 2002; Köhler et al., 2012; Ngugi and Brune, 2012). It has been hypothesized that the [FeFe] hydrogenases and putative [FeFe] hydrogen sensors of Treponema primitia, whose homologs are abundantly represented in the metagenomes of Nasutitermes corniger (Warnecke et al., 2007), are not only involved in reductive acetogenesis but also assist spirochetes in locating optimal positions within hydrogen gradients (Ballor et al., 2012).

Another factor that can rapidly change the microbial profile in termite gut is diet (David et al., 2014; Mikaelyan et al., 2015). Progressive changes in the availability and complexity of polysaccharides in lignocellulose during gut passage (Tokuda et al., 2018), together with a longer retention time of digesta in the hindgut, contributes to the relative abundance of specialized taxa like Bacteroidetes (Mikaelyan et al., 2017). Furthermore, variations in the chemical composition of the digesta along the gut passage (e.g., predominant oxidation of lignin in the foregut and midgut) also influence the structure of the microbial communities (Coy et al., 2010). These observations strongly support the notion that niche heterogeneity within microbial environments is a key determinant of community structure (Yang et al., 2005) and aligns with the unique diversity observed in different gut fractions of C. formosanus.

Coptotermes formosanus being a socio-economic pest of wood, efficiently degrades the lignocellulose-based diets in collaboration with gut symbionts including bacteria (Shinzato et al., 2005; Geng et al., 2018b; Dar et al., 2022). Previous studies have reported that Proteobacteria and Actinobacteria, significantly contribute to the lignocellulose digestion in the hindgut of Holotrichia parallela larvae (Huang et al., 2012). Consequently, the predominance of these bacteria in the foregut and hindgut of C. formosanus suggests their potential roles in plant cell wall digestion, thereby supplying energy and nutrients to the host (Huang et al., 2012). Similarly, the occurrence of Bacteroidetes throughout the gut system, with higher abundance in the hindgut of C. formosanus (Figure 12), aligns with their recognized involvement in fermentative metabolism and hydrolysis of plant-derived oligosaccharides (Hongoh et al., 2008b; Ghanbari et al., 2015). The higher abundance of Bacteroidetes in the hindgut also corresponds to their obligately anaerobic nature, underscoring their proficiency in polysaccharide fermentation, generating essential molecules such as acetate, and butyrate, absorbed by the host (Pan et al., 2023). Moreover, Spirochaetes are lauded for complementing reductive acetogenesis and nitrogen fixation activities in termites (Tokuda, 2021). The gram-positive, Leuconostoc is advantageous to the host by fermenting a variety of polysaccharides, mannitol, vitamins K, and bacteriocins, and catalyzes the hydrolysis of α-galactosides (Weymarn et al., 2002; Sybesma et al., 2003; Hemme and Foucaud-Scheunemann, 2004). Recently Leuconostoc pseudomesenteroides has been found to restore intestinal disorders induced by high-fat diets (Sun et al., 2020). The elevated abundance of these phylotypes in the anterior gut of C. formosanus signposts their involvement in lignocellulose digestion and fermentation, alongside potential roles in maintaining intestinal homeostasis. Further, the higher abundance of bacterial lineages like uncultured members of the “Treponema I” clade, Cluster_IV, Dysgonomonas, Candidatus Armantifilum, Candidatus Azobacteroides, Ruminococcus_2, and the Termite_cockroach_cluster, etc. that represent the lignocellulolytic community (Tokuda et al., 2018) throughout the gut system signifies the continuous degradation of the lignocellulose by the termite. Nonetheless, the low abundance or absence of some lineages among the gut-regions could be attributed particularly to the changes in the structure and composition of the processed lignocellulose-based diet besides the microenvironmental conditions and presence of flagellates in the termite gut system (Benjamino et al., 2018).

The gut microbiota not only aids in the digestion of plant fiber but also plays a crucial function in the nitrogen economy and immunity of the termites. Hitherto, numerous nitrogen-fixing bacterial strains have been identified from termite guts, though a majority of them are unculturable. The enormous diversity of nifH genes observed in termite guts stipulates that nitrogen fixation capabilities are attributed to Spirochaetes, Clostridiales, Bacteroidetes, and other gut-bacterial symbionts (Lilburn et al., 2001; Warnecke et al., 2007; Yamada et al., 2007; Hongoh et al., 2008b; Desai and Brune, 2012), predominantly observed in the gut-metagenomes of C. formosanus (Figure 12). Our results are congruent with previous findings indicating that the endosymbiont, Candidatus Azobacteroides pseudotrichonymphae (Noda et al., 2007; Hongoh et al., 2008b) serves as the most significant diazotroph in the gut of C. formosanus, that possess nif genes. The presence of the diazotroph, Candidatus Azobacteroides, and Spirochetes like Treponema spp. throughout the gut system of C. formosanus also signposts its importance in nitrogen metabolism of the termite (Figure 9) as these symbionts are known to express paralogs of nifH genes (Desai and Brune, 2012; Tokuda, 2021). Recent identifications of many nitrogenase-encoding genes from Treponematales and Enterobacteriaceae in higher termite gut metagenomes further substantiate these findings (Arora et al., 2022). Moreover, termite guts contain nitrogenous residues derived from the cell biomass of bacteria (He et al., 2013; Liu et al., 2019), therefore, multiple KOs relevant to protein degradation, amino acid metabolism, cofactors, and vitamin metabolisms were enriched in the metagenomes of termite gut, particularly in the foregut and midgut. The anterior gut is mainly responsible for the elimination of environmental and allochthonous bacteria. Noteworthy, the dominance of Firmicutes in the anterior gut (foregut and midgut) of C. formosanus signifies its possible involvement in protein degradation, maintenance of gut homeostasis, and the development of host immunity (Colston and Jackson, 2016). Furthermore, the higher abundance of biofilm-forming bacteria in the anterior gut regions confers advantages to the termite by fortifying host defenses, augmenting nutrient exchange, and preventing the colonization of pathogens (Tytgat et al., 2019). The unique occurrence of the Deinococcus and Armatimonadetes in the anterior gut (foregut and midgut) could be augmenting the metabolism of terpenoids and polyketides and preventing stress from the secondary metabolites present in the wood (Gerber et al., 2015).