Introduction

Stink bugs under the genus Tessaratoma (Hemiptera: Tessaratomidae), having about 26 species, out of which Tessaratoma javanica (Thunberg), Tessaratoma papillosa (Dury), and Tessaratoma quadrataDistant (Hemiptera: Tessaratomidae) are reported in litchi from different parts of the world and are commonly referred to as litchi stink bugs1. In India, T. javanica is reported as an economically important sucking pest of litchi (Litchi chinensisSonn.)2,3,4. This species of bug is also an important pest of Rambutan (Nephelium lappaceum), Longan (Dimocarpus longan), Pomegranate, (Punica granatum), Kusum (Schleichera oleosa), champak (Michelia champaca), and many trees, including Eucalyptus and mulberry. T. javanicawas reported as an economically important—litchi pest in India and its neighbouring countries, where numerous outbreaks have been documented3,5. T. javanicainfests litchi with the onset of the floral development phase in the last week of February and continues until fruits are ready for harvest. Throughout this duration, gregarious nymphs and adults penetrate and extract sap from floral buds, panicles, juvenile fruit peduncles, and delicate shoots. This feeding reduces the growth and development of flowers and fruits, leading to heavy crop loss in litchi3. Most of the studies related to T. javanicahave been restricted to its biology, distribution, molecular characterization, and management through biological and chemical methods6,7. Diversity and the predicted functions of associated bacterial communities have been investigated in many groups of insects, including stink bugs for planning effective management strategies8,9,10 however, limited studies have been carried out on developmental stage-associated bacterial communities with T. javanica11.

Insects possess diverse microorganisms in their hemocoel, gut, reproductive tracts, cells, and exoskeleton12,13. The connections between bacteria and insects are equally significant for both the insects and microbes associated. These bacteria are universal in insect populations and play a role in immunological response, growth, reproduction, feeding, detoxification, stress protection, and the creation of chemo-signals for the synthesis of semiochemicals and other vital biological activities14,15,16,17. Besides their significance in particular insects, these microbiota influence the overall dynamics of interactions between insect pest species and their host plants18, aid in the development of natural defense systems against predators19, and help in the breakdown of complex molecules to simple nutritional compounds20,21. Certain microbes act as primary symbionts to fulfil the dietary imbalances, whereas secondary symbionts enhance the host’s fitness22; reproductive parasites markedly affect their hosts’ reproductive strategy. Herbivorous insects rely on symbiotic bacteria for the degradation of plant components, including cellulose23. Consequently, insects have gained from this mutualism by enhancing their ability to digest a wider variety of plants, as they receive improved nutrients and assistance in metabolizing harmful flora18,24. The essential involvement of microorganisms in insect growth and ecological processes necessitates a thorough understanding of these intricate relationships to formulate efficient strategies for biological control and pest management12. I microbiota changes with metamorphosis11,25,26,27, leading to changes in molecular and phenotypic characters that provide adaptative advantages to associated insects in any environment28,29.

The present study was conducted to assess the diversity and distribution pattern of bacterial communities across developmental stages of litchi stink bug, T. javanica, an economically and ecologically significant pest of litchi with the hypothesis that it may have influence on growth, development, reproduction, immunity, digestion, and general adaptability of the insect. We examined the bacterial communities associated with T. javanica using comprehensive 16 S rRNA amplicon sequencing and predicted functional activities of associated bacterial communities, which will be a basis for microbial ecosystem-linked novel management strategies.

Materials and methods

Sample collection

All the developmental stages of litchi stink bug, T. javanica, were directly collected from the ‘Shahi’ cultivar of litchi at the ICAR-National Research Centre on Litchi situated at Mushahari, Muzaffarpur district of Bihar state, India (26.2217° N latitude and 85.1194° E longitude; 60 m ASL) during 2023. Randomly collected live 10 days-old eggs, 5 days 1 st and 2nd instar, 10 days 3rd, 4th, and 5 th instar, and 15 days old male and female of T. javanica were placed in the freezer at −20 °C for 30 min for euthanization and then stored in absolute ethanol at −20 °C for further processing. Differentiation between different nymphal stages was done based on morphological characters described by Parveen et al.4. Three replicates were maintained for each developmental stage, viz., egg, 1 st instar nymph, 2nd instar nymph, 3rd instar nymph, 4 th instar nymph, 5 th instar nymph, adult female, and adult male, and DNA was isolated from the entire eggs, and the guts of individual nymph and adult under aseptic conditions in a laminar airflow hood for further processing. Three replicates, each comprising 10 eggs and 5 guts for the developmental stages, viz., egg, 1 st instar nymph, 2nd instar nymph, 3rd instar nymph, 4 th instar nymph, 5 th instar nymph, adult female, and adult males were maintained for DNA extraction. Nymphs and adults were surface sterilized in 0.2% sodium hypochlorite (NaOCl) in 70% ethanol and washed in cold water for 2 min each.

DNA extraction and amplicon sequencing of 16 S rRNA gene

Total DNA was extracted from each samples following the methodology outlined by Choudhary et al.7 with some modifications. The quality, purity, and integrity of the extracted DNA were performed by Qubit fluorometer 4.0.

The DNA samples were PCR-amplified using the primer targeting the V3-V4 hypervariable region of the 16S rRNA gene of bacteria (Forward primer 314 F = 5’ CCTACGGGNGGCWGCAG 3’; reverse primer 805R = 5’ GACTACHVGGGTATCTAATCC 3’)30. The PCR was performed with the following conditions: an initial denaturation phase at 98 oC for 30 s, followed by 27 cycles consisting of a 10 s denaturation at 98 oC, annealing at 55 oC for 30 s, extension at 72 oC for 30 s, and a final extension step for 10 min at 72 oC. For 16 S rRNA based V3-V4 sequencing, the DNA was extracted using KAPA NGS DNA Extraction Kit and the library was prepared using Xploregen 16 S rRNA v3-v4 library preparation kit. The sequencing protocol utilized for the paired-end run of DNA samples, with a read length of 300 base pairs, was performed using the MiSeq Reagent kit V3, which offers 600 cycles. The sequence read archives (SRA) obtained in the study have been deposited in the NCBI SRA database under the accession number PRJNA1142079 (https://www.ncbi.nlm.nih.gov/sra/.PRJNA1142079).

Sequencing and statistical analysis of data

The quality of raw sequences was assessed using FastQC31, which provided an overview of GC content, base quality, and sequence composition. TrimGalore32 was used to remove low-quality bases (Phred score < 20), adapter sequences, and primers from the 3ʹ end of the reads. Illumina adapters at the 3’ end of the sequence read were removed using Cutadapt33. After filtering, reads shorter than 50 bp were discarded. The high-quality reads were imported into QIIME2 v. 2024.5.0 for bioinformatics analyses34. Raw sequence data were filtered with q2-dada2plugin to eliminate any sequences containing phiX reads and chimeric sequences35. Subsequently, the sequences were denoised and combined using the dada2 method to predict Amplicon Sequence Variants (ASV)31. To maintain the quality of sequences, forward reads were shortened to a length of 290 bp, while the reverse reads were shortened to a length of 270 bp from the 3’-end, and 20 base pairs of reads were eliminated from the 5’-end using the dada2 denoise-single method trimmed off. Removed positions were chosen based on visual inspection of plotted quality scores from demultiplexed reads. The Naive Bayes classifier implemented in QIIME2 was utilized to assign taxonomy to the Amplicon Sequence Variants (ASV). All ASVs were aligned with the mafft program (via q2-alignment plugin), SEPP, implemented in the q2-fragment-insertionplugin in QIIME2, was used to construct the phylogeny tree36. The trained classifier aligned with feature-classifier classifysklearn(threshold value 80%) against the Greengenes2 database v. 202237 on the Illumina 16 rRNA gene primers targeting the V3-V4 region.

The downstream analyses of generated biome files were performed in the MicrobiomeAnalyst, a user-friendly web-based platform (https://www.microbiomeanalyst.ca/)38, Statistical Analysis of Metagenomic Profiles (STAMP) software (Parks et al., 2014), and in R (https://cran.r-project.org/bin/windows/base/). Default parameters of MicrobiomeAnalyst, such as ASV with a minimum count of four (4) per library, low count filter based on 20% prevalence, and rarefied to the minimum library size (1155 reads) across the samples, were kept before any statistical comparison. Alpha diversity indices such as Ace (Abundance-based Coverage Estimator), Chao1, Shannon, and Simpson, were calculated using the Bray-Curtis dissimilarity index. The Kruskal-Wallis H test, in SPSS (version 22.0), was conducted to assess the significant variations across the samples in alpha diversity indices. The box plots of diversity indices were created using the vegan package implemented in R. The beta diversity was investigated through Principal Coordinate Analysis (PCoA) using the Bray-Curtis dissimilarities index39. The relative abundances of top hierarchy levels were represented by bar blots using R, and grouping among samples was performed by SPSS (version 22.0) with a one-way ANOVA analysis followed by Tukey’s test (p ≤ 0.05). A cluster dendrogram was constructed using the hclust function from the stats package in R software (version 4.4.1) to identify patterns among the microbial communities associated with different developmental stages of T. javanica (Fig. 1). The heatmap visualizes data as a matrix where each cell’s color corresponds to its value. The warmer colors (e.g., red) indicate higher values, and cooler colors (e.g., green) represent lower values. While the dendrograms showed hierarchical relationships among clusters. They were displayed alongside the rows and/or columns of the heatmap to illustrate how data points were grouped. Rows and columns are reordered based on the hierarchical clustering results to place similar items closer together, enhancing interpretability.

The dissimilarities in the makeup of bacterial communities between developmental stages were assessed using linear discriminant analysis (LDA) effect size (LEfSe) in R40. The co-occurrence relationship among phyla was established using Pearson correlation coefficient (SparCC) analysis with a correlation threshold value of 0.3 and p ≤0.05 and visualized in Gephi (Version 0.10.1)38. Bipartite analysis and modular bipartite network were prepared to illustrate the relationships among bacterial genera associated with T. javanica41,42. To analyse the gut microbiome data at the genus level, bipartite networks were constructed using R with a R/igraph package for constructing and analysing the bipartite network and Gephi for visualizing the network structure. The top 20 genera identified from the gut samples were represented as nodes, while their associations with other microbial taxa were represented as edges. During Network Construction, each Node represented Each unique genus identified from the sequencing data, while Edges represented significant correlations between genera based on co-occurrence patterns (Spearman correlation coefficient ≥ 0.75). Finally, the networks were visualized to identify key genera and their interactions within the gut microbiome. The functional prediction of 16 S rRNA data for each sample was predicted using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States2 (PICRUSt2) plugin for QIIME2. The ASV matrix and ASV sequences were utilized as input. The 16 S rRNA gene copy number and reference genome database were utilized in the background of PICRUSt2 to acquire predictions of Enzymes (EC) and Pathways (Metcyc database). The Kyoto KEGG pathway abundances43,44,45 from the predicted KEGG ORTHOLOGY (KO) abundances were made following the default protocols of the PICRUSt2 GitHub page (https://github.com/picrust/picrust2/wiki). The NSTI score, which represents the distance to the nearest reference genome in the phylogeny, was utilized to identify the reference genome for PICRUSt2. Functional pathways were compared using Welch’s t-test in STAMP46. The disparities in routes were noted with a false discovery rate (FDR)-adjusted p-value of less than 0.05.

Results

16 S rRNA amplicon sequencing profile

The rarefaction curve between sequencing depth and the number of ASVs detected showed the sequencing depth was sufficient for further analyses while reaching the plateau of curves. A total of 621 ASVs were observed in all the developmental stages. The maximum number (182) of ASVs were produced in the fifth instar nymph of T. javanica, while the first instar nymph produced the least number of ASVs (81). The variations in ASV counts between developmental stages indicate the complexity of the microbial communities associated with T. javanica.The cluster dendrogram illustrates the grouping of microbial communities into four distinct clusters (Fig. 1) each cluster represents a group with similar microbial communities associated. First nymphal instar was grouped with 3rd nymphal instar, while the egg was grouped with 4 th instar. The 2nd instar exhibits more resemblance to the adult female of T. javanica, whereas the 5 th instar shows higher similarity to the adult male, resulting in their respective classifications.

Fig. 1
figure 1

A UPGMA clustering and heatmaps showing the relative abundance of 16 S rRNA gene amplicon sequence variants (ASVs) between different developmental stages. The colors indicate relative abundance ranging from green (lower abundance) to red (higher abundance) (indicated on a scale of 0 to 8).

Full size image

The taxonomic analysis of bacterial communities of different developmental stages (egg, 1–5 th instar nymphs, adult female, and adult males) of T. javanica from phylum to species level showed significant differences. A total of 14 phyla, 24 classes, 69 orders, 89 families, 134 genera and 192 species were characterized among all the developmental stages. The predominant bacterial phyla associated with T. javanica comprised Proteobacteria, Firmicutes, Bacteroidetes, Actinobacteria, Patescibacteria, and Nitrospirota (Fig. 2). The dominance of Proteobacteria was more in the 1 st, 4 th instar nymph, and adult female, while Firmicutes were dominant in the 3rd and 5 th nymphal instar of T. javanica (F = 0.58; p < 0.001). Actinobacteria dominated the egg, 3rd and 4 th nymphal instars. Further, Armatimonadota was found to dominate in 3rd, 4 th and 5 th nymphal instar; Chloroflexota, Fusobacteriota and Patescibacteria dominated the egg and 5 th nymphal instar.

Fig. 2
figure 2

Relative abundance of bacterial diversity at phylum level across different developmental stages of litchi stink bug, T. javanica. Taxa with (*) marks are statistically differed among different developmental stages (Kruskal-Wallis test; p ≤ 0.05).

Full size image

Class gammaproteobacteria dominated among different bacterial communities, followed by bacilli, alphaproteobacteria, and bacteriodia across all developmental stages of T. javanica. Gammaproteobacteria was significantly higher in 1 st, 4 th nymphal instar and adult females, while, in 3rd, 4 th, and 5 th nymphal instar, the Bacilli was found to be significantly higher (F = 0.93; p < 0.001) (Supplementary Fig. 1). Alphaproteobacteria dominated the gut bacterial community of egg and adult females of T. javanica, while bacteroidia and actinomycetia were found to be dominant in the 2nd nymphal instar.

Pseudomonadales, lactobacillales, staphylococcales, enterobacterales, pseudomonadales, aneurinibacillales, rhizobiales, bacillales, and xanthomonadales were identified as the most prevalent bacterial orders during distinct developmental stages of T. javanica (Supplementary Fig. 2). Pseudomonadales, enterobacterales were found to dominate the gut of adult females and 1 st instar respectively, while lactobacillales and staphylococcales were prevalent in the gut of 5 th nymphal instar of T. javanica. Bacteria from family pseudomonadaceae was dominant in adult males and females, 4 th instar and egg stages, while enterobacteriaceae in 1 st and 3rd instar nymphs and lactobacillaceae in the 5 th instar nymphs (Supplementary Fig. 3).

Genera Pseudomonas was the most prevalent in eggs (16.51 ± 0.30), followed by first (6.66 ± 0.17), second (15.08 ± 0.40), fourth (21.31 ± 0.50), adult male (21.47 ± 0.11), and females (24.44 ± 0.25a) (Table 1). It was observed that Ligilactobacillus (16.42 ± 0.32), Acinetobacter (6.72 ± 0.09), Sinorhizobium (4.77 ± 0.14) and Bacillus (3.79 ± 0.13) had higher dominance in 5 th instar nymphs. Ligilactobacillus apodemi, Staphylococcus xylosus, and Pseudomonas furukawaii were found to predominate in the gut of T. javanica. In the 1 st, 2nd, and 4 th nymphal stages and females of T. javanica, P. furukawaii was the most dominant bacterial species. L. apodemi had higher dominance in the 5 th instar (16.51 ± 0.25) followed by the egg (14.18 ± 0.15), male adult (13.47 ± 0.19) and 4 th instar (13.13 ± 0.44) (Supplementary Table 1).

Table 1 Top 10 genus-level relative abundance of bacterial communities across different developmental stages of litchi stink bug, T. Javanica.
Full size table

Diversity of bacteria across the developmental stages of T. javanica

The Ace and Chao1 indices were observed to be highest in the 5 th nymphal istar and the lowest in the 1 st nymphal instar (Table 2). Significantly higher values of Simpson and Shannon diversity indices were observed in the egg stage and the lowest was found in the 1 st nymphal instar (Table 2). The Bray-Curtis dissimilarity method result indicated that bacterial communities of the egg stage, and adult male stages, overlapped together, which suggests that bacterial communities shared among these developmental stages, while, other stages were clustered separately, indicating distinct bacterial communities associated with them (ANOSIM, R = 0.92, p < 0.001) (Fig. 3).

Fig. 3
figure 3

Principal coordinate analysis (PCoA) plot visualizes the beta diversity based on the Bray-Curtis dissimilarity index among developmental stages of the litchi stink bug, T. javanica (ANOSIM, R = 0.92, p < 0.001). Ellipses indicate clusters of developmental stages which are classified based on principal component analysis (PCA) score.

Full size image
Table 2 Alpha diversity indices provide insights into the species richness and diversity indices of bacterial communities associated with different developmental and growth stages of litchi stink bug, T. Javanica.
Full size table

Thirty clades were consistently linked to the developmental stages of T. javanica as determined by the LEfSe analysis (Fig. 4). The phylum Proteobacteria was markedly elevated in all embryonic stages. The dissimilarities from bacterial phylum to genus level were noted and the LDA score exceeded 2 (Fig. 4). The bacterial taxa at the phylum, class, order, family, and genus levels exhibited significant differences across each developmental stage of T. javanica. The co-occurrence network suggested that Bacteroidia was positively and strongly correlated with Fusobacteria but had a strong negative association with Fimbrimonadia. Further, Methanobacteria, Paceibacteria, Gracilibacteria and Verrucomicrobia had strong positive associations with each other (Supplementary Fig. 4).

Fig. 4
figure 4

LEfSe analysis (A) to identify and characterize bacterial communities from phylum to genus level associated with developmental stages of the litchi stink bug, T. javanica. Red nodes indicate no significant difference. The nodes with different colours denote that the bacterial community significantly increased at different developmental stages. LDA (Linear discriminate analysis) (B) score was greater than 2 in most bacterial taxa and indicated family-level diversity in bacterial communities across developmental stages of T. javanica.

Full size image

Bipartite networks illustrate that the genus Pseudomonas was the most abundant across all life stages, particularly in adults (Fig. 5) and Ligilactobacillus showed relatively high abundance in the egg. Acinetobacter, Staphylococcus, and Bacillus were abundant in other developmental stages (Fig. 6). Modular bipartite network as analysed by a bipartite network of the corresponding bacterial communities across various life cycle phases of the insect elucidated the intricate interactions among bacterial genus associated with the litchi stink bug (T. javanica) (Fig. 6). This analysis observed the richness and evenness of bacterial communities. Pseudomonas dominated in all developmental stages while Ligilactobacillus in egg, 4 th, 5 th instar, male and female and Staphylococcus, Kluyvera were dominant in the 3rd nymphal instar of T. javanica.

Fig. 5
figure 5

Functional prediction on KEGG ORTHOLOGY analysis of bacteria associated with adult females compared with other developmental stages of litchi stink bug, T. javanica with p-value < 0.05.

Full size image
Fig. 6
figure 6

Bipartite (A) networks illustrate the relationships across 20 dominated bacterial genera associated with different developmental stages of the litchi stink bug, Tessaratoma javanica (Photos were taken from ICAR-National Research Centre on Litchi orchard, Muzaffarpur, Bihar, India). The upper nodes (colour) denote the developmental stages of T. javanica while the lower node (black) denotes the bacterial genera. The length and width of each node are scaled to the total number of interactions for each developmental stage. Modular bipartite (B) matrix of identified modules based on the bipartite network analysis of shared bacterial genera among different developmental stages of T. javanica. The intensity of the blue colour in each box indicates the number of interactions identified between the modules.

Full size image

A total of 9789 KEGG Orthology (KO) groupings were predicted. The pathway predicted that bacteria participate in critical biological processes such as amino acid biosynthesis and catabolism, glycerol and glucose metabolism, energy metabolism, membrane transport, DNA replication and repair, as well as translation and transcription. The results revealed that the bacterial population during the egg stage of T. javanica are more involved in metabolic processes related to fundamental requirements for initial development (Supplementary Fig. 5). Conversely, bacterial communities are plentiful and metabolically active in the synthesis of fatty acids and amino acids at distinct nymphal stages characterized by elevated growth rates (Supplementary Fig. 6). Furthermore, in adult females, bacterial participation escalates in the production of fatty acids, amino acids, and glucose (Fig. 5). Conversely, adult males exhibit increased engagement in the pentose phosphate pathway and Diacylglycerol production, indicating that energy metabolism and lipid synthesis are crucial for activity and reproductive functions (Supplementary Fig. 7).

Discussion

The present findings underscore the variability in microbial communities associated with different developmental stages of the litchi stink bug, T. javanicasuggesting that as the nymphs mature, their gut microbiota becomes more diverse and potentially more specialized. This complexity may be influenced by factors such as diet, habitat, and physiological changes that occur during development44,45,46. The significant differences in ASV counts across developmental stages highlight the dynamic nature of microbial associations in this pest species and could have implications for understanding pest management strategies and ecological interactions within their environment.The bacterial communities present within insect guts significantly influence the growth and development of host organisms, particularly in holometabolous and hemimetabolous insects like T. javanica. This necessitates substantial alterations in the structure that lead to modifications in the microbial profile from the egg to the adult stage. Proteobacteria was identified as the predominant bacterial phylum inhabiting T. javanicaacross all developmental stages47,48,49, followed by Firmicutes. This pattern aligns with other hemipteran species, including T. papillosa and Adelphocoris suturalisindicating that the microbial interactions among these species are analogous48,50. Variations in microbial composition during different developmental stages of T. javanica, encompassing the egg, nymphal, and adult stages are also noted11which is due to food factors, environmental conditions and physiological changes that occur before, during, and after metamorphosis. The comparable bacterial load in male adults and eggs may suggest vertical transmission of bacteria, as observed in other insects51. The commensal bacteria associated with T. javanicaare deemed to play a role in many activities, including nutrient consumption, toxin removal and immunomodulation46. These functions are crucial for the insect’s survival and adaptation to prevailing conditions52,53. Proteobacteria and Firmicutes were identified in all embryonic stages and considered crucial to the ecological characteristics of T. javanica. The interactions between microbiota and hosts are intricate and sensitive54. Furthermore, T. javanicaeggs are dominated by Proteobacteria, which might help to adapt to their environment. Erwinia and Enterobacteriaceae might benefit host development by breaking down polysaccharides and fixing nitrogen55. Bacteria in plants help with nutrient intake, growth, and development. They also influence the microbes on insects associated with those plants, showing complex relationships in these systems.

Proteobacteria are the most common bacteria in T. javanica at all stages, especially nymphs. Other common bacteria are Bacteroidetes, Actinobacteria, and Cyanobacteria. Erwinia is prevalent in all T. javanica stages. These bacteria work together to support T. javanica growth and development, helping to use plant nutrients and co-evolve with its host plants. The presence of Erwinia suggests that this genus may play a vital role in the life cycle of T. javanica. Further research is necessary to comprehend the specifics of Erwinia activity and its position within the host insect microbiome. Previous research suggests that symbiotic bacteria may contribute to pesticide resistance by facilitating the biodegradation of pesticides in insect pests. Numerous data indicate that Pseudomonas species, including P. putida, Pseudoxanthomonas indica, Pseudomonas sp., and Pseudomonas sp.RPT 52, could metabolize imidacloprid56. The understanding of Proteobacteria community evolution during insect development including litchi stink bugs improves through comparisons with model insects such as Spodoptera frugiperda and Drosophila melanogaster. The bacterial community of S. frugiperda consists mainly of Proteobacteria and Firmicutes from the larval to pupal to adult stages but eggs contain diverse Ralstonia and Sediminibacteriumspecies along with other bacteria57,58. The changes in microbial communities follow metabolic needs that include nutrient absorption during feeding periods and energy consumption for the reproduction and migration activities of adults. The gut microbiome in S. frugiperda and Cnaphalocrosis medinalis contains Enterococcus (Firmicutes) which both breaks down plant toxins and facilitates digestion according to Yang et al.59. The gut segments (foregut, midgut, hindgut) of Protohermes xanthodesexhibit different microbial compositions that support carbohydrate metabolism through the phosphotransferase system in its larval stage60. Proteobacteria serves two essential functions in the host by synthesizing nutrients and fixing nitrogen while also detoxifying plant substances. The digestion process and energy absorption in S. frugiperda is facilitated by Enterobacter. The gut homeostasis and dietary adaptation capabilities of arthropods are supported by Proteobacteria which maintain stable metabolic functions. The bacterial populations in D. melanogaster surface microbiomes rise after eclosion to help prevent fungal infections through Gilliamellaspecies61. The litchi stink bug (Tessaratoma papillosa) presents the most diverse bacterial community which contains Pantoea(Proteobacteria) as its dominant member while its bacterial communities progress through three stages from eggs to early nymphs (instars 1–3) to late nymphs/adults and these bacterial communities in later stages enable nutrient synthesis and detoxification processes necessary for litchi and longan plant feeding62. The process demonstrates how microorganisms adjust their composition according to environmental forces while developing. The microbial community structure of insects develops based on their diet of insects and their host range and their specific physiological requirements during each life stage. Each stage of an organism requires unique microbial activities to support its development. The detoxification and plant material digestion requires microbial assistance for larvae while adult insects need microbial support to perform energy-intensive reproductive activities.

Ligilactobacillusmay participate in the fermentation of complex polysaccharides, aiding insects in nutrient extraction from their meals63. This is particularly true for sap-feeding insects belonging to Homoptera. Some Staphylococcus bacteria can change insect immune responses64,65, affecting their health and disease resistance66,67,68,69. Several studies have examined variations in gut microbiota throughout different developmental stages of the host. Sudakaran et al.70. found that Pyrrhocoris apterus has a very stable midgut bacterial community with six main species that stay the same throughout its life. Conversely, in Bombus pascuorum, the gut bacteria of larvae and adults are different71. The predominant phyla in sensitive populations devoid of Beauveria brongniartii infection have been identified as Proteobacteria, Bacteroidetes, and Firmicutes. Nonetheless, uninfected resistant populations are primarily composed of Proteobacteria and Actinobacteria. Within 1–3 days of B. brongniartiiinfection, Proteobacteria and Actinomyces emerged as the predominant taxa in the two populations, with Proteobacteria exhibiting a particularly substantial dominance, indicating their potential significance in either the avoidance or detoxification of pesticides56,72. Firmicutes and Proteobacteria were prominent in the midgut of black soldier fly larvae fed with fish, whereas Bacteroidetes were dominating in the midgut of BSFL fed with a regular dipteran diet or a mixed vegetable diet73. Thus, nutrition influenced distribution of microorganisms in the insect gut, which serves various purposes at several developmental stages52. In T. javanicaat various developmental phases, there was a difference in gut bacteria, each serving distinct tasks according to the specific stages in which they were active11. Bacteria of family Enterobacteriaceae under the class Gammaproteobacteria and order Enterobacteriales are more prevalent in the first nymphal stage of stink bugs11. The enterobacterial community is essential for medfly (Ceratitis capitata) nitrogen and carbon metabolism, development, and reproductive success74; it also aids in safeguarding the host from harmful bacteria. The predominant species in wild and laboratory medfly populations were identified as Klebsiella oxytoca and Enterobacter agglomerans, respectively, while gammaproteobacteria is a crucial symbiont necessary for the fitness of Hemiptera hosts75. Recently, Kashkouli et al.76 elucidated the mutualistic relationship between the pistachio stinkbug, Acrosternum heegeri, and a gammaproteobacterium in the posterior midgut crypts. No notable difference was observed in the roaming activity of the initial nymphal stages, but the insects devoid of symbionts exhibited stunted development and decreased survival rates74. The microbial composition of the adult and egg stages is more analogous than that of the larval and pupal stages9,77,78. There exists a significant congruence between the bacteria present in adult males and the eggs of T. javanica, potentially attributable to factors such as the insects incomplete metamorphosis, horizontal transmission of microbes from the mother, and variations in the microenvironments from which the microbes originate, as well as the roles these microbes play in the insect’s health and development. The egg stage exhibited greater bacterial richness, so corroborating prior study conducted of T. papillosaand other hemipteran insects11,79,80. Consequently, a significant bacterial variety during the egg stage may be a prevalent occurrence in direct field-isolated samples, as demonstrated in the current investigation. Eggs may exhibit greater environmental dependence and complexity than the host feeding phases, as the egg stage interacts with surrounding environments, including soil and plant nectar81. The allocation of high egg stage variability among females on the egg surface is essential for providing sustenance to freshly emerging larvae and facilitating their complete growth that is less mobile82. The bacterial burden in the second and fourth nymphal stages was approximately equivalent, while the bacterial density in the adult stage exceeded that of the nymphal stage. However, in previous studies by Xue et al.50, during the transition period from fifth instar nymph to adult in Adelphocoris suturalis, there was no significant change in alpha diversity, because A. suturalis is an incomplete metamorphosis insect. Nymphs and adults have small changes in their living environment, feeding habits, and food sources, which lead to changes in the intestinal bacterial community.

The nymphs and male adult bacterial communities exhibit similarity and are clustered together, as demonstrated in other species of the litchi stink bug, T. papillosa83. For instance, different studies9,26,50,62,84,85,86,87,88have delineated variations in bacterial composition across different stages of the insect life cycle. Conversely, Gupta and Nair18demonstrated that bacterial richness is greater in different stages of other insects. The makeup of bacterial communities in various developmental stages is contingent upon the insect species and the habitat or food sources accessible to that specific insect. The diversity of intestinal bacteria in nymphs and adults of insects exhibiting incomplete metamorphosis may be less distinct89. Variations in environment, feeding behaviour, and food can elucidate the disparities in bacterial concentrations between nymphs and adults. These variables result in substantial alterations in the gut bacterial composition90,91. The ongoing discourse pertains to two sets of experiments conducted by Hu et al.92 and Xue et al.50 found that Female adults exhibited more species abundance than males, hence, the differentiation in the clusters is somewhat distinct. The current study hypothesizes that females possess a larger and more diversified gut bacterial population than males due to their greater versatility and elevated reproductive capacity. This suggests that females may require additional resources for reproduction and ovarian function. Thus, dietary modifications are significantly associated with the dynamic variations of bacterial communities in insects, as indicated by Li et al.93. The facts reported herein elucidate the intricate symbiotic link between insects and their microbial partners within the gut. Examining the bacterial functional profiles of several stink bug species may uncover analogous trends or adaptations related to their gastrointestinal systems. Microorganisms provide several functions in the metabolic processes of insects.

The predominant bacterial communities play a crucial role in numerous metabolic pathways94, including Fatty Acid Biosynthesis with beta-oxidation95, fatty acid elongation, and oleate biosynthesis14; amino acid biosynthesis (production of essential amino acids such as L-lysine, isoleucine, L-serine, L-valine, and glycine)29, ribonucleotide biosynthesis67,91, respiration, fermentation of pyruvate to isobutanol96, and Diacylglycerol biosynthesis50,95,97. The functional study indicated that amino acid metabolism, production of various secondary metabolites, and carbohydrate metabolism were much more enriched during the egg stage compared to other stages. This discovery demonstrates that metabolic activity is crucial for an insects survival from the egg stage to the larval stage. Nonetheless, not all bacteria residing within a host are necessarily beneficial59.

Current study can generate scientific interest and augment understanding of microbial existence and interactions with hosts and other organisms. This research will aid in formulating novel techniques for pest eradication that utilize microorganisms, including symbiont-mediated RNA interference and paratransgenesis98. Nevertheless, this current study has many drawbacks, as detailed below. The sequences generated by the next-generation sequencing platform in this study cannot be classified at the bacterial species level, necessitating further investigation with enhanced gut bacterial profiling through advanced informatics and superior sequencing technologies such as PacBio SMRT sequencing systems99. The bacterial community analysis was conducted at the genus level due to limitations of the sequencing platform; expected functions of the bacterial community were indicated in the classification using PICRUSt2 thus, further research employing metagenomic and metatranscriptomic methodologies will be necessary to uncover potential microbial activities. Symbiotic connections with bacteria at the interdomain level allow insects to inhabit underexplored ecological niches, which underpins their evolution83,100.

Conclusions

The results demonstrate that bacterial communities’ composition and relative abundance are significantly dynamic and vary across different life stages, implying that various bacterial phyla and genera may fulfil distinct roles in specific life stages, including metabolism, nutrition, detoxification, and reproduction. This study introduces the community structure and ecological interactions of symbiotic bacteria throughout the life cycle of T. javanica, which may be offering novel strategies for biological control.