The mosquito strain used in this study was the Ae. albopictus GT strain, derived from the wild-type GUA strain through tetracycline treatment to eliminate Wolbachia infection^[19]. This strain has been mass-reared at the mosquito facility of Sun Yat-sen University (SYSU) since December 2020. The egg collection, hatching, larval rearing, adult rearing and blood-feeding procedures were the same as those described in a previous study^[19]. Adult females were blood-fed using commercially obtained sterile defibrinated sheep blood (CHINOOK, Ararat Biotechnology Co., Guangzhou, China). Rearing conditions were maintained at 27 ± 2 °C, with a 12:12 h light:dark (L:D) photoperiod and 75% ± 10% relative humidity (RH).

Male pupae (24-36-hour-old) were transported from the mosquito facility to the North Campus of SYSU in a refrigerated vehicle within 2 h. The pupae were irradiated using a RadSource 2000 Pro X-ray irradiator (160 kV, 25 mA) at an absorbed dose of 78 Gy (95% CI: 74.1-81.9 Gy). The irradiation was performed without copper filtration. This absorbed dose exceeds the fully sterilizing dose reported for Ae. albopictus males (30 Gy in our conditions). After irradiation, male pupae were allowed to emerge, and adults were maintained on a 10% glucose solution sterilized by autoclaving (121 °C, 20 min). Non-irradiated males served as controls. All mosquitoes were maintained at 28 ± 1 °C, 70% ± 10% RH, under a 12:12 (L:D) cycle. Over 99.0% sterility was induced between irradiated males and normal females under such exposure conditions as previously reported^[21].

To evaluate the effect of irradiation on adult emergence, 30 irradiated or control pupae were placed in small plastic cups inside a rearing cage. After 48 h, emerged adults (fully escaped from the exuviae) were counted. Individuals that failed to emerge or died during emergence were considered as non-viable. Six replicates were performed for each group.

To assess flight ability, 100 anesthetized male mosquitoes from each group were randomly selected and transferred to a sterile Petri dish. Mosquitoes were anesthetized by placing each group in a 4 °C cold chamber; within 30 min, individuals were gently transferred into Petri dishes using a fine brush. After removal from the cold chamber, mosquitoes were maintained at room temperature (the flight assay room) for 2 h to allow full recovery prior to testing. A transparent cylindrical flight chamber (10 cm diameter, 80 cm height, 2 mm wall thickness) was placed over the Petri dish, and mosquitoes were allowed to escape for 6 minutes at room temperature. Further details of the flight ability assay are provided in supplementary materials [Supplementary Figure 1]. Flight ability (%) was calculated as the percentage of mosquitoes that successfully escaped from the flight chamber: Flight ability (%) = [100 - (m + n)] / 100 × 100%, where m and n represent the numbers of mosquitoes remaining in the dish and on the chamber wall, respectively. Eighteen replicates were performed per group.

To measure male longevity, 30 irradiated or control males were initially placed in separate cages, with three replicate cages per treatment group (90 males per group). Mortality was recorded daily, and dead mosquitoes were removed from the cages during each observation. During routine daily handling, a small number of live mosquitoes were inadvertently lost from the cages. Because the exact time of loss could not be determined retrospectively, these individuals were not assigned artificial death or censoring times. The numbers of mosquitoes with traceable outcomes were therefore slightly lower than the initial sample size. Two feeding regimes were used: (1) 10% glucose solution (standard condition) and (2) sterilized distilled water (stress condition), with three replicates per group. To account for the potential non-independence of mosquitoes within the same cage, an additional cage-level sensitivity analysis was performed using replicate cages as the experimental unit.

16S rRNA gene sequencing was used to compare the gut microbiota between irradiated and control males. Samples were collected from 40 individuals, including pupae and 1-, 4-, and 7-day-old adults, for both irradiated and non-irradiated treatments. Each group consisted of five biological replicates, with each replicate comprising 20 dissected guts. Adults were collected using an aspirator, anesthetized with CO₂, and maintained on a 10% sugar solution. After collection, the mosquitoes were surface-disinfected by dipping in 70% ethanol (1 min), rinsed in sterile 1 × phosphate-buffered saline (PBS), and stored at -80°C prior to DNA extraction. Gut dissections were performed using sterilized instruments under aseptic conditions to minimize potential contamination during sample preparation. Sterile consumables were used throughout sample processing.

DNA was extracted using the QIAamp Fast DNA Stool Mini Kit (Qiagen, USA) and purified with the DNA Gel Extraction Kit (Axygen, USA) according to the manufacturer’s protocols. The V3-V4 regions of the 16S rRNA gene were amplified using primers 338F (ACTCCTACGGGAGGCAGCA) and 806R (GGACTACHVGGGTWTCTAAT). Amplicons were pooled in equimolar concentrations and paired-end sequenced on an Illumina NextSeq 2000 platform (Illumina, San Diego, USA) by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China).

Raw FASTQ files were demultiplexed using a custom Perl script, quality-filtered with fastp (version 0.19.6), and merged using FLASH (version 1.2.7). These analyses were performed via the above-mentioned platform of Majorbio. Details on quality filtering were as follows: (i) reads truncated when the average quality score dropped below Q20 over a 50 bp sliding window; (ii) only overlapping sequences longer than 10 bp were assembled based on their overlapping region, with a maximum mismatch ratio of 0.2; (iii) samples were distinguished based on barcodes and primers, with exact barcode matching and a 2-nucleotide mismatch allowed in primer matching. The optimized sequences were clustered into operational taxonomic units (OTUs) at a 97% sequence similarity using UPARSE version 7.1. Taxonomic classification of each OTU representative sequence was performed using the RDP Classifier (version 2.2) against the Silva v138 database at a 0.7 confidence threshold. Functional predictions were made using PICRUSt2 (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States)^[22].

To evaluate irradiation effects on the cultivable gut-associated bacterial species, the entire gut (from foregut to hindgut) was dissected and collected from males at 4 stages: pupae (24-36 h post-pupae) and adults at 1, 4, and 7 days post-emergence. Three replicates were performed for each age group, with 20 guts per replicate. Samples were homogenized and plated on selective media: Columbia blood agar (CBA), nutrient broth agar (NA), LB agar (LB), tryptic soy broth agar (TSA), and brain-heart infusion broth agar (BHI). Bacterial colonies were identified by MALDI-TOF MS (Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry). For species that could not be identified by MALDI-TOF MS, sequencing was performed using universal primers: 27F (AGAGTTTGATCMTGGCTCAG), 1492R (GGTTACCTTGTTACGACTT), 909F (ACTCAAAKGAATWGACGG), and 1391R (GACGGGCGGTGWGTRCA).

DNA was extracted from the cultured bacterial colonies using the FastPure Cell/Tissue DNA Isolation Mini Kit (Vazyme, China). Concentrations were measured using a NanoDrop 3000 spectrophotometer and standardized to > 15 ng/μL. PCR reactions (25 μL) included: 12.5 μL of 2×Rapid Taq Master Mix, 1 μL of primers (10 μmol/L), and 1 μL of DNA template. Thermocycling: 95 °C for 3 min, followed by 30 cycles of 15 s at 95 °C, 15 s at 55/52 °C, 5 s at 72 °C, and a final extension at 72 °C for 5 min. PCR products were visualized on a 1% agarose gel stained with 1 μg/mL ethidium bromide.

qPCR was performed to validate the relative abundance of eight gut bacterial taxa that were both significantly altered in the 16S rRNA sequencing analysis and successfully isolated through culture-based methods. These taxa included eight genera: Delftia, Achromobacter, Stenotrophomonas, Leucobacter, Pseudomonas, Deinococcus, Comamonas, and Sphingobacterium.

Total DNA was extracted from dissected gut samples using a DNA extraction Kit (Vazyme Biotech Co., Ltd., Nanjing, China; Cat. No. DC102) according to the manufacturer’s instructions. The Ae. albopictus ribosomal protein S6 (rpS6) gene served as the internal reference. Primer sequences for the eight bacterial taxa are presented in Supplementary Table 1.

qPCR amplification was performed using SYBR Green Supermix (Vazyme, Nanjing, China) on a LightCycler 96 real-time PCR system (Roche, Basel, Switzerland). Each 20 μL reaction contained 10 μL of 2×Supermix, 2 μL of DNA template, 0.4 μL of each primer (10 μmol/L), and 7.2 μL of nuclease-free water. The thermocycling protocol consisted of an initial denaturation at 95 °C for 3 min, followed by 40 cycles of 95 °C for 15 s, primer-specific annealing temperature for 30 s, and 72 °C for 30 s. Fluorescence signals were collected during the extension stage. A melting curve (65-95 °C) was generated at the end of each run to verify amplification specificity. Relative bacterial abundance was calculated using the 2^-∆Ct method. ∆Ct was calculated as the difference between the Cq values of the target bacterial taxon and the internal reference gene, and relative abundance was expressed directly as 2^-ΔCt for each individual sample.

Based on microbial function predictions from 16S rRNA data, five metabolic indicators in mosquitoes were assessed to evaluate irradiation-induced physiological changes, including glutamine synthetase (GS) activity, glucose content, malondialdehyde (MDA) (a marker of lipid peroxidation), protein carbonyl content, and hydrogen peroxide levels. The rationale for selecting these indicators is as follows: glutamine synthetase activity was included because functional prediction suggested differences in glutamine metabolism between irradiated and control groups; glucose content was measured as a key indicator of carbohydrate metabolism and energy supply^[23]; protein carbonyl and MDA were determined as classical markers of oxidative damage to proteins and lipids, respectively^[24,25]; and hydrogen peroxide levels were assessed because previous studies have shown that irradiation can significantly affect reactive oxygen species (ROS) metabolism in insects^[26]. The following kits were used for the metabolic indicator measurement: GS activity (Bioleaper, BSY-BR5000167-100T), MDA (CheKine^TM Micro, Abbkine, KTB1050-96T), glucose content (CheKine^TM Micro, KTB1300-96T), protein carbonyl content (CheKine^TM Micro, Abbkine, KTB1200-96T), and hydrogen peroxide content (CheKine^TM Micro, Abbkine, KTB1041). Measurements were made using a microplate reader (Agilent Technologies, SH1MF-SN) at the wavelengths specified in each protocol.

Bioinformatics analyses of 16S rRNA gene sequencing data were performed using the Majorbio Cloud platform (https://cloud.majorbio.com). OTU-level alpha diversity indices (observed OTUs, Chao1, Shannon index, and Good’s coverage) were computed using Mothur v1.30.1. Beta diversity was assessed via principal coordinate analysis (PCoA) based on weighted UniFrac distance and non-metric multidimensional scaling (NMDS) using Bray-Curtis dissimilarity (Vegan v2.5-3). Group differences were tested with the permutational multivariate analysis of variance (PERMANOVA) test (Vegan v2.5-3). The linear discriminant analysis (LDA) effect size (LEfSe) (http://huttenhower.sph.harvard.edu/LEfSe) was used to identify the differentially abundant taxa (LDA score > 2, P < 0.05, representing statistically significant difference) across groups^[27]. Additional statistical analysis and data visualizations were performed in GraphPad Prism v7.0 and R v3.6.3. Student’s t-test or Wilcoxon rank-sum test was used for two-group comparisons. One-way analysis of variance (ANOVA) with Tukey’s post-hoc test was used for multi-group comparisons. Emergence rate and flight ability were analyzed using generalized linear mixed models (GLMMs) with binomial error distribution, with treatment as a fixed effect and replicate as a random effect. Survival curves were generated using the Kaplan-Meier method and compared with the log-rank test. As a sensitivity analysis, replicate cages were treated as the experimental unit, and cage-level restricted mean survival time (RMST) was compared between treatment groups. The two-tailed Mann-Whitney test was used to compare the relative abundance of eight gut bacterial taxa between irradiated and control groups. Metabolic indicator values between irradiated and control groups were analyzed using the Mann-Whitney test. A significance threshold of P < 0.05 was considered statistically significant.

Irradiation significantly reduced both adult emergence rate and flight ability (GLMMs, P < 0.001 for both comparisons). The emergence rate of irradiated pupae was 78.33% ± 4.0%, significantly lower than 96.67% ± 3.0% in control groups [Figure 1A]. Flight ability was also markedly reduced in irradiated males (35.17% ± 19.32%) compared to controls (85.17% ± 6.89%) [Figure 1B]. Under 10% sugar solution feeding, irradiated males had reduced lifespans compared to control groups (Figure 1C, Log-rank test, χ² = 171.2, P < 0.0001). Notably, while 45.56% of control males survived until the 30th day, irradiated males reached 0% survival by the 23^rd day. Under water-only conditions, survival also differed significantly between irradiated and control males (Figure 1D, χ² = 5.109, P = 0.0238), with a survival pattern distinct from that observed under sugar feeding. A cage-level sensitivity analysis using replicate cages as the experimental unit yielded results consistent with the primary Kaplan-Meier analysis [Supplementary Table 2].

Comparison of alpha diversity metrics (Chao and Shannon indices) revealed no significant differences between irradiated and control groups at any developmental stage [Figure 2A-E]. Though diversity tended to decrease with age in both groups, PERMANOVA confirmed no statistically significant effects of treatment or age (all P > 0.05, Table 1).

NMDS analysis based on Bray-Curtis dissimilarity suggested a separation trend between irradiated and control samples at the pupal and 1-day-old stages, whereas samples from 4-day-old and 7-day-old males clustered more closely [Figure 2F]. These clustering patterns suggest that irradiation induces stage-specific shifts in gut microbiota composition, with pronounced differences at earlier developmental stages. As mosquitoes age, their gut microbiota appears to stabilize, resulting in reduced variation between the 4-day-old and 7-day-old groups.

The Circos diagram [Figure 3A], illustrating species relationships at the phylum level, shows that the gut microbiota of GT males is primarily composed of Bacteroidota, Pseudomonadota, and Bacillota. Comparative analysis showed that both irradiated and control males harbored overlapping as well as unique phyla. Specifically, 22 phyla were consistently present in both groups [Figure 3B]. Two phyla, GAL15 (58.3%) and Fibrobacterota (41.7%), were unique to the control group [Figure 3C], while seven phyla, including Elusimicrobiota (30.7%), Dependentiae (21.1%), Armatimonadota (14.9%), Caldatribacteriota (11.4%), NB1-j (7.9%), Deferribacterota (7.0%), and Latescibacterota (7.0%), were unique to the irradiated group [Figure 3D].

Across 40 gut microbiota samples, the relative abundance profile at the genus level revealed that the microbiota composition was strongly affected by mosquito age. The gut microbiota of pupae and 1-day-old mosquitoes were generally similar, dominated by Enterococcus, Methylobacterium-Methylorubrum, and Achromobacter. In contrast, 4-day-old and 7-day-old males shared a different profile, with a predominance of Elizabethkingia, Achromobacter, and Chryseobacterium.

Irradiation primarily affected the relative abundance of bacterial genera. In irradiated pupae, Enterococcus abundance decreased, while Methylobacterium-Methylorubrum and Achromobacter increased. In 1-day-old irradiated males, Comamonas increased, while Methylobacterium-Methylorubrum and Achromobacter decreased. Among 4-day-old irradiated males, Elizabethkingia increased, whereas Achromobacter and Leucobacter decreased. In 7-day-old irradiated males, Elizabethkingia again increased, accompanied by a decline in Achromobacter and Sphingomonas [Figure 3E].

Dynamic analysis of the four most abundant genera revealed the following distinct trends. Elizabethkingia showed the highest relative abundance in 4-day-old and 7-day-old males [Figure 3F], while Enterococcus reached its peak at the 1-day-old stage before declining. Methylobacterium-Methylorubrum declined significantly with age, eventually disappearing. The abundance of Achromobacter varied notably between irradiated and control groups, depending on developmental stage.

A total of 487 cultivable microbial isolates were analyzed, 204 from the control group and 283 from the irradiated group, to investigate compositional differences in the cultivable microbiota of GT male mosquitoes. At the family level, the gut microbiota of both groups was predominantly composed of Enterococcaceae (control: 55.3%; irradiated: 21.1%), Yersiniaceae (control: 15.2%; irradiated: 9.5%), Weeksellaceae (control: 7.4%; irradiated: 18.5%), and Comamonadaceae (control: 5.9%; irradiated: 5.8%) [Figure 4A and B]. Notably, Microbacteriaceae, Aeromonadaceae, and Flavobacteriaceae were only detected in the irradiated group [Supplementary Table 3]. Additionally, three fungal families, Aspergillaceae, Debaryomycetaceae, and Rhynchogastremaceae, were also cultured from the gut microbiota of both groups, further highlighting the microbial diversity in the mosquito gut [Supplementary Table 3].

Genus-level analysis across developmental stages revealed distinct age- and treatment-related patterns. In the control male group, dominant genera included Rahnella (54.5%), Enterococcus (27.3%), and Serratia (18.2%) [Figure 4C]. In contrast, irradiated pupae harbored Acinetobacter (37.5%), Enterococcus (37.5%), Serratia (12.5%), and Microbacterium (12.5%) [Figure 4D].

At 1 day old, the control group was dominated by Rahnella (28.2%), Enterococcus (28.2%), Comamonas (14.1%), and Chryseobacterium (12.8%). In irradiated 1-day-old groups, Enterococcus (23.7%), Acinetobacter (21.9%), Flavobacterium (2.7%), and Comamonas (9.1%) were dominant.

At 4 days old, the microbiota of the control samples was mainly composed of Enterococcus (63.2%), Enterobacter (11.8%), and Asaia (8.8%). In the irradiated group, the community was more diverse, including Asaia (23.3%), Enterococcus (21.1%), Chryseobacterium (12.2%), Leucobacter (11.1%), Acinetobacter (7.8%), and Rahnella (7.8%).

At 7 days old, Enterococcus dominated the control group (95.8%), while the irradiation group showed a more heterogeneous profile, comprising Elizabethkingia (25.4%), Asaia (20.9%), Enterococcus (14.9%), and Serratia (10.4%).

In the control group, Enterococcus increased with age, while Rahnella and Serratia decreased. In contrast, irradiation resulted in greater bacterial diversity across stages, though Enterococcus remained a consistently abundant genus.

To identify further significant differences in gut-associated bacterial communities between groups, we performed an in-depth analysis using the Linear Discriminant Analysis Effect Size (LEfSe) method. An LDA score threshold of > 2 was used to determine significantly enriched taxa [Figure 5A]. These results were combined with inter-group comparison tests [Figure 5B], revealing distinct microbial signatures in the irradiated group. Notably, the genera Delftia and Stenotrophomonas exhibited significant shifts in abundance. Further age-specific analysis showed that Delftia was significantly more abundant in irradiated pupae compared to controls [Figure 5C]. In 1-day-old males, Achromobacter significantly decreased post-irradiation, whereas Stenotrophomonas increased [Figure 5D]. No significant genus-level differences were observed in 4-day-old irradiated and control groups. The abundance of Comamonas and Sphingobacterium was significantly higher in the irradiated group compared to the control group [Figure 5E]. To further explore temporal patterns, we examined the relative abundance of selected culturable genera that were previously identified as differentially abundant between groups. The abundance of these genera was visualized across developmental stages [Figure 5F]. While no statistical comparisons were performed for these temporal profiles, several genera, including Stenotrophomonas, Comamonas, and Sphingobacterium, showed apparent increases following irradiation and subsequent declines over time. Similarly, Delftia and Achromobacter displayed higher relative abundance at early stages, followed by gradual decreases as mosquitoes aged. These observations provide a descriptive overview of potential temporal dynamics of irradiation-associated bacterial taxa.

To further validate the irradiation-associated shifts detected in the 16S rRNA sequencing analysis, the relative abundance of the eight culturable genera was quantified by qPCR [Figure 6]. In pupae, Delftia [Figure 6A] exhibited a higher relative abundance in the irradiated group than in the control group, consistent with its sequencing-derived enrichment at this stage. In one-day-old males, the relative abundance of Achromobacter [Figure 6B], Leucobacter [Figure 6C], Pseudomonas [Figure 6D], and Deinococcus [Figure 6E] was markedly reduced in irradiated individuals, supporting the sequencing-based depletion of these taxa shortly after adult emergence. Conversely, Stenotrophomonas [Figure 6F] showed an increased relative abundance in irradiated one-day-old males, consistent with its irradiation-associated elevation observed in the sequencing analysis. In contrast, in seven-day-old males, Comamonas [Figure 6G] and Sphingobacterium [Figure 6H] were significantly more abundant in irradiated mosquitoes, matching the stage-specific increases observed in the sequencing dataset. Overall, the qPCR results confirmed the irradiation-responsive patterns identified by the 16S rRNA gene analysis and validated these eight taxa as key differential genera affected by pupal irradiation.

Functional prediction based on 16S rRNA gene sequencing revealed a decreased abundance of enzymes 6.3.5.7 (glutaminyl-tRNA synthetase, glutamine hydrolyzing) and 6.3.5.6 (asparaginyl-tRNA synthetase, glutamate hydrolyzing) in irradiated male mosquito groups [Figure 7A]. In response, glutamine synthetase (GS) activity was measured across four age groups. Compared to the control, GS activity was significantly reduced in irradiated groups at the pupal stage (57.37 ± 3.44 vs. 38.34 ± 5.19, μmol/h/g, P = 0.0022), four-day-old group (36.36 ± 3.41 vs. 29.72 ± 3.17, μmol/h/g, P = 0.0108), and the seven-day-old group (13.50 ± 3.84 vs. 8.61 ± 2.57, μmol/h/g, P = 0.0260), but not at one-day-old group (44.64 ± 2.73 vs. 45.12 ± 4.32, μmol/h/g, P = 0.8733) [Figure 7B].

Protein carbonyl content showed age-dependent responses to irradiation across developmental stages. No significant difference was observed between irradiated and control groups at the pupal stage (3.56 ± 0.81 vs. 3.77 ± 0.83, P = 0.6147) or in the one-day-old group (0.076 ± 0.020 vs. 0.079 ± 0.026, μmol/g, P = 0.9242). However, content significantly increased in the irradiated four-day-old group (0.055 ± 0.019 vs. 0.098 ± 0.024, μmol/g, P = 0.0130) and significantly decreased in the seven-day-old group (4.81 ± 0.78 vs. 1.50 ± 1.08, μmol/g, P = 0.0022) [Figure 7C].

Glucose content was significantly elevated in irradiated males across most age groups, indicating disrupted glucose metabolism. Compared to controls, irradiated mosquitoes had higher glucose levels at the pupal stage (97.64 ± 33.41 vs. 169.71 ± 32.07, mg/g, P = 0.0022), four-day-old (647.00 ± 35.65 vs. 844.23 ± 41.04, mg/g, P = 0.0022), and seven-day-old stages (169.09 ± 16.35 vs. 250.09 ± 45.51, mg/g, P = 0.0022). Interestingly, glucose content was significantly lower in irradiated one-day-old males (263.37 ± 19.26 vs. 243.01 ± 12.10, mg/g, P = 0.0368) [Figure 7D].

Lipid peroxidation levels did not differ significantly between irradiated and control groups at any developmental stage [Figure 7E].

Hydrogen peroxide (H₂O₂) levels were significantly elevated in irradiated males at most developmental stages. Compared to controls, irradiated individuals exhibited higher H₂O₂ levels at the pupal stage (9.36 ± 2.12 vs. 16.16 ± 2.99, nmol/g, P = 0.0087), one-day-old stage (13.49 ± 1.89 vs. 18.82 ± 1.23, nmol/g, P = 0.0022), and seven-day-old group (22.99 ± 2.31 vs. 29.04 ± 1.81, nmol/g, P = 0.0022). However, no significant difference was observed in the four-day-old group (17.57 ± 0.66 vs. 17.79 ± 0.60, nmol/g, P = 0.5108) [Figure 7F].