Abstract Amur ide (Leuciscus waleckii), which inhabits Lake Dali, a soda lake in Northeast China with extremely high alkalinity (~ 53.57 mmol/L) and pH value (~ 9.6), is considered to be an ideal model for elucidating alkaline adaption mechanisms. To uncover the molecular mechanisms underlying this adaptation, we conducted a comparative study between the alkaline water ecotype (JY) and freshwater ecotype (DY). Both groups were exposed to a gradient of NaHCO[3] stress levels (0, 10, 30, and 50 mmol/L), and their responses were systematically assessed through integrated multi-omics analyses alongside physiological assays. Our results revealed that under low and moderate alkaline stress (10 and 30 mmol/L), JY group significantly upregulated the gene anpep, facilitating the hydrolysis of cysteinyl-glycine to release l-cysteine, thereby enhancing antioxidant capacity. Under high stress conditions (50 mmol/L), JY further synergistically upregulated gpx to activated the glutathione peroxidase (GPx) pathway to eliminate excess ROS. In contrast, the DY group predominantly relied on upregulating chac1-mediated γ-glutamyltransferase activity to facilitate glutathione cycling. Notably, while cysteinyl-glycine content significantly increased in the alkaline water ecotype (JY) under moderate and high alkalinity stress (30 and 50 mmol/L), the expression of its upstream gene chac1 was significantly downregulated. This paradox suggests alternative sources or regulatory mechanisms for cysteinyl-glycine accumulation in JY. Microbial tracing analysis revealed a positive correlation between cysteinyl-glycine levels and the gut microbiota genus Stenotrophomonas in JY, whose relative abundance increased progressively with elevated alkalinity. It is speculated that Stenotrophomonas may modulate host glutathione metabolism by regulating cysteinyl-glycine levels, thereby facilitating alkaline adaptation. Supplementary Information The online version contains supplementary material available at 10.1186/s42523-025-00452-6. Keywords: Gut microbiota, Leuciscus waleckii, Alkaline adaption, Stenotrophomonas Introduction Amur ide (Leuciscus waleckii), belonging to the order Cypriniformes, family Cyprinidae, subfamily Leuciscinae, and genus Leuciscus, is primarily distributed in freshwater rivers and inland lakes across Russia, Mongolia, and China [[36]1]. This species can survive in both freshwater and alkaline water. Notably, the population (alkaline water ecotype) inhabiting Dali Lake in Inner Mongolia of China, exhibits remarkable tolerance to extreme alkaline conditions, thriving in environments with an alkalinity of 53.57 mmol/L (pH 9.6) and salinity of 6‰ [[37]2]. Due to its extremely alkaline adaptability, Amur ide has been recognized as an ideal model for studying alkaline tolerance mechanisms. Extensive research has been conducted in China to elucidate the extremely alkaline tolerance of Amur ide, encompassing physiological and biochemical characteristics [[38]3], genome [[39]4, [40]5], transcriptome [[41]6, [42]7], and metabolomic studies [[43]8]. Acute toxicity assays of bicarbonate alkalinity have demonstrated that the alkaline water ecotype exhibits 1.5 times greater alkaline tolerance than the freshwater ecotype [[44]9]. Further genomic and transcriptomic analyses have identified the genetic and molecular mechanisms underlying alkaline adaptation in L. waleckii. Comparative transcriptomic analyses have shown significant differential expression of genes involved in ion homeostasis, antioxidant defense, and urea metabolism, which are subject to positive selection [[45]10]. Moreover, genome-wide selective sweep analyses have identified genomic regions under selection that are enriched in pathways related to ion transport and reactive oxygen species (ROS) scavenging, suggesting their crucial roles in adaptation to high alkaline-saline environments [[46]11]. However, previous studies have primarily focused on the host's regulatory mechanisms of alkaline tolerance, while the role of gut microbiota in the adaptation of L. waleckii to alkaline environments remains largely unexplored. Given the critical role of gut microbiota in host metabolism, nutrient absorption, and environmental adaptation, investigating host-microbiota interactions will provide deeper insights into the alkaline tolerance mechanisms of L. waleckii. The gut not only serves as the primary site for digestion and nutrient absorption but also represents the largest and most complex immune organ [[47]12]. Within this complex environment resides a highly diverse and functionally dynamic microbial community, the intestine often regarded as an "extra organ" due to its profound influence on host physiology [[48]13]. The host provides a nutrient-rich environment for the microbiota, while gut microbes contribute to host energy homeostasis by regulating feeding behavior, digestion, metabolism, and immune responses [[49]14, [50]15]. For instance, in Central stoneroller (Campostoma anomalum), gut microbiota modulate alanine aminotransferase activity in the intestine, facilitating ammonia detoxification [[51]16]. In zebrafish (Danio rerio), commensal gram-negative bacteria produce low levels of lipopolysaccharides (LPS), which activate intestinal alkaline phosphatase and subsequently reduce gut inflammation [[52]17]. Additionally, Pseudomonas sp. GP21 and Psychrobacter sp. GP12, isolated from the gut of Atlantic cod (Gadus morhua), exhibit amylase, chitinase, and cellulase activities, aiding in carbohydrate digestion [[53]18]. Furthermore, bacteria belonging to the genus Citrobacter enhance host energy acquisition from high-lipid diets by increasing triglyceride absorption efficiency and re-esterification [[54]19]. To explore the ecological adaptive evolution mechanisms of L. waleckii under long-term alkaline stress and to uncover the adaptive differences between the alkaline water ecotype and freshwater ecotype Amur ide. This study investigated the alkaline water ecotype Amur ide from Dali Lake and the freshwater ecotype from the Songhua River freshwater basin. By setting up different alkalinity gradient stress experiments, we analyzed their adaptive responses to different alkaline environments and explored the underlying biological and evolutionary mechanisms. Firstly, using metabolomic analysis, we identified differential serum metabolites and enriched pathways under alkaline stress, providing insights into the phenotypic differences in alkaline adaptation between these two ecotypes. Secondly, transcriptomic analysis of intestine was conducted to further investigate the gene regulatory mechanisms driving these metabolic changes. Finally, gut microbiome data were integrated to assess the potential role of microbiota in metabolic regulation. In addition, this study elucidated the molecular and physiological mechanisms underlying the alkaline tolerance of L. waleckii, providing new evidence for understanding the biological mechanisms of fish adaptation to extreme environments. It also offers a theoretical basis and technical support e targeted microbial improvement of alkaline tolerance in non-tolerant fish species, thereby enhancing the ecological diversity of fish in alkaline water. Materials and methods Experimental fish source and management The alkaline water ecotype Amur ide (hereafter abbreviated JY) was derived from the F[3] generation of artificially cultured and bred fish from Dali Lake in Inner Mongolia, while the freshwater ecotype (hereafter abbreviated DY) was obtained from the F[3] generation of artificially cultured and bred fish collected from the Suibin section of the Songhua River. Both forms of Amur ide were maintained at the Hulan Experimental Station (126.63°E, 45.97°N) of the Heilongjiang Fisheries Research Institute, Chinese Academy of Fishery Sciences. A total of 200 fish (100 individuals per form) with uniform size and normal appearance were randomly selected, with average body weights of (119.0 ± 22.5) g for JY and (199.7 ± 21.3) g for DY. The fish were transferred from outdoor earthen ponds to indoor recirculating aquariums for a one-week acclimation period. During this time, they were fed twice daily, and residual feed and feces were promptly removed. Half of the water was replaced daily to maintain optimal water quality. Stress experiment of Amur ide under different NaHCO[3] alkalinity conditions This study established three NaHCO[3] alkalinity gradient groups—low (10 mmol/L, AW-10), medium (30 mmol/L, AW-30), high (50 mmol/L, AW-50), and a freshwater control group (FW-0). A total of 72 alkaline water ecotype Amur ide (JY-FW-0, JY-AW-10, JY-AW-30, JY-AW-50) and 72 freshwater ecotype of Amur ide (DY-FW-0, DY-AW-10, DY-AW-30, DY-AW-50) were fin-clipped for identification and evenly distributed into four 400 L indoor recirculating aquariums (177 cm × 57 cm × 40 cm). Each aquarium was partitioned into three replicates, with each replicate containing 12 fish (JY = 6, DY = 6). The 28-day culture experiment was conducted under different alkalinity conditions. NaHCO[3] (analytical grade, Tianjin Kaitong Chemical Reagent Co., Ltd.) was used to prepare the respective alkalinity levels. During the experiment, feeding was performed at 3% of the fish's body weight twice daily (8:00 and 16:00). Due to reduced feed intake in both fish species across different alkalinity groups, residual feed was promptly removed daily. Continuous aeration was maintained, and half of the water was replaced regularly while monitoring key water parameters. Alkalinity was measured via titration, pH was determined using a pH meter, dissolved oxygen (DO) was measured with a DO meter, and temperature was recorded using a standard thermometer. Throughout the experiment, the measured alkalinity levels (mean ± SD) for the control and treatment groups (from low to high) were (0.51 ± 0.04) mmol/L, (10.51 ± 0.21) mmol/L, (30.86 ± 1.74) mmol/L, and (49.15 ± 1.25) mmol/L, respectively. The corresponding pH values were 7.06 ± 0.07, 8.2 ± 0.09, 8.6 ± 0.10, and 8.93 ± 0.11, while salinity levels were (0.1 ± 0.02) g/L, (1.1 ± 0.33) g/L, (2.4 ± 0.46) g/L, and (3.6 ± 0.52) g/L. The dissolved oxygen concentration was (6.53 ± 0.41) mg/L, and the water temperature was maintained at (18.56 ± 1.28) °C. Sample collection At the end of the alkalinity gradient culture experiment, six fish per replicate (totaling 36 fish: JY = 18, DY = 18) were sampled from each group. Fish were anesthetized using MS-222 (Sigma-Aldrich, USA), dried, weighed, and measured for body length. Approximately 2.5 mL of blood was collected from the caudal vein using a 5 mL syringe and immediately aliquoted into two clean 2 mL centrifuge tubes. The tubes were incubated overnight at 4 °C to allow clotting, followed by centrifugation at 3000 rpm for 10 min to separate serum. The supernatant serum was carefully transferred and stored at − 80 °C for subsequent broad-targeted metabolomic profiling and enzyme activity assays. For 16S rRNA sequencing, intestinal contents were separated into 2 mL cryotubes, flash-frozen in liquid nitrogen, and subsequently stored at − 80 °C. For RNA-seq and RT-qPCR, intestinal tissues (after content removal) were finely dissected, wrapped in aluminum foil, flash-frozen in liquid nitrogen, and transferred to − 80 °C for long-term storage. All samples were collected during winter in December. Serum widely-targeted metabolomics Serum samples were collected from both the JY and DY across all experimental groups, with six biological replicates per group (totaling 24 samples per ecotype, JY = 24 and DY = 24). This study employed LC–MS/MS-based metabolomics analysis. For sample preparation, frozen specimens stored at − 80 °C were thawed on ice and mixed with 300 μL of cold extraction solvent (acetonitrile: methanol = 1:4, v/v) containing internal standards. After vortexing for 3 min and centrifugation at 12,000 rpm for 10 min at 4 °C, 200 μL of supernatant was collected and kept at − 20 °C for 30 min before a second centrifugation to obtain 180 μL of clarified extract for injection. Chromatographic separation was performed using two complementary modes: reversed-phase chromatography utilized a Waters ACQUITY UPLC HSS T3 C18 column (1.8 μm, 2.1 × 100 mm) with mobile phases consisting of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B) under gradient elution; hydrophilic interaction chromatography was conducted on an ACQUITY UPLC BEH HILIC column (1.7 μm, 1 × 150 mm) using an acetonitrile–water-methanol system containing 20 mM ammonium formate. Mass spectrometric detection was performed using both QTOF-MS/MS and QTRAP-MS/MS systems. The QTOF-MS/MS operated in information-dependent acquisition (IDA) mode with a mass scan range of 50–1000 m/z, ion source temperature of 550 °C, and spray voltage of ± 5000/4000 V (positive/negative mode). The QTRAP-MS/MS worked in multiple reaction monitoring (MRM) mode with ion source temperature of 500 °C and spray voltage of ± 5500/4500 V (positive/negative mode). All analytical parameters including gas pressure and collision energy were optimized to ensure detection sensitivity. Gut transcriptomic profiling We performed comparative transcriptomic analysis of intestinal tissues from JY and DY under different alkalinity conditions. For each ecotype, 36 intestinal samples were collected (9 biological replicates per treatment group and control). Samples were homogenized in liquid nitrogen, with three biological replicates pooled equally to generate one technical replicate (resulting in 12 pooled samples per strain, n = 3 pools per treatment group). Total RNA was extracted using Trizol reagent, followed by mRNA enrichment with oligo (dT)-coupled magnetic beads. RNA fragmentation and cDNA library preparation were performed using standard protocols, including first-strand synthesis with random hexamers and M-MuLV reverse transcriptase, second-strand synthesis, and PCR amplification. Library quality was assessed by Qubit 2.0 Fluorometer quantification and Agilent Bioanalyzer 2100 size distribution analysis before normalization to 1.5 ng/μL. Paired-end sequencing (150 bp) was conducted on the MGISEQ-2000 platform. Raw reads were processed using fastp (v0.23.2) for quality control and Hisat2 (v2.2.1) for alignment to our recently assembled L. waleckii reference genome. All sequencing and preliminary bioinformatics analyses were performed by Novogene Co., Ltd (Beijing). This experimental design provided sufficient biological and technical replication for robust differential gene expression analysis across treatment groups. Gut microbiota 16S rRNA sequencing We conducted 16S rRNA gene sequencing to analyze the gut microbiota composition in both JY and DY under different alkalinity conditions. A total of 144 intestinal content samples (72 per form, comprising 18 biological replicates per treatment group and control) were collected from − 80 °C storage. Three biological replicates were pooled to generate one technical replicate, resulting in 24 pooled samples per strain (6 pools per treatment group). Genomic DNA was extracted using the DNeasy PowerSoil Kit, with quality assessed by UV spectrophotometry (concentration and OD260/280 ratio) and 1% agarose gel electrophoresis. The V3-V4 hypervariable regions of bacterial 16S rRNA genes were amplified using universal primers 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′), with triplicate PCR reactions pooled per sample and products purified by gel extraction. Libraries were prepared following Illumina's standard protocols using the TruSeq DNA Sample Prep Kit and sequenced on the MiSeq platform (2 × 250 bp). Bioinformatic processing was performed using QIIME2 (v1.7.0): raw paired-end reads were demultiplexed, quality-filtered (dada2 denoising algorithm), merged, and chimera-checked to generate amplicon sequence variants (ASVs). Taxonomic classification was achieved by aligning ASVs against the SILVA database (v138.1). All laboratory procedures, from library preparation to sequencing, were conducted by Novogene Co. Ltd (Beijing). This experimental design with pooled replicates ensured robust characterization of core microbiota while maintaining sufficient biological representation across treatments. Intestinal total RNA extraction and RT-qPCR Based on the RNA-seq and serum metabolomics results, key regulatory genes from significantly enriched pathways of differentially expressed genes and metabolites were selected. The mRNA sequences of these genes were retrieved, and specific primers were designed using Primer-BLAST and subsequently synthesized by Beijing Novogene Bioinformatics Technology Co., Ltd. (Table [55]1). Total RNA was extracted from all intestinal tissue samples following the TRIzol reagent protocol (Invitrogen, USA). RNA concentration was measured using a NanoDrop 8000 spectrophotometer (Thermo Fisher Scientific, USA), and RNA integrity was verified by 1% agarose gel electrophoresis. First-strand cDNA was synthesized using the PrimeScript™ RT Reagent Kit with gDNA Eraser (TaKaRa, Japan) according to the manufacturer's instructions and stored at − 20 °C for subsequent use. Quantitative real-time PCR (RT-qPCR) was performed using SYBR® Premix Ex Taq™ II (TaKaRa, Japan) on an ABI 7500 Fast Real-Time PCR System (Applied Biosystems, USA). The β-actin gene was used as an internal control. The PCR amplification protocol consisted of an initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 34 s. A final dissociation step was performed at 95 °C for 15 s, 60 °C for 1 min, and 95 °C for 15 s to assess amplification specificity. Table 1. Primer information for real-time quantitative PCR Gene Sequence of primers Tm/℃ Product size/bp anpep F: TGATCCGTAAACAGGACGCT 60 128 bp R: GGACCCGACACCATACTCTG chac1 F: GGTTACGGCTCGTTGGTTTG 60 138 bp R: CACCACTCTTCCGGGCATT gpx1 F: TTCGGACATCAGGAGAACTGC 60 200 bp R: GGATCACCCATCAGGGACAC β-actin F: CTGGCCCCTAGCACAATGAA 59 148 bp R: AGACTCGTCGTACTCCTGCT [56]Open in a new tab Measurement of serum antioxidant-related parameters Serum samples from both ecotypes of Amur ide (JY = 24, DY = 24, 6 samples per treatment group) were analyzed for the following parameters according to manufacturer protocols. Oxidative stress markers: Total antioxidant capacity (T-AOC), malondialdehyde (MDA), glutathione S-transferase (GST), reduced glutathione (GSH), oxidized glutathione (GSSG), glutathione peroxidase (GPx), and catalase (CAT) were measured using kits from Suzhou Comin Biotechnology Co., Ltd. Nitrogen metabolism markers: Blood ammonia (AMON) and urea nitrogen (BUN) levels were quantified using kits from Nanjing Jiancheng Bioengineering Institute. Absorbance values were determined using a Bio-Tek Synergy 2 multifunctional microplate reader (BioTek, USA). Total protein (TP) content was measured for each sample, and final parameter values were normalized to protein concentration. Data statistics Differentially expressed genes (DEGs) were identified using DESeq2 with thresholds of |log₂(FoldChange)|> 1 and adjusted p < 0.05 (Benjamini–Hochberg method). Differential metabolites (DEMs) were selected based on variable importance in projection (VIP) > 1 (PLS-DA model) and p < 0.05 (Student's t-test). KEGG pathway enrichment of DEGs was analyzed via clusterProfiler (v3.8.1). Mass spectrometry data were processed using MassHunter 8.0. PLS-DA was performed with SIMCA 14.1. Metabolic origin tracing used MetOrigin 2.0 ([57]https://metorigin.met-bioinformatics.cn/). All the data are presented as mean ± SD. RT-qPCR results were calculated by the 2^−ΔΔCt method and validated against RNA-seq FPKM values. One-way ANOVA with Duncan’s post-hoc test (SPSS 19.0) determined significance. Graphs were generated using GraphPad Prism 9, with dual-axis plots combining RT-qPCR (bar) and RNA-seq (line) data. Results and analysis Selection of serum differential metabolites (DEMs) and KEGG functional analysis The results of the analysis of the composition ratio of JY and DY metabolites (Fig. [58]1A) showed metabolites with a relative abundance ≥ 1% included alcohols and amines (3.81%), aldehydes, ketones, and esters (3.97%), amino acids and their metabolites (20.08%), benzene and substituted derivatives (11.42%), carbohydrates and its metabolites (4.05%), fatty acids (10.36%), glycoproteins (6.64%), heterocyclic compounds (8.34%), hormones and hormone related compounds (1.78%), nucleotides and its metabolites (5.75%), organic acids and its derivatives (15.63%). The numbers of differential metabolites (DEMs) in DY-AW-10, DY-AW-30, and DY-AW-50 groups were 228 (75 up, 153 down), 263 (101 up, 162 down), and 298 (113 up, 185 down). The numbers of differential metabolites (DEMs) in JY-AW-10, JY- AW-30, and JY-AW-50 were 184 (110 up, 74 down), 159 (96 up, 63 down), and 237 (123 up, 114 down,) (Fig. [59]1B). Multivariate statistics of JY (Fig. [60]1C) and DY (Fig. [61]1D) were used to perform PLS-DA analysis on the samples including quality control samples, which confirmed significant intergroup metabolic divergence between JY and DY (Q^2 = 0.829 and 0.895, p < 0.05). Fig. 1. [62]Fig. 1 [63]Open in a new tab Serum metabolomic profiling of Amur ide. A Circular taxonomy plot of identified metabolites. B Number of differentially expressed metabolites (DEMs) between alkalinity-treated groups and controls. C–D Partial least squares-discriminant analysis (PLS-DA) of JY and DY, respectively. E–F Venn diagrams showing shared/unique DEMs between different alkalinity treatments and controls in JY and DY. (G-H) Top 10 enriched KEGG pathways for DEMs in JY and DY groups Among the common DEMs, a total of 32 shared metabolites were identified across the alkaline-treated groups compared to the control group in JY group (Fig. [64]1E, Table S1). Notably, urea, a biomarker associated with alkaline adaptation, exhibited a significant decrease in the DY-AW-10, DY-AW-30, and DY-AW-50 groups, with DY-AW-50 being significantly higher than DY-AW-10 and DY-AW-30. Similarly, in JY-AW-10, JY-AW-30, and JY-AW-50 groups, urea levels significantly decreased, with the JY-AW-10 group displaying a significantly higher level compared to the JY-AW-30 group (Fig. [65]2P). Arginine, a direct precursor of urea synthesized through the hydrolysis of arginase in the urea cycle, showed no significant difference at low and moderate alkalinity levels in JY and DY. However, at high alkalinity, arginine levels significantly decreased in the DY-AW-50 group, while a significant increase was observed in the JY-AW-50 group (Fig. [66]2Q). These findings suggest that although urea levels decreased significantly under alkaline conditions in both JY and DY, the opposite trends in arginine levels in the DY-AW-50 and JY-AW-50 groups indicate distinct regulatory responses in urea synthesis pathways. Fig. 2. [67]Fig. 2 [68]Open in a new tab Content of LPCs (A–I), LPEs (J–L), PCs (M), phosphatidylethanolamine lyso alkenyl 18:3 (N), and cis-8, 11, 14, 17-eicosatetraenoic acid (O) in DY group under different alkalinity treatments. Content of urea (P) and arginine (Q) in DY and JY groups under different alkalinity treatments In DY group, a total of 88 shared DEMs were identified across the three alkalinity treatment groups, among which 27 were glycerophospholipids (Fig. [69]1F, Tab S2), including phosphatidylcholines (PCs, Fig. [70]2M), lysophosphatidylcholines (LPCs, Fig. [71]2 A-I), and lysophosphatidylethanolamines (LPEs, Fig. [72]2 J-L), accounting for 30.7% of the total shared DEMs. Specifically, in DY group the levels of PC and LPC were significantly decreased in the DY-AW-10, DY-AW-30, and DY-AW-50 groups. The LPE level in the DY-AW-30 group was significantly higher than that in the DY-FW-0 group, whereas no significant differences were observed when compared with the DY-AW-10 and DY-AW-50 groups. Additionally, the level of ω-3 arachidonic acid (Cis-8,11,14,17-eicosatetraenoic acid) was significantly increased in the DY-AW-10, DY-AW-30, and DY-AW-50 groups (Fig. [73]2O). KEGG functional annotation and enrichment analysis of the top 10 pathways for differential metabolites (Fig. [74]1G) revealed that the JY-AW-10 group exhibited the highest enrichment in glycerophospholipid metabolism and choline metabolism in cancer. Other enriched pathways were primarily related to lipid metabolism, including linoleic acid metabolism, α-linolenic acid metabolism, arachidonic acid metabolism, as well as steroid hormone metabolism, specifically cortisol synthesis and secretion, aldosterone-regulated sodium reabsorption, and aldosterone synthesis and secretion. In the JY-AW-30 group, significantly enriched pathways were predominantly associated with genetic material repair (pyrimidine metabolism), lipid metabolism (primary bile acid biosynthesis), and xenobiotic metabolism by cytochrome P450. The JY-AW-50 group was mainly enriched in amino acid metabolism (arginine and proline metabolism) and xenobiotic metabolism by cytochrome P450. In the DY group (Fig. [75]1H), serum metabolites exhibited high similarity across different alkalinity stress conditions. The most significantly enriched KEGG pathways in the low, medium, and high alkalinity groups were glycerophospholipid metabolism and choline metabolism in cancer, similar to the regulatory mechanisms observed in the JY-AW-10 group. Selection of intestine differential expressed genes (DEGs) and KEGG functional analysis Differential gene expression analysis (|log₂(FoldChange)|≥ 1, adjusted p < 0.05) revealed 3217 DEGs (1464 up, 1807 down) in JY-AW-10, 2426 (1070 up, 1356 down) in JY-AW-30, and 676 (381 up, 295 down) in JY-AW-50, respectively. In contrast, the DY-AW-10, DY-AW-30, and DY-AW-50 groups exhibited 2399 (1135 up, 1264 down), 2307 (842 up, 1465 down), and 2054 (1176 up, 878 down) DEGs, respectively (Fig. [76]3A). Comparative analysis with JY-FW-0 identified 73 shared DEGs across the JY-AW-10, JY-AW-30, and JY-AW-50 groups, accounting for 2.2%, 3.0%, and 10.8% of the total DEGs in each respective group. The number of uniquely expressed DEGs in JY-AW-10, JY-AW-30, and JY-AW-50 was 1840 (56.2%), 985 (40.6%), and 431 (63.7%), respectively (Fig. [77]3B). Similarly, in DY, 165 DEGs were shared among the DY-AW-10, DY-AW-30, and DY-AW-50 groups relative to DY-FW-0, constituting 6.9%, 7.2%, and 8.0% of total DEGs, respectively. The number of uniquely expressed DEGs in DY-AW-10, DY-AW-30, and DY-AW-50 was 1311 (54.6%), 1074 (46.5%), and 1410 (68.6%), respectively, demonstrating a similar trend to the alkaline water ecotype (Fig. [78]3 C). Fig. 3. [79]Fig. 3 [80]Open in a new tab Intestinal transcriptome analysis of Amur ide. (A) Volcano diagram of the number of DEGs in each alkalinity group compared with the control group (B–C) JY and DY, the number of common and endemic DEMs in different alkalinity groups compared with the control group, the Venn diagram (D–E) DEGs enrichment in the JY group and the DY group, and the analysis of the KEGG pathway (top10) KEGG pathway enrichment analysis of DEGs (top10) revealed that JY-AW-10 was predominantly enriched in pathways associated with xenobiotic metabolism and antioxidative defense, including metabolism of xenobiotics by cytochrome P450, glutathione metabolism, drug metabolism–cytochrome P450, drug metabolism–other enzymes, as well as pathways related to genetic material replication and repair, such as DNA replication and nucleotide excision repair (Fig. [81]2D). In JY-AW-30, significantly enriched pathways were primarily associated with xenobiotic metabolism and antioxidative responses, including glutathione metabolism and drug metabolism–other enzymes. Similarly, JY-AW-50 exhibited enrichment in pathways related to xenobiotic metabolism, including metabolism of xenobiotics by cytochrome P450, drug metabolism–cytochrome P450, and drug metabolism–other enzymes (Fig. [82]3 D). In DY, KEGG enrichment analysis (top10) revealed that DY-AW-10 was primarily enriched in pathways involved in genetic material replication and repair, including DNA replication, nucleotide excision repair, mismatch repair, and base excision repair. DY-AW-30 exhibited significant enrichment in carbohydrate metabolism pathways, such as starch and sucrose metabolism, galactose metabolism, and amino sugar and nucleotide sugar metabolism. DY-AW-50 showed enrichment in pathways related to energy metabolism, including amino acid metabolism (e.g., tyrosine metabolism, glycine/serine/threonine metabolism), amino acid biosynthesis (e.g., phenylalanine metabolism, tryptophan metabolism), carbohydrate metabolism (e.g., starch and sucrose metabolism, galactose metabolism), and central energy metabolism (e.g., pyruvate metabolism) (Fig. [83]3 E). These findings suggest that Amur ide employs distinct transcriptomic strategies to cope with alkaline stress in different ecotypes. However, three metabolic pathways—ribosome, cardiac muscle contraction, and oxidative phosphorylation—were consistently enriched in the DY-AW-10, DY-AW-30, JY-AW-10, and JY-AW-30 groups, indicating that Amur ide broadly regulates protein synthesis and energy metabolism under low-to-moderate alkalinity conditions as a common adaptive response. Correlation analysis between DEMs and DEGs Metabolomic analysis revealed that the metabolism of xenobiotics by cytochrome P450 pathway was significantly enriched in the JY-AW-30, JY-AW-50, and DY-AW-50 groups. Notably, within this pathway, the differential metabolite 2-S-glutathionyl acetate (GTA) exhibited a significant decrease in the DY-AW-50 group, whereas its levels were markedly elevated in the JY-AW-30 and JY-AW-50 groups (Fig. [84]4[85]I). This finding suggests that JY and DY groups may employ distinct metabolic adaptation mechanisms to maintain physiological homeostasis under alkaline stress. As a glutathione conjugate, GTA level fluctuations may directly reflect differences in the regulation of the glutathione metabolism pathway. Another key metabolite in this pathway, cysteinyl-glycine, was significantly upregulated in the JY-AW-30 and JY-AW-50 groups but showed no significant changes in the DY groups (Fig. [86]4 J). These results indicate that the glutathione metabolism pathway plays a crucial role in the alkaline adaptation of JY and DY groups, and its differential regulation may underly their distinct alkaline tolerance capacities. Consistently, transcriptomic data showed that the glutathione metabolism pathway was significantly enriched only in the JY-AW-10 and JY-AW-30 groups, while no significant enrichment was observed in the DY groups. This further supports the hypothesis that the glutathione metabolism pathway may have a specific regulatory role in the alkaline adaptation mechanism of the JY group. Fig. 4. [87]Fig. 4 [88]Open in a new tab Cytochrome P450 metabolism of exogenous substances and glutathione metabolism pathway in Amur ide. (A) Pattern diagram of the metabolic pathway of P450 to exogenous substances (B) Pattern diagram of glutathione metabolic pathway (C–E) Histogram and line chart with dual axes of anpep, gpx, chac1 genes in different alkalinity groups of JY (F–H), histogram and line chart with dual axes of anpep, gpx, chac1 gene in different alkalinity groups of DY (I) Changes in 2-S-glutathione acetate content in different alkalinity groups of JY and DY (J) Changes in cysteinyl-glycine content in different alkalinity groups of JY and DY Differential gene analysis within the glutathione metabolism pathway revealed that the aminopeptidase N (anpep) gene was significantly upregulated in the JY-AW-10, JY-AW-30, and JY-AW-50 groups, with expression levels in JY-AW-10 being higher than in JY-AW-30, whereas no significant changes were observed across different treatment groups in DY (Fig. [89]4C, [90]F). Additionally, the glutathione peroxidase gene, gpx was significantly upregulated in the JY-AW-50 group but showed no significant changes in the DY groups (Fig. [91]4D, [92]G). The γ-glutamyltransferase (chac1) gene was significantly downregulated in the JY-AW-30 and JY-AW-50 groups but exhibited significant upregulation in the DY-AW-50 group (Fig. [93]4 E, H). Notably, despite the downregulation of chac1 in the JY groups under alkaline conditions, its downstream metabolite, cysteinyl-glycine, exhibited a significant increase. This suggests that the accumulation of cysteinyl-glycine may not be entirely dependent on the regulation of chac1, highlighting the presence of alternative regulatory mechanisms contributing to its synthesis. Serum antioxidant-related indicators analysis As shown in Fig. [94]5 a–i, blood ammonia levels in the JY group exhibited a progressive increase with rising alkalinity. Specifically, blood ammonia levels in the JY-AW-10 and JY-AW-30 groups were significantly higher than those in the JY-FW-0 group, while the JY-AW-50 group had significantly higher blood ammonia levels than all other treatment groups (p < 0.05). In contrast, serum urea nitrogen levels showed a decreasing trend with increasing alkalinity, with the JY-FW-0 group exhibiting significantly higher levels than all alkalinity-treated groups. Regarding antioxidant indicators, the reduced glutathione (GSH) content initially increased and then decreased with increasing alkalinity. The GSH level in the JY-AW-10 group was significantly higher than that in the JY-FW-0 group, whereas GSH levels in the JY-AW-30 and JY-AW-50 groups were significantly lower than in the control group. Oxidized glutathione (GSSG) levels were significantly reduced in all alkalinity-treated groups. No significant differences in glutathione peroxidase (GPx) activity were observed among the groups. The activity of glutathione S-transferase (GST) significantly increased in the JY-AW-30 group but significantly decreased in the JY-AW-50 group. Catalase (CAT) activity initially increased and then decreased with increasing alkalinity, with the JY-AW-10 group showing significantly higher levels than the control group, the JY-AW-30 group showing no significant difference from the control, and the JY-AW-50 group displaying a significant decrease. Malondialdehyde (MDA) levels did not show significant differences among the groups. The total antioxidant capacity (T-AOC) was significantly higher in all alkalinity-treated groups compared to the control, reaching its peak in the JY-AW-10 group. Fig. 5. [95]Fig. 5 [96]Open in a new tab Serum physiological indexes of alkaline water species (lowercase letters) and freshwater species (uppercase letters) of different alkalinity groups. (a, A) blood ammonia (b, B) urea nitrogen (c, C) reduced glutathione (d, D) oxidized glutathione (e, E) Glutathione peroxidase (f, F) glutathione-S-transferase (g, G), catalase (h, H) malondialdehyde (i, I) total antioxidant capacity As shown in Fig. [97]5A–I, changes in blood ammonia and urea nitrogen levels in the DY group followed a similar trend to those in the JY group. Blood ammonia levels in the DY-AW-10 and DY-AW-30 groups were significantly higher than those in the DY-FW-0 group, while the DY-AW-50 group exhibited significantly higher blood ammonia levels than all other treatment groups (p < 0.05). Serum urea nitrogen levels decreased with increasing alkalinity, with the DY-FW-0 and DY-AW-10 groups showing significantly higher levels than the DY-AW-30 and DY-AW-50 groups. For antioxidant indicators, GSH levels showed a decreasing trend with increasing alkalinity, with the DY-FW-0 group exhibiting the highest GSH content, which was significantly higher than in the alkalinity-treated groups. The GSSG level was significantly increased in the DY-AW-50 group compared to the other three groups. GPx activity was significantly higher in the DY-FW-0 and DY-AW-10 groups than in the DY-AW-30 and DY-AW-50 groups. GST levels decreased in the DY-AW-50 group, significantly lower than in the DY-AW-30 group but not significantly different from those in the DY-FW-0 and DY-AW-10 groups. CAT activity initially increased and then decreased with rising alkalinity, with the DY-AW-10 group showing significantly higher levels than the control, while the DY-AW-30 and DY-AW-50 groups exhibited no significant differences from the control group. MDA levels showed no significant differences among the groups. In terms of total antioxidant capacity, the DY-AW-10 and DY-AW-30 groups displayed significantly higher T-AOC than the control and DY-AW-50 groups. Microbiota tracing analysis of the DEMs in the glutathione metabolism pathway The analysis of microbial community composition revealed that the top 10 most abundant genera in the JY group were Chryseobacterium, Stenotrophomonas, Ralstonia, Delftia, Aeromonas, ZOR0006, Mycobacterium, Legionella, Escherichia-Shigella, and Proteocatella. Among these, genera with a relative abundance of ≥ 1% and sample coverage of ≥ 80% included Chryseobacterium, Stenotrophomonas, and Ralstonia. The abundance of Stenotrophomonas and Ralstonia increased under alkaline conditions, whereas Chryseobacterium showed a decreasing trend in alkaline environments. By tracing the differential metabolites in the JY group and conducting a correlation analysis with their gut microbiota, it was found that cysteinyl-glycine, a key metabolite in the glutathione metabolism pathway, exhibited a negative correlation with genus Pseudomonadota, Legionella, Aeromonas, Bosea, Chryseobacterium, Flavobacterium, and Acinetobacter, while showing a positive correlation with Stenotrophomonas, Cupriavidus, Ralstonia, Delftia, and Legionella (Fig. [98]6). Fig. 6. [99]Fig. 6 [100]Open in a new tab Traceability analysis of differential metabolites in alkaline ecotype Amur ide gut microbiota composition and glutathione metabolic pathway (a) Top10 relative abundance of gut microbes (b) BIO-Sankey network diagram of glutathione metabolic pathway ko00480 Discussion Common adaptive mechanisms in two ecotypes of Amur ide under alkalinity stress Stress response in fish is a nonspecific physiological reaction triggered by one or more adverse environmental factors, which initiates almost immediately upon perceiving the stress [[101]20]. Mild stress responses may have positive effects, whereas excessive stress can negatively impact the organism. During stress responses, fish allocate essential resources such as energy and oxygen to vital physiological activities [[102]20]. Through the analysis of commonly enriched pathways in the intestinal transcriptomes of JY and DY, it was found that the JY-AW-10 and DY-AW-10 groups were significantly enriched in DNA replication and repair-related pathways, including DNA replication, nucleotide excision repair, base excision repair, and mismatch repair pathways. Additionally, the JY-AW-10, JY-AW-30, DY-AW-10, and DY-AW-30 groups were significantly enriched in oxidative phosphorylation, ribosome, and cardiac muscle contraction pathways. Notably, the differentially expressed genes (DEGs) in DNA replication and repair pathways, as well as in oxidative phosphorylation and ribosome pathways, were significantly downregulated, which may be associated with the energy allocation strategies of Amur ide intestines under alkaline conditions. Many genes and proteins involved in the cardiac muscle contraction pathway, such as calmodulin, myosin, actin, and Ca^2⁺-regulatory components, play crucial roles in regulating intestinal smooth muscle contraction [[103]21]. Intestinal peristalsis depends on smooth muscle activity, which is significantly influenced by Ca^2⁺ regulation. The opening of voltage-gated or ligand-gated calcium channels on smooth muscle cell membranes allows extracellular Ca^2⁺ to enter the cytoplasm. Additionally, endogenous calcium stores (e.g., the endoplasmic reticulum) release Ca^2⁺ via calcium release channels in the sarcoplasmic reticulum, leading to an intracellular Ca^2⁺ increase, which serves as the trigger signal for smooth muscle contraction [[104]22, [105]23]. Ca^2⁺ binds to calmodulin (CaM) to form a Ca^2⁺-CaM complex, which activates myosin light chain kinase (MLCK). MLCK catalyzes the phosphorylation of myosin light chain (MLC) using ATP, allowing phosphorylated MLC to interact with actin, ultimately inducing smooth muscle contraction [[106]22, [107]23]. Further analysis of commonly differentially expressed genes in the cardiac muscle contraction pathway across the JY-AW-10, JY-AW-30, DY-AW-10, and DY-AW-30 groups revealed that the myosin light chain 2 gene (myl2) was significantly upregulated in all four groups. The primary function of intestinal smooth muscle is to mix and propel luminal contents, facilitating efficient digestion, nutrient absorption, and waste excretion. Under low-to-moderate alkaline stress, increased intestinal peristalsis may serve as a crucial adaptation strategy in Amur ide, enhancing nutrient uptake, providing energy for the host, and regulating gut microbial metabolism. In summary, the significant enrichment of these pathways suggests that the intestines of Amur ide exhibit a highly specialized energy allocation strategy under low-to-moderate alkaline stress. While pathways related to cell growth are suppressed, the regulation of intestinal smooth muscle activity appears to be a fundamental physiological adaptation that plays a critical role in alkaline tolerance. Specific alkaline adaptative mechanisms in alkaline water ecotype of Amur ide Glycerophospholipids play an essential role in maintaining cell structure, organelle function, and energy homeostasis [[108]24]. The main phospholipids in biological membranes are phosphatidylethanolamine (PE) and phosphatidylcholine (PC), which help improve membrane fluidity and integrity, crucial for maintaining normal cell functions [[109]25]. ROS induced by abiotic stress can attack membrane lipids, leading to lipid peroxidation, reducing membrane permeability, and affecting membrane function [[110]26]. At the same time, glycerophospholipids can be hydrolyzed by phospholipases to produce unsaturated fatty acids and lysophospholipids (LPLs), with unsaturated fatty acids, such as arachidonic acid (AA), acting as inflammatory factors that trigger immune responses [[111]27]. The formation of LPLs can lead to membrane rupture or cell necrosis [[112]28]. In this study, DY showed a significant decrease in PC and PE content across all alkaline stress groups (DY-AW-10, DY-AW-30, DY-AW-50), with an increase in lysophosphatidylethanolamine (LPE) and a significant decrease in lysophosphatidylcholine (LPC) levels. The ω-3 arachidonic acid content significantly increased. It is speculated that under alkaline stress, DY may reduce energy consumption by inhibiting cell proliferation and growth while increasing PE hydrolysis and ω-3 arachidonic acid content, thus triggering immune responses to mitigate alkaline damage. However, the generation of large amounts of LPLs may lead to cell membrane rupture or necrosis, indicating that although DY induces innate immune responses under varying alkaline stress, the integrity of its cell membrane structure faces significant risk. Ammonia (including gas NH[3] and [MATH: NH4+ :MATH] ) is the primary metabolic end product of undigested proteins or amino acids and is toxic to all vertebrates [[113]29]. Therefore, timely and effective ammonia excretion is a critical mechanism for maintaining physiological homeostasis [[114]30]. Physiological results showed that blood ammonia levels in both the JY and DY groups significantly increased with rising alkalinity, reaching the highest levels in the AW-50 group, suggesting that high alkalinity may inhibit ammonia metabolism and excretion. Urea content significantly decreased in all alkaline groups of JY and DY. Fish urea synthesis primarily comes from the last step of the ornithine-urea cycle (OUC), where l-arginine is converted into l-ornithine and urea via arginase. The significant decrease in urea content indicates that Amur ide reduces the complex and energy-intensive arginine degradation and urea synthesis pathways under alkaline conditions, primarily detoxifying ammonia via other pathways like the gills. However, in the JY-AW-50 group, arginine and proline biosynthesis pathways were activated, and N-acetyl-ornithine and l-arginine levels significantly increased. N-acetyl-ornithine, an important intermediate in the arginine biosynthesis pathway, may suggest increased arginine synthesis. Arginine degradation was inhibited under high alkaline stress, but synthesis increased, resulting in significant accumulation of l-arginine. Research indicates that alkaline stress may cause extensive physiological damage to fish, such as slowing blood flow, affecting heart function, and inducing local or systemic hypoxia [[115]31, [116]32]. l-arginine, as a precursor of NO, induces vasodilation, reducing heart load and oxygen demand [[117]26]. Key genes uniquely and significantly expressed in the JY group include protein phosphatase 1 regulatory subunit 12a (ppp1r12a), a key player in the balance of contraction and relaxation in vascular smooth muscle (Tab.S3). Thus, the significant accumulation of l-arginine in the JY-AW-50 group may relate to increased oxygen demand under alkaline stress. Different responses of glutathione metabolism and cytochrome P450 pathway to alkaline stress in two ecotypes of Amur ide Reactive oxygen species (ROS) serve as crucial signaling molecules, playing a significant role in cellular physiological functions and immune regulation [[118]33]. At low concentrations, ROS act as signaling molecules to maintain cellular homeostasis, whereas excessive ROS production under stress conditions often leads to oxidative damage [[119]34]. Studies have shown that oxidative stress induced by alkaline stress triggers antioxidant defense systems in aquatic organisms to mitigate oxidative damage [[120]35]. Generally, under alkaline stress, fish exhibit a simultaneous activation of repair mechanisms and inflammatory responses. When the repair mechanisms prevail, physiological functions remain stable; however, if the damage caused by inflammation exceeds the repair capacity, physiological imbalance occurs, potentially resulting in irreversible damage or death [[121]36]. Glutathione (GSH) is one of the most critical antioxidants, protecting cells from oxidative damage through its thiol group (-SH), which reacts with ROS and free radicals to neutralize them into non-toxic or low-toxicity compounds [[122]37, [123]38]. In this study, the GSH content in the freshwater ecotype (DY) significantly decreased with increasing alkalinity, while oxidized glutathione (GSSG) levels markedly increased in the DY-AW-50 group. This indicates that fish in the DY group experienced severe oxidative stress under high-alkalinity conditions, leading to substantial GSH consumption and GSSG accumulation. The impaired antioxidant system could further exacerbate oxidative damage. Conversely, in the alkaline water ecotype (JY), although GSH levels also declined in the JY-AW-30 and JY-AW-50 groups, GSSG levels significantly decreased across all alkalinity groups, suggesting a stronger antioxidant regulation capacity in JY fish. This efficient ROS scavenging mechanism likely prevented excessive GSSG accumulation, contributing to oxidative stress resistance. Glutathione metabolism involves its degradation by γ-glutamyl transpeptidase (γ-GT), which cleaves GSH into glutamate and cysteinyl-glycine. Cysteinyl-glycine is further hydrolyzed by peptidase to produce l-cysteine and glycine, completing the GSH metabolism cycle [[124]39]. In this study, the expression of the γ-GT gene chac1 was significantly downregulated in the JY-AW-30 and JY-AW-50 groups, while it was markedly upregulated in the DY-AW-50 group. As a major GSH degradation enzyme, γ-GT is a key regulator of intracellular GSH levels [[125]40]. This suggests that the freshwater ecotype enhances GSH degradation and recycling to maintain GSH homeostasis under high-alkalinity stress. In contrast, the alkaline water ecotype did not rely on this GSH degradation and regeneration strategy. After GSH is degraded into cysteinyl-glycine, peptidase hydrolyzes it into l-cysteine and glycine [[126]41]. In this study, anpep exhibited significantly higher expression across all alkalinity groups in JY group but showed no significant differences in the DY group, indicating that the alkaline water ecotype accelerates l-cysteine production under alkaline conditions. This aligns with previous findings that anpep is highly expressed in the kidney and liver tissues of alkaline water ecotype and is under positive selection [[127]10]. Although l-cysteine comprises only around 2% of cellular amino acids, its distinct physicochemical properties, including a large sulfur atom and high polarizability, endow it with significant detoxification potential. l-cysteine not only participates in intracellular redox reactions but also contributes to liver phospholipid metabolism, offering hepatoprotective and liver function-restoring effects [[128]42]. Additionally, l-cysteine can be catalyzed by cystathionine γ-lyase (CSE) and cystathionine β-synthase (CBS) to produce hydrogen sulfide (H₂S) and cysteine persulfide (CysSSH). Notably, CysSSH exhibits superior antioxidative capacity compared to l-cysteine, effectively scavenging reactive oxygen and nitrogen species (RONS) [[129]43]. Therefore, the accelerated hydrolysis of cysteinyl-glycine and increased l-cysteine and glycine production likely enhance the JY adaptation to alkaline environments. Following detoxification, GSH is oxidized to GSSG by glutathione peroxidase (GPx), maintaining cellular redox balance [[130]44]. Transcriptome analysis revealed that gpx expression was significantly upregulated in the JY-AW-50 group, whereas no significant changes were observed in the DY group. This further demonstrates that the alkaline water ecotype utilizes the GPx pathway to mitigate oxidative damage under high-alkalinity stress. Based on the transcriptional and metabolic responses of the two ecotypes to different alkalinity levels, it can be inferred that the alkaline water ecotype prefers a more energy-efficient strategy by hydrolyzing cysteinyl-glycine to generate l-cysteine for oxidative stress resistance at low to moderate alkalinity. However, under high-alkalinity stress, the alkaline ecotype activates both the cysteinyl-glycine hydrolysis and GPx pathways for detoxification. In contrast, the freshwater ecotype relies predominantly on GSH degradation and recycling, lacking the flexible regulation strategy of the alkaline water ecotype. This is consistent with the observed decline in the total antioxidant capacity in the DY-AW-50 group, while the JY group maintained stable antioxidant capacity. Moreover, glutathione S-transferase (GST) catalyzes the conjugation of GSH with electrophilic compounds, enhancing their hydrophilicity for subsequent excretion [[131]45]. In both the JY-AW-50 and DY-AW-50 groups, GST activity was significantly reduced, suggesting that high alkalinity impairs GSH utilization for detoxification. Therefore, although the DY group attempted to compensate for oxidative stress by accelerating GSH degradation, the decline in GST activity further limited the effective use of GSH for detoxification. Finally, malondialdehyde (MDA), a toxic byproduct of lipid peroxidation, is widely used as a biomarker of oxidative stress [[132]46]. MDA levels showed no significant differences between JY and DY groups under various alkalinity conditions, indicating no substantial increase in lipid peroxidation. In addition, catalase (CAT), a key antioxidant enzyme that decomposes hydrogen peroxide into water and oxygen [[133]47], exhibited significantly increased activity in the JY-AW-10 group but decreased in the JY-AW-50 group. In the DY group, CAT activity only increased in the DY-AW-10 group, suggesting that the alkaline water ecotype upregulates CAT activity to mitigate oxidative stress at low to moderate alkalinity, but its antioxidant system is suppressed under severe stress. Interestingly, while chac1 was highly expressed in the JY-AW-10 group, its expression showed no significant changes in the JY-AW-30 and JY-AW-50 groups. However, the downstream metabolite cysteinyl-glycine significantly accumulated in the JY-AW-30 and JY-AW-50 groups, suggesting that cysteinyl-glycine accumulation in the alkaline water ecotype may not be entirely dependent on chac1 regulation. The underlying mechanisms of this phenomenon require further investigation. Stenotrophomonas involved in glutathione metabolism enhances alkaline adaptation in alkaline water ecotype of Amur ide During the biological activation of the glutathione metabolic pathway to combat oxidative stress, progressive oxidative stress may lead to sustained depletion of intracellular glutathione, potentially resulting in glutathione exhaustion [[134]48]. In this study, elevated levels of cysteinyl-glycine were observed in the blood of alkaline water ecotype of Amur ide. Given that chac1-mediated glutathione degradation serves as the primary endogenous pathway for cysteinyl-glycine generation, the intestinal uptake of exogenous cysteinyl-glycine may represent a compensatory mechanism for this species to prevent glutathione exhaustion during oxidative stress. The host and its gut microbiota form a close symbiotic relationship through long-term co-selection and co-evolution [[135]49, [136]50]. The host shares nutrients with the symbiotic microbes while selecting suitable microbial populations. In return, the gut microbiota helps the host by breaking down indigestible food and participating in metabolism [[137]51]. Studies have shown that the microbiome plays a critical role in shaping host metabolism, and microbial communities can influence blood metabolites [[138]52]. For example, a study analyzing 930 metabolites from the blood of over 1500 individuals found that 69% of these metabolites were driven by the microbiome [[139]53]. In this study, three bacterial genera (Chryseobacterium, Stenotrophomonas, and Ralstonia) ranked among the top 10 most abundant gut microbiota in alkaline water ecotype of Amur ide exhibited significant abundance differences between freshwater control and alkaline water treatment groups. Notably, the alkalinity-responsive abundance patterns of Stenotrophomonas and Ralstonia paralleled the fluctuations in blood cysteinyl-glycine concentrations. Furthermore, Using the MetOrigin 2.0 platform to predict the source of cysteinyl-glycine, it was found that Stenotrophomonas positively regulates cysteinyl-glycine, while Chryseobacterium negatively regulates it. The positive correlation between cysteinyl-glycine and Stenotrophomonas in the JY group indicates that the microbial community contributes to the host's metabolic adaptation under alkaline conditions. Stenotrophomonas is ubiquitously distributed across aquatic systems, terrestrial soils, and animal hosts. It plays significant roles in nitrogen and sulfur cycling and represents a well-established plant rhizosphere microbe [[140]54, [141]55]. Stenotrophomonas may serve as a potential probiotic to enhance environmental stress resistance in aquatic animals. During arsenic intoxication in tilapia (Oreochromis spp.), this genus becomes a dominant member of the gut microbiota and facilitates host biotransformation of toxic compounds [[142]56, [143]57]. What’s more, dietary supplementation with antimicrobial peptides, hydrolysable tannins, or glycerol monolaurate increased the abundance of Stenotrophomonas in the host intestine and enhanced host resistance against bacterial pathogens and oxidative stress [[144]58, [145]59]. These findings align with our results. Collectively, the evidence suggests that Stenotrophomonas potentially modulates host glutathione metabolism to enhance alkali stress resistance in the alkaline water ecotype of Amur ide, though the precise mechanisms necessitate further mechanistic validation. Supplementary Information [146]Additional file1 (XLSX 9 KB)^ (9.4KB, xlsx) [147]Additional file2 (XLSX 11 KB)^ (11KB, xlsx) Acknowledgements