Abstract Agarwood, a traditional Chinese medicine, has been widely used in the treatment of gastrointestinal diseases. The antibacterial and antioxidant properties of agarwood essential oils (EOs) have been well documented. High-altitude (> 2500 m) regions attract tens of millions of visitors worldwide each year; however, the hypoxic environment poses a threat to the health of the body’s organ systems, including the digestive system. Additionally, hypoxia has been reported to alter the gut microbiota and metabolites. Our previous study demonstrated that hypoxia exposure triggered ferroptosis in the gastric and small intestinal mucosa. This study aimed to explore the therapeutic effects and potential mechanisms of EOs in hypoxia-induced gastric and small intestinal mucosal injury. EO effects were evaluated based on clinical manifestations, histopathological assessments, and lipid peroxidation as determined by reactive oxygen species (ROS) and 4-hydroxynonenal (4-HNE) levels. We also assessed microbiota changes through 16 S rRNA gene amplicon sequencing and analyzed metabolites using untargeted liquid chromatography–mass spectrometry in the gastric and small intestinal contents of mice. EO treatment significantly alleviated hypoxia-triggered mucosal damage in the stomach and small intestine. Notably, EOs reduced hypoxia-induced lipid peroxidation and partially recovered the microbiota and metabolite disruptions induced by hypoxia. Specifically, Candidatus_Saccharimonas and Akkermansia may contribute to mucosal repair via regulating xanthoxic acid and aspartylglycosamine, and guanosine, respectively. EOs may provide a promising approach for treating hypoxia-induced gastric and small intestinal damage by repressing lipid peroxidation and regulating the microbiota and metabolites. Graphical abstract [36]graphic file with name 13568_2025_1928_Figa_HTML.jpg Keywords: Agarwood essential oils (EOs), Hypoxia, Microbiota, Metabolites, Lipid peroxidation, Gastrointestinal mucosa Introduction The gastrointestinal tract is colonized by trillions of microbiota, which regulate homeostasis and the host immune response (Smet et al. [37]2022; Wang et al. [38]2023). The stomach, ileum, and colon contain around 10^2–10^4, 10^3–10^9, and 10^10–10^22 bacteria colony forming units (CFU)/mL, respectively (Walker et al. [39]2014; Santacroce et al. [40]2021). The gut microbiota interacts with the host through its metabolites, which impact host health, energy metabolism, and epithelial integrity (Kayama et al. [41]2020; Debnath et al. [42]2021). Gastrointestinal microbial dysbiosis and metabolite disruption have been identified as pivotal factors contributing to various diseases of the digestive system, including atrophic gastritis, gastric cancer, inflammatory bowel diseases, and colorectal cancer (Lavelle et al. [43]2020; Stewart et al. [44]2020; Chen et al. [45]2021; Conti et al. [46]2021; Dai et al. [47]2021; Nagata et al. [48]2021; Cai et al. [49]2022). Modulating the gut microbiota and metabolites is a feasible approach to treating gastrointestinal disease (Mo et al. [50]2022; Gowen et al. [51]2023). Tens of millions of people worldwide visit high-altitude regions (> 2500 m) each year for tourism, sports, business, military, and other activities. However, individuals exposed to high-altitude environments usually experience gastrointestinal discomfort, including anorexia, nausea, vomiting, and diarrhea (Anand et al. [52]2006), which can be explained by hypoxic and oxidative stress induced by splanchnic hypoperfusion and decreased blood oxygen levels (McKenna et al. [53]2022). Hypoxia exposure has been reported to alter the microbiota and metabolites. The gut microbiota composition of lowlanders gradually converged toward that of high-altitude residents as their altitude of the lowlanders increased (Han et al. [54]2024). Decreased α- and β-diversity of the gut microbiota was observed in the initial stage of hypoxia exposure, as reported by Su et al. (Su et al. [55]2024). According to a study by Adak et al. (Adak et al. [56]2013), hypoxic conditions caused a significant reduction in the total aerobes in the gut microbiota, while the levels of total and facultative Anaerobes increased. A total of 33 significantly altered metabolites were identified in healthy participants exposed to high altitude for 3 days. Notably, a small fraction of these metabolites remained altered for 14 days after returning to low altitude, according to a study by Gao et al. (Gao et al. [57]2023a). Our previous study found that hypoxia-induced gastrointestinal damage primarily occurred in the stomach and small intestine (Wang et al. [58]2024). Consistently, gastric mucosal lesions were observed in Mountaineers who traveled from 490 to 4559 m above sea level (Fruehauf et al. [59]2020). Exposure to altitudes of 3,842 m and 4,767 m for 3 days led to swollen, shortened, and structurally disorganized villi in the small intestines of Wistar rats, which worsened with altitude (Zhang et al. [60]2015). Ferroptosis is a form of non-apoptotic cell death characterized by increased reactive oxygen species (ROS) levels and excessive lipid peroxidation (Gao et al. [61]b). According to our previous study, ferroptosis contributed to hypoxia-induced gastrointestinal mucosal injury (Wang et al. [62]2024). Agarwood, known as “the king of all incense”, is a resinous substance derived from the trunks and branches of Aquilaria trees (Li et al. [63]2021). Agarwood, recognized for its high pharmacological value, is used in the treatment of gastrointestinal diseases, including gastric ulcer and fluorouracil-induced intestinal injury (Wang et al. [64]2019, [65]2021). Agarwood was also shown to alleviate bile reflux gastritis by repressing gastric mucosal cell apoptosis through Wnt/β-catenin pathway regulation (Ma et al. [66]2025). Agarwood has been broadly used in traditional Chinese medicine as a carminative remedy in gastric diseases and as a qi-modulating medicine to alleviate vomiting (Alamil et al. [67]2022). Agarwood essential oils (EOs) were shown to have antibacterial and antioxidant properties (Xie et al. [68]2024). A total of 522 bacterial strains were shown to be sensitive to EOs, with Bacillus species the most sensitive, as determined by a disc diffusion assay (Singh et al. [69]2020). Sesquiterpenes and 2-(2-phenethyl) chromones, with antimicrobial and antioxidant activity, are the primary components of EOs (Chen et al. [70]2022; Xie et al. [71]2024). Given the microbiota-modulating and antioxidant properties of EOs and the importance of the microbiota and metabolites, we investigated the effects of EOs on hypoxia-induced damage of gastric and intestinal mucosa in mice, examined peroxide levels in mucosal lesions, and analyzed the microbiota and metabolites in the contents of the lesions, to investigate the potential therapeutic mechanisms of EOs. Materials and methods EO extraction The wood of Aquilaria yunnanensis (1 kg, coarse powder) underwent hydrodistillation in 10 L of water at 100 °C for 7 h, which yielded 502.4 mg of EOs. Analysis of gas chromatography-mass spectrometry (GC-MS) The composition of EOs was Analyzed by Agilent 6890–5977 A GC/MSD system (Agilent Technologies, Santa Clara, CA, USA) equipped with a HP-5 ms column (30 m × 250 μm × 0.25 μm). Compounds were identified by comparing with the NIST database (Linstrom [72]2022). EO samples (10 µL) were diluted in methanol (1 mL), and centrifuged at 12,000 rpm for 5 min, then the supernatant was used for subsequent analysis. The conditions for GC were as follows: injection volume, 1 µL; split ratio, 10:1; flow rate, 1.0 mL/min; injector temperature, 220 °C; detector temperature, 280 °C. The temperature programs were as follows: 60 °C (initial), increased to 160 °C at a rate of 20 °C/min, then to 163 °C at 3 °C/min (Maintained for 3 min), to 170 °C at 1 °C/min, and finally to 270 °C at 50 °C/min (Maintained for 3 min). The parameters of MS were as previously reported (Li et al. [73]2024b). Animal experiments Male and female C57BL/6 wild-type mice (6 weeks old) were purchased from Charles River Laboratories (Beijing, China). The mice were housed in a specific-pathogen-free Animal facility at Tsinghua University under a controlled 12-h light/dark cycle, with ad libitum access to sterilized food and water. A hypoxic chamber (LP-1500, Shanghai, China), purchased from Yuyan Instruments, was used to mimic high-altitude hypoxia that induced gastrointestinal injury in mice. Following a one-week acclimation, mice were paired by sex and age and randomly allocated into five groups (n = 18 per group, with n = 6 in each group were specifically used to detect gastrointestinal permeability). Each group consisted of an equal number of males and females with individuals housed in separate cages based on gender. All experimental days were standardized using day 0 as the reference point for hypoxia initiation. The groups were as follows: normoxic control group, in which mice were exposed to normoxic conditions for 3 days and gavaged with 100 µL of PBS on day 1; hypoxic control group, in which mice were housed in the hypoxic chamber for 3 days (72 h) gavaged with 100 µL of PBS on day 1 (24 h post-hypoxia); low-dose EO treatment group, in which mice were placed in the hypoxic chamber for 3 days and gavaged with 1 µg of EOs emulsified in 100 µL of PBS on day 1; medium-dose EO treatment group, in which mice were housed in the hypoxic chamber for 3 days and gavaged with 5 µg of EOs emulsified in 100 µL of PBS on day 1; and high-dose EO treatment group, in which mice were placed in the hypoxic chamber for 3 days and gavaged with 10 µg of EOs emulsified in 100 µL of PBS on day 1. The mice were euthanized on day 3 (72 h post-hypoxia), and the gastric and small intestinal contents were collected for Subsequent 16 S rRNA sequencing (n = 6 per group) and metabolite analysis (n = 6 per group). The gastric and small intestinal samples were collected for hematoxylin and eosin (H&E) staining (n = 12 per group), and the mucosa were scraped and harvested to analyze ROS and 4-HNE levels (n = 12 per group). All experimental procedures followed the criteria approved by the Animal Care and Use Committee of Tsinghua University (Protocol # 23-CZJ1). Gastrointestinal permeability in vivo The mice were fasted for 4 h, and then received an oral gavage of FITC-dextran (Molecular weight 70,000; MedChemExpress, Saint Louis, MO, USA) at a dose of 600 mg/kg body weight. 4 h post administration, mice were placed in a closed induction chamber and exposed to medical oxygen containing 5% isoflurane (v/v) until deep anesthesia was achieved. Retro-orbital blood was then collected in K3-EDTA-coated tubes followed by cervical dislocation. The blood samples were centrifuged (4 °C, 15 min, 1000 g), and serum was collected in clear Eppendorf tubes (Voetmann et al. [74]2023). The serum FITC-dextran concentrations were quantified using fluorometry. H&E staining Gastric and small intestinal tissue of the mice were fixed in 4% paraformaldehyde and embedded in paraffin. Sections were cut 5 μm thick and underwent H&E staining using standard protocols (Cardiff et al. [75]2014). Histological scores of the stomach and small intestine were determined following previously reported criteria (Liu et al. [76]2016). ROS and 4-HNE analysis The levels of ROS and 4-hydroxynonenal (4-HNE) in the gastric and small intestinal mucosa were determined using kits according to the manufacturer’s protocol (ROS, E-BC-K138-F; 4-HNE, E-EL-0128c; Elabscience, Wuhan, Hubei, China). 16 S rRNA sequencing Gastric and intestinal contents were collected, immediately frozen in liquid nitrogen, and stored at − 80 °C until 16 S rRNA sequencing. Sequencing was conducted as previously described (Wang et al. [77]2024). The samples underwent DNA extraction and 16 S rRNA gene amplification. QIIME platform scripts ([78]www.qiime.org) were used to analyze the sequencing data. Operational taxonomic units (OTUs) were clustered with 97% consistency using USEARCH v7.0. The α-diversity was evaluated using Chao1, Shannon, and Simpson diversity indices. The β-diversity was evaluated using non-metric multidimensional scaling (NMDS) and principal component analysis (PCoA). Analysis of metabolites in the gastric and small intestinal contents Metabolites in the gastric and small intestinal contents were analyzed using untargeted liquid chromatography-mass spectrometry (LC-MS) on a Vanquish UHPLC system (Thermo Fisher Scientific, Waltham, MA, USA) and ACQUITY UPLC ^® HSS T3 (Waters, Milford, MA, USA), according to a previously described protocol (Wang et al. [79]2024). KEGG pathway enrichment analysis of metabolites KEGG pathway enrichment analysis ( Kanehisa et al. [80]2017) was performed using the hypergeometric distribution, implemented in the hypergeom.sf function of the scipy.stats package of the Python 3.9 (Oliphant [81]2007). Correlation analysis of microbiota and metabolites Data preprocessing was performed using Python 3.9. Pearson correlation coefficients and P-values for the differential microbiota OTUs and metabolites were calculated using the corr.test function from the “psych” R package 4.3.0. The “method” was set to “Pearson”, and the “adjust” was set to “none”. Correlation heatmaps were subsequently generated by the “pheatmap” package, with both “cluster_row” and “cluster_col” options set to “T”. Statistical analysis One or two-way analysis of variance (ANOVA) was performed to compare the means of three or More groups using GraphPad Prism 10.0.2 (GraphPad Software, San Diego, CA, USA). P < 0.05 was identified as statistically significant. Results EO administration significantly ameliorated hypoxia-induced gastric and small intestinal mucosal damage and lipid peroxidation GC-MS analysis identified forty-six components in EOs, including thirty-two sesquiterpenes (relative abundance 62.04%), eleven aromatic compounds (relative abundance 30.44%), one fatty acid (relative abundance 0.29%), and two other constituents (relative abundance 7.23%) (Fig. [82]1A). The details regarding the molecular formula, name, peak area, and classification were provided in Table [83]1. The mice were administered EOs according to the experimental timeline shown in Fig. [84]1B. High-dose EO supplementation effectively ameliorated hypoxia-induced disorders, including reduced food intake, weight loss, diarrhea, hematochezia, and increased gastrointestinal permeability (Fig. [85]1C–G). H&E staining revealed that hypoxia exposure led to stomach and small intestine mucosa thinning and shedding and structurally disorganized intestinal villi with injured crypts, while EO administration significantly alleviated these morphological changes (Fig. [86]1H and I). EO administration significantly reduced the elevated levels of ROS and 4-HNE in the gastric and small intestinal mucosa induced by hypoxia (Fig. [87]1J and K). Overall, EOs alleviated hypoxia-induced mucosal damage, potentially by inhibiting lipid peroxidation. Fig. 1. [88]Fig. 1 [89]Open in a new tab EO treatment effectively alleviated hypoxia-triggered gastric and small intestinal mucosal injury and lipid peroxidation. A Base peak ion flow chart of EOs. B The flowchart of the mouse experiments. Mice were exposed to either normoxia or hypoxia, which mimicked the conditions at 4,500 m above sea level, for 3 days. The mice in the EO treatment groups were orally administered EOs (1 µg, 5 µg, or 10 µg) on day 1, and euthanized on day 3 (n = 18 per group, with n = 6 in each group specifically for serum FITC-dextran detection). Each group consisted of an equal number of males and females. The food consumption (C), weight changes (D), fecal consistency score (E), and hematochezia score (F) of mice. (G) Gastrointestinal permeability of the indicated group was assessed by serum FITC-dextran concentrations. H Representative images of hematoxylin and eosin (H&E) staining of gastric and small intestinal tissue of mice (40× magnification, scale bar: 500 μm) and the corresponding histological score (I). The levels of ROS (J) and 4-HNE (K) in the stomach and small intestine. The dose of EOs corresponding to the results shown in G–K was 10 µg. Error bars denote mean ± standard errors of the mean (SEM). Two-way ANOVA was used in C–F and one-way ANOVA was performed in G, I–K. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; ns, not significant Table 1. The composition of EOs identified by GC-MS No T (min) Molecular formula Molecular weight Name CAS number Score Peak area Classification 1 3.35 C[7]H[6]O 106.04 Benzaldehyde 100-52-7 97.60 1,533,214 Aromatic compound 2 3.96 C[7]H[6]O[2] 122.04 2-hydroxybenzaldehyde 32,912 92.17 214,575 Aromatic compound 3 4.11 C[8]H[8]O 120.06 1-phenylethanone 98-86-2 93.16 824,683 Aromatic compound 4 4.74 C[9]H[10]O 134.07 1-ethenyl-4-methoxybenzene 637-69-4 93.95 757,489 Aromatic compound 5 4.82 C[9]H[10]O 134.07 Benzenepropanal 104-53-0 89.52 18,831 Aromatic compound 6 4.85 C[8]H[8]O[2] 136.05 4-Hydroxyacetophenone 99-93-4 88.07 59,502 Aromatic compound 7 5.08 C[8]H[8]O[3] 152.05 Methyl salicylate 119-36-8 90.59 61,657 Aromatic compound 8 5.46 C[10]H[12]O 148.09 4-Phenylbutan-2-one 2550-26-7 93.99 8,135,929 Aromatic compound 9 6.37 C[10]H[20]O[2] 172.15 n-Decanoic acid 334-48-5 86.09 120,517 Fatty acid 10 6.87 C[10]H[12]O[2] 164.08 m-Ethylphenyl acetate 3056-60-8 83.47 371,567 Aromatic compound 11 7.39 C[15]H[24] 204.19 α-Guaiene 654,486 90.83 7188 Sesquiterpene 12 7.91 C[15]H[22] 202.17 α-Curcumene 644-30-4 90.97 217,595 Sesquiterpene 13 8.00 C[15]H[24]O 220.18 α-Cedrene epoxide 29597-36-2 87.03 1,727,361 Sesquiterpene 14 8.17 C[11]H[14]O[2] 178.1 Anisylacetone 104-20-1 84.53 414,876 Aromatic compound 15 8.27 C[15]H[24] 204.19 Isolongifolene 1135-66-6 90.5 103,259 Sesquiterpene 16 8.38 C[15]H[24] 204.19 α-Bulnesene 3691-11-0 88.34 97,861 Sesquiterpene 17 8.44 C[15]H[26]O 222.2 Dihydroagarofuran 5956-09-2 90.44 254,880 Sesquiterpene 18 8.72 C[15]H[24]O 220.18 8,9-Epoxycedrane 29597-36-2 83.47 657,803 Sesquiterpene 19 9.1 C[15]H[26]O 222.2 Elemol 639-99-6 92.11 431,018 Sesquiterpene 20 9.77 C[15]H[24]O 220.18 α-Santalol 115-71-9 84.11 2,544,015 Sesquiterpene 21 10.06 C[15]H[24]O 220.18 trans-Longipinocarveol 889109-69-7 84.18 365,877 Sesquiterpene 22 10.56 C[15]H[24] 204.19 γ-himachalene 53111-25-4 87.29 27,420 Sesquiterpene 23 10.76 C[15]H[26]O 222.20 10-epi-γ-eudesmol 15051-81-7 95.14 1,225,261 Sesquiterpene 24 10.96 C[15]H[26]O 222.20 γ-Eudesmol 1209-71-8 94.3 1,111,510 Sesquiterpene 25 11.06 C[15]H[26]O 222.20 Guaiol 489-86-1 89.54 271,160 Sesquiterpene 26 11.17 C[15]H[24] 204.19 Longifolene 475-20-7 89.74 90,757 Sesquiterpene 27 11.32 C[15]H[24] 204.19 (-)-Aristolene 6831-16-9 90.82 1,806,731 Sesquiterpene 28 11.54 C[15]H[26]O 222.20 α-Eudesmol 473-16-5 91.16 3,167,352 Sesquiterpene 29 11.62 C[15]H[26]O 222.20 Elemol 639-99-6 87.97 448,710 Sesquiterpene 30 11.7 C[15]H[24] 204.19 β-Guaiene 88-84-6 85.31 400,239 Sesquiterpene 31 11.91 C[15]H[24]O 220.18 Aromadendrene epoxide 85760-81-2 82.22 1,069,046 Sesquiterpene 32 12.09 C[12]H[18] 162.14 1,4,4a,5,6,7-Hexahydro-2,3-dimethylnaphthalene 107914-92-1 85.48 2,866,478 Other constituent 33 12.20 C[15]H[22] 202.17 (-)-Dehydroaromadendrene 698388-95-3 85.33 381,099 Sesquiterpene 34 12.49 C[15]H[24]O[2] 236.18 1,2,3,4,4a,5,6,8a-Octahydro-6-hydroxy-4a,8-dimethyl-β-methylene-2-napht haleneethanol 1005276-32-3 87.87 86,856 Sesquiterpene 35 12.63 C[15]H[24]O 220.18 (-)-Spathulenol 77171-55-2 84.16 207,261 Sesquiterpene 36 12.68 C[15]H[28] 208.22 4-β-H,-5-α-Eremopholine 15404-63-4 79.00 2,615,816 Sesquiterpene 37 12.78 C[15]H[24]O 220.18 Aromadendrene oxide 2 85710-39-0 81.52 190,300 Sesquiterpene 38 13.56 C[15]H[24]O 220.18 Longifolenaldehyde 66537-42-6 87.86 203,394 Sesquiterpene 39 13.99 C[15]H[22]O 218.17 Nootkatone 4674-50-4 82.68 255,148 Sesquiterpene 40 14.41 C[15]H[22]O 218.17 α-Cyperone 473-08-5 88.20 321,258 Sesquiterpene 41 14.53 C[15]H[22]O 218.17 Germacrone 6902-91-6 83.08 3,850,959 Sesquiterpene 42 14.95 C[15]H[22]O 218.17 3,5,6,7,8,8a-Hexahydro-4,8a-dimethyl-6-(1-methylethenyl)-2(1 H)-naphtha lenone 725240-70-0 87.35 197,906 Sesquiterpene 43 16.32 C[15]H[22]O 218.17 Dehydrofukinone 19598-45-9 92.82 1,433,881 Sesquiterpene 44 16.63 C[16]H[24] 216.19 Tetrakis(1-methylethylidene)-Cyclobutane 88919-66-8 87.86 137,483 Other constituent 45 16.82 C[15]H[24]O[2] 236.18 Baimuxinal 86408-21-1 88.80 4196 Sesquiterpene 46 19.31 C[17]H[16]O 236.12 1,5-diphenyl-1-Penten-3-one 62510-08-1 87.61 253,229 Aromatic compound [90]Open in a new tab α- and β-Diversity of the microbiota in the gastric and small intestinal contents partially recovered following EO administration To explore whether EOs regulated the microbiota, 16 S rRNA sequencing was conducted to evaluated the microbiota composition of the stomach and small intestine. As illustrated in Fig. [91]2A and D, hypoxia exposure reduced the α-diversity of the gastric and small intestinal contents as indicated by Chao1, Shannon, and Simpson curves, whereas EOs partially restored the α-diversity. There were 538, 33, and 319 unique OTUs in the gastric microbiota in the normoxia, hypoxia, and EO treatment groups, respectively. Meanwhile, these three indicated groups possessed 78 shared OTUs (Fig. [92]2B). The small intestinal microbiota of the normoxia, hypoxia, and EO treatment groups had 173, 13, and 188 unique OTUs, respectively. Additionally, 88 shared OTUs were identified in the indicated group (Fig. [93]2E). EO treatment also partially restored the hypoxia effects on the β-diversity in the stomach and small intestine. The β-diversity, as determined by PCoA and NMDS, revealed a clear distinction in the gastric and small intestinal microbiota between the normoxia and hypoxia groups, while EO treatment reduced this separation (Fig. [94]2C and F). These results suggest that EO treatment partially restored the microbial diversity in the gastric and small intestinal contents following hypoxia exposure. Fig. 2. [95]Fig. 2 [96]Open in a new tab EO administration partially reversed the α- and β-diversity of the microbiota of the gastric and small intestinal contents. Mice were housed in a hypoxic chamber for 3 days to trigger gastric and small intestinal injury. The mice were orally administered PBS or EOs (10 µg) on day 1 and euthanized on day 3 (n = 6 per group, 3 Males and 3 females). α-Diversity analysis (Chao1, Shannon, and Simpson) of microbiota in the gastric (A) and small intestinal (D) contents. A Venn diagram displaying the unique and overlapping operational taxonomic units (OTUs) of the microbiota of the gastric (B) and small intestinal (E) contents. β-Diversity analysis of the microbiota of the gastric (C) and small intestinal (F) contents by using non-metric multidimensional scaling (NMDS) and principal component analysis (PCoA) The microbial composition of the gastric and small intestinal contents was partially restored following EO supplementation As shown in Fig. [97]3A and E, EO treatment recovered the abundances of the phyla Firmicutes, Bacteroidetes, Actinobacteria, Cyanobacteria, and Proteobacteria in the stomach that were affected by hypoxia exposure (Fig. [98]3A and E). The abundances of the genera Lactobacillus, Bifidobacterium, chloroplast-unclassified, and mitochondria-unclassified in the phyla Firmicutes, Actinobacteria, Cyanobacteria, and Proteobacteria, respectively, were also altered. Additionally, the genera Bacteroidales-unclassified and S24–7-unclassified were attributed to changes in the abundance of the phylum Bacteroidetes (Fig. [99]3C and E). In the small intestine, EOs recovered the levels of the phyla Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria, Candidate division TM7, and Verrucomicrobia that were altered by hypoxic conditions (Fig. [100]3B and F). Specifically, changes in the abundances of the genera Lactobacillus, Streptococcus, and Enterococcus altered the abundance of the phylum Firmicutes. Changes in the abundances of the genera S24–7-unclassified, Bifidobacterium, Desulfovibrio, Candidatus_Saccharimonas, and Akkermansia contributed to altered abundances of the phyla Bacteroidetes, Actinobacteria, Proteobacteria, Candidate division TM7, and Verrucomicrobia, respectively (Fig. [101]3D and F). Notably, the abundances of Candidatus_Saccharimonas and Akkermansia following EO treatment under hypoxia was higher than that in both the hypoxic control and normoxic control groups. These findings indicate that Candidatus_Saccharimonas and Akkermansia may ameliorate hypoxia-induced small intestinal mucosal injury. Overall, EOs improved the composition and abundance of microbiota affected by hypoxia in the stomach and small intestine. Fig. 3. [102]Fig. 3 [103]Open in a new tab EO supplementation partially recovered the microbiota composition of the gastric and small intestinal contents. Microbial compositional profiling of the indicated groups at the phylum and genus levels of the gastric (A and C) and small intestinal (B and D) contents. The differential microbiota of the indicated groups at the phylum and genus levels of the gastric (E) and small intestinal (F) contents (the genera corresponding to the phyla are displayed below the phyla). Error bars denote mean ± standard errors of the mean (SEM). One-way ANOVA was performed in E and F. *P < 0.05, **P < 0.01, and ***P < 0.001; ns, not significant EO administration partially recovered the metabolite levels in the gastric and small intestinal contents Levels of the following metabolites in the gastric contents improved after EO supplementation compared to those of the hypoxia group (Fig. [104]4A): glycerophosphocholine, cis-zeatin, 4-hydroxy-2-quinolone, gamma-tocotrienol, N1-acetylspermine, pelargonidin, 8-hydroxyquinoline, N-nitroso-pyrrolidine, tyrosol, 3,4-dihydroxymandelaldehyde, gemfibrozil, DHHA, deoxycholic acid, 6-keto-prostaglandin F1a, 2-phenylethanol, traumatic acid, 4,5,6,7-tetrahydroisoxazolo(5,4-c) pyridin-3-ol, sotalol, N5-methyl-L-glutamine, N-formyl-L-methionine, gluconic acid, ectoine, tetracosanoic acid, N-methylhydantoin, 1-palmitoyl-dihydroxyacetone-phosphate, aldosterone, lysoPA(16:0/0:0), gamma-L-glutamyl-D-alanine, retinoyl b-glucuronide, 3beta,5beta-ketodiol, L-lyxonate, (2R,3R)-3-methylglutamyl-5-semialdehyde-N6-lysine, paspalicine, 2-ketobutyric acid, L-arogenate, gluconolactone, palmitoyl-L-carnitine, and 3-(2-hydroxyphenyl) propanoic acid. Levels of the following metabolites in the small intestinal contents improved after EOs administration in comparison with the hypoxia group (Fig. [105]4B): xanthoxic acid, nervonic acid, guanosine, alpha-linolenoyl ethanolamide, aspartylglycosamine, and 3-ketosphingosine. Fig. 4. [106]Fig. 4 [107]Open in a new tab EO treatment promoted metabolic recovery of the gastric and small intestinal contents. Mice were housed in a hypoxic chamber for 3 days. PBS or EOs (10 µg) were administered via oral gavage on day 1 (n = 6 per group, 3 Males and 3 females). Relative abundance of altered metabolites in the gastric and small intestinal contents among the indicated groups. One-way ANOVA was conducted in A and B. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; ns, not significant KEGG pathway enrichment analysis of the metabolites improved by EOs in the gastric and small intestinal contents Key enrichment pathways and their corresponding metabolites in the gastric contents included gluconolactone and gluconic acid in the pentose phosphate pathway; aldosterone in glucocorticoid and mineralocorticoid receptor agonists/antagonists; 2-ketobutyric acid and ectoine in glycine, serine and threonine metabolism; 3-(2-hydroxyphenyl)propanoic acid and 2-phenylethanol in phenylalanine metabolism; aldosterone in aldosterone − regulated sodium reabsorption; 2-ketobutyric acid and N-formyl-L-methionine in cysteine and methionine metabolism; tyrosol and 3,4-dihydroxymandelaldehyde in tyrosine metabolism; glycerophosphocholine in choline metabolism in cancer; cis-zeatin in plant hormone signal transduction; gluconic acid and gluconolactone in carbon metabolism (Fig. [108]5A). Metabolites in the small intestinal contents showed distinct enrichment pathway, including guanosine in nucleotide metabolism, purine metabolism, and ABC transporters; nervonic acid in biosynthesis of unsaturated fatty acids; and xanthoxic acid in carotenoid biosynthesis (Fig. [109]5B). Fig. 5. [110]Fig. 5 [111]Open in a new tab KEGG pathway enrichment analysis of metabolites improved by EOs in the gastric and small intestinal contents. Enriched pathways and corresponding metabolites in the gastric (A) and small intestinal (B) contents. Hypergeometric teat was used in A and B Correlation analysis between the altered microbiota and metabolites in the gastric contents A Spearman association analysis was performed to determine the correlations between the altered microbiota and metabolites in the gastric contents. The metabolites that decreased under hypoxia and increased after EO administration were negatively correlated with the microbiota phyla Firmicutes, Acidobacteria, Cyanobacteria, and Proteobacteria, while showing a positive correlation with Candidate division TM7, Tenericutes, Bacteroidetes, and Verrucomicrobia. Genus-level analysis revealed that the microbiota Lactobacillus, Chloroplast_unclassified, and Mitochondria_unclassified correlated negatively with these metabolites, whereas Bacteroidales_unclassified and S24 − 7_unclassified showed a positive correlation (Fig. [112]6A). The metabolites showing hypoxia-elevated and EO-suppressed profiles were negatively associated with the microbiota phyla Bacteroidetes and Tenericutes, while showing a positive association with the phyla Cyanobacteria, Proteobacteria, Firmicutes, and Acidobacteria. At the genus level, these metabolites were positively associated with the microbiota Lactobacillus and Chloroplast_unclassified, whereas they were negatively associated with Bacteroidales_unclassified and S24 − 7_unclassified (Fig. [113]6B). The alteration of the above microbiota and metabolites potentially either contribute to or result from gastric mucosal healing. Fig. 6. [114]Fig. 6 [115]Open in a new tab Association analysis of altered microbiota and metabolites in the gastric contents. Association analysis between altered microbiota abundances (at the phylum and genus levels) and metabolite levels in the gastric contents, with the metabolites displaying a decrease under hypoxic conditions and an increase following EOs administration (A), and the metabolites with increased levels under hypoxic conditions and reduced levels after EO supplementation (B). *P < 0.05, **P < 0.01 Association analysis of differential microbiota and metabolites in the small intestinal contents The heatmap indicated that the metabolites for which levels decreased under hypoxic conditions and increased following EO treatment in the small intestinal contents showed positive correlations with the microbiota phyla Candidate division TM7, Tenericutes, Verrucomicrobia, Chloroflexi, Actinobacteria, and Planctomycetes. At the genus level, these metabolites were negatively associated with the microbiota Streptococcus and positively associated with Candidatus_Saccharimonas and Akkermansia. Specifically, the metabolites xanthoxic acid and guanosine positively correlated with the microbiota Candidatus_Saccharimonas and Akkermansia, respectively (Fig. [116]7A). The metabolites exhibiting hypoxia-elevated and EO-suppressed profiles displayed a negative association with the microbiota phyla Candidate division TM7 and Planctomycetes, while showing a positive association with the phylum Firmicutes. Additionally, these metabolites further showed a positive correlation with the microbiota genera Enterococcus but a negative correlation with genera Candidatus_Saccharimonas. Notably, the metabolite aspartylglycosamine was negatively associated with the microbiota Candidatus_Saccharimonas (Fig. [117]7B). Collectively, the metabolites xanthoxic acid, guanosine, and aspartylglycosamine were potentially associated with the partial recovery of hypoxia-triggered small intestinal mucosal injury. Fig. 7. [118]Fig. 7 [119]Open in a new tab Correlation analysis of differential microbiota and metabolites in the small intestinal contents. Pearson correlation analysis between differential microbiota (at the phylum and genus levels) and metabolites in the small intestinal contents, with the metabolites for which levels decreased under hypoxic conditions and increased after EO administration (A), and with the metabolites for which levels increased under hypoxia and decreased after EO supplementation (B). *P < 0.05, **P < 0.01 Discussion Agarwood is formed upon exposure to environmental stressors, including wounding and fungal infection (Li et al. [120]2021). The output of agarwood has significantly increased due to large-scale Aquilaria spp. plantations and improved artificial inducing techniques. Given the high relative abundance of sesquiterpenes (62.04%) in EOs as revealed by GC-MS analysis and their antioxidant and microbiota-modulating effects (Chen et al. [121]2022; Xie et al. [122]2024), we speculate that sesquiterpenes may be the primary bioactive components of EOs. In future studies, we will investigate which specific sesquiterpenes exert these effects. Ferroptosis, triggered by an imbalance between oxidant and antioxidant levels (Tang et al. [123]2021), was considered to be pivotal in the development of hypoxia-induced gastrointestinal damage (Wang et al. [124]2024). In this study, EOs reduced the levels of oxidants, including ROS and 4-HNE levels that were enhanced following hypoxia exposure, partially recovering the redox balance and thus alleviating ferroptosis in the gastric and small intestinal mucosa. Additionally, EO treatment reshaped the microbiota and metabolite levels in the small intestine, which were disrupted by hypoxia exposure. In this process, EOs changed the abundances of certain bacteria that exhibited the potential to repair the mucosa. EOs enhanced the abundances of Candidatus_Saccharimonas and Akkermansia, which were decreased by hypoxia exposure. Moreover, levels of these bacteria were increased by EO treatment compared to normoxic conditions. Thus, Candidatus_Saccharimonas and Akkermansia may contribute to the attenuated hypoxia-induced gastrointestinal mucosal injury. Candidatus_Saccharimonas and Akkermansia are commonly regarded as beneficial gut microbiota (Qian et al. [125]2022; Feng et al. [126]2024). β-Acetylaminohexosidase, produced by Akkermansia, was reported to upregulate the mRNA expression of tight junction proteins and remodel the gut microbiota (Qian et al. [127]2022). Certain bacteria may still undergo changes that could either contribute to or result from gastrointestinal mucosal healing. The abundances of the pernicious bacteria Streptococcus and Enterococcus in the phylum Firmicutes were partially recovered in the small intestine after EO treatment. Xiong et al. found that Enterococcus and Streptococcus were more prevalent in infants infected with human norovirus (HNoV) than in healthy infants (Xiong et al. [128]2021). Furthermore, Li et al. confirmed that Enterococcus produced tyramine via tyrosine decarboxylase that inhibited intestinal stem cell proliferation, thereby impairing epithelial regeneration (Li et al. [129]2024a). The phyla Firmicutes and Bacteroidetes constitute the most abundant microbiota in the human gut (Hidalgo-Cantabrana et al. [130]2017), and the Firmicutes/Bacteroidetes (F/B) ratio was reported to influence intestinal function (Stojanov et al. [131]2020). Additionally, Bacteroides-derived sphingolipids were found to be pivotal in maintaining gut homeostasis and symbiosis (Brown et al. [132]2019). In the current study, EOs restored the F/B ratio and the abundance of Bacteroides to some extent. Although Actinobacteria was only a small percentage of the gut microbiota, it was shown to be essential for sustaining intestinal homeostasis (Binda et al. [133]2018). Bifidobacterium, a dominant genus of the phylum Actinobacteria in this study, had a decreased abundance under hypoxic conditions and an increased abundance following EO supplementation. Notably, Bifidobacterium was shown to improve the intestinal barrier function (Krumbeck et al. [134]2018). Cyanobacteria and Proteobacteria abundances were also restored following EO treatment. Cyanobacteria was found to be abundant in HNoV-infected infants (Xiong et al. [135]2021) and colorectal adenoma patients (Lu et al. [136]2016). Preterm infants suffering from necrotizing enterocolitis (NEC) exhibited an increased abundance of Proteobacteria prior to the onset of NEC (Pammi et al. [137]2017). Desulfovibrio is the primary genus in the phylum Proteobacteria. Desulfovibrio with sulfate-reducing activity was shown to increase tight junction permeability by increasing snail protein expression and activating nuclear translocation of snail, according to a study of Singh et al. (Singh et al. [138]2022). Moreover, metabolites with levels that were reduced by hypoxia exposure and increased after EO administration have antioxidant and antimicrobial properties and can regulate mucosal blood flow. Certain metabolites enhanced by EOs have been linked to food consumption and microbial diversity. Specifically, gamma-tocotrienol, pelargonidin, traumatic acid, nervonic acid, and guanosine demonstrated potent antioxidant activities by scavenging superoxide radicals (Gudkov et al. [139]2006; Nagy et al. [140]2007; Jabłońska-Trypuć et al. [141]2016; Seo et al. [142]2020; Liu et al. [143]2021). Tyrosol, derived from olive oil, was capable of reducing ROS levels and enhancing glutathione levels (Muriana et al. [144]2017). Gemfibrozil mitigated stress conditions by enhancing superoxide dismutase and glutathione peroxidase activity and reducing malondialdehyde levels (Hakimizadeh et al. [145]2022). Ectoine was demonstrated to assist microorganisms in overcoming stress exposure (Kadam et al. [146]2024). The abundance of glycerophosphocholine was reported to decrease under oxidative stress conditions (Rosas-Rodríguez et al. [147]2010). 2-Phenylethanol, with a rose-like odor, was reported to exhibit antimicrobial properties (Zhu et al. [148]2011) that may improve the microbiota composition. Deoxycholic acid was a byproduct of intestinal metabolism, and N5-methylglutamine production was maintained by microorganisms (Villar et al. [149]2023). Decreased levels of deoxycholic acid and N5-methylglutamine may result from a reduced microbiota abundance under hypoxia. Consistently, EOs increased the abundance of the microbiota, and thus enhanced the levels of deoxycholic acid and N5-methylglutamine. 6-Keto-prostaglandin F1a served as an indicator of prostaglandin-I-2, which enhanced mucosal blood flow and maintained mucosal integrity (Konturek et al. [150]1986). In the present study, EOs enhanced the abundance of 6-keto-prostaglandin, which potentially increased mucosal blood flow and maintained mucosal integrity in the stomach. Cis-zeatin and gluconic acid, both derived from food, decreased under hypoxic conditions, which may be attributed to a reduction in food consumption under hypoxia. EOs reduced the abundance of certain metabolites with oxidative properties. Levels of palmitoyl-L-carnitine, which were reduced by EOs, have been linked to enhanced oxidative phosphorylation rates in mitochondria (Lenartowicz et al. [151]1976). Overall, EOs increased the abundance of metabolites with antioxidant properties and decreased the abundance of metabolites with oxidative properties. The metabolites improved by EOs were significantly enriched in the pentose phosphate pathway, glucocorticoid and mineralocorticoid receptor agonists/antagonists, nucleotide metabolism, and biosynthesis of unsaturated fatty acids. The pentose phosphate pathway Supplies ribose 5-phosphate and nicotinamide adenine dinucleotide phosphate (NADPH) for nucleotide synthesis and redox homoeostasis, respectively (TeSlaa et al. [152]2023). Mineralocorticoid receptors regulate fluid and electrolyte homeostasis (Agarwal et al. [153]2021), while glucocorticoid receptors modulate energy homeostasis and stress responses (Gomez-Sanchez et al. [154]2014). Nucleotide synthesis is crucial for cell proliferation (Cao et al. [155]2023), and unsaturated fatty acids provide efficient energy substrates. Collectively, these metabolic alterations may mitigate hypoxia-induced gastric and small intestinal mucosa injury by suppressing oxidative stress, accelerating mucosal repair, and modulating energy metabolism. An association analysis between the microbiota and metabolites revealed that Candidatus_Saccharimonas positively correlated with xanthoxic acid levels and negatively associated with aspartylglycosamine levels, while Akkermansia positively correlated with guanosine levels. Candidatus_Saccharimonas was shown to provide therapeutic effects, potentially by enhancing levels of xanthoxic acid, an apo carotenoid sesquiterpenoid characterized by its antimicrobial and antioxidant properties (PubChem [156]2025). Aspartylglycosamine was the only compound significantly elevated in patients with N-glycanase 1 (NGLY1) deficiency (Haijes et al. [157]2019). NGLY1 Suppressed ferroptosis by enhancing the level of glutathione peroxidase 4 (Forcina et al. [158]2022). The reduction of aspartylglycosamine levels after EO administration may represent the ability of EOs to prevent ferroptosis through NGLY1; notably, Candidatus_Saccharimonas potentially upregulated NGLY1 to resist ferroptosis after EO supplementation. Akkermansia relieved hypoxia-induced mucosal injury, potentially by upregulating the antioxidant guanosine. In conclusion, EOs effectively ameliorated hypoxia-induced gastric and small intestinal injury, potentially by inhibiting lipid peroxidation and improving the microbiota and metabolite levels. EOs may be regarded as an alternative or supplementary approach for treating hypoxia-induced damage of the gastric and small intestinal mucosa. Acknowledgements