Abstract Hashimoto’s Thyroiditis (HT) is an autoimmune disorder characterized by metabolic disturbances. However, a comprehensive metabolic and lipid profile of HT has not been reported. The metabolic and lipid profile of HT can be comprehensively analyzed through metabolomics and lipidomics technologies, providing a detailed understanding of the disease’s biochemical alterations. Plasma samples were obtained from 20 mice, comprising 10 from the control group and 10 from the HT group. Plasma metabolomics and lipidomics were analyzed via LC-MS/MS. PCA, PLS-DA, and OPLS-DA of the preprocessed data matrix were conducted using the ropls package (V.1.6.2) in R. The identification of significant differential metabolites was based on the VIP scores in the OPLS-DA model and p-values (Student’s t-test). Metabolites (VIP exceeding 1 and p < 0.05) were classified as significantly different. Pathway annotation of these metabolites was carried out using the KEGG database to identify associated metabolic pathways. Pathway enrichment analysis was performed using the Python scipy. Stats package, with Fisher’s exact test employed to identify biological pathways most relevant to the experimental conditions. Metabolomics identified 6384 metabolites (2943 in positive ion mode and 3441 in negative), with 195 differential metabolites, comprising 114 upregulated and 81 downregulated in the HT group. Lipidomics analysis revealed 1054 lipid metabolites (695 detected in positive ion mode and 359 in negative), and 247 differentially expressed lipids were identified, including 165 upregulated and 82 downregulated in the HT group. KEGG enrichment analysis indicated that metabolites upregulated in the HT group were primarily associated with pathways such as Autophagy, Choline metabolism, PPAR signaling, and Glycerophospholipid metabolism. In contrast, pathways involved in Apoptosis, Cholesterol metabolism, Sphingolipid metabolism, EGFR tyrosine kinase inhibitor resistance, Th1 and Th2 cell differentiation, viral infection, and Chemokine signaling were suppressed. Metabolic and lipidomic dysregulation was observed in HT animal models, with pronounced alterations in Phospholipids, Eicosanoids, and Carnitines. Choline, Glycerophospholipid, and Linoleic acid metabolism pathways exhibited significant enrichment in HT. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-025-07905-7. Keywords: Hashimoto’s thyroiditis, Metabolomics, Lipidomics, LC-MS/MS, Eicosanoids, Phospholipids, Carnitines Subject terms: Lipidomics, Metabolomics Introduction Hashimoto’s thyroiditis (HT), an autoimmune disorder, is a primary cause of hypothyroidism^[34]1. Characterized histologically by lymphoplasmacytic infiltration, germinal center development within lymphoid follicles, and parenchymal atrophy, its diagnosis is established through clinical assessment, serological detection of thyroid antibodies (thyroid peroxidase and thyroglobulin), and cytological evidence of lymphocytic infiltration. Standard treatment focuses on thyroid hormone replacement therapy for managing hypothyroidism^[35]2. While the precise etiology remains unclear, HT is thought to result from the interplay of genetic predispositions, environmental exposures, and epigenetic modifications^[36]3. Epidemiological data indicate an incidence rate exceeding 10%, with a higher prevalence among middle-aged women, typically presenting between 30 and 50 years of age. Approximately 20% of patients progress to symptomatic hypothyroidism^[37]4,[38]5. A 2020 survey conducted across 31 Chinese provinces reported a 14.19% prevalence of thyroid antibody positivity, with significantly higher rates in women compared to men (20.35% vs. 8.16%)^[39]6. Metabolic alterations are typically present in HT, as multi-omics analysis by Zheng et al.^[40]7 identified a link between dysregulated gene expression in HT and Graves’ Disease with disrupted metabolic pathways. Specifically, changes were observed in phenylalanine, tryptophan, and tyrosine metabolism. Lipids play a vital role in maintaining cellular membrane integrity, regulating energy homeostasis, and participating in key biological pathways. Both thyroid dysfunction and inflammatory processes have significant effects on plasma lipoprotein metabolism. Lipid metabolism disorders frequently co-occur with HT, with evidence suggesting that lipid profile deterioration correlates with the progression of hypothyroidism^[41]8. Certain phospholipids, including Sphingomyelin (SM), Lysophosphatidylcholine (LPC), and Phosphatidylcholine (PC), have been implicated in HT onset, while pathways related to fatty acid and lysine degradation show variation across clinical stages of HT^[42]5. Furthermore, the analysis of small-molecule metabolites in autoimmune thyroid disease reveals altered serum concentrations of lipids and amino acids, such as PC, LPC, SM, and Glycodeoxycholic Acid (GDCA)^[43]9. The robust HT animal models must accurately replicate both serological markers and pathological features found in human patients, with model development closely aligned to the disease progression seen in humans. Among the few organ-specific autoimmune disease models, Nonobese Diabetic (NOD) mice, particularly the NOD.h-2h4 variant, stand out. These mice, which express H-2KK and I-AK genes on a NOD-like genetic background^[44]10, do not manifest diabetes but naturally develop autoimmune thyroiditis. The modeling of NOD.h-2h4 mice via high-iodine water exposure offers a straightforward and consistently reproducible approach. HE staining revealed that the extent of thyroid lymphocyte infiltration escalated as the duration of iodine exposure increased, leading to a progressive rise in HT incidence. Within approximately 8 weeks, the model stabilized and was suitable for probing HT pathogenesis and evaluating treatments from both traditional Chinese and Western medicine. Research demonstrates that administering drinking water containing 0.05% sodium iodide (NAI) over a 6–8 week period induces an HT incidence rate in NOD.h-2h4 mice nearing 100%^[45]11. Metabolomics, an emerging omics discipline following genomics, transcriptomics, and proteomics, has advanced significantly in recent years, offering promising avenues for identifying specific and sensitive biomarkers for HT disease. This study aims to evaluate the abnormalities of HT plasma metabolites and lipids by LC-MS/MS non-targeted metabolomics and targeted lipidomics, providing insights into the changes in key metabolic pathways and lipid regulation associated with HT. Methods Experimental animals NOD.H-2h4 mice were obtained from Cyagen Biotechnology Co., Ltd. and housed at the Animal Core Facility of Nanjing Medical University. Ethical approval for this study was granted by the Laboratory Animal Ethics Committee of Nanjing Medical University (K-2024–006-K01). This study confirmed that all methods were carried out under relevant guidelines and regulations. All methods were carried out in the 20 items of the ARRIVE (Animals in Research In Vivo Experiments) guidelines. Twenty NOD.H-2h4 mice (4 weeks old, equally distributed by sex) were randomly assigned to a control group and an HT group. The control group received sterile water, while the HT group was administered sterile water containing 0.05% sodium iodide for 8 weeks. The two groups of mice were placed in a small animal anesthesia machine and inhaled 3% isoflurane. After 3 min, when the mice were fully anesthetized, they were euthanized by cervical dislocation and sampled. Hematoxylin-eosin (HE) staining of mouse thyroid tissue Fresh thyroid tissues from mice were collected and fixed in EP tubes containing 4% paraformaldehyde. Tissue dehydration was performed through a graded alcohol series, followed by transparency treatment in xylene: absolute ethyl alcohol (1:1) solution and xylene transparent sample for 30 min per step. Samples were then fully immersed in paraffin at 60 °C for 2 h. The paraffin-embedded tissue blocks were prepared using an embedding machine, cooled for solidification, and trimmed before being mounted on a microtome. Continuous sections of 5 μm thickness were cut and floated on 42 °C water before being mounted onto slides. After drying at 40 °C, sections were subjected to HE staining. The slides were immersed in xylene and ethanol at decreasing concentrations for 10 min each. Subsequent steps involved xylene:100% ethanol (1:1), 100% ethanol, 50% ethanol, and distilled water for 2 min each, followed by hematoxylin staining for 5–10 min, water rinsing, and differentiation in 1% hydrochloric acid ethanol for 5 s. After washing in running water for 10 min, sections were treated with 1% ammonia for blueing, followed by successive immersions in 50%, 70%, 80%, 90%, and 95% ethanol for 2 min each. Eosin staining was performed for 5 min, followed by dehydration in 95% ethanol, 100% ethanol I and II, xylene:100% ethanol (1:1), and xylene I for 2 min each. Finally, the sections were sealed using neutral resin. Assay of TgAb and TPOAb antibodies by ELISA The samples stored at -80℃ were retrieved to allow them to equilibrate to room temperature before conducting the Enzyme-Linked Immunosorbent Assay (ELISA) assay. The washing buffer was prepared as per the anti-Thyroglobulin Antibodies (TgAb) and anti-Thyroid Peroxidase Autoantibody (TPOAb) kit instructions. All reagents should reach room temperature (18–25℃) for a minimum of 30 min prior to use. The ELISA plate was removed to allocate duplicate wells for each standard point, and 50 µl of the appropriate standard was added to each. And 50 µl of the test sample was introduced directly into the remaining wells. The 50 µl of biotin-conjugated marker was added to all wells except the blank control; then, mix thoroughly, cover with an adhesive seal, and incubate at 37℃ for 1 h. Perform manual plate washing, discarding the well contents, then fill each well with washing solution, allow it to stand for 10 s, decant, repeat three times, and dry thoroughly. Then, 50 µl of horseradish peroxidase-conjugated avidin was added to each well, excluding the blank control, mix well, reseal, and incubate at 37℃ for 30 min. Repeat the manual washing steps. Following this, add 50 µl each of color developers A and B to every well, shake gently to mix, incubate at 37℃ in darkness for 15 min, and terminate the reaction with 50 µl of stop solution. OD at 450 nm was measured using a microplate reader. Sample preparation Metabolomics samples The 100 µL plasma sample was added to a 1.5 mL centrifuge tube with 400 µL solution (acetonitrile: methanol = 1:1(v: v)) containing 0.02 mg/mL internal standard (L-2-chlorophenylalanine) to extract metabolites. The samples were mixed by vortex for 30 s and low-temperature sonicated for 30 min (5 °C, 40 kHz). The samples were placed at -20 °C for 30 min to precipitate the proteins. Then the samples were centrifuged for 15 min (4 °C, 13000 g). The supernatant was removed and blown dry under nitrogen. The sample was then re-solubilized with 100 µL solution (acetonitrile: water = 1:1) and extracted by low-temperature ultrasonication for 5 min (5 °C, 40 kHz), followed by centrifugation at 13,000 g and 4 °C for 10 min. The supernatant was transferred to sample vials for LC-MS/MS analysis. Lipidomics samples For plasma extraction, 200 µL of liquid sample was precisely pipetted into an EP tube, 80 µL of methanol and 400 µL of Methyl Tert-Butyl Ether (MTBE) were added, vortexed, and mixed for 30 s, and then extracted by ultrasonication at 5 ℃ and 40 kHz for 30 min. Subsequently, the samples were allowed to stand for 30 min at -20 °C and then centrifuged at 13,000 g for 15 min at 4 °C. A total of 350µL of the supernatant was taken and dried in a vacuum concentrator, and re-dissolved by adding 100 µL of the reagent solution (isopropanol: acetonitrile = 1:1). After vortex mixing for 30 s, the sample was sonicated at 40 kHz for 5 min in an ice-water bath. Extracted lipids were spun for 15 min at 13,000 g at 4 °C on a bench-top centrifuge and cleared supernatant was transferred to sample vials, then 2µL portions of each sample were injected into the UHPLC-MS/MS system. UHPLC-MS/MS analysis Metabolomics The LC-MS/MS analysis of the sample was conducted on a Thermo UHPLC-Q Exactive HF-X system equipped with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm; Waters, USA) at Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The mobile phases consisted of 0.1% formic acid in water: acetonitrile (95:5, v/v) (solvent A) and 0.1% formic acid in acetonitrile: isopropanol: water (47.5:47.5, v/v) (solvent B). The flow rate was 0.40 mL/min, and the column temperature was 40℃. The mass spectrometric data were collected using a Thermo UHPLC-Q Exactive HF-X Mass Spectrometer equipped with an electrospray ionization source operating in positive mode and negative mode. The optimal conditions were set as follows: source temperature at 425℃; sheath gas flow rate at 50 arb; Aux gas flow rate at 13 arb; Ion-Spray Voltage Floating (ISVF) at -3500 V in negative mode and 3500 V in positive mode, respectively; Normalized collision energy, 20-40-60 V rolling for MS/MS. Full MS resolution was 60,000, and MS/MS resolution was 7500. Data acquisition was performed with the Data Dependent Acquisition (DDA) mode. The detection was carried out over a mass range of 70–1050 m/z. Lipidomics The LC-MS/MS analysis of the sample was conducted on a Thermo UHPLC-Q Exactive HF-X Vanquish Horizon system equipped with an Accucore C30 column (100 mm × 2.1 mm i.d., 2.6 μm; Thermo, USA) at Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The mobile phases consisted of 10 mM ammonium acetate in ACN: H2O (1:1, v/v) (0.1% (v/v) formic acid) (solvent A), and 2 mM ammonium acetate in ACN: IPA: H2O (10:88:2, v/v/v) (0.02% (v/v) formic acid). The standard sample injection parameters were: volume: 2 µL, flow rate: 0.4 mL/min, and column temperature: 40 ℃, 20 min of total chromatographic separation. The solvent gradient changed according to the following conditions: linear gradient from 35–60%B, 0–4 min; from 60 to 85% B, 4–12 min; from 85 to 100% B, 12–15 min; hold at 100% B, 15–17 min; from100–35% B, 17–18 min; hold at 35% B, until the end of separation. During the analysis, all these samples were stored at 4 °C. The mass spectrometric data were collected using a Thermo UHPLC-Q-Exactive HF-X Benchtop Orbitrap Mass Spectrometer equipped with a heated-electrospray ionization source operating in positive and negative ion mode. The optimal conditions were set as follows: Sheath gas flow rate at 60 psi; Aus gas flow rate at 20 psi; Aus gas heater temperature at 37 ℃; ISVF at -3000 V in negative mode and at + 3000 V in positive mode, respectively; Normalized collision energy, 20-40-60 V rolling for MS/MS. Data acquisition was performed with the DDA mode. The detection was carried out over a mass range of m/z 200–2000. Quality control sample As a part of the system conditioning and quality control process, a pooled Quality Control (QC) sample was prepared by mixing equal volumes of all samples. The QC samples were disposed of and tested in the same manner as the analytical samples. It helped to represent the whole sample set, which would be injected at regular intervals (every 10 samples) to monitor the stability of the analysis. Metabolomics substance identification and analysis Following completion of the computational process, import LC-MS raw data to Progenesis QI (Waters Corporation, Milford, USA) for baseline filtration, peak recognition, integration, Retention Time (RT) alignment, and peak alignment. This yielded a data matrix consisting of RT, mass-to-charge ratio, and peak intensity. Concurrently, MS and MS/MS spectral data were matched against public metabolic databases, including Human Metabolome Database (HMDB) ([46]http://www.hmdb.ca/)^[47]12 and Metlin ([48]https://metlin.scripps.edu/)^[49]13, along with Majorbio’s proprietary database, to identify metabolite information. Then, upload the data matrix to the platform (cloud.majorbio.com) for further analysis. Preprocessing of the matrix followed specific steps: the data matrix applied the 80% threshold to address missing values, retaining variables with non-zero values in at least 80% of one or more sample sets. Missing values were imputed by replacing them with the smallest value from the original matrix. To minimize errors from sample preparation and instrumental fluctuations, mass spectrometry peak intensities were normalized using the sum-normalization method, resulting in a standardized data matrix. Then, variables from QC samples with Relative Standard Deviation (RSD) exceeding 30% were removed; then, perform log10 transformation was performed to generate the final data matrix for later analysis. Lipidomics data preprocessing and annotation After UPLC-MS/MS analyses, the raw data were imported into the LipidSearch (Thermo, CA) for peak detection, alignment and identification. The lipids were identified by MS/MS fragments, Mass tolerance for precursor and fragment was both set to 10 ppm. The displayed m-score threshold was set as 2.0 and grades A, B ,C, D were all used for ID quality filter. The preprocessing results generated a data matrix that consisted of the lipid class, RT, mass-to-charge ratio(m/z) values, and peak intensity. The data were analyzed through the free online platform of majorbio choud platform. Lipidomic features detected at least 80% in anyset of samples were retained. After filtering, minimum metabolite values were imputed for specific samples in which the metabolite levels fell below the lower limit of quantitation and each Metabolic feature was normalized by sum. To reduce the errors caused by sample preparation and instrument instability, the response intensity of the sample mass spectrum peaks was normalized by the sum normalization method, and the normalized data matrix was obtained. At the same time, variables with RSD > 30% of QC samples were removed, and logl0 logarithmization was performed to obtain the final data matrix for subsequent analysis. Statistical analysis The ropls package (V 1.6.2) in R was employed for Principle Components Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA), and Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) (OPLS-DA) on the preprocessed data matrix, followed by seven rounds of interactive validation to evaluate model robustness. PLS-DA was used to analyze the overall difference between the two groups, and OPLS-DA was used to analyze the VIP value of metabolites and Fold change and p-value in univariate analysis to screen differential metabolites. The identification of significant differential metabolites was based on the VIP score derived from the OPLS-DA model, combined with p-values from the Student’s t-test. Metabolites with VIP > 1 and p < 0.05 were classified as significantly different. Differential metabolites were mapped to metabolic pathways using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database ([50]https://www.kegg.jp/kegg/pathway.html)^[51]14–[52]16 to determine their involvement in specific pathways. Pathway enrichment analysis was conducted using the Python package scipy. Stats ([53]https://docs.scipy.org/doc/scipy/), applying Fisher’s exact test to determine the biological pathways most associated with the experimental conditions. PCA, PLS-DA, OPLS-DA, KEGG pathway enrichment, and Visualization of data were performed using the free online platform of majorbio choud platform (cloud.majorbio.com)^[54]17. Results Mouse HT model Figure [55]1A and B illustrated elevated serum levels of thyroid autoantibodies (anti-TPO and anti-Tg) in the HT group compared to the control group. Histopathological analysis revealed intact and homogeneously distributed thyroid follicles in the control group (Fig. [56]1C-E), with minimal lymphocyte infiltration. Conversely, the HT group (Fig. [57]1F-H) exhibited significant follicular destruction accompanied by extensive lymphocytic infiltration within the thyroid tissue. These observations confirm the successful establishment of the HT mouse model in this study. Fig. 1. [58]Fig. 1 [59]Open in a new tab Plasma thyroid antibodies and thyroid tissue morphology of HT mice. (A,B) Comparison of plasma thyroglobulin and thyroid peroxidase antibody concentrations between HT and control mice by t-test. (C–E) HE staining of thyroid glands of control mice. (F–H) HE staining of thyroid glands of HT mice. * P < 0.05, *** P < 0.001. scale bar: 100 μm. HT metabolic profile A total of 6384 ion peaks (2943 in positive ion mode and 3441 in negative) were identified through non-targeted high-resolution metabolomics analysis of plasma samples from both mouse groups. After comparison with the database and excluding metabolites with over 20% missing values, 698 metabolites remained for statistical evaluation (377 in positive ion mode and 321 in negative) (Supplementary Table 1). The quality of the data was assessed using QC sample evaluation plots for both ion modes, as shown in Fig. [60]2A and B. The RSD of the QC samples was < 0.3, and the cumulative peak proportion exceeded 70%, confirming the reliability of the data. Subsequently, PLS-DA was applied to identify differential metabolomic profiles between the HT and control groups across both ionization modes. The score plots revealed a clear distinction between the HT and control groups (Fig. [61]2C and E), with R2 = 0.9213, Q2 = -0.0558 in the positive ion mode (Fig. [62]2D), and R2 = 0.8527, Q2 = -0.2302 in the negative ion mode (Fig. [63]2F). Fig. 2. [64]Fig. 2 [65]Open in a new tab Metabolomics QC sample evaluation graphs in positive and negative ion modes, and PLS-DA score graphs for the control group and HT group. (A,B) QC sample evaluation graphs in positive and negative ion modes, respectively; the horizontal axis is the RSD (%) value, i.e., standard deviation/mean, and the vertical axis is the cumulative proportion of ion peaks. (C) PLS-DA score graphs for the control group and HT group in positive ion mode. (D) 7-fold cross-validation and response permutation tests of PLS-DA in positive ion mode. (E) PLS-DA score graphs for the control group and HT group in negative ion mode. (F) 7-fold cross-validation and response permutation tests of PLS-DA in negative ion mode. Following validation of the overall data quality, metabolite analysis was conducted for both the HT and control groups. HMDB served as the primary reference for metabolomics, recognized for its comprehensive metabolite identification and query resources in human metabolic studies. Figure [66]3A illustrated the secondary classification of all identified metabolites from HMDB, revealing that over 80% of the identified metabolites belonged to categories such as Lipids and lipid-like molecules, Organic acids and derivatives, Organoheterocyclic compounds, and Benzenoids. A Venn diagram (Fig. [67]3B) displayed the distribution of metabolites between the groups: 2 metabolites were unique to the HT group, 5 to the control group, with 691 metabolites shared between both. Fig. 3. [68]Fig. 3 [69]Open in a new tab Visualization of the overall profile of plasma metabolites detected in the HT group and the control group. (A) Percentages of metabolites identified in HMDB superclass according to HMDB chemical taxonomy. (B) Venn diagram showing the number of metabolites overlapping between the HT group and the control group. (C) Volcano plot showing differential plasma metabolites detected in HT and control samples, with up-regulated and down-regulated metabolites indicated in red and blue, respectively. (D) Violin plot showing the top 1–5 metabolites with the highest significant differences between the two groups. (E) Violin plot showing the top 6–10 metabolites with the highest significant differences between the two groups. *** P < 0.001. Differential metabolites of HT and control Due to the high-dimensional and large-scale nature of metabolomics data, statistical methods were employed to identify differential metabolites between the two biological groups. Metabolite selection was based on VIP values from OPLS-DA (Supplementary Fig. 1) analysis, along with Fold change and p-value derived from univariate analysis, visualized through a volcano plot (Fig. [70]3C). The selection criteria included Fold Change > 1, VIP ≥ 1 in the OPLS-DA model, and p-value < 0.05, yielding 195 differential metabolites. Of these, 114 were upregulated and 81 downregulated in the HT group. The 10 most significant differential metabolites between the groups were (1 S)-(-)-Camphanic Acid, (5R)-5-Hydroxyhexanoic acid, TetraHCA, 15-hydroxyicosanoic acid, LysoPC(18:3(6Z, 9Z,12Z)/0:0), 3-Hydroxyoctadecanoylcarnitine, Pipecolic Acid, Palmitoylcarnitine, 2,2’-(3-methylcyclohexane-1,1-diyl)dacetic acid, and MG(0:0/20:4(5Z, 8Z,11Z,14Z)/0:0) (Fig. [71]3D-E). Notably, 8 metabolites were upregulated in the HT group, while 2 showed lower expression. These metabolites were involved in pathways related to lipid metabolism, amino acid metabolism, signaling molecules and interactions, and nervous system function. Cluster analysis of differential metabolites Metabolites exhibiting similar expression patterns are often functionally correlated. Clustering differentially expressed metabolites enables the visualization of expression trends across different groups, identifying those metabolites significantly up-or down-regulated in distinct experimental conditions. In Fig. [72]4A, cluster analysis of plasma differential metabolites from the HT and control groups yielded 10 subclusters. Among these, subcluster 6 comprised 10 metabolites consistently overexpressed in the HT group (Fig. [73]4B), specifically including 1,2-Dipalmitoylphosphatidylcholine, 3-Hydroxyoctadecanoylcarnitine, Palmitoylcarnitine, Tetradecanoylcarnitine, Hydroxytetradecadienyl-l-carnitine, Hydroxyoctadecenoylcarnitine, O-Acetylcarnitine, Indolelactic acid, TetraHCA, and 3-Hydroxybutanoic acid (Fig. [74]4C). Notably, the majority of metabolites in subcluster 6 were carnitines, integral to the fatty acid metabolism pathway. Fig. 4. [75]Fig. 4 [76]Open in a new tab Cluster analysis of differential metabolites between the HT group and the control group. (A) Cluster analysis of the top 50 plasma differential metabolites between the HT group and the control group. (B) Line chart showing subcluster 6 with uniformly high expression in HT. (C) Box plot showing the metabolite expression levels in subcluster 6. *** P < 0.001. Pathway enrichment analysis of differential metabolites identified HT metabolic dysregulation and abnormal metabolic signals Previously, 114 metabolites were identified as upregulated in the HT group, while 81 were downregulated. These metabolites are strongly associated with functional disruptions, such as alterations in lipid, amino acid, and bile acid metabolism caused by disease infection. To elucidate the roles of these differential metabolites, relevant metabolic pathways were mapped. Figure [77]5A and B presented the KEGG pathway enrichment analysis of the metabolites in HT and control groups, using histograms to depict enrichment levels. The y-axis displayed the ratio of enriched metabolite number to the background number annotated in each pathway. A higher ratio indicated a stronger enrichment. The color gradient of the bars reflected enrichment significance, with darker colors representing greater significance. Metabolites overexpressed in the HT group were predominantly enriched in pathways such as Autophagy, Choline metabolism in cancer, PPAR signaling, and Glycerophospholipid (GP) metabolism. In contrast, metabolites with lower expression in the HT group were primarily involved in Apoptosis, Cholesterol metabolism, Sphingolipid metabolism, and Sphingolipid signaling. Figure [78]5C and D illustrated the network relationships between metabolites and pathways through KEGG network diagrams. Figure [79]5C represented the upregulated metabolites in HT, while Fig. [80]5D focused on downregulated ones. Green triangular nodes denoted metabolites, and blue circular nodes represented KEGG pathways, with the size of the circles indicating the metabolite number within each pathway. This network visualization clearly revealed the interactions between enriched pathways and their associated metabolites. For instance, GP metabolism was linked to metabolites such as Glycerophosphocholine, Glycerylphosphorylcholine, and 1,2-Dipalmitoylphosphatidylcholine. Fig. 5. [81]Fig. 5 [82]Open in a new tab KEGG pathway analysis of significantly different metabolites between the HT group and the control group. (A) The bar graph shows the KEGG enrichment pathways of metabolites upregulated in the HT group. (B) The bar graph shows the KEGG enrichment pathways of metabolites downregulated in the HT group. The vertical axis Enrichment ratio represents the ratio of the number of enriched metabolites (Mebolite number) to the number of metabolites (Background number) annotated to the pathway. The column color gradient represents the significance of enrichment. The darker the color, the more significantly enriched the KEGG entry. (C) The KEGG network diagram shows the relationship between the pathway and the metabolites upregulated in the HT group in the form of a network. (D) The KEGG network diagram shows the relationship between the pathway and the metabolites downregulated in the HT group in the form of a network. Green triangular nodes represent metabolites; other circular nodes of different colors and sizes represent KEGG pathways. The color represents the P value, and the size represents the number of metabolites or genes in the pathway. HT lipid profile This study analyzed the plasma lipidomics of HT mice, identifying 1054 lipid metabolites, with 695 detected in positive ion mode and 359 in negative ion mode. Following data preprocessing, which excluded lipid metabolites with over 20% missing values, 604 metabolites remained in positive ion mode and 328 in negative ion mode (Supplementary Table 2). Figure [83]6A and B presented the RSD evaluation of QC samples in both ionization modes. Results indicated RSD values below 0.3 for both modes, with a cumulative peak proportion exceeding 70%, confirming data reliability (dotted lines represented values before preprocessing; solid lines, after preprocessing; raw data was shown with a single solid line). The PLS-DA model was then employed to assess sample differences between the HT and control groups. Score plots in Fig. [84]6C-F revealed distinct separation between these groups, with R2 = 0.9953 and Q2 = 0.3877 in positive ion mode, and R2 = 0.9458 and Q2 = -0.0019 in negative ion mode. Additionally, lipidomics PCA and OPLS-DA models are also provided (Supplementary Fig. 2). Fig. 6. [85]Fig. 6 [86]Open in a new tab Evaluation graphs of lipid group QC samples in positive and negative ion modes, and PLS-DA score graphs of control and HT groups. (A,B) Evaluation graphs of QC samples in positive and negative ion modes; the horizontal axis is the RSD (%) value, i.e., standard deviation/mean, and the vertical axis is the cumulative proportion of ion peaks. (C) PLS-DA score graphs of control and HT groups in positive ion mode. (D) 7-fold cross-validation and response permutation tests of PLS-DA in positive ion mode. (E) PLS-DA score graphs of control and HT groups in negative ion mode. (F) 7-fold cross-validation and response permutation tests of PLS-DA in negative ion mode. Differential lipid metabolites of HT and control The volcano plot in Fig. [87]7A illustrated the differential expression of lipid metabolites between the HT and Con groups. Based on the criteria of Fold Change > 1, VIP ≥ 1, and p-value < 0.05, 247 lipid metabolites were identified as differentially expressed. Of these, 165 metabolites were upregulated, while 82 were downregulated in the HT group. These metabolites were further categorized, and the classification along with the differential expression levels was visualized using a scatter plot (Fig. [88]7B). Each point represented a lipid metabolite, with the x-axis depicting the fold change in expression between groups and the y-axis indicating the lipid subclass. Different colors corresponded to distinct lipid classifications. The analysis revealed that Triglyceride (TG) exhibited the most significant upregulation, while Phosphatidylserine (PS) showed the most pronounced downregulation in the HT group. A heat map was employed to highlight the top 30 metabolites with the highest VIP scores (Fig. [89]7C). The data demonstrated that PS (18:2/22:6) experienced the largest downregulation in the HT group, with a fold change of 4. Several TG metabolites, including TG(20:1/18:2/21:0), TG(18:2/12:1/22:6), TG(16:0e/18:1/20:1), TG(20:0e/17:0/18:1), TG(16:0e/18:1/18:2), TG(18:2e/18:2/24:0), TG(18:0e/16:0/18:1), TG(15:0/16:1/18:2), and TG(16:0e/16:0/18:2), were significantly upregulated in the HT group, all exceeding a two-fold increase. Fig. 7. [90]Fig. 7 [91]Open in a new tab Comparison of differential lipid expression between the HT group and the control group. (A) The volcano plot shows the differential plasma lipid metabolites between the HT group and the control group, with up-regulated and down-regulated lipid metabolites represented in red and green, respectively. (B) The scatter plot shows the classification results and differential status of differential lipid metabolites; different dots represent a metabolite, the abscissa is the fold change value of the difference in expression of lipid metabolites between the two groups, and the ordinate represents the subclass classification of lipids. (C) The cluster heat map and VIP bar chart show the expression patterns and metabolites of differential lipid metabolites in each sample; the left side is the metabolite cluster dendrogram, and the right side is the metabolite VIP bar chart. * P < 0.05, ** P < 0.01, *** P < 0.001. Pathway enrichment analysis of differential lipid metabolites Histogram in Fig. [92]8 illustrated the enrichment analysis of lipid metabolite pathways differentially expressed between the HT and control groups. Figure [93]8A highlighted up-regulated lipid metabolites, while Fig. [94]8B presented down-regulated lipid metabolites. The x-axis labeled the pathway names, and the y-axis denotes the enrichment rate, calculated as the ratio of metabolite number enriched in a given pathway to the background number annotated in that pathway. A higher ratio reflected a stronger degree of enrichment. The color gradient of the bars signified the statistical significance of pathway enrichment, with darker shades indicating higher significance. Results revealed that metabolic pathways such as Choline metabolism, Thermogenesis, and Insulin resistance were up-regulated in HT mice. In contrast, pathways including Sphingolipid metabolism, EGFR tyrosine kinase inhibitor resistance, Th1 and Th2 cell differentiation, viral infection, and Chemokine signaling were down-regulated. Fig. 8. [95]Fig. 8 [96]Open in a new tab KEGG pathway analysis of lipid metabolites with significant differences between the HT and the control group. (A) The bar graph shows the KEGG enriched pathways of lipid metabolites that were upregulated in the HT group. (B) The bar graph of the KEGG enriched pathways of lipid metabolites that were downregulated in the HT group. The vertical axis enrichment ratio represents the ratio of the number of enriched metabolites to the number of metabolites annotated to the pathway. The column color gradient represents the significance of enrichment. The darker the color, the more significantly enriched the KEGG entry. Discussion The study sought to examine the potential involvement of circulating metabolites in the inflammatory processes associated with Systemic Sclerosis (SSc). Significant differences in plasma metabolites and lipids were observed between HT mice and controls, with 195 altered metabolites and 247 altered lipids identified in the HT group. Notably, the majority of the differential metabolites in HT showed increased concentrations, regardless of the metabolic or lipid panel. Lipid analysis revealed the most pronounced alterations in HT group, particularly in Phospholipids, Eicosanoids, and Fatty acids. Further examination of lipid metabolites highlighted substantial variations in TG and PE classes, particularly involving Glyceride (GL) and GP. A recent study has indicated elevated levels of phospholipids, including SM, LPC, and PC, in the serum of HT patients, aligning with the present findings. As fundamental components of cellular membranes, phospholipid metabolism and transport disruptions are strongly associated with autoimmune disorders. Research has demonstrated that the accumulation of oxidized phospholipids can trigger systemic immune dysregulation, potentially contributing to the onset of autoimmune diseases^[97]18. Clinically, antiphospholipid syndrome is an autoimmune disorder where the immune system mistakenly targets proteins bound to phospholipids^[98]14. In patients with autoimmune thyroid disease, the prevalence of antiphospholipid antibodies is significantly higher compared to healthy controls, reaching up to 43%^[99]19. Eicosanoids, bioactive oxygenated polyunsaturated fatty acids containing 20 carbons derived from arachidonic acid and similar polyunsaturated fatty acids, play a regulatory role in inflammatory processes linked to various diseases^[100]20. The inflammatory effects of eicosanoids, however, remain context-dependent. Research by Jakob et al. has predominantly classified eicosanoid signaling as pro-inflammatory, particularly within the innate immune response. Prostaglandins synthesized via cyclooxygenase 1 during inflammasome activation induce severe vascular leakage and fatality in murine models, representing a key aspect of the pro-inflammatory eicosanoid response to infection^[101]21. Conversely, other studies indicate that eicosanoids, along with related behenic acids, exhibit anti-inflammatory and resolution-promoting properties^[102]22. Beyond their role in inflammation regulation, specific eicosanoid mediators have also been implicated in enhanced pathogen elimination, neutrophil clearance, and antibody-mediated immune responses^[103]23. Figure [104]4 demonstrated a marked elevation of Carnitine in the HT group. As a key cofactor in fatty acid metabolism, Carnitine is integral to energy metabolism and fatty acid oxidation^[105]24. Research indicates that carnitine contributes to metabolic imbalances in autoimmune diseases. Mahmoud et al.^[106]25 reported that metabolomics revealed altered acylcarnitine levels in patients with Graves’ disease before and after treatment, with thyroid function normalizing post-treatment. Andrea et al. observed disrupted carnitine metabolism in both plasma and dendritic cells of SSc patients^[107]26. Their findings, validated across three plasma measurements from two distinct cohorts, highlighted fatty acid dysregulation in SSc patients. Similarly, carnitine metabolism is significantly disrupted in psoriasis, another autoimmune condition. Chao et al.‘s metabolomic analysis identified 40 significantly altered carnitines, with palmitoylcarnitine notably downregulated, and hexanoylcarnitine and 3-OH-octadecenoylcarnitine significantly upregulated in psoriasis^[108]27. Furthermore, carnitine plays a critical role in other autoimmune diseases, including multiple sclerosis^[109]28, pemphigus vulgaris^[110]29, and arthritis^[111]30. Metabolic pathways are recognized as key regulators of immune function, influencing the onset and progression of autoimmune diseases. These pathways modulate immune cell growth, differentiation, survival, and activation by supplying energy and essential biosynthetic precursors, making metabolic targeting a potential therapeutic strategy for Autoimmune Diseases (ATID)^[112]31,[113]32. To elucidate the dysregulated metabolic pathways in HT disease, KEGG pathway enrichment analysis was conducted on both up- and down-regulated metabolites and lipids. The analysis revealed that HT primarily involved alterations in Sphingolipid metabolism, Autophagy, and GP metabolic pathways, when compared to the control group. Similarly, Pang et al.^[114]33 employed metabolomics to investigate the therapeutic mechanisms in rheumatoid arthritis, identifying disruptions in three inflammation-related metabolic pathways—arachidonic acid, linoleic acid, and sphingolipid metabolism—which largely align with our findings. Electronic supplementary material Below is the link to the electronic supplementary material. [115]Supplementary Material 1^ (198.1KB, csv) [116]Supplementary Material 2^ (299.7KB, csv) [117]Supplementary Material 3^ (13.8KB, xlsx) [118]Supplementary Material 4^ (2.6MB, tif) [119]Supplementary Material 5^ (3.8MB, tif) [120]Supplementary Material 6^ (20.3KB, docx) Acknowledgements