Abstract Background Previous studies found that Polygonatum sibiricum polysaccharide (PSP) exhibited anti-inflammatory and anti-oxidative effects on type 2 diabetes (T2DM). However, the roles and detailed mechanisms of PSP on the anti-diabetic osteoporosis are still poorly defined. The purpose of this study was to identify the detailed mechanisms of PSP-induced anti-diabetic osteoporosis. Methods The roles of PSP on diabetic osteoporosis were evaluated on a T2DM zebrafish model. Furthermore, the changes of related physiological indicators and metabolites in adult zebrafish (Danio rerio) were investigated by pharmacology and metabolomic analyses. Results Results showed that PSP significantly reduced the blood glucose content and induced caudal fin regeneration. Metabolomic analysis found that the key metabolites involved in arachidonic acid metabolism, PPAR signaling pathway, linoleic acid metabolism, steroid hormone biosynthesis, amino acid biosynthesis and metabolism, alpha-Linolenic acid metabolism, sphingolipid signaling pathway were affected by PSP treatment. Among these, PSP reversed the trend changes of several metabolites, such as L-aspartic acid, choline, and propionylcarnitine. Besides, further changes of TNF-α and IL-1β transcripts suggested that the proinflammatory cytokines were also involved in the PSP-triggered anti-diabetic osteoporosis. Conclusion Above findings provided the theoretical basis for the use to plant PSP as food ingredients or bioactive compounds in natural health products on diabetic osteoporosis. Supplementary Information The online version contains supplementary material available at 10.1186/s12906-025-04978-9. Keywords: Anti-diabetic, Metabolomics, Osteoporosis, Polysaccharide, Zebrafish Introduction Diabetes mellitus (DM) is a common disease that stands as a serious threat to people’s health [[42]1]. It has gradually increased over the past few decades, and posed a major public health challenge of the 21st century [[43]1]. There were 537 million people with diabetes worldwide in 2021, and this number is expected to rise to 783 million by 2045 [[44]2]. There are three main types of diabetes mellitus, including type 1 diabetes (T1DM), type 2 diabetes (T2DM), and gestational diabetes. Among these, T2DM accounted for about 95% patients with diabetes [[45]3].It was also estimated that almost 21.3 million live births to women were affected by hyperglycaemia in pregnancy [[46]1]. T2DM is a chronic disease that required lifelong management, and it can lead to serious complications, such as cardiovascular disease, nephropathy, retinopathy, neuropathy, and infertility [[47]4, [48]5]. Diabetic osteoporosis (DOP) is a common complication in people with diabetes that involves a loss of bone mass and an increased risk of fracture [[49]6]. It seriously affects the quality of life and health of patients. Metabolic abnormalities in people with diabetes due to long-term high blood sugar can affect bone health, especially in people with diabetes [[50]7, [51]8]. Long-term hyperglycemia induce abnormal glucose metabolism, which is the main cause of diabetic osteoporosis, through disordering the metabolism of calcium and phosphorus in bones [[52]9]. Besides, insufficient oxygen supply to the bones and neuropathy may also lead to osteoporosis in diabetic patients [[53]4]. The diagnosis and treatment of DOP are a multi-faceted and integrated management process, there includes blood glucose control, lifestyle adjustments, intake of essential bone health supplements, and appropriate pharmacological interventions [[54]9, [55]10]. Therefore, there is an urgent need to explore the methods to relieve DOP. Polysaccharide are complex carbohydrates that composed of multiple monosaccharides, and widely exist in plants. Numerous studies demonstrated that polysaccharide exhibited a variety of important physiological functions in living organisms, such as antioxidant [[56]11], anti-tumor [[57]12], anti-inflammatory [[58]13, [59]14], immune-enhancing effects [[60]15, [61]16], and anti-ageing [[62]17]. In recent years, researchers have discovered that polysaccharide played a critical role in preventing diabetes through various mechanisms, including enhancing insulin sensitivity, increasing anti-inflammatory, modulating lipid metabolism, reducing starch digestion, and regulating gut microbiota [[63]18–[64]20]. For example, Dendrobium officinale polysaccharide regulated the dysfunctional glucolipid metabolism, lipopolysaccharide leakage, and inflammation levels in T2DM mouse models [[65]18]. Polysaccharide from small black soybean also modulated the gut microbiota and serum metabolism to alleviate the diabetes in streptozotocin (STZ)-induced diabetic rats [[66]19]. Several findings also showed potential application of polysaccharide in the treatment of osteoporosis [[67]21, [68]22]. It was confirmed that Achyranthes bidentata polysaccharide possessed anti-osteoporosis effects via increasing the biomarkers accumulation and regulating the lipid metabolism [[69]21]. Arabinogalactan from Phellodendron Chinense Schneid. reduced the accumulation of advanced glycation and products by down-regulated the expression of receptor, and protected osteoporosis associated with hyperglycemia [[70]22]. Previous preliminarily studies found that polysaccharide could regulate the diabetic osteoporosis, however, the complex molecular mechanisms remain to be fully deciphered. Polygonatum sibiricum is a medicinal and edible perennial plant, and has various functions, such as nourishing yin, moistening the lungs, invigorating qi, and tonifying the kidney and spleen [[71]13, [72]23, [73]24]. It has shown that Polygonatum sibiricum was rich in polysaccharides, amino acids, vitamins, saponins, flavonoids, trace elements and other nutrients [[74]23, [75]25]. Among these bioactive compounds, numerous studies have focused on the biological functions of Polygonatum sibiricum polysaccharide (PSP) [[76]23], such as anti-inflammatory [[77]26], antioxidant [[78]27], anti-diabetic [[79]28, [80]29], and protective on brain ageing [[81]17], as well as the kidney-protective and immunoregulatory effects [[82]26, [83]30]. For example, PSP lowered the levels of fasting blood glucose, and alleviated hyperglycemia and reduced the oxidative stress in rats with streptozotocin-induced diabetes mellitus [[84]27]. Similarly, polysaccharide from Polygonatum kingianum regulated the serum lipids metabolism and insulin tolerance, and significantly improved the anti-hyperglycemic activity on STZ-induced mice [[85]14]. Besides, Polygonatum cyrtonema Hua polysaccharide acted an vital component in ameliorating long-term glucose metabolism and regulating glycemia in T2DM mice [[86]31]. Moreover, it has been revealed that PSP could induce osteoblast formation and promote block osteoclastogenesis, and enhance bone health and prevent the osteoporosis [[87]32, [88]33]. Several studies found that polysaccharides derived from natural sources contributed to the prevention of osteoporosis by mitigating oxidative stress, decreasing inflammation, and regulating the bone metabolism [[89]34]. However, the impact of PSP on diabetic osteoporosis has never been addressed. Therefore, whether or how PSP acts as a modulator in alleviating the diabetic osteoporosis remains to be fully elucidated. Previous studies have extensively explored the anti-diabetic effects of PSP, including their roles in regulating glucose metabolism, improving insulin sensitivity, and modulating lipid metabolism [[90]14, [91]28, [92]29, [93]31], while the specific impact of PSP on diabetic osteoporosis has not been elucidated. In this study, we first demonstrated that PSP effectively reduced the blood glucose levels in STZ-induced diabetic zebrafish. Moreover, it significantly induced the caudal fin regeneration in a zebrafish hyperglycemia model. The underlying mechanisms by metabolomics analysis were further explored. Several metabolites related to the arachidonic acid metabolism, PPAR signaling pathway, linoleic acid metabolism, steroid hormone biosynthesis, amino acid biosynthesis and metabolism, alpha-Linolenic acid metabolism, phospholipase D signaling pathway, sphingolipid signaling pathway, and vascular smooth muscle contraction were obviously regulated by PSP. Results also showed that it significantly modulated the expression levels of TNF-α and IL-1β cytokines. This study elucidates the potential application of PSP in controlling blood sugar and improving osteoporosis caused by diabetes, providing a new direction for developing new functional foods. Materials and methods Materials and reagents Polygonatum sibiricum polysaccharide (PSP) was provided by Anhui Academy of Science and Technology (Hefei, Anhui, China). The hot water extraction and ethanol precipitation (85%) methods were used. Firstly, 100 mL of petroleum ether was added to 10 g dried Polygonatum sibiricum powder, and refluxed at 80 °C for 2 h. After filtration, defatted Polygonatum sibiricum powder was obtained and washed by 80% ethanol for 3 times. Then, adding the hot distilled water to the residue at a solid-to-liquid ratio of 1:20, heating at 90 °C for 2 h. The filtrate was collected, and the residue was extracted once more times under the same conditions, then the filtrates were combined and concentrated to 100 mL. Adding 95% ethanol for 12 h, filter, after filtering and vacuum drying, crude PSP were obtained. Furthermore, purified PSP were obtained by using 5% trichloroacetic acid and the Sevage method and dialyzed in a dialysis bag to remove proteins and small-molecule impurities. Streptozotocin (STZ) was obtained from Guangzhou Saiguo Biotechnology Co., LTD (Guangzhou, Guangdong, China). Citric acid was bought from Sinopharm Group Chemical Reagent Co., LTD (Shanghai, China). Sodium citrate was purchased from Xilong Science Co., LTD (Shantou, Guangdong, China). QuantiChromTM Glucose Assay Kit Glucose test box was obtained from Qunxiao Keyuan Biotechnology Co., LTD (Beijing, China). Animal experiments Zebrafish maintenance and growth conditions Adult AB strain zebrafish (5–6 month-old, approximately 0.3 g) were commercially obtained from Shanghai Feixi Fair Co., Ltd (Shanghai, China). Zebrafish were raised in a laboratory standard zebrafish farming system with a 26/28.5℃ (room temperature / water temperature) and a 14/10 h day/night regime. Zebrafish grouping and drug concentration screening The healthy adult zebrafish were selected as experimental subjects and randomly assigned to conical flasks (n = 6/bottle). Zebrafish were anesthetized by placing them in a tank containing 500 ppm of 2-phenoxyethanol for 2 min. The experiment setup included four groups: control group, model group, metformin group (Met_1, Met_2, Met_3, Met_4, Met_5, Met_6), and PSP group (PSP_1, PSP_2, PSP_3, PSP_4, PSP_5, PSP_6). The control and model groups were maintained in breeding water, the metformin group received 0, 50, 100, 200, 300, 400, and 500 µg/mL metformin, respectively, in breeding water. The PSP group received 0, 75, 150, 300, 400, 500, and 600 µg/mL PSP, respectively, in breeding water. Each flask contained 250 mL of solution, with three replicates per group, and the solution was replaced every 24 h. The mortality of zebrafish was recorded. The lethal concentration 50% (LC50) for each medication group was calculated using the software Graphpad Prism 8.0. Based on these results, the low, medium, and high concentrations for subsequent drug administration were determined. Construction of hyperglycemia model in zebrafish After anesthetizing the zebrafish, STZ was injected into the abdominal cavity of adult zebrafish. The dosage of the injection depended on the body weight of the zebrafish itself, and control fish were injected with the same volume of citric acid buffer solution. Before blood collection, the fish were fasted for 24 h and euthanized with 1:500 2-phenoxyethanol. From each group, five fish were selected and the process was repeated three times. After decapitating the zebrafish at the pre-heart level, blood was collected using a micropipette, collecting approximately 5 µL per group into EP tubes. A glucose assay kit was then used to measure the blood glucose levels, determining the fasting blood-glucose (FGB) on the 3rd, 7th, and 14th days to assess the stability of the model and the fluctuations in blood sugar levels. Construction of zebrafish diabetic osteoporotic model The detailed experiment procedure was shown in Fig. [94]1, on the third day after STZ injection, the zebrafish were anesthetized, their tail fins were observed and photographed using the EVOS imaging system for live cells. A portion of the tail fin was then cut off according to a standard procedure. The normal zebrafish were divided into two groups: blank group and model group, and returned to the breeding water. The zebrafish injected with STZ were divided into four groups: placed in drug-containing culture medium with Met (300 µg/mL) and PSP (75, 150, and 300 µg/mL), respectively. Each group contained 18 fish. On the 6th day after STZ injection, we recorded the regeneration of the zebrafish tail fin and measured the length of the regenerated tissue using Image Pro Plus 6.0. The regenerated tail fin was cut off and stored it at -80℃ for subsequent experiments. Zebrafish were euthanized to check for the changes in fasting blood glucose levels. Fig. 1. [95]Fig. 1 [96]Open in a new tab Experimental design Gene expression analysis Total RNA was extracted from the collected and preserved regenerative tissue using a TransZol Up Kit (Nanjing Nuoweizan Biotechnology, Nanjing, China) according to the manufacturer’s instructions. RNA concentration and quality were checked using a NanoDrop 2000 (Thermo Fisher Scientific, Wilmington, DE, USA). cDNAs were synthesized from 1 µg of total RNA using the reverse transcription reagent (Nanjing Nuoweizan Biotechnology, Nanjing, China). By using the gene-specific primers (Table [97]S1), real-time quantitative reverse transcription RT-qPCR was conducted using QuantStudio 6 Flex System (Thermo Lifetech, United States) with Real-time quantitative PCR SuperMix (Nanjing Nuoweizan Biotechnology, Nanjing, China) according to the manufacturer’s instructions. The expression levels of corresponding genes were normalized against β-actin, an internal control gene. Data were based on at least three independent biological replicates, and each sample was prepared in triplicate. Sample Preparation for metabolomics analysis Briefly, 0.1 g sample were ground and adequately vortexed with 1 mL precooling methyl alcohol/acetonitrile/water (2:2:1, v/v). Then, kept at 4 °C for 30 min, samples were incubated for 10 min at -20 °C to precipitate the protein. After centrifugating at 14,000 g for 20 min at 4 °C, the supernatants were collected and dried under vacuum. Samples were redissolved in 100 µL acetonitrile/water (1:1, v/v), and adequately vortexed, and then centrifuged at 14,000 g for 10 min at 4 °C. The supernatants were passed through a 0.22 μm filter membrane to the sample bottle for further UPLC-MS/MS analysis. LC-MS/MS analysis Sample separation was carried out on Agilent 1290 Infinity LC system (Agilent Technologies). The mobile phases consisted of buffer A (water + 25 mM ammonium acetate + 25 mM ammonium hydroxide) and buffer B (acetonitrile). The gradient was 95% B for 1 min and was linearly reduced to 65% in 13 min, and reduced to 40% in 2 min, then kept at 40% for 2 min, increased to 95% in 0.1 min, with a 5 min re-equilibration period employed. Samples were detected in both electrospray ionization (ESI) positive and negative modes. Data acquisition was performed using an UHPLC coupled to a quadrupole time-of-flight (AB SCIEX TripleTOF 6600). In the extracted ion features, only the variables having more than 50% of the nonzero measurement values in at least one group were kept. Principal Component Analysis (PCA) was performed using SIMCA-P 14.1 (Umetrics, Umea, Sweden) [[98]35]. Metabolites with VIP > 1, P < 0.05, were considered significantly different metabolites based on the variable importance in projection (VIP) values and the P value generated by the Student’s t-test. Differential metabolites between two groups were mapped to their corresponding biochemical pathways using metabolic enrichment and pathway analysis based on the KEGG database ([99]http://www.genome.jp/kegg/). Enrichment analysis was performed using the Python ([100]https://docs.scipy.org/doc/scipy/) to identify the most relevant biological pathways. Statistical analysis Statistical analysis was performed using SPSS 18.0 software. Means and standard errors were calculated from at least three independent experiments with at least three biological replicates for each. For statistical analysis, data were analyzed by one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test, and P < 0.05 were considered statistically significant. Results Protective effect of polygonatum sibiricum polysaccharide (PSP) on the fasting blood glucose of zebrafish In order to explore the role of Polygonatum sibiricum polysaccharide (PSP) on the fasting blood glucose of zebrafish, a dose-response study of PSP was performed. Metformin (Met) has been the well-known mainstay of therapy for diabetes mellitus. The death rates of zebrafish were tested by various Met and PSP treatments for 9 days post administration (Fig. [101]2). Results revealed that 300 µg/mL Met could bring about 33.33% death rate of zebrafish, and high concentrations (400 and 500 µg/mL) induced high rates (56.56% and 72.22%) at 6 days (Fig. [102]2A). Then, 300 µg/mL Met was used as positive control. Time-course and concentration analysis revealed that the death rates of zebrafish were steady value by various concentrations of PSP treatments at the 6th day over a 9-day period. Different concentrations 75, 150, 300, 400, 500, and 600 µg/mL PSP caused the death rates by 22.22%, 33.33%, 50.00%, 66.67%, 72.22%, and 100% at 6 days post administration in zebrafish, respectively (Fig. [103]2B). Then, 75, 150, and 300 µg/mL PSP were applied in the following experiments. Fig. 2. [104]Fig. 2 [105]Open in a new tab Effects of metformin (Met) and Polygonatum sibiricum polysaccharide (PSP) on mortality of zebrafish. (A-B) Healthy adult zebrafish were transferred to liquid medium containing 0, 50, 100, 200, 300, 400, and 500 µg/mL Met, or 0, 75, 150, 300, 400, 500, and 600 µg/mL PSP for 9 days. Death rates of zebrafish were recorded. Data were means ± SE of three independent experiments with at least three biological replicates for each individual experiment To build a model for detecting diabetes, streptozocin (STZ) was applied by microinjection needle on the healthy adult zebrafish. After the injection, zebrafish were put back into liquid medium for 14 days, and the fasting blood were detected. Results showed that the fasting blood glucose maintained a high level at 3 and 7 days (Fig. [106]3A). After STZ treatment for 3 days, zebrafish were transferred to the medium contained 300 µg/mL Met, or 75, 150, 300 µg/mL PSP for another 3 days. It was found that Met and PSP significantly reduced the fasting blood glucose (Fig. [107]3B). These results suggested that PSP might act as a protective effect on diabetes in zebrafish. Fig. 3. Fig. 3 [108]Open in a new tab Effects of metformin (Met) and Polygonatum sibiricum polysaccharide (PSP) on the fasting blood glucose of zebrafish under model treatment. (A) Healthy adult zebrafish were injected with streptozocin (STZ) by microinjection needle. Control zebrafish were injected with the citric acid buffer. After the injection, zebrafish were put back into liquid medium for 14 days, and the fasting blood was detected. (B) Healthy adult zebrafish were injected with STZ for 3 days, and then transferred to liquid medium containing 300 µg/mL Met, or 75, 150, or 300 µg/mL PSP for another 3 days. Afterwards, the fasting blood were detected. Data were means ± SE of three independent experiments with at least three biological replicates for each individual experiment. Within each set of experiments, columns with different letters denoted significant differences at P < 0.05 according to Duncan’s multiple range test Protective effect of PSP on the diabetic osteoporosis in zebrafish To unravel the specific role of PSP, the caudal fin regeneration was tested to simulate the detrimental effects of diabetic osteoporosis in zebrafish. Healthy adult zebrafish were injected with STZ by microinjection needle for 3 days, then part of the tail fin was cut off according to uniform standards. These samples were transferred to liquid medium containing with or without 300 µg/mL Met, or 75, 150, or 300 µg/mL PSP for another 6 days. Microscopic analysis showed that the caudal fin regeneration was inhibited by STZ, which set to the Model group, compared with the control condition (Fig. [109]4). Moreover, Met and PSP significantly promoted the caudal fin regeneration compared with the Model group (Fig. [110]4). These results suggested a beneficial role of PSP in the regulation of diabetic osteoporosis in zebrafish. Fig. 4. [111]Fig. 4 [112]Open in a new tab Effects of metformin (Met) and Polygonatum sibiricum polysaccharide (PSP) on the caudal fin regeneration of zebrafish under model treatment. Healthy adult zebrafish were injected with streptozocin (STZ) by microinjection needle for 3 days. Then, the tail fin was cut off according to uniform standards, and transferred to liquid medium containing 300 µg/mL Met, or 75, 150, or 300 µg/mL PSP for another 3 days. (A) The caudal fin pictures of zebrafish were respectively taken by EVOS cell imaging system. (B) The caudal fin regeneration length was measured by Image Pro Plus 6.0. Scale bars = 400 μm. Control zebrafish were injected with the citric acid buffer. Within each set of experiments, columns with different letters denoted significant differences at P < 0.05 according to Duncan’s multiple range test Metabolites identification and multivariate statistical analysis on metabolite composition of the anti-diabetic osteoporosis induced by PSP in zebrafish To explore the mechanism underlying effective function of PSP on the caudal fin regeneration diabetes complications of zebrafish, metabolomics were performed. A total of 319 distinct annotated metabolites were identified, including carboxylic acids and derivatives (21.6%), fatty acyls (17.4%), steroids and steroid derivatives (8.7%), benzene and substituted derivatives (6.2%), organonitrogen compounds (5.0%), organooxygen compounds (4.6%), phenol lipids (4.6%), phenols (2.9%), organic oxides (2.1%), azacyclic compounds (1.7%), and others (25.3%) (Fig. [113]5A). Principal component analysis (PCA) was subsequently conducted to reduce the dimensionality of the histological data. Results showed a greater degree of correlation in group samples, and the two principal components together accounted for 31.9% of the total variance. The first principal component (PC1) and second principal component (PC2) accounted for 17.4% and 14.4%, respectively (Fig. [114]5B). Additionally, hierarchical clustering analysis also showed the change trends of metabolite levels in the experimental groups (Fig. [115]5C). Fig. 5. [116]Fig. 5 [117]Open in a new tab Classification, principal component analysis (PCA), and hierarchical cluster analysis (HCA) of the metabolites in the four groups of zebrafish. (A) Classification; (B) PCA score plot; (C) HCA heatmap of all metabolic profiles. The upregulated and downregulated metabolites were indicated by different shades of red and blue, respectively Furthermore, differential metabolites were screened by OPLS-DA VIP > 1 and P value < 0.05 conditions. As shown in Fig. [118]6A, there were 142 (58 up-regulated and 84 down-regulated), 157 (48 up-regulated and 109 down-regulated), 138 (59 up-regulated and 79 down-regulated), 129 (48 up-regulated and 81 down-regulated), 127 (66 up-regulated and 61 down-regulated) differential metabolites in Model/Control, Met/Control, PSP/Control, Met/Model, PSP/Model comparison groups, respectively. Then, to investigate the numbers of unique differential metabolites, a Venn diagram was constructed by each pairwise comparison (Fig. [119]6B). A total of 28, 24, and 29 significantly different metabolites were distinctively showed in Model/Control, PSP/Control, and Met/Control comparisons, respectively. Besides, there were a total of 49 and 51 significantly different metabolites in PSP/Model and Met/Model, respectively. It also showed the overlap of 78 significantly different metabolites obtained for these comparison groups. These results suggested that the mechanisms of PSP and Met on the alleviation of the caudal fin regeneration diabetes complications existed differences in zebrafish, at least partially. Fig. 6. [120]Fig. 6 [121]Open in a new tab Venn diagram analysis and number of metabolites of five comparisons. (A) Venn diagram illustrating the overlapping and specific differential metabolites for the five comparisons. (B) Number of specific differential metabolites of five comparisons (Model/Control, Met/Control, PSP/Control, Met/Model, PSP/Model) Metabolic pathway enrichment analysis on metabolite composition of the caudal fin regeneration diabetes complications in PSP-treated zebrafish model Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed for the differential metabolites between the Model and Control group, the Met and Model group, and the PSP and Model group, respectively. Several metabolic pathways were enriched, and the diagrams for the top 20 ranked metabolic pathways were presented (Fig. [122]7). The differential metabolites between the Model and Control group were mainly involved in ABC transporters, arachidonic acid metabolism, serotonergic synapse, steroid hormone biosynthesis, amino acid biosynthesis and metabolism, synaptic vesicle cycle, cortisol synthesis and secretion, mTOR signaling pathway, FoxO signaling pathway, glutathione metabolism (Fig. [123]7A). Differential metabolites between the Met and Model group were mainly involved in PPAR signaling pathway, serotonergic synapse, arachidonic acid metabolism, ABC transporters, linoleic acid metabolism, amino acid biosynthesis and metabolism, cAMP signaling pathway, cGMP-PKG signaling pathway, biotin metabolism, sphingolipid signaling pathway, vascular smooth muscle contraction (Fig. [124]7B). Moreover, differential metabolites between the PSP and Model group were mainly involved in serotonergic synapse, arachidonic acid metabolism, PPAR signaling pathway, ABC transporters, linoleic acid metabolism, steroid hormone biosynthesis, amino acid biosynthesis and metabolism, alpha-Linolenic acid metabolism, ovarian steroidogenesis, ferroptosis, phospholipase D signaling pathway, synaptic vesicle cycle, sphingolipid signaling pathway, and vascular smooth muscle contraction (Fig. [125]7C). Fig. 7. [126]Fig. 7 [127]Open in a new tab Kyoto Encyclopedia of Genes and Genomes (KEGG) annotations and enrichment results of the differentially accumulated in the comparisons Model/Control (A), Met/Model (B), PSP/Model (C). The p value calculated based on the hypergeometric distribution, which the number of differential substances KEGGID in the enrichment table of this pathway and the number of metabolites of the pathway itself To explore the key differential metabolites, 39 overlapping differential metabolites were regulated by both Met and PSP, including 6 amino acid and derivatives, 5 steroids and steroid derivatives, and 3 fatty acyls (Table [128]1). Amino acid and derivatives and steroids and steroid derivatives accounted for 15.3% and 12.8% of the key differential metabolites, respectively. For example, both Met and PSP treatments induced L-Aspartic acid accumulation in the caudal fin regeneration diabetes zebrafish. Besides, we also found 16 differential metabolites were specifically regulated by Met, including 4 amino acids, peptides, and analogues, 3 organonitrogen compounds (Table [129]2). There were 20 differential metabolites which reversely regulated by PSP compared with Model and/or Met (Table [130]3). For example, choline and propionylcarnitine were particularly upregulated and downregulated by PSP, respectively. Moreover, level of propionylcarnitine was inhibited by both Model and Met, while obviously induced by PSP treatment in the caudal fin regeneration diabetes zebrafish. Table 1. Key metabolites were significantly regulated by both Met and PSP Class Compound log2FoldChange Control_vs_Model Model_vs_Met Model_vs_PSP Alcohols and polyols L-Olivosyl-oleandolide -0.65 0.87 0.97 Antibiotic G-418 -0.41 0.80 0.68 Alkyl hydroperoxides 9(S)-HPETE -0.56 0.65 0.58 Amines (2 S)-2-{[1^®Carboxyethyl]amino}pentanoate -0.22 0.19 0.18 Amino acids, peptides, and analogues L-Aspartic acid -0.70 0.23 0.54 Glycylleucine -0.66 0.20 0.25 4-Guanidinobutanoic acid -0.25 0.30 0.58 Oxidized glutathione 1.79 -1.52 -1.44 Aspartame 2.10 -0.94 -1.43 L-Homoserine 2.29 -1.36 -1.44 Bile acids, alcohols and derivatives Taurocholic acid -2.53 0.90 1.72 Deoxycholic acid -1.35 1.63 1.07 Carbohydrates and carbohydrate conjugates Mannitol 0.18 -0.43 -0.44 Carbonyl compounds 14,15-EET -0.84 0.67 0.94 Hydroxykynurenine -2.14 4.13 3.49 Cresols m-Cresol 0.32 -0.40 -0.35 Diphenylmethanes Cyclizine -0.19 0.21 0.23 Eicosanoids 8-HETE -1.86 1.41 1.28 Prostaglandin B1 -0.78 0.40 0.47 Fatty acids and conjugates Dethiobiotin -0.33 0.19 0.17 Hydroxycinnamic acids and derivatives 4-Hydroxycinnamoylagmatine 0.82 -0.75 -0.59 Hydroxyindoles N-Acetylserotonin -0.50 0.14 0.45 Cortisone -1.33 1.44 1.93 17alpha,21-Dihydroxypregnenolone 1.01 -1.42 -1.20 Phenols and derivatives 3-Methyl-1-(2,4,6-trihydroxyphenyl)-1-butanone 0.27 -0.51 -0.59 Pregnane steroids Pregnenolone 0.43 -1.07 -0.91 Pyrimidines and pyrimidine derivatives 5,6-Dihydro-5-fluorouracil 0.36 -0.45 -0.39 Sesquiterpenoids Caryophyllene epoxide -0.41 0.47 0.69 alpha-Cedrene -0.21 0.11 0.29 Myristicin 0.45 -0.48 -0.45 Cafestol -0.49 1.04 1.24 Guanosine 2.50 -0.91 -2.35 1-Pyrroline-2-carboxylic acid 0.44 -0.81 -1.11 Alprenolol -1.22 0.28 0.30 Qing Hau Sau -1.21 0.74 0.74 (S)-beta-Tyrosine -1.17 0.88 0.70 N5-(L-1-Carboxyethyl)-L-ornithine -0.88 0.37 0.36 13(S)-HpOTrE 0.26 -0.49 -0.32 3-Dehydroshikimate 3.11 -1.88 -2.89 [131]Open in a new tab Table 2. Key metabolites were specificly regulated by Met Class Compound log2FoldChange Control_vs_Model Model_vs_Met Model_vs_PSP Amines Spermidine 0.07 -0.19 0.04 Porphobilinogen 0.30 -0.52 0.11 Amino acids, peptides, and analogues Pyrrolidonecarboxylic acid 1.13 -0.80 -0.33 (2E)-Decenoyl-ACP -0.15 0.16 -0.05 L-Isoleucine -0.76 0.72 0.53 N-Acetylhistidine 0.28 -0.31 0.50 Benzylisoquinolines (S)-N-Methylcoclaurine -3.98 2.38 0.77 Carbonyl compounds (S)-Methylmalonic acid semialdehyde 0.99 -0.63 -0.36 Estrane steroids Levonorgestrel -0.88 1.03 0.23 Hydroxysteroids Cortexolone 0.99 -1.08 0.02 Pyridinecarboxylic acids and derivatives Nicotinic acid 0.42 -0.27 -0.13 Quaternary ammonium salts Acetylcholine 0.73 -0.27 0.09 (2E,6E)-Farnesol 0.56 -0.37 -0.11 (9Z,11E,13E)-Octadecatrienoic acid -1.07 0.65 0.44 9(S)-HPOT -0.55 0.26 0.18 Isotretinoin 2.57 -1.01 -0.22 [132]Open in a new tab Table 3. Key metabolites were specificly regulated by PSP Class Compound log2FoldChange Control_vs_Model Model_vs_Met Model_vs_PSP 1-hydroxy-2-unsubstituted benzenoids 4-Hydroxyphenylglyoxylate -0.80 0.39 0.77 Amines N6-Acetyl-L-lysine -0.65 -0.10 0.63 Sphingosine -0.68 0.13 0.47 Amino acids, peptides, and analogues Pipecolic acid -2.25 0.44 1.40 Carbohydrates and carbohydrate conjugates N-Acetyl-a-neuraminic acid -0.71 -0.03 0.68 Dibenzazepines 10-Hydroxycarbazepine 0.39 -0.76 -0.69 Fatty acid esters Propionylcarnitine -0.34 -0.39 0.35 Phenols D-synephrine -1.60 0.22 0.76 Phenothiazines Chlorpromazine -1.10 0.31 0.85 Pregnane steroids 5a-Pregnane-3,20-dione -0.54 0.07 0.32 Purine ribonucleotides AMP 3.47 -0.31 -1.81 Quaternary ammonium salts Choline 0.83 -0.08 -0.55 Indoleglycerol phosphate -1.54 0.15 1.51 Glutathione 1.17 0.92 -1.02 5-(2-Hydroxyethyl)-4-methylthiazole -0.74 -0.18 0.58 12-Hydroxydodecanoic acid -0.95 0.76 1.00 1-Hexadecylthio-2-hexadecanoylamino-1,2-dideoxy-sn-glycero-3-phosphocho line -1.64 -1.08 1.51 4-Quinolinecarboxylic acid -0.55 -0.11 0.36 Niaprazine -1.60 -0.01 1.40 [133]Open in a new tab Changes of related gene expression To further investigate the corresponding mechanisms, the transcript levels of these two genes related to the caudal fin regeneration, namely TNF-α and IL-1β, were examined in PSP-treated caudal fin regeneration diabetes zebrafish. As expected, expression of TNF-α and IL-1β was induced significantly in Model zebrafish, and that both effects were obviously reversed when Met or PSP was added (Fig. [134]8). Combined with the changes of the caudal fin regeneration of zebrafish and enriched metabolic pathway (Figs. [135]4 and [136]7), it can be deduced that PSP might act a key role in promoting tissue regeneration of zebrafish by modulating the expression of proinflammatory cytokines. Fig. 8. Fig. 8 [137]Open in a new tab Changes in the expression levels of TNF-α and IL-1β genes related serum inflammatory cytokines. Healthy adult zebrafish were injected with STZ for 3 days, and then transferred to liquid medium containing 300 µg/mL Met, or 75, 150, or 300 µg/mL PSP for another 3 days. Afterwards, the expression levels of TNF-α and IL-1β genes were analyzed using RT-qPCR. Relative TNF-α and IL-1β expression in the control samples was set as 1. The expression levels of corresponding genes were normalized against β-actin, an internal control gene. Data were means ± SE of three independent experiments with at least three biological replicates for each individual experiment. Within each set of experiments, columns with different letters denoted significant differences at P < 0.05 according to Duncan’s multiple range test Discussion As a traditional source of natural products, Polygonatum sibiricum contains many components, such as polysaccharides, saponin, alkaloids, and amino acids. Numerous studies have shown that Polygonatum sibiricum polysaccharide (PSP) are the main active component and has pharmacological effects, such as antioxidant, immune regulation, antitumor, hypoglycemic, and e.t.c [[138]23, [139]26, [140]27, [141]30]. Previous studies showed that polysaccharide could improve the insulin resistance, and regulate the immune and oxidative stress responses to diabetes [[142]15, [143]19, [144]36–[145]38]. Meanwhile, they are also widely accepted as anti-osteoporosis ingredient [[146]20, [147]21, [148]39]. The pathogenesis of osteoporosis involves a multifactorial etiology, including the nutritional status, physical activity, medication use, age-related hormonal changes, genetic susceptibility, and e.t.c [[149]6, [150]7]. In addition, oxidative stress serves as a critical mediator in osteoporosis pathogenesis. Excessive reactive oxygen species (ROS) generation suppressed the osteoblastic activity and stimulates osteoclast differentiation under cadmium (Cd) exposure [[151]40]. To date, diabetic osteoporosis (DOP) has been increasingly deemed as an important diabetes complication, and imposes a major threat to human health and the social economy [[152]6]. Over 9 million osteoporotic fractures are reported globally, with the majority being associating with DOP. Despite these known actions, the precise mechanisms of polysaccharide in addressing diabetic osteoporosis remain unclear. In this study, our results aimed to explore the impact of the amelioration of PSP on the diabetic osteoporosis. As an ideal model, zebrafish were increasingly used to study the effects of extracts from medicinal plants on diabetic osteoporosis in vivo [[153]41]. Thus, cutting off part of the tail fin of STZ-injected zebrafish mimicked the diabetic osteoporosis (Fig. [154]1). Results demonstrated that PSP can reduce the fasting blood glucose and promote the caudal fin regeneration compared with the model zebrafish (Figs. [155]3B and [156]4), suggesting that PSP acted as a key role in the treatment of diabetic osteoporosis. Complex pathways were involved in polysaccharide-mediated diabetes or/and osteoporosis [[157]36, [158]42]. Dendrobium officinale polysaccharide restored the gut microbiota via the LPS/TLR4/TRIF/NF-kB Axis to alleviate the T2DM [[159]36]. In skeletal muscle cells, insulin signaling was restored by Polygonatum sibiricum polysaccharides (PSP) via GLUT4-involved glucose metabolism, which reduced the pro-inflammatory cytokines levels [[160]42]. In this study, we further explored the potential targets and signaling pathways of PSP in the treatment of diabetic osteoporosis. By using pharmacological and metabolomics analysis, we discovered the key metabolic pathways and compounds might be involved in PSP-alleviated diabetic osteoporosis. Firstly, KEGG pathway enrichment analysis showed that PPAR signaling pathway, serotonergic synapse, arachidonic acid metabolism, linoleic acid metabolism, amino acid biosynthesis and metabolism, sphingolipid signaling pathway, and vascular smooth muscle contraction were enriched both in Met/Model group and PSP/Model group. These results indicated that these pathways might be the common approaches to relieve the diabetic osteoporosis (Fig. 7BC). Accumulation evidences have also demonstrated that the metabolic pathway of linoleic acid metabolism, biosynthesis of unsaturated fatty acids, and arachidonic acid metabolism were regulated by polysaccharides of small black soybean in T2DM rats [[161]19]. In addition, arachidonic acid, tyrosine and phenylalanine amino acid metabolism, and sphingolipid metabolism were the major pathways, which involved in Portulaca oleracea L. polysaccharide-induced anti-osteoporotic effects in zebrafish [[162]43]. Glucocorticoids belong to the class of steroid hormones that have an important impact on bone cells secretion, and disturb the osteoblastogenesis and osteoclastogenesis processes [[163]36]. Clinical studies have shown that α-linolenic acid could be used to treat diabetes [[164]44]. In this study, we also found that steroid hormone biosynthesis, alpha-Linolenic acid metabolism, ovarian steroidogenesis, ferroptosis, and phospholipase D signaling pathway were specifically regulated by PSP (Fig. [165]7C). Interestingly, Tenconi et al. found that phospholipase D (PLD) pathway, especially PLD1 and PLD2, mediated the high glucose-induced ERK and NFκB activation and cell viability loss in response to diabetic retinopathy [[166]45]. Besides, ferroptosis was a critical mechanism in regulating osteocyte death through iron-dependent programmed cell death in murine models of diabetic osteoporosis [[167]6]. These results suggested that steroid hormone biosynthesis, alpha-linolenic acid metabolism, ferroptosis, and phospholipase D signaling pathways regulated by PSP might play key roles in zebrafish of diabetic osteoporosis, and should be further explored in future. Numerous studies have reported that low concentration of aspartic acid, hydroxy-eicosatetraenoic acids (HETE), prostaglandin B1, and high concentration of aspartame were found in responses to diabetes or/and osteoporosis [[168]46–[169]49]. For example, diabetic KK-Akita mice accumulated lower level of L-aspartic acid [[170]47]. Its application reduced the oxidative stress and improved albuminuria, and thereby leading to alleviate diabetic kidney disease [[171]47]. In rats, excessive doses of aspartame caused an elevation in vitamin D levels and led to negative effects on bone formation [[172]46]. Consistent with theses literatures, similar results were found in the STZ-model zebrafish (Table [173]1). Moreover, PSP treatment revised the change trends of aspartic acid, hydroxy-eicosatetraenoic acids (HETE), prostaglandin B1, N-acetylserotonin, caryophyllene epoxide, and aspartame (Table [174]1). These changes might contribute to the alleviation of diabetic osteoporosis in zebrafish. Furthermore, some other key metabolites, such as choline and propionylcarnitine, were particularly regulated by PSP (Table [175]3). Emerging evidences revealed fascinating new insights into the roles of choline on diabetes and osteoporosis [[176]50, [177]51]. It was shown that high level of choline was significantly associated with diabetes through regulating the blood glucose, while its low level was positively associated with the increased risk of osteoporosis [[178]50, [179]51]. Therefore, it is important to control the balance of choline in PSP-alleviated diabetic osteoporosis in zebrafish. Moreover, propionylcarnitine exhibited protective effects on heel bone mineral density [[180]52]. It is worth to establish the roles of these metabolites involved in anti-diabetic osteoporosis in zebrafish. Elevated blood glucose was the important feature of diabetes, and it was closely related to osteoporosis. Excessive accumulation of advanced glycation end products (AGEs) can bind to the surface receptors on monocytes and macrophages, and produce various inflammatory factors, such as tumor necrosis factor-α (TNF-α), interleukin (IL)-1β, e.t.c [[181]53]. Overproduction of TNF-α inhibited the proliferation of mesenchymal stem cells (MSCs), and eventually impeded the fracture healing outcome [[182]51]. Previous result showed that chondroitin sulfate significantly decreased the levels of TNF-α and IL-1β serum inflammatory cytokines, and exhibiting the anti-osteoporotic effect [[183]54]. Furthermore, emerging studies have indicated that the expression level of TNF-α significantly increased in the early stage of caudal fin regeneration, which was crucial for the proliferation and regeneration process of germ cells [[184]55]. TNF-α indirectly promoted regeneration by regulating the number of neutrophils and the level of IL-1β [[185]56]. In the present study, PSP significantly decreased the blood glucose level (Fig. [186]3). Besides, the expression levels of TNF-α and IL-1β serum inflammatory cytokines were significantly modulated (Fig. [187]8). Hence, we speculated that PSP exerted an effect on anti-diabetic osteoporosis via its anti-inflammatory function, at least partially. Conclusion In conclusion, PSP ameliorated high blood glucose level, and induced the caudal fin regeneration in T2DM zebrafish. Interestingly, metabolomic assay proved that above efficacies of PSP were related to its modulative effect on the key differential metabolites involved in arachidonic acid metabolism, PPAR signaling pathway, linoleic acid metabolism, steroid hormone biosynthesis, amino acid biosynthesis and metabolism, alpha-Linolenic acid metabolism, sphingolipid signaling pathway, and vascular smooth muscle contraction. PSP reversed the trend of changes of L-aspartic acid, choline, and propionylcarnitine in Model zebrafish. Combined with the changes of TNF-α and IL-1β transcripts, and thereby revealing the functions of PSP on the anti-diabetic osteoporosis. This study provided the theoretical basis for the use of PSP as a promising probiotic in functional foods or a dietary supplement for diabetic osteoporosis. Electronic supplementary material Below is the link to the electronic supplementary material. [188]Supplementary Material 1^ (16.9KB, docx) Acknowledgements