Metabolomic and lipidomic studies on the intervention of taurochenodeoxycholic acid in mice with hyperlipidemia

Bile acids are the main component of animal bile and are directly involved in the metabolic process of lipids in vivo. Taurochenodeoxycholic acid (TCDCA) is the primary biologically active substance in bile acids and has biological functions such as antioxidant, antipyretic, anti-inflammatory, and analgesic activities and improves immunity. In the present study, we assessed the impact of TCDCA on hyperlipidemia development in mouse models. Mice were fed a high-fat diet (HFD) to induce hyperlipidemia and orally administered different doses of TCDCA orally for 30 days. Then, indicators such as triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) in mice were detected. Using HE and ORO staining techniques, the morphology of the mice’s liver tissue was detected. Based on metabolomic and lipidomic analyses, we determined the mechanism of TCDCA in treating hyperlipidemia. The results showed that TCDCA had a significant ameliorating effect on dietary hyperlipidemia. In addition, it exerted therapeutic effects through glycerophospholipid metabolism.


Introduction
With the aging of the global population and improving living standards, the number of people suffering from metabolic diseases is gradually increasing.Common conditions include hyperlipidemia, diabetes mellitus, and coronary heart disease (Jia et al., 2021).Among them, hyperlipidemia has many pathogenic factors.Excessive intake of fat (Sumiyoshi et al., 2006), abnormal lipoprotein synthesis, and metabolism can lead to dyslipidemia (Kaysen, 1992).Dyslipidemia can lead to lipid metabolism disorders and is the main pathogenic factor of hyperlipidemia (Koopal et al., 2017).Since various metabolic enzymes are secreted by the liver, improper functioning of the liver results in fat buildup that can easily cause excessive lipid synthesis and secretion, leading to hyperlipidemia over time (Michels et al., 2022).
Taurochenodeoxycholic acid (TCDCA) is a bile acid derivative that combines two molecules of taurine and chenodeoxycholic acid and is found in high concentrations in the bile of poultry and livestock (Zhong et al., 2021;Nguyen and Bouscarel, 2008).It is clear that bile acids facilitate digestion, absorption, and excretion of dietary lipids and interact with numerous cellular signaling pathways (Liu et al., 2011).TCDCA is also used as a nutritional fortifier added to foods.TCDCA is one of the main active ingredients of bear bile powder (Shi et al., 2017).It has been shown that TCDCA is used to treat diseases such as astrocytic neuroinflammation (Xu et al., 2023a), pulmonary fibrosis (Zhou et al., 2013), and gastric cancer (Zhang et al., 2022), with inhibitory effects on both acute and chronic inflammation as well as modulating immune responses (Li et al., 2019;Bao et al., 2021).Recent studies have reported that TCDCA can reduce intestinal fat transport and treat non-alcoholic fatty liver disease in mice (Wang et al., 2018).Moreover, natural products play a crucial role in treating metabolic disorders in human diseases.
In this study, we focused on the effect of TCDCA on dietary HFD.TCDCA improves lipid metabolism abnormalities by altering the glycerophospholipid metabolism pathway and ameliorates HFD-associated dyslipidemia and hepatic injury.TCDCA is expected to be a promising lipid-lowering active ingredient, which opens new avenues for studying its therapeutic use in treating metabolic diseases.In this regard, TCDCA has become the current research hotspot.

Animals and ethics statement
This study followed the Chinese Guide for the Care and Use of Laboratory Animals guidelines.All animal experiments were approved by the Laboratory Animal Ethics Committee of Guangdong Medical Laboratory Animal Center (Approval No. C202211-1).Forty-eight adult male C57BL/6 mice were obtained from Guangdong Medical Laboratory Animal Center (Guangdong Province, China, SCXK (Guangdong) 2022-0002.).The mice were housed in a standard room (12/12-h light/dark cycle, 23 °C ± 2 °C; relative humidity: 55% ± 2%) with food and water provided ad libitum.Mice were randomly divided into six groups: control group, model group, fenofibrate group, high-dose group (TCDCA-H), medium-dose group (TCDCA-M), and low-dose group (TCDCA-L).

Efficacy evaluation
After the last dose and fasting for 12 h, the body weight of the mice was recorded.The levels of AST, ALT, TC, TG, LDL-C, and HDL-C in mouse serum were detected again, and the liver index was calculated using the following formula: (wet weight of the liver/ mouse body weight) × 100%.

HE staining
After mice were deeply anesthetized with sodium pentobarbital, livers were removed and fixed with 4% paraformaldehyde.After fixation, they were washed with different concentrations of ethanol and water and finally stained with hematoxylin for 5 min.The procedure for staining is as follows: first, the samples were washed with varying concentrations of ethanol, rinsed with water, and stained with hematoxylin for 5 min; then, the samples were differentiated with 1% hydrochloric acid in alcohol for 30 s, rinsed with water for 5 min, and stained with 0.5% eosin solution for 1 min and again rinsed with water for 3 min.Finally, the samples were dehydrated until they became transparent, mounted, and photographed using a microscope (Nikon ECLIPSE Ci-L) (magnification: ×400).The images of the samples were captured.

ORO staining
Fresh liver tissue sections were fixed, washed, and dried.The dried samples were stained with Oil Red O for 10 min, removed, and immersed in 60% isopropanol twice for differentiation and immersed in pure water twice for washing, 10 s each time.The slices were removed and then stained with hematoxylin for 3-5 min.They were immersed in sterile water three times.Subsequently, they were differentiated with 60% ethanol for 8 s and washed with distilled water two times.The slices were washed with hematoxylin blue returning solution to return the blue color for 1 s and washed with sterile water three times.Then, the slices were differentiated with 60% ethanol for 8 s and washed with distilled water two times.The slices were sealed with glycerin, and the images were acquired.

LC-MS metabolomic and lipidomic analyses 2.7.1 Sample preparation
Sample preparation for untargeted metabolomics: Serum or liver tissue homogenate was mixed with acetonitrile (1:3, v/v) and centrifuged at 13,000 rpm for 10 min at 4 °C.The supernatant was evaporated to dryness, and the residue was re-dissolved in acetonitrile-water solution (1:1, v/v).After centrifugation at high speed, the supernatant was collected and assayed.A measure of 2 μL filtrate was injected into the instrument for analysis.Quality control (QC) samples were injected after every seven samples to monitor the reproducibility of the metabolomic analysis results.

Multivariate data analysis
The raw metabolomic data acquired from UHPLC-Orbitrap Fusion MS were treated for peak picking using Orbitrap Fusion Tune and for peak alignment using Xcalibur 4.3 software (Thermo Fisher, USA).We obtained potential and critical information from raw data files via Orbitrap Fusion Tune using Progenesis QI software (Waters, USA).Metabolic data were obtained for health and disease through principal component analysis and dimensionality reduction.Orthogonal partial least squares discriminant analysis (OPLS-DA) was used to determine potential metabolite differences.Objective numerical values screened out the components that significantly impact the grouping.

Statistical analysis
All data are presented as mean ± standard deviation (SD).Data were subjected to t-test using GraphPad Prism 7 (GraphPad Software, United States).A p-value < 0.05 was considered a significant difference.A p-value < 0.01 was supposed to be highly effective.Histograms were drawn using GraphPad Prism 7 software.

Pathway analysis
Results from serum metabolomics were analyzed using MetaboAnalyst (www.metaboanalyst.ca)to find critical metabolic pathways.KEGG (www.kegg.jp)was used to find connections between pathways.

Body weight and liver index in mice
Compared with the control group, the liver index and body weight of mice were significantly increased in the HFD group.Compared with the HFD group, the liver index and body weight of the mice treated with TCDCA-L were significantly reduced (Figure 1).

Physiological and biochemical indicators of mice
We established a hyperlipidemia mouse model to study the therapeutic effect of TCDCA.During HFD feeding, we observed that the levels of TC, TG, and LDL-C in the blood of the HFD group were significantly increased.At the same time, the concentration of HDL-C was significantly decreased compared to the control group.The dramatic changes in lipid levels in the serum samples of the mice indicated that the hyperlipidemia model was successfully established.Subsequently, different doses of TCDCA and fenofibrate were administered to treat the hyperlipidemia model mice for 30 days, respectively, and we observed that the TCDCA-L group showed a more striking effect compared to the HFD group in which the levels of TC, TG, and HDL-C were significantly downregulated.However, LDL-C levels showed a downward trend with no significant difference.The ALT level in the fenofibrate group was considerably higher than that in the HFD group.There was no significant difference in ALT and AST levels in the other groups (Figure 2).

HE staining
The changes in the liver tissue of hyperlipidemic mice are shown in Figure 3. HE-stained liver sections of the HFD group showed more rounded lipid droplets and large vesicular steatosis than those of the control group, and the structure of  the liver tissue was unclear, with a large number of hepatocytes with granular degeneration, sparse cytoplasm, eosinophilic granules, and a small number of lymphocytes with mildly focal infiltration in the periphery.At the end of the trial, TCDCA-L treatment significantly attenuated fatty liver and lipid deposition, and hepatic lobular and confluent area lesions were relieved.

ORO staining
The hepatocyte deposits of neutral fat were red in ORO staining.At the same time, the control group showed no evident fat deposits, and the HFD group showed more intracellular neutral fat deposits with a more extensive distribution.In the TCDCA group, the deposition of neutral fat in hepatocytes was reduced to varying degrees, and the distribution range was narrowed, especially in the TCDCA-L group (Figure 4).

Multivariate data analysis
In this study, UHPLC-Orbitrap Fusion MS was used for metabolomic analysis.Chromatographic peaks from Orbitrap Fusion Tune raw files were extracted and compared using Xcalibur 4.3 and Progenesis QI software programs.To maximize the collection of metabolomic and lipidomic information and fingerprints of the hyperlipidemia model, unsupervised principal component analysis (PCA) of serum and hepatic metabolites of the six groups (control, HFD, fenofibrate, TCDCA-L, TCDCA-M, and TCDCA-H) was performed in both HESI-positive and HESInegative modes (Figure 5) The PCA results showed a clear separation between the six groups.The TCDCA group moved toward the control group to varying degrees; particularly, the TCDCA-L group was closer to the control group.The results showed that TCDCA-L had a significant positive effect on hyperlipidemic mice.To search for potential metabolite markers [variable importance in projection (VIP) > 1, p < 0.05], we conducted partial least squares discriminant analysis (PLS-DA).Again, it showed that different groups between the components were more prominent, and there was a significant difference between the control and HFD groups, indicating that the overall metabolic state of the organism changed after 4 weeks.After administration of TCDCA, all positive-and negative-ionized hyperlipidemic mice showed different degrees of improvement, with TCDCA-L showing the most significant performance.

Analysis of metabolic pathways
To further understand the metabolic pathways in TCDCAtreated mice, metabolic pathway analyses of different metabolites enriched were performed using MetaboAnalyst 5.0 (www.MetaboAnalyst.ca).In liver and serum metabolomics and lipidomics, to compare the expression levels of metabolites between samples, based on fold-change values, we created FC value histograms (Figures 6A; 6B; 6C; 6D) and input metabolites into MetaboAnalyst to explore the metabolic pathways of TCDCA in the treatment of hyperlipidemia.The results showed that TCDCA might affect the glycerophospholipid metabolism and has the most significant impact, proving that it is one of the essential metabolic pathways (Figures 6E-H).The Venn diagram showed one common differential metabolite PC (16:0/16:0) in the four groups of serum, liver, serum lipids, and liver lipids (Figure 6I).

Correlation analysis
We performed Spearman's correlation tests to explore the relationship between metabolites and physiological characteristics.A total of 15 metabolites were correlated with physiological traits associated with hyperlipidemia (HDL-C, LDL-C, TC, TG, ALT, and AST; Figure 7A).Among these 15 metabolites, we observed stronger relationships between metabolites and lipid indicators (HDL-C, LDL-C, and TC) than liver parameters (TG, AST, and ALT).Four metabolites were positively correlated with LDL-C and TC, three metabolites were negatively correlated with LDL-C and TC and negatively correlated with HDL-C; three metabolites were positively correlated with HDL-C; and two metabolites were negatively correlated with HDL-C.For example, PI (16:1(9Z)/16: 0) was significantly positively correlated with LDL-C and TC and negatively correlated with HDL-C.A total of 20 lipid metabolites were correlated with physiological characteristics associated with hyperlipidemia (HDL-C, LDL-C, TC, TG, ALT, and AST; Figure 7B) for correlation analysis.Metabolites and hyperlipidemia indicators were analyzed in a correlation network (positive correlation threshold ≥0.5, negative correlation threshold ≤0.5, and a p-value threshold ≤0.05; Figure 7D).Among these eight lipid metabolites, we observed stronger relationships between lipid metabolites and lipid metrics (HDL-C, LDL-C, TC, and TG) than liver parameters (AST and ALT).Four metabolites were positively correlated with LDL-C, TC, and TG, four metabolites were negatively associated with LDL-C, TC and TG; one metabolite was positively correlated with HDL-C; two metabolites were positively correlated with HDL-C; and two metabolites were negatively correlated with HDL-C.For example, LysoPC (0:0/18:0) had a significant positive correlation with LDL-C and TC and a significant negative correlation with HDL-C.Lipid metabolites and hyperlipidemia indicators were analyzed in a correlation network (positive correlation threshold ≥0.5, negative correlation threshold ≤0.5, and a p-value threshold ≤0.05; Figure 7E).Finally, a correlation analysis was performed between metabolites and lipid metabolites (Figure 7C).Lipid metabolites and metabolite indicators were analyzed in a correlation network (positive  correlation threshold ≥0.5, negative correlation threshold ≤0.5, and a p-value threshold ≤0.05; Figure 7F).

Discussion
The pathogenesis of hyperlipidemia is characterized by metabolic dysfunction and abnormal lipid metabolism in vivo (Kim and Scherer, 2021).By identifying pathological metabolic changes, metabolomic approaches point to common endogenous markers of diet-induced hyperlipidemia (Xu et al., 2023c).After pharmacological intervention, target differential metabolites are sought.It can be used for early disease diagnosis (Zhang et al., 2012), drug target discovery (Rabinowitz et al., 2011), disease mechanism studies, and disease diagnosis (Beyoğlu and Idle, 2020).
Many studies have confirmed that bile acids promote lipid digestion and absorption, inhibit cholesterol precipitation in bile, reduce cholesterol saturation, and prevent the formation of gallstones (de Aguiar Vallim et al., 2013;Zhou et al., 2023).Bile acids affect cholesterol levels by influencing lipid metabolism in the liver and intestines and also promote the expression of the LDL receptor, which helps lower blood LDL levels.In some patients with hyperlipidemia, bile acids may be used to regulate lipid metabolism and lower cholesterol levels.Chenodeoxycholic acid tablets have been used as gallstone solubilizers to inhibit cholesterol supersaturation, limit hydroxymethylglutaryl coenzyme activity, reduce cholesterol synthesis and secretion, and prevent and dissolve gallstones (Ahlberg et al., 1981).In the studies of animals modeling lipid disorders, drug efficacy was accompanied by an increased abundance of some bile acids, including TUDCA, TCDCA, and CDCA (Li et al., 2022b), and it may be that these increased bile acids enhanced the efficacy of the drugs.
A high-fat diet (HFD) affects the gut microbiota of mice, and changes in gut microbiota structure may affect serum and liver lipids (Kim et al., 2012).In the present study, it was found that in the course of daily dietary HFD in mice, the weight gain after the administration was significantly slower in terms of the mice's behavior than that in the HFD group.From the four clinical indicators of TC, TG, LDL-C, and HDL-C (Gong et al., 2016), the TCDCA-L group all had a callback effect.Meanwhile, there was no significant difference in the levels of TC, TG, LDL-C, and HDL-C in the TCDCA-H group, but the efficacy was more pronounced in the TCDCA-M group than in the TCDCA-H group.Another symptom of hyperlipidemia is liver enlargement (Bay et al., 2017), which may be caused by excessive fat intake and disturbance of lipid metabolism (Xenoulis and Steiner, 2010).The HE and ORO staining results also demonstrated that TCDCA-L could significantly reduce granular hepatocyte deformation and neutral fat deposition.Therefore, TCDCA was shown to have a lipid-lowering effect.
The liver is the center of lipid metabolism and secretes various enzymes related to lipid metabolism (Wu et al., 2021).In hyperlipidemia, the liver may synthesize excess VLDL, which contains more glycerophospholipids, leading to elevated levels of triacylglycerol and cholesterol in the blood (Tsuji et al., 2021).In addition, the main carrier of cholesterol is low-density lipoprotein (LDL), which transports cholesterol from the body to individual cells (Du et al., 2022).However, abnormalities in glycerophospholipid metabolism caused by hyperlipidemia affect the structure and stability of LDL, which may lead to oxidation or glycosylation of LDL, which in turn creates a burden for cholesterol transport and makes it more likely to accumulate in the lining of the blood vessels, leading to cardiovascular diseases such as atherosclerosis (hardening of the arteries) (Morita, 2016).
Glycerophospholipid metabolism is the common pathway.The key metabolite of this pathway is PC.Previous studies have shown that PC is an important signaling molecule in atherosclerosis and an essential component of cell membranes (Zeng et al., 2017).PC enhances the assembly of VLDL in hepatocytes and the production of bile acids in the liver, thereby promoting lipid export and regulating lipid metabolism (Cole et al., 2012).Therefore, in the present study, we noted the metabolic characteristics of liver and hepatic lipids while observing the changes in serum lipids.The analysis of metabolomic and lipidomic results showed that the efficacy of the TCDCA-L group was higher than that of the other two groups.By enrichment analysis of differential metabolites, we found that TCDCA achieves beneficial effects and regulates glycerophospholipid metabolism by modulating metabolites.Therefore, we concluded that the critical pathway of TCDCA in treating hyperlipidemia may be glycerophospholipid metabolism.

Conclusion
Our data suggest that TCDCA ameliorates diet-induced hyperlipidemia in mice.Its mechanism may be used to regulate lipid metabolism pathways such as glycerophospholipid metabolism.These findings indicate that TCDCA may be an effective alternative for hyperlipidemia treatment.
writing-review and editing.QW: funding acquisition and writing-review and editing.

FIGURE 1
FIGURE 1 Behavioral indicators of mice in each group: (A) body weight and (B) liver index.Each value represents the mean ± SD (n = 6).#p < 0.05 compared with the control group; ***p < 0.001 and *p < 0.05 compared with the HFD group.

FIGURE 5
FIGURE 5 Multivariate data of UPLC-MS/MS: (A,B) PCA and (C,D) PLS-DA of each group in ESI-negative and -positive ion modes in serum metabolomics; (E,F) PCA and (G,H) PLS-DA of each group in ESI-positive and -negative ion modes in serum lipidomics; (I,J) PCA and (K,L) PLS-DA of each group in the ESIpositive and -negative ion modes of liver metabolomics; (M,N) PCA and (O,P) PLS-DA of each group in ESI-negative and -positive ion modes of liver lipidomics.

FIGURE 6
FIGURE 6 FC value histograms and bubble plots of serum and liver metabolomics and lipidomics: (A,E) serum metabolomics; (B,F) liver metabolomics; (C,G) serum lipidomics; (D,H) liver lipidomics; (I) Venn diagrams of serum and liver metabolomic and lipidomic metabolites.represents an important pathway.

FIGURE 7
FIGURE 7 Spearman's correlation analysis: (A,D) serum biochemical indexes with serum and liver metabolomic endogenous metabolites.(B,E) Serum biochemical indexes with serum and liver lipidomic endogenous metabolites.(C,F) Serum and liver metabolomic endogenous metabolites with serum and liver lipidomic endogenous metabolites.

TABLE 1
Detection of hyperlipidemia-related metabolites in serum metabolomics.Trend 1 is the control group compared with the HFD group; Trend 2 is the HFD group compared with the TCDCA-L group.

TABLE 2
Detection of hyperlipidemia-related metabolites in liver metabolomics.Trend 1 is the control group compared with the HFD group; Trend 2 is the HFD group compared with the TCDCA-L group.

TABLE 3
Detection of hyperlipidemia-related metabolites in serum lipidomics.Trend 1 is the control group compared with the HFD group; Trend 2 is the HFD group compared with the TCDCA-L group.

TABLE 4
Detection of hyperlipidemia-related metabolites in liver lipidomics.Trend 1 is the control group compared with the HFD group; Trend 2 is the HFD group compared with the TCDCA-L group.