Skip to main content
NCBI home page
Toxicology Research logoLink to Toxicology Research
. 2016 May 24;5(4):1148–1159. doi: 10.1039/c6tx00082g

Protection of baicalin against lipopolysaccharide induced liver and kidney injuries based on 1H NMR metabolomic profiling†

Shanting Liao a,‡, Pei Li a,‡, Junsong Wang b,✉, Qian Zhang a, Dingqiao Xu a, Minghua Yang a, Lingyi Kong a,✉
PMCID: PMC6060722  PMID: 30090421

graphic file with name c6tx00082g-ga.jpgLPS disturbed the metabolomic profiles in the serum, livers and kidneys of mice, and baicalin ameliorated these disturbances.

Abstract

Severe sepsis and septic shock are common and lethal conditions characterized by a systemic inflammatory response that is activated by invasive infection. In this study, a lipopolysaccharide (LPS) induced sepsis mice model was established to investigate the toxicities of LPS and the therapeutic effect of baicalin. Sera for clinical biochemistry and NMR metabolomic investigation, and liver and kidney tissues for histopathological examination, molecular biology measurement and NMR metabolomic profiling were collected. Multivariate analysis of metabolic profiles of the serum, liver and kidney extracts of mice revealed the occurrence of a severe inflammatory response, oxidative stress, and perturbances in energy and amino acid metabolism in LPS induced sepsis mice, which could be greatly ameliorated by baicalin treatment. This integrated 1H NMR based metabolomics approach gave us a new insight into the pathology of LPS induced sepsis, and helped in understanding the therapeutic effects of baicalin in a holistic view.

1. Introduction

Bacterial endotoxin lipopolysaccharide (LPS) played an important role in the development of sepsis by eliciting the release of an array of inflammatory mediators.1–3 LPS could lead to an imbalanced, dysregulated immune response, triggering sepsis, causing multiple organ failure.1 Baicalin, a flavonoid from Scutellariae radix, has various activities including anti-inflammatory, antioxidant, antibacterial, anti-apoptosis, anti-viral, and anti-allergic effects.4–6 Baicalin could inhibit LPS evoked inducible nitric oxide synthase (iNOS) protein expression and showed protection against LPS-induced injury.4–7 However, the mechanisms underlying the therapeutic effect of baicalin on sepsis induced tissue damage were still unclear.

Metabolomics monitors the disturbance of endogenous small molecule metabolites in cells, tissues, and biofluids of the body in response to a toxicant or environmental change. It has been successfully used in medicine, pharmacy, toxicology, agriculture, food, environmental science and so on.8–11 1H NMR has become one of the main techniques for the metabolomics study owing to its inherent advantages of being rapid, non-destructive and bias free for a wide range of metabolites, while at the same time, providing rich structural information for metabolite characterization.12

In this work, a 1H NMR based metabolomics approach complemented with histopathological examination, clinical chemistry and molecular biology methods was used to explore the metabolic disturbance of LPS induced sepsis mice and to assess the therapeutic effect of baicalin for the first time.

2. Materials and methods

2.1. Chemicals and reagents

Baicalin was purchased from Dalian Meilun Biology Technology Co., Ltd (Dalian, China). LPS from Escherichia coli 055:B5 (Sigma L2880) and 3-trimethylsilylpropionic acid (TSP) were bought from Sigma Chemical Co. (St Louis, MO, USA). Acetonitrile was bought from Merck KGaA (Darmstadt, Germany). Deuterium oxide (D2O, 99.9%) was provided by Sea Sky Bio Technology Co. Ltd (Beijing, China). Ultra-pure distilled water was prepared using a Milli-Q purification system. All reagents were of analytical grade.

2.2. Animal preparation and drug administration

Seventy-five male ICR mice (25–28 g in weight) were purchased from the Experimental Animal Center of Yangzhou University (Yangzhou, China). All animals had ad libitum access to standard diet and water, and were group-housed in polysulfone cages with bedding material. They were housed in a well-ventilated room (5 mice in one cage) at a constant room temperature (25 ± 2 °C) and controlled humidity (50 ± 5%) with a light–dark cycle (12–12 h).

All mice were acclimatized for one week and then randomly divided into three groups (25 animals each) as follows: normal control group (NC), LPS model group (LPS), and baicalin-treated group (Bai). The mice in the NC and LPS groups were orally administered with vehicle (0.5% CMC-Na), and the Bai group received baicalin (74 mg per kg per day, weight ratio between crude drug and mice). Mice were orally administrated with vehicle or baicalin once a day for one week. Mice in LPS and Bai groups were then intraperitoneally (i.p.) injected with 3 mg kg–1 LPS dissolved in normal saline to establish a mice sepsis model, and the mice of the NC group were i.p. injected with an equivalent amount of normal saline.

2.3. Sample collection and preparation

At 24 h after i.p. injection of LPS, mice were fasted overnight and sacrificed after anesthetization by chloral hydrates (400 mg kg–1, i.p.). Blood was taken from abdominal aorta of mice, and then centrifuged at 13 000g to afford serum, kept at –80 °C before analysis. The livers and kidneys were quickly removed from the body and rinsed with cold normal saline. Parts of the liver and kidney tissues were immersed in 10% neutral buffered formaldehyde for histological examination, and the others were frozen and stored at –80 °C for 1H NMR recording and gene expression analysis.

All procedures for animal handling were approved by the Animal Care and Use Committee of China Pharmaceutical University and were in accordance with the National Institute of Health guidelines for the Care and Use of Laboratory Animals.

2.4. Serum clinical biochemistry and histopathology

Levels of nitric oxide (NO), malondialdehyde (MDA), alanine aminotransferase (ALT) and creatinine (Cr) in serum were measured using commercially available kits (Nanjing Jianchen Biotech Inc., China).

The liver and kidney tissues for histological examination were embedded in paraffin, and were cut into serial thin sections of 5 μm thickness and stained with hematoxylin and eosin (HE).

2.5. Quantitative real-time PCR (qRT-PCR)

Liver and kidney samples were thawed for RNA isolation. A tris reagent (Sigma-Aldrich, St Louis, MO) was used to extract total RNA according to manufacturer's instructions. For each sample, 1 μg of total RNA was reversely transcribed using a PrimeScript™ RT Master Mix reagent Kit (Takara, China). The real time fluorescence quantitation was performed on a Light Cycler 480 (Roche Molecular Biochemicals, Mannheim, Germany). The Δ cycle threshold method was used to calculate the relative levels of assayed mRNAs, and the relative expression level of each target gene was normalized to that of the endogenous control GAPDH. The sequences of primers for GAPDH, TNF-α, IL-1β, IL-10, IL-6, PK and CS are shown in Table S1.†

2.6. 1H NMR spectroscopy

After thawing, serum samples (300 μl) were added to 300 μl D2O (containing 0.2 mol per l Na2HPO4 and 0.2 mol per l NaH2PO4, and 0.05% TSP), vortexed and then centrifuged at 13 000g for 10 min at 4 °C to obtain the supernatant. The transparent supernatant solution was transferred into 5 mm NMR tubes for NMR analysis. Liver and kidney tissues, ca. 200 mg, were reconstituted in a mixture of volumetric equivalent acetonitrile and water, and then centrifuged at 13 000g for 10 min at 4 °C to remove any precipitates. The supernatant was collected and then lyophilized. Dried tissue extracts were homogenized in 600 μl D2O containing 0.05% TSP as an internal standard and phosphate buffer salt (0.2 mol per l Na2HPO4 and 0.2 mol per l NaH2PO4). The samples were vortexed and centrifuged at 13 000g, and the collected supernatants (550 μl) were placed in 5 mm NMR tubes.

All 1H NMR spectra were recorded at 298 K using a Bruker AV 500 MHz spectrometer. For serum samples, a Carr–Purcell–Meibom–Gill (CPMG) spin-echo pulse sequence (90-(τ-180-τ)n-acquisition) with a total spin-echo delay (2 nτ) of 10 ms was adopted to attenuate the broad signals from macro molecules (i.e. proteins or lipoproteins), whereupon the signals of micro molecule metabolites were clearly observed. 1H NMR spectra were collected with 128 scans into 32 000 data points over a spectral width of 10 000 Hz. For tissue samples, modified nuclear Overhauser enhancement spectroscopy with a presaturation (NOESYPR) pulse sequence (relaxation delay-90°-μs-90°-tm-90°-acquire-FID) was used to suppress the residual water signal. Prior to the Fourier transformation, a line broadening of 0.3 Hz was applied to all spectra. An exponential weighting factor corresponding to a line broadening of 0.3 Hz was used to all acquired free induction decays prior to Fourier transformation and phase correction.

2.7. Data processing and analysis

All 1H NMR spectra were manually phased, baseline corrected and referenced to TSP (1H, δ 0.00) using Bruker Topspin 3.0 software (Bruker GmbH, Karlsruhe, Germany), and were exported to ASCII files using MestReNova (Version 8.0.1, Mestrelab Research SL, Santiago de Compostela, Spain), which were then imported into an open-source software “R” (; http://cran.r-project.org/). The spectra were aligned based on least squares minimization, then binned using an adaptive intelligent algorithm13 with the average bin width of 0.015 ppm between 0.2 and 10 ppm, probability quotient normalized, mean-centered and pareto scaled with “R” before multivariate statistical analysis.14 The noisy and residual water affected regions (4.40–9.70 ppm for serum, 4.65–5.25 ppm for liver, and 4.55–5.00 ppm for kidney) were removed.

Orthogonal projection to latent structures discriminant analysis (OPLS-DA), which is a supervised method, was used to maximize covariance between the measured data (peak intensities in NMR spectra) and the response variable (predictive classifications).15 The parameters of R2 and Q2 were used to judge the model, where R2 indicates the total variation and Q2 is a cross-validation parameter that represents the predictability of the model. The model was validated by repeated two-fold cross-validation and a 2000 times permutation test.16 Permutation testing is based on the comparison of the predictive capabilities of an OPLS-DA model using real class assignments to a number of models calculated after random permutation of the class labels. The statistical P-values obtained via permutation testing were all less than 0.05, thus suggesting the validity of the OPLS-DA model.

The fold change values of metabolites between groups were calculated by the ratio of the integral area through the integral of interval of metabolites. The Benjamini and Hochberg method17 was applied to adjust the associated P-values for controlling the false positive rate in multiple comparisons using scripts written in R language. The parametric (Student's t-test) or nonparametric Mann–Whitney test was applied to evaluate the difference of metabolites that are increased or decreased between groups. Data were expressed as mean ± SD and P < 0.05 was considered statistically significant.

3. Result

3.1. Liver and kidney histopathological evaluation

Tissues of the liver and kidney were inspected after H&E staining (Fig. S1†). Compared with the NC group, the LPS group exhibited histopathological changes such as glycogen cavity and inflammatory cell infiltration in livers, and congestion and renal tubular epithelial cell necrosis in kidneys. Mice in the Bai group displayed no obvious histopathological changes in livers and kidneys, which indicated that baicalin could markedly alleviate the LPS induced injury.

3.2. Biochemical analysis

Compared with the NC group (Fig. 1), the concentrations of NO, MDA, ALT and Cr in the serum of the LPS group were significantly increased, which could be markedly attenuated by baicalin (Bai group).

Fig. 1. Boxplots for serum NO (A), MDA (B), ALT (C), Cr (D) levels in sepsis mice at 24 hours after i.p. injection of LPS. The bottom of each box, the line drawn in the box and the top of the box represent the 1st, 2nd, and 3rd quartiles, respectively. The whiskers extend to ±1.5 times the interquartile range (from the 1st to 3rd quartile). *P < 0.05, **P < 0.01 and ***P < 0.005 for the LPS group vs. NC group; #P < 0.05, ##P < 0.01 and ###P < 0.005 for the Bai group vs. LPS group.

Fig. 1

Open in a new tab

3.3. Effects of baicalin on the expressions of relevant genes

Characterized by a systemic inflammatory response, sepsis was triggered by LPS, which, as potent pro-inflammatory molecules, induced the release of an array of inflammatory mediators.18 We detected mRNA levels of inflammatory cytokines, including TNF-α, IL-1β, IL-6 and IL-10 (Fig. 2). They were significantly augmented in the livers and kidneys of mice at 24 h after the i.p. injection of LPS, which could be obviously weakened by baicalin treatment. As the key enzyme of the tricarboxylic acid (TCA) cycle, citrate synthase (CS) catalyzed the synthesis of citrate. The conspicuously down-regulated expression of CS in both livers and kidneys after the treatment of LPS might demonstrate an inhibition of the TCA cycle (Fig. 3a and c). Pyruvate kinase (PK) was the key regulator of pyruvate synthesis, and its expression was markedly decreased in the livers and kidneys of the LPS group, possibly suggesting an inhibition of pyruvate synthesis (Fig. 3b and d). The expressions of CS and PK were notably increased in the livers and kidneys of the baicalin treatment group compared with the LPS group (Fig. 3).

Fig. 2. Boxplots for mRNA expression analysis by RT-PCR in the liver and kidney of NC, LPS and Bai groups: TNF-α (a and e), IL-1β (b and f), IL-6 (c and g), IL-10 (d and h); a, b, c and d for liver; c, d, e and f for kidney. *P < 0.05, **P < 0.01 and ***P < 0.005 for the LPS group vs. NC group; #P < 0.05, ##P < 0.01 and ###P < 0.005 for the Bai group vs. LPS group.

Fig. 2

Open in a new tab

Fig. 3. Boxplots for mRNA expression analysis by RT-PCR in the liver and kidney of NC, LPS and Bai groups: CS (a and c), PK (b and d); a and b for liver; c and d for kidney. *P < 0.05, **P < 0.01 and ***P < 0.001 for the LPS group vs. NC group; #P < 0.05, ##P < 0.01 and ###P < 0.001 for the Bai group vs. LPS group.

Fig. 3

Open in a new tab

3.4. Metabolite identification

Typical 1H NMR spectra of the serum samples, and liver and kidney extracts are presented in Fig. 4. Metabolites were assigned by Chenomx NMR suite 7.7 (Chenomx Inc., Edmonton, Canada), and aided by querying publicly accessible metabolomics databases such as Human Metabolome Database (HMDB, ; http://www.hmdb.ca) and Madison-Qingdao Metabolomics Consortium Database (MMCD, ; http://mmcd.nmrfam.wisc.edu/): 18 metabolites in the serum, 23 metabolites in the liver, 27 metabolites in the kidney were assigned and listed in Tables S2–S4.†

Fig. 4. Representative 500 MHz 1H NMR spectra of serum (a), liver (b) and kidney extracts (c) with the metabolites labeled. Metabolites in the serum: 1. LDL/VLDL; 2. 3-hydroxybutyrate (3-HB); 3. lactate (Lac); 4. alanine (Ala); 5. acetate (Ace); 6. N-acetylglucosamine (NAGS); 7. N-acetylglycoprotein (NAGP); 8. O-acetylglycoprotein (OAGP); 9. α-oxoglutarate (2-OG); 10. pyruvate (Pyr); 11. citrate (Cit); 12. nicotinamide adenine dinucleotide phosphate (NADPH); 13. creatinine (Cre); 14 taurine (Tau); 15. betaine (Bet); 16. TMAO; 17. acetoacetate (Acet); 18. glucose (Glu). Metabolites in liver extracts: 1. LDL/VLDL; 2. 3-HB; 3. isoleucine (Iso); 4. leucine (Leu); 5. valine (Val); 6. Lac; 7. Ala; 8. Ace; 9. Acet; 10. NADPH; 11. creatine (Cr); 12. Cre; 13. choline (Cho); 14. phosphocholine (Pco); 15. Bet; 16. Tau; 17. TMAO; 18. Glu; 19. uridine (Ude); 20. tyrosine (Tyr); 21. phenylalanine (Phe); 22. adenosine (Ade); 23. inosine (Ino). Metabolites in kidney extracts: 1. LDL/VLDL; 2. 3-HB; 3. Lac; 4. Ala; 5. Acet; 6. 2-OG; 7. sarcosine (Sar); 8. NADPH; 9. Cr; 10. Cre; 11. Cho; 12. Pco; 13. TMAO; 14. Tau; 15. myo-inositol (Myo); 16. Bet; 17. Ino; 18. Lact; 19. Suc; 20. Mal; 21. Ans; 22. Tyr; 23. tryptophan (Trp); 24. Phe; 25. Nin; 26. Ude; 27. Ade.

Fig. 4

Open in a new tab

3.5. Multivariate analysis of 1H NMR spectral data

OPLS-DA analysis was performed on the 1H NMR spectral data of the serum, liver and kidney extracts to obtain an overview of metabolite variations among groups and investigate the therapeutic effects of baicalin. In the score plots, the showcased clusters correspond to metabolic patterns in different groups with each point representing one sample. In the OPLS-DA score plots, the serum, liver and kidney extracts of the NC, LPS and Bai groups were completely separated with the LPS group the furthest away from the NC group and the Bai group in between (Fig. S3†), demonstrating severe metabolic disturbance induced by LPS and amelioration of the disturbance after baicalin treatment. To further investigate the metabolic perturbations induced by LPS and baicalin, the 1H NMR data of the LPS group was compared with the NC and baicalin groups by OPLS-DA analysis, respectively.

3.5.1. Metabolic variation in the serum of mice

NC and LPS groups exhibited clear separation in the OPLS-DA score plots of the serum (Fig. 5a) with satisfactory goodness of fit (R2 = 0.89, Q2 = 0.76) (Fig. 5g) and statistical significance (P = 0.0045) (Fig. S2a†). Metabolite variations were visualized in the loading plots, which were color-coded according to the absolute correlation coefficient of each variable to grouping, a hot-colored signal (red) indicated more significant contribution to class separation than a cold-colored one (blue). S-plots were another way to identify significantly altered metabolites, which should be farther away from the origin and located in the upper right or lower left quadrant. The color coded loading plots (Fig. 5c) and S-plots (Fig. 5e) showed obvious increases of LDL/VLDL, NAGS, NAGP, OAGP, NADPH, creatinine, and significant decreases of 3-HB, lactate, alanine, acetate, pyruvate, citrate, taurine, betaine, TMAO, acetoacetate, and glucose in the LPS group compared with the NC group.

Fig. 5. OPLS-DA analysis of the serum 1H NMR data of the NC, LPS and Bai groups after the removal of water signals: score plots (a and b), the corresponding loading plots (c and d) and S-plots (e and f). OPLS-DA scatter plot from the serum (g and h) of the statistical validations obtained by 2000 times permutation tests, with R2 and Q2 values in the vertical axis, the correlation coefficients (between the permuted and true class) in the horizontal axis, and the OLS line representing the regression of R2 and Q2 on the correlation coefficients.

Fig. 5

Open in a new tab

OPLS-DA analysis of the NMR data of the LPS and Bai groups was performed to investigate the therapeutic effects of baicalin on LPS induced sepsis. The LPS and Bai groups showed a clear separation (R2 = 0.86, Q2 = 0.74) (Fig. 5h) with statistical significance (P = 0.0425) (Fig. S2b†) in the score plots (Fig. 5b). The loading plots (Fig. 5d) and S-plots (Fig. 5f) indicated that most of the metabolites disturbed by LPS could be reversed after baicalin treatment.

3.5.2. Metabolic changes in the liver of mice

The score plots (Fig. 6a) of the liver presented a well separation between NC and LPS groups (R2 = 0.95, Q2 = 0.91) (Fig. 7e) with statistical significance (P < 5 × 10–4) (Fig. S2b†). The corresponding loading plots (Fig. 6b) and S-plots (Fig. 7a) revealed the variation of metabolites in the liver extracts, including increased levels of LDL/VLDL, 3-HB, isoleucine, leucine, valine, lactate, alanine, acetate, acetoacetate, pyruvate, succinate, NADPH, creatine, creatinine, TMAO, choline, phosphocholine, tyrosine, phenylalanine, adenosine, and inosine, and decreased levels of betaine, taurine, glucose, and uridine.

Fig. 6. OPLS-DA analysis of liver and kidney extract 1H NMR data of NC, LPS and Bai groups. Score plots (a and c for liver, e and g for kidney) and the corresponding loading plots of OPLS-DA (b and d for liver, f and h for kidney) color-coded with the absolute value of correlation coefficients.

Fig. 6

Open in a new tab
Fig. 7. The S-plots (a and b for liver extracts, c and d for kidney extracts) and the OPLS-DA scatter plot (e and f for liver extracts, g and h for kidney extracts) of the 1H NMR data for NC, LPS and Bai mice.

Fig. 7

Open in a new tab

The loading plots (Fig. 6d) and S-plots (Fig. 7b) of the LPS and Bai groups also showcased some differential metabolites: increases of betaine and glucose, and decreases of isoleucine, leucine, 3-HB, acetoacetate, pyruvate, NADPH, creatine, creatinine, and TMAO in baicalin treated mice, suggesting the amelioration of baicalin on the disturbed metabolisms in sepsis.

3.5.3. Metabolic changes in the kidney of mice

The score plots (Fig. 6e) for the kidney NMR data of the NC and LPS groups showed a clear clustering of each group (R2 = 0.93, Q2 = 0.72) (Fig. 7g) with statistical significance (P = 0.0165) (Fig. S2c†). The associated loading plots (Fig. 6f) and S-plots (Fig. 7c) exhibited conspicuously elevated levels of creatine, creatinine, choline, phosphocholine, TMAO and myo-inositol, and markedly decreased levels of alanine, glutamine, taurine, betaine, lactose, sucrose, maltose, tyrosine, anserine, tryptophan and niacinamide. Most of the disturbed metabolites were ameliorated after baicalin treatment (Fig. 6h and 8d). The corresponding fold change score plots and color table are shown in Fig. S4† and Table 1, respectively.

Fig. 8. Schematic diagram of the disturbed metabolic pathways induced by LPS and the effects of baicalin, showing the interrelationship of the identified metabolic pathways. Metabolites in red and blue represent an increase and a decrease in the LPS group compared with the NC groups, respectively. The metabolites in the serum, liver and kidney of the NC, LPS and Bai groups which were detected by kits and qRT-PCR are presented by box-plots with values expressed as mean ± SD. The metabolites observed from 1H NMR are presented by the heat map, and superscript “s” means observed from the serum; “L” means observed from the liver; “k” means observed from the kidney. Color key indicates the metabolite expression value: red represents the highest and blue represents the lowest. *P < 0.05, **P < 0.01 and ***P < 0.005 for the LPS group vs. NC group; #P < 0.05, ##P < 0.01 and ###P < 0.005 for the Bai group vs. LPS group.

Fig. 8

Open in a new tab
Table 1. Identified metabolites from different groups with the fold change and P value a , b .
graphic file with name c6tx00082g-u2.jpg
Open in a new tab

aThe FC and P means the fold change and P value respectively (*P < 0.05, **P < 0.01 and ***P < 0.001); the N_L and L_B means LPS vs. NC and Bai vs. LPS respectively. Color bar Inline graphic.

bColor coded according to log2 (fold change) using the color bar labeled at the bottom.

The metabolomics alterations were visualized in fold change plots (Fig. S4†) where the corresponding p-values were adjusted by the Benjamini–Hochberg method using scripts written in R language, and the results are listed in Table 1.

4. Discussion

In our present work, a 1H NMR based metabolomics approach along with histopathological evaluation, biochemical assays and molecular biology methods was performed to holistically assess the potential toxic effects of LPS on mice and the therapeutic effect of baicalin. LPS elicited inflammatory response and excessive production of reactive oxygen species (ROS), producing disturbances in energy metabolism and amino acid metabolism, and a series of injuries. A schematic diagram of the perturbed metabolic pathways is shown in Fig. 8. Baicalin administration markedly suppressed inflammatory response, attenuated oxidative stress, ameliorated energy metabolism and amino acid metabolism.

4.1. Inflammatory response

Bacterial LPS is a potent pro-inflammatory molecule that elicits an innate immune response, triggering the release of inflammatory cytokines and inducing sepsis.19,20 In the serum, compared with the NC group, the LPS group showed a significant increase of NAGP and OAGP, which as the acute phase protein response, are markers of inflammation and tissue damage.21 Therefore, the increased levels of serum NAGP and OAGP indicated an enhanced inflammatory response induced by LPS. Baicalin markedly attenuated the increase of NAGP and OAGP induced by LPS, showing good performance on the inhibition of LPS-induced inflammatory response. TNF-α and IL-1β, the most important pro-inflammatory cytokines, caused the generation of more inflammatory cytokines, such as IL-6, which is largely responsible for the clinical features of sepsis.19,22 Inflammatory stimuli activate the production of anti-inflammatory cytokines such as IL-10, indicative of a widespread inflammation in the body.23 The expressions of these inflammatory cytokines in the liver and kidney tissues were quantified by qRT-PCR. Compared with the NC group, mRNA expressions of these cytokines were significantly up-regulated in the LPS group, which could be markedly inhibited after baicalin treatment.

4.2. Oxidative stress

The release of inflammatory cytokines trigger the generation of ROS in response to physiological or pathological stimuli.24,25 We measured several biochemical parameters to evaluate the status of oxidative stress, including NO and MDA. NO is normally formed by arginine under the catalysis of endothelial nitric oxide synthase (eNOS) in the vasculature. In the inflammatory status, inducible NOS can be expressed in macrophages and smooth muscle cells, facilitating the production of NO. Excessive endogenous NO could react with ROS to produce neurotoxic peroxynitrite. In addition, ROS can attack polyunsaturated fatty acids of biomembrane, and cause lipid peroxidation. MDA is one of the end-products of polyunsaturated fatty acid peroxidation whose production is accelerated by oxidative stress.26,27 The levels of NO and MDA were markedly increased in LPS mice, demonstrating the accumulation of ROS and the occurring of oxidative stress. Baicalin administration could ameliorate LPS induced oxidative stress as evidenced by its significant inhibition on the LPS induced increase of NO and MDA.

Phospholipids, the major components of cell membranes, consisting of choline, phosphocholine and ethanolamine, were essential for maintaining the integrity of cell membranes.28,29 The obvious increase of choline and phosphocholine in both livers and kidneys, and ethanolamine in kidneys were observed in LPS mice compared with NC mice, reflecting the accelerated degradation of phospholipid and breakdown of cellular membranes. Excessive generation of ROS disrupted the ROS–antioxidant balance, causing a status of oxidative stress. Dramatically decreased levels of taurine and betaine were observed in the serum, kidney and liver of LPS mice. Taurine could scavenge ROS and protect the body from ROS induced injury.30,31 Betaine also played an important role in protecting cells from oxidative damage by improving the antioxidant status and decreasing lipid peroxidation.32 Therefore, the decrease of taurine and betaine in the LPS group manifested a self-protection mechanism of the body to alleviate ROS induced oxidative damage. Taking together, the conspicuous increase of choline and phosphocholine, and decrease of betaine and taurine indicated severe oxidative stress induced by LPS. Baicalin treatment significantly lowered the contents of choline, phosphocholine, and ethanolamine in septic mice, and enhanced the levels of taurine and betaine, manifesting its ability to ameliorate the oxidative stress.

4.3. Energy metabolism

Notable decreases of citrate in the serum and 2-OG in both livers and kidneys of the LPS group were observed compared with the NC group. As an important intermediate of the TCA cycle,33,34 their decrease suggested an inhibited TCA cycle by LPS, which was also verified by the significantly lowered expressions of citrate synthase (CS) and pyruvate kinase (PK) in the LPS group. As a crucial enzyme of the TCA cycle, CS catalyzed the irreversible conversion of oxaloacetic acid to citrate.35 Its down-regulation led to a reduced supply of citrate, thus a hampered TCA cycle. Pyruvate kinase (PK) catalyzed the conversion of phosphoenolpyruvate to pyruvate, an intermediate metabolite of glucose metabolism. The down-regulation of PK in livers and kidneys, and as a result, the decrease of pyruvate in the serum, suggested that the production of pyruvate from phosphoenolpyruvate was inhibited. Lactate and alanine, the anaerobic products of pyruvate,36 were notably decreased in the serum and kidneys, demonstrating an inhibited glycolysis. The TCA cycle and glycolysis are the two major means of glucose consumption, however, glucose was markedly decreased in the serum and liver of the LPS group, suggesting the acceleration of other means of glucose metabolism. Glucose could be converted into glucose-6-phosphate by phosphorylation, and finally produced NADPH in the pentose phosphate pathway.37–39 NADPH provides an electron during the reducing reaction, playing an important role in oxidative stress defense.40 Therefore, the increased levels of NADPH in the serum, liver and kidney suggested an enhanced NADPH synthesis to counteract the overproduction of ROS induced by LPS, reflecting self-protection of the body.

As the two major means of energy production, the inhibition of the TCA cycle and glycolysis definitely resulted in energy deficiency. Therefore, other means of energy supply have to be facilitated to replenish the unmet energy need, such as metabolism of ketone bodies , whereby ketone bodies, e.g. acetoacetate and 3-HB were oxidized to afford energy.41,42 Acetoacetate and 3-HB were metabolites of fatty acid β-oxidation in the liver, and could be transferred through the serum to muscles and other organs for their utilization for energy supply,43 which were well matched with the results in the LPS group: their levels were increased significantly in livers and slightly in kidneys, but markedly decreased in the serum. Markedly increased levels of LDL/VLDL were observed in the serum and livers of LPS mice, which might also indicate acceleration in fat mobilization to afford fatty acids for subsequent oxidation.

Compared with the NC group, a significant increase of adenosine, inosine in livers and kidneys were also found in the LPS group. Catabolism of ATP produced adenosine, which was further decomposed into inosine to produce extra energy.44 The increased adenosine and inosine evidenced the overconsumption of energy in LPS mice. The marked increase of creatine and creatinine in livers and kidneys, and a slight increase of serum creatine in LPS dosed mice suggested an accelerated utilization of creatine–phosphate,45,46 which acted as an energy reservoir by transferring high-energy phosphate to ADP to afford ATP for energy demand. These alternative energy producing mechanisms helped the maintenance of the ATP level, being essential for the survival of the body suffering from LPS induced energy crisis. After baicalin treatment, these disordered energy metabolism related metabolites were reversed, revealing a good improvement in energy supply by baicalin.

4.4. Amino acid metabolism

The levels of branched-chain amino acids (BCAAs), leucine, isoleucine and valine, were markedly increased in the livers of the LPS group. BCAAs are essential amino acids that have to be absorbed from food and are important precursors for protein synthesis.47,48 Their increases indicated enhanced protein decomposition due to the attack of the over-generated ROS induced by LPS.

Liver and kidney impairments induced by LPS were apparent in the histopathological inspection, and also being supported by the significantly increased ALT and Cr levels. Tryptophan was reported to treat liver disease, due to its ability to reduce the levels of pro-inflammatory cytokines.49 The significant decrease of tryptophan in livers thus indicated its over-consumption to protect the liver from LPS induced injury. Phenylalanine is an essential amino acid and can be converted into tyrosine under the catalysis of phenylalanine hydroxylase (PAH).50 LPS induced kidney injury could be evidenced by the increased level of phenylalanine and the decreased level of tyrosine in the kidneys of LPS mice, which has been observed in chronic renal damage.51 The levels of BCAAs, tryptophan, phenylalanine and tyrosine in LPS dosed mice were ameliorated to the levels close to those in the NC group after baicalin treatment, showing great protection of baicalin on LPS induced damage.

5. Conclusion

In this study, LPS-induced injury of sepsis and the treatment effects of baicalin were investigated using a 1H NMR based metabolomics approach combined with histopathological evaluation, clinical chemistry and molecular biological methods for the first time. The metabolomic profiles of the serum, liver and kidney extracts were analyzed by multivariate and univariate statistical analyses. Baicalin could ameliorate the abnormal metabolic status induced by LPS towards normal by ameliorating the inflammatory response, oxidative stress, energy and amino acid metabolism disturbance. This integrated metabolomics approach might help the development of a systematic vision of LPS-induced injury of sepsis and its treatment. These results showcased the ability of metabolomics as a holistic approach to characterize the global metabolic features of organisms, to understand pathological mechanisms, and to assess the therapeutical effects of medication.

Conflict of interest

We confirm that this manuscript has not been published by another journal. All authors have approved the manuscript and agree with its submission to your journal. The authors have no conflicts of interest to declare.

Supplementary Material

Click here for additional data file. (1.1MB, pdf)

Acknowledgments

The research work was financially supported by the Key Project of the National Natural Science Foundation of China (no. 81430092), the National Natural Science Foundation of China (no. 81173526), and the Program for Changjiang Scholars and Innovative Research Team in University (IRT_15R63).

Footnotes

†Electronic supplementary information (ESI) available. See DOI: 10.1039/c6tx00082g

References

  1. Anderson S. T., Commins S., Moynagh P. N., Coogan A. N. Brain, Behav., Immun. 2015;43:98–109. doi: 10.1016/j.bbi.2014.07.007. [DOI] [PubMed] [Google Scholar]
  2. Rivers E. P., Coba V., Whitmill M. Curr. Opin. Anesthesiol. 2008;21:128–140. doi: 10.1097/ACO.0b013e3282f4db7a. [DOI] [PubMed] [Google Scholar]
  3. Deng M., Scott M. J., Loughran P., Gibson G., Sodhi C., Watkins S., Hackam D., Billiar T. R. J. Immunol. 2013;190:5152–5160. doi: 10.4049/jimmunol.1300496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Shen M., Wang L., Yang G., Gao L., Wang B., Guo X., Zeng C., Xu Y., Shen L., Cheng K., Xia Y., Li X., Wang H., Fan L., Wang X. PLoS One. 2014;9:e88389. doi: 10.1371/journal.pone.0088389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Gao N., Zou D., Qiao H.-L. PLoS One. 2013;8:e53038. doi: 10.1371/journal.pone.0053038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Zhu J., Wang J., Sheng Y., Zou Y., Bo L., Wang F., Lou J., Fan X., Bao R., Wu Y., Chen F., Deng X., Li J. PLoS One. 2012;7:e35523. doi: 10.1371/journal.pone.0035523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen Y. C., Shen S. C., Chen L. G., Lee T. J. F., Yang L. L. Biochem. Pharmacol. 2001;61:1417–1427. doi: 10.1016/s0006-2952(01)00594-9. [DOI] [PubMed] [Google Scholar]
  8. Fiehn O., Putri S. P., Saito K., Salek R. M., Creek D. J. Metabolomics. 2015;11:1036–1040. [Google Scholar]
  9. Salek R. M., Arita M., Dayalan S., Ebbels T., Jones A. R., Neumann S., Rocca-Serra P., Viant M. R., Vizcaino J.-A. Metabolomics. 2015;11:782–783. [Google Scholar]
  10. Lamichhane S., Yde C. C., Schmedes M. S., Jensen H. M., Meier S., Bertram H. C. Anal. Chem. 2015;87:5930–5937. doi: 10.1021/acs.analchem.5b00977. [DOI] [PubMed] [Google Scholar]
  11. Powers R. J. Med. Chem. 2014;57:5860–5870. doi: 10.1021/jm401803b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Rankin N. J., Preiss D., Welsh P., Burgess K. E. V., Nelson S. M., Lawlor D. A., Sattar N. Atherosclerosis. 2014;237:287–300. doi: 10.1016/j.atherosclerosis.2014.09.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. De Meyer T., Sinnaeve D., Van Gasse B., Tsiporkova E., Rietzschel E. R., De Buyzere M. L., Gillebert T. C., Bekaert S., Martins J. C., Van Criekinge W. Anal. Chem. 2008;80:3783–3790. doi: 10.1021/ac7025964. [DOI] [PubMed] [Google Scholar]
  14. Dieterle F., Ross A., Schlotterbeck G., Senn H. Anal. Chem. 2006;78:4281–4290. doi: 10.1021/ac051632c. [DOI] [PubMed] [Google Scholar]
  15. Jung J. Y., Lee H. S., Kang D. G., Kim N. S., Cha M. H., Bang O. S., Ryu do H., Hwang G. S. Stroke. 2011;42:1282–1288. doi: 10.1161/STROKEAHA.110.598789. [DOI] [PubMed] [Google Scholar]
  16. van der Kloet F. M., Tempels F. W. A., Ismail N., van der Heijden R., Kasper P. T., Rojas-Cherto M., van Doorn R., Spijksma G., Koek M., van der Greef J., Makinen V. P., Forsblom C., Holthofer H., Groop P. H., Reijmers T. H., Hankemeier T. Metabolomics. 2012;8:109–119. doi: 10.1007/s11306-011-0291-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. De Meyer T., Sinnaeve D., Van Gasse B., Tsiporkova E., Rietzschel E. R., De Buyzere M. L., Gillebert T. C., Bekaert S., Martins J. C., Van Criekinge W. Anal. Chem. 2008;80:3783–3790. doi: 10.1021/ac7025964. [DOI] [PubMed] [Google Scholar]
  18. Iiott N. E., Heward J. A., Roux B., Tsitsiou E., Fenwick P. S., Lenzi L., Goodhead I., Hertz-Fowler C., Heger A., Hall N., Donnelly L. E., Sims D., Lindsay M. A. Nat. Commun. 2014;5:1–13. doi: 10.1038/ncomms4979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. De Meyer T., Sinnaeve D., Van Gasse B., Tsiporkova E., Rietzschel E. R., De Buyzere M. L., Gillebert T. C., Bekaert S., Martins J. C., Van Criekinge W. Anal. Chem. 2008;80:3783–3790. doi: 10.1021/ac7025964. [DOI] [PubMed] [Google Scholar]
  20. Chen J., Xu J., Li J., Du L., Chen T., Liu P., Peng S., Wang M., Song H. Int. Immunopharmacol. 2015;26:147–152. doi: 10.1016/j.intimp.2015.03.025. [DOI] [PubMed] [Google Scholar]
  21. Martin J. C., Canlet C., Delplanque B., Agnani G., Lairon D., Gottardi G., Bencharif K., Gripois D., Thaminy A., Paris A. Atherosclerosis. 2009;206:127–133. doi: 10.1016/j.atherosclerosis.2009.01.040. [DOI] [PubMed] [Google Scholar]
  22. Kim Y. K., Na K. S., Myint A. M., Leonard B. E. Prog. Neuro-Psychopharmacol. 2016;64:277–284. doi: 10.1016/j.pnpbp.2015.06.008. [DOI] [PubMed] [Google Scholar]
  23. da Silva M. D., Bobinski F., Sato K. L., Kolker S. J., Sluka K. A., Santos A. R. S. Mol. Neurobiol. 2015;51:19–31. doi: 10.1007/s12035-014-8790-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Illeperuma R. P., Kim D. K., Park Y. J., Son H. K., Kim J. Y., Kim J., Lee D. Y., Kim K.-Y., Jung D.-W., Tilakaratne W. M., Kim J. Int. J. Cancer. 2015;137:2545–2557. doi: 10.1002/ijc.29636. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Loehr J. A., Abo-Zahrah R., Pal R., Rodney G. G. Biophys. J. 2015;108:424A. [Google Scholar]
  26. Griendling K. K., FitzGerald G. A. Circulation. 2003;108:1912–1916. doi: 10.1161/01.CIR.0000093660.86242.BB. [DOI] [PubMed] [Google Scholar]
  27. Aziz I. A., Yacoub M., Rashid L., Solieman A. Acta Parasitol. 2015;60:735–742. doi: 10.1515/ap-2015-0105. [DOI] [PubMed] [Google Scholar]
  28. Cullis P. R., de Kruijff B. Biochim. Biophys. Acta. 1979;559:399–420. doi: 10.1016/0304-4157(79)90012-1. [DOI] [PubMed] [Google Scholar]
  29. Weiss D. E. Subcell. Biochem. 1973;2:201–235. [PubMed] [Google Scholar]
  30. Shimada K., Jong C. J., Takahashi K., Schaffer S. W. Adv. Exp. Med. Biol. 2015;803:581–596. doi: 10.1007/978-3-319-15126-7_47. [DOI] [PubMed] [Google Scholar]
  31. Zhang H., Hu C.-A. A., Kovacs-Nolan J., Mine Y. Amino Acids. 2015;47:2127–2141. doi: 10.1007/s00726-014-1886-9. [DOI] [PubMed] [Google Scholar]
  32. Shao H.-B., Chu L.-Y., Lu Z.-H., Kang C.-M. Int. J. Biol. Sci. 2008;4:8–14. doi: 10.7150/ijbs.4.8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Abdel-Salam O. M., Youness E. R., Mohammed N. A., Morsy S. M., Omara E. A., Sleem A. A. J. Med. Food. 2014;17:588–598. doi: 10.1089/jmf.2013.0065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Liang Q., Liu H., Wang C., Li B. Sci. Rep. 2015;6:203–208. [Google Scholar]
  35. Haslbeck M., Schuster L., Grallert H. J. Chromatogr., B: Biomed. Appl. 2003;786:127–136. doi: 10.1016/s1570-0232(02)00716-x. [DOI] [PubMed] [Google Scholar]
  36. Rogatzki M. J., Ferguson B. S., Goodwin M. L., Gladden L. B. Front. Neurosci. 2015;9:1–7. doi: 10.3389/fnins.2015.00022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nikel P. I., Chavarria M., Fuhrer T., Sauer U., de Lorenzo V. J. Biol. Chem. 2015;290:25920–25932. doi: 10.1074/jbc.M115.687749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wasylenko T. M., Ahn W. S., Stephanopoulos G. Metab. Eng. 2015;30:27–39. doi: 10.1016/j.ymben.2015.02.007. [DOI] [PubMed] [Google Scholar]
  39. Gebregiworgis T., Nielsen H. H., Massilamany C., Gangaplara A., Reddy J., Illes Z., Powers R. J. Proteome Res. 2016;15:659–666. doi: 10.1021/acs.jproteome.5b01111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lei R., Yang B., Wu C., Liao M., Ding R., Wang Q. Toxicol. Res. 2015;4:351–364. [Google Scholar]
  41. Stemmer K., Zani F., Habegger K. M., Neff C., Kotzbeck P., Bauer M., Yalamanchilli S., Azad A., Lehti M., Martins P. J. F., Mueller T. D., Pfluger P. T., Seeley R. J. Diabetologia. 2015;58:2414–2423. doi: 10.1007/s00125-015-3668-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tan Y., Li J., Liu X., Ko J., He X., Lu C., Liu Z., Zhao H., Xiao C., Niu X., Zha Q., Yu Z., Zhang W., Lu A. J. Proteome Res. 2013;12:513–524. doi: 10.1021/pr300965d. [DOI] [PubMed] [Google Scholar]
  43. Liu X., Liu Y., Qu Y., Cheng M., Xiao H. Toxicol. Res. 2015;4:948–955. [Google Scholar]
  44. Lin C.-Y., Huang F.-P., Ling Y. S., Liang H.-J., Lee S.-H., Hu M.-Y., Tsao P.-N. PLoS One. 2015;10:e0120429. doi: 10.1371/journal.pone.0120429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sakita M., Murakami S., Fujino H. J. Phys. Ther. Sci. 2014;26:263–267. doi: 10.1589/jpts.26.263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Stitt M., Zhu X.-G. Plant, Cell Environ. 2014;37:1985–1988. doi: 10.1111/pce.12290. [DOI] [PubMed] [Google Scholar]
  47. Cruzat V. F., Krause M., Newsholme P. J. Int. Soc. Sports Nutr. 2014;11:1–13. doi: 10.1186/s12970-014-0061-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Pechlivanis A., Kostidis S., Saraslanidis P., Petridou A., Tsalis G., Veselkov K., Mikros E., Mougios V., Theodoridis G. A. J. Proteome Res. 2013;12:470–480. doi: 10.1021/pr300846x. [DOI] [PubMed] [Google Scholar]
  49. Celinski K., Konurek P. C., Slomka M., Cichoz-Lach H., Brzozowski T., Konturek S. J., Korolczuk A. J. Physiol. Pharmacol. 2014;65:75–82. [PubMed] [Google Scholar]
  50. Fernandez-Canon J. M., Baetscher M. W., Finegold M., Burlingame T., Gibson K. M., Grompe M. Mol. Cell. Biol. 2002;22:4943–4951. doi: 10.1128/MCB.22.13.4943-4951.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Huang Q., Zhang J., Luo L., Wang X., Wang X., Alamdar A., Peng S., Liu L., Tian M., Shen H. Toxicol. Res. 2015;4:939–947. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file. (1.1MB, pdf)

Articles from Toxicology Research are provided here courtesy of Oxford University Press

RESOURCES