Metabolic analysis of the effect of rheum on a taurocholate-induced acute pancreatitis rat model

The effects of rheum on serum parameters in a taurocholate-induced acute pancreatitis (AP) rat model were investigated using pathological and biochemical tests, and a proton nuclear magnetic resonance (1H NMR)-based metabonomic strategy. Healthy rats and rats with AP were either treated with rheum (7.5% at a dose of 1.5 g/kg) or left untreated. Serum samples were collected from the AP and rheum-treated groups at 6, 12, and 24 h after treatment. The effect of rheum on pathological changes in the pancreatic was investigated to validate the AP model. We obtained 1H NMR spectra and analyzed the results using the partial least squares discriminant method. The results of the pathological and metabolic analyses revealed an amelioration of multiple metabolic abnormalities and an increase in the aerobic respiration ratio after treatment, compared with the AP groups. These results were attributed to improvements in energy supply and the elimination of metabolic products. The study also promoted NMR-based metabonomic analysis as a feasible method of assessing traditional Chinese drugs.


INTRODUCTION
Pancreatitis is inflammation of the pancreas, including aseptic, bacterial, and chronic non-specific inflammation, and specific inflammation caused by tuberculosis or parasites.Acute pancreatitis (AP), chronic pancreatitis, and specific pancreatitis are the most common clinical types (Crocker, 2012), of which AP is associated with the highest morbidity and mortality (Barreto, Rodrigues, 2008).Evidence indicates that pancreatin-induced autodigestion of the pancreatic tissue can lead to acute chemical inflammation in AP (Markus, 2013).However, there is currently no effective clinical treatment for AP owing to its complex pathophysiology and pathogenesis (Braganza, 2001;Gaisano, Gorelick, 2009;Kostyrnoi, 2012).
Rheum is a perennial herbaceous plant of the Polygonaceae family.According to the Pharmacopoeia of the People's Republic of China 2010 (Chinese Pharmacopoeia Commission, 2010), the dry roots or rhizomes of Rheum palmatum L., Rheum tangguticum Maxim.ex Balf, and Rheum officinale Baill are the standard medicinal portions of rheum.Rheum has been used widely in traditional Chinese medicine to treat numerous commonly encountered diseases such as acute abdominal disorders (Chinese Pharmacopoeia Commission, 2010).Zhang et al. (2004) reported that patients who received integrated traditional Chinese medicine-western medicine (TCM-WM) presented with lower mortality (10.77%) than those who received monotherapy (40.52%).Moreover, classic ancient books and records mention the use of rheum as therapy for AP (Jiangsu New Medical College, 1997;Xiao, 2002).With the development of integrated TCM-WM, rheum has become one of the main components of traditional Chinese medicine.The role of rheum in a variety of effective treatments has been explored in several studies (Ge, Chen, Wang, 2008;Gu, Wang, 2009;Gao, He, 2007).
Metabolomics is a systemic approach to metabolism research by analysis of metabolite profiles in the body fluids of an organism, followed by comparison of the variations induced by external stimuli or genovariation (Nicholson, Lindon, Holmes, 1999).Nuclear magnetic resonance (NMR), which has been very useful for characterizing physiological metabolites in biofluids, is one of the techniques used in metabolomics (Chen et al., 2012).However, analysis by NMR is difficult owing to the complexity of NMR spectral profiling arising from thousands of signals from endogenous molecules.Usually, principal component analysis (PCA), cluster analysis, and neuron analysis are coupled with NMR data analysis; PCA is the most popular technique.Partial least squares projection to latent structure discriminant analysis (PLS-DA), which is a PCA technique, is a type of supervised analysis related to unsupervised analysis.It can involve dividing the samples into groups before analysis, and ignoring random differences in existing groups (Zhang et al., 2013).Since 1999, metabonomics has been widely used to assess the pharmacological safety and toxicological mechanism of drugs in animal models.(Shi et al., 2007;Wei et al., 2008).
In the present study, we used a metabolomic approach involving proton nuclear magnetic resonance ( 1 H NMR) spectroscopy with PCA to analyze samples from AP rat serum.PLS-DA was used to explore the differences among groups and to characterize the metabonomic metabolites.Furthermore, the effect of rheum on pancreatic pathology was also investigated to validate the AP rat model.The objectives of this research were: to demonstrate a new method of assessing traditional Chinese drugs; to investigate the oral potency of rheum in rat metabolism; to characterize alterations in metabolites after administering rheum to AP rat models; and to illustrate the underlying mechanisms.

Test compound preparation
Dried rheum root was acquired from Derentang Pharmacy (Chengdu, China) in August, 2009.To investigate the main constituents of the rheum, the root was ground into a powder and passed through a sieve (60 mesh).The contents of the sample powder were then investigated using high-performance liquid chromatography (HPLC).The powder was stored in a desiccated environment at room temperature until required.Aloe-emodin (110795-200806), Physcion (110758-200610), Emodin (110756-200710), Chrysophanol (0796-200705), and Rhein (0757-200802) standardization controls were from the National Institute for the Control of Pharmaceutical and Biological Products.

Mass detection
The rheum was made into a suspension solution before our experiment commenced.The rheum powder (7.50 g) was added to approximately 100 mL boiled vehicle (0.5% saline plus sodium carboxymethyl cellulose), mixed, and 100.0 mL was measured out.Finally, we obtained a well-distributed brownish yellow 7.5% rheum suspension.Afterwards, the rheum was analyzed in accordance with the Pharmacopoeia of the People's Republic of China 2010 (Chinese Pharmacopoeia Commission, 2010).The HPLC system, comprising a Hypersil ODS2 column (150 × 4.6 mm) (Thermo Hypersil Co., United Kingdom) and methanol:0.1% phosphate solution (85:15), operated at a volumetric flow rate of 1.0 mL•min -1 , a wavelength of 254 nm, and a temperature of 25 °C, was used to validate the mass of the crude drug.The HPLC chromatograms are shown in Figure 1.
By matching the retention times and peak area, the contents of the five principal components in rheum, i.e., Aloe-emodin, Physcion, Emodin, Chrysophanol, and Rhein, were determined to be 2.57, 3.65, 1.91, 6.93, and 1.01 mg•g -1 , respectively.All these five major components constituted 1.61% of the crude mixture, which met the quality requirements (not less than 1.5%).Rheum was administered to the rats in this study.

Animals
The study was approved by the Institutional Animal Use and Care Committee of Sichuan University, and the protocol was compliant with the Guide for the Care and Use of Laboratory Animals.Male Wistar rats (250-300 g; free from specified pathogens) were supplied by Chengdu Dashuo Biotechnology Company Ltd.The animals were kept at the Animal Laboratory of Sichuan University under the following conditions: a 12 h light/dark cycle; free access to standard laboratory feed and water; 20-24 °C; and 50-60% humidity.
The rats were all fasted for 12 h but had free access to water before the operation.Two hours before the operation, rats from the control-medication group, the 6-h medication group, the 12-h medication group, and the 24-h medication group were administered 7.5% rheum suspension (2 mL/100 g, intragastric administration (ig)), and the other groups were administered 0.5% sodium carboxymethyl cellulose in physiological saline solution (2 mL/100 g, ig).

Operation
The rats were anesthetized by intraperitoneal injection of 10% chloralhydrate (0.3 mL/100 g).An incision (approximately 1 cm long) was made along the midline of the abdomen and the duodenum was clamped.A needle was inserted into the anterior wall of the duodenum close to the bile duct.When the needle had penetrated 1 cm into the choledoc, the biliopancreatic duct on the porta of the hepatic side was clamped using a small artery clamp, and physiological saline solution (0.1 mL/100 g) (control and control-medication group) or 0.3% sodium taurocholate (0.1 mL/100 g) (other AP and medication groups) was injected uniformly.After 5 min of injection, the artery clamp was removed and the abdomen was sutured.

Sample collection
The rats were sacrificed by femoral artery exsanguination under anesthesia at 6 h (6-h AP, 6-h medication, control, and control-medication groups), 12 h (12-h AP and 12-h medication groups), and 24 h (24-h AP and 24-h medication groups) after the operation.The blood samples were collected in centrifuge tubes and centrifuged (4°C, 3000 rpm, 10 min) to remove the precipitates.The serum samples were separated and stored at -80°C for biochemical and metabonomic analysis.The pancreas was removed to provide specimens for pathological examination following blood collection.

H NMR spectroscopic measurement and analysis
After the biochemical tests, serum samples (400 μL) were centrifuged (4°C, 14000 rpm, 10 min), and the supernatant was mixed with 50 μL D 2 O (99.8%, 0.5 mL) and 50 μL phosphate buffer solution (0.2 mol/L Na 2 HPO 4 and 0.2 mol/L NaH 2 PO 4 , pH 7.4) to minimize variations in the pH of the serum samples.The samples were transferred to 5-mm NMR tubes, and investigated using an 1 H NMR spectrometer at 600.13 MHz (Bruker Av II 600 spectrometer).The water-suppressed Carr-Purcell-Meiboom-Gill pulse (CPMG pulse) sequence method was used.The 64 free induction decay (FID) signals were represented by 64 K data points with a spectral width of 7002.8Hz, an acquisition time of 4.68 s, and a total pulse recycle delay of 6.68 s at a temperature of 298 K.
All NMR spectra were manually re-phased and baseline-corrected.The data were then reduced to 225 integrated regions of equal width (0.04 ppm) corresponding to the δ9.5-δ0.5 region using the MestRe-C4.7 software package.The δ5.0-δ4.6 regions of the CPMG spectra were excluded to eliminate artifacts caused by the resonance of residual water.All the remaining spectral segments were scaled to the total integrated area of the spectrum to reduce variations in concentration.
The 1 H NMR spectra were all obtained by the same procedure, and all segments were imported into Microsoft Excel to obtain a (7 × 215)-dimension matrix (the number of samples was 7; the number of segment regions outside δ5.0-δ4.6 was 215).The output files were fed into SIMCA-P11.0software for PLS-DA to maximize the difference between groups.A plot of the scores was used to establish significance in the data, and the loading plots were used to identify different chemical shifts.Metabolites with different chemical shifts were assigned based on the human metabolome database (HMDB) (Wishart et al., 2007), and certain other related research articles (Yang et al., 1999;Bloomer, Stagliano, Gazzillo, 1993).

Effect of rheum on pancreatic pathology in AP rats
Pancreatic pathology was determined using the Schmidt scoring criteria.The Schmidt scores for pancreatic edema, acinar necrosis, hemorrhage, fat necrosis, inflammation, and perivascular infiltration were calculated.The rats in the control and control-medication groups were assigned a score of 0 on average.As illustrated in Figure 2, the rats in the AP groups scored higher than those in the medication groups for each pathological category at the same time-point, and the Schmidt score increased as time progressed after AP modeling in all groups.Similar pathological alterations were revealed by microscopic examination (Figure 3).

Biochemical analysis of serum samples
The results of the biochemical analyses are reported in Table I.Compared with the control group, there were reduced glucose and increased fatty acid levels in the AP and AP medication groups (6 h, 12 h, and 24 h).For both valine and histidine, increases only occurred in the AP groups (6 h, 12 h, and 24 h).Serum creatinine levels in all the AP and AP medication groups (6 h, 12 h, and 24 h) were much higher than in the control and controlmedication groups.

PLS-DA analysis of serum samples
We used PLS-DA to classify the differences among the groups and to characterize the metabolites that contributed to disparities.The δ9.5-δ0.5 ppm region was chosen for calculation of PLS-DA.According to the analysis of serum samples, the control group, the controlmedication group, the AP groups (6 h, 12 h, and 24 h), and the medication groups (6 h, 12 h, and 24 h) showed significant differences in chemical shifts, especially in the δ3.00-δ4.00and δ4.50-δ5.20 regions.The analyses are described below.

Summary
The detailed variations in the endogenous metabolites detected in the serum samples indicated on the loadings plots are shown in Tables II and III.

DISCUSSION
In recent decades, various animal models have been established for AP research; the four most common models are induced by cerulein (Dembinski et al., 2008), l-arginine (Ji et al., 2003), dl-ethionine (Sewing, Albinus, 1969), FIGURE 9 -Loading plot of the proton nuclear magnetic resonance ( 1 H NMR) spectra of serum samples from the control and medication groups (6-h, 12-h, and 24-h).and sodium taurocholate (Lombardi, Estes, Longnecker, 1975).Of those, the sodium taurocholate-induced AP rat model exhibits high repeatability and an achievement ratio of 100%.Moreover, topical administration of sodium taurocholate causes limited impairment of the other organs.Furthermore, the pathological changes after modeling are similar to those in pancreatitis caused by autodigestion of the pancreas tissue.In this study, we replicated the stability of a sodium taurocholate-induced AP rat model.PLS-DA, which is a PCA technique, is a type of supervised analysis related to unsupervised analysis; it can involve dividing the samples into groups before analysis, and ignoring random differences between existing groups.Allowing for the grouping of animal serum samples in our test, the PLS-DA method was used to determine the differences between the metabonomic values found with rheum and the physiological metabonomic parameters among the other groups, and to characterize the metabolites on score and loading plots.

Effect on pancreatic pathological changes
According to the pancreatic tissue slices, pathological changes were not noted in the control and control-medication groups.However, pancreatic dropsy, acinus necrosis, hemorrhage, steatonecrosis, and inflammatory cell infiltration were observed in the other groups, which indicated that sporadic necrosis occurred and protein infiltrate was markedly alleviated.Infiltration of inflammatory cells was obvious in the AP medication groups.There was also a reduction in the pathologic score in the AP medication groups.The results described above indicate that rheum might ameliorate the pathological changes in rats with AP.

Effect on metabolism of the major components of rheum
Emodin, chrysophanol, rhein, and 2,5-dimethyl-7methoxy chromone are the main components of rheum (Yang et al., 1999).A study by Bloomer, Stagliano, and Gazzillo (1993) indicated that the chemical shifts of emodin, chrysophanol, rhein, and 2,5-dimethyl-7methoxy chromone were δ7. 10, δ7.34, δ7.42, and (δ7.82, δ7.9), respectively.In the test, chrysophanol, rhein, and 2,5-dimethyl-7-methoxy chromone increased in the control-medication group compared with the control group.Similarly, there were increased levels of emodin and 2,5-dimethyl-7-methoxy chromone in the medication groups, which indicated that rheum is efficiently absorbed via oral gavage.As well as emodin and 2,5-dimethyl-7methoxy chromone, chrysophanol and rhein were present at the same levels in the control and medication groups.It is possible that the serum concentrations of chrysophanol and rhein decreased as the drug was metabolized.However, residual emodin and 2,5-dimethyl-7-methoxy chromone were detected in the serum because their contents in rheum were much higher than those of chrysophanol and rhein (Bloomer, Stagliano, Gazzillo, 1993).Moreover, Shia et al. (2009) suggested that rhein, emodin, and chrysophanol glucuronides were predominant in the blood via oral administration; in our studies, emodin, chrysophanol, rhein, and 2,5-dimethyl-7-methoxy chromone were still significantly distinguishable on the loading plots compared with the other metabolites, which might be attributed to the shortening of persistence time and pathological changes.
Effect on carbohydrates, lipids, protein metabolism, and nucleic acid metabolism α-ketoglutaric acid, succinic acid, malic acid, and citric acid are the main intermediate products in the tricarboxylic acid (TCA) cycle.In the TCA, acetyl CoA condenses with oxaloacetic acid to produce citric acid.Isocitric acid produces α-ketoglutaric acid by oxidative decarboxylation, succinyl CoA transforms to succinic acid by releasing adenosine triphosphate (ATP), and fumaric acid produces malic acid by hydration.In our research, the levels of α-ketoglutaric acid, succinic acid, malic acid, and citric acid increased in the control-medication groups, but no significant differences were noted in the other groups compared with the control group.However, in the AP groups, succinic acid decreased significantly.As a result, probably related to a negative chemical equilibrium shift in TCA that led to the accumulation of intermediate metabolites, the activation of TCA in the AP model rats presumably occurred to a small extent, and rheum was able to significantly reduce this activation in normal rats but not in AP rats.
During glycolysis, 2-phosphoglyceric acid produces phosphoenolpyruvate by dehydration.The phosphoric acid from phosphoenolpyruvate is transferred to adenosine diphosphate (ADP), and pyruvic acid and ATP are produced.The process ultimately produces lactic acid and [NAD + ] ( Murray et al., 2003).In our test, an increase in the level of phosphoenolpyruvate and lactic acid occurred in the AP groups compared with the control group.This suggested an acceleration of glycolysis in the AP rats, which was consistent with reports of anoxia and decreased aerobic respiration ratio caused by pulmonary injury and dehydration in AP patients (Surbatovic et al., 2005;Xie, Xu, 2008).In the medication groups, phosphoenolpyruvate was present at the same level as in the control group, although an increase in lactic acid was also observed.This result indicates that rheum might increase the level of glycolysis by decreasing phosphoenolpyruvate and accelerating the process.
Compared with the control group, the level of lipids decreased and the level of β-hydroxybutyric acid increased in the control-medication group.This might have been induced by the acceleration of lipid metabolism, because fatty acids are decomposed to ketones and transferred to acetyl CoA in the liver (Berg, Tymoczko, Stryer, 2002).In the AP groups, both acetoacetic acid and β-hydroxybutyric acid increased rapidly, indicating that the hydrolysis of fatty acids in the AP rats was accelerated.In combination with the acceleration of glycolysis in the AP rats, aerobic respiration might not have been able to provide sufficient energy, and lipid metabolism acted as a supplement.In the medication groups, there was no change in ketone metabolites compared with the control group, which might be attributed to a reverse in the aerobic respiration ratio, indicating that rheum slows energy metabolism.
Moreover, the lactic acid/pyruvic acid (L/P) ratio was directly proportional to the level of tissue anoxia, and the L/P balance was always affected by the [NADH]/ [NAD + ] ratio (Shia et al., 2009).The [NADH]/[NAD + ] ratio decreased in the AP groups and increased in the control-medication and AP-medication groups.Based on the results, the rate of glycolysis was normal, and rheum probably prevented inhibition of the energizing process, which can produce a higher [NADH]/[NAD + ] ratio through a negative equilibrium shift, along with a deceleration of lipid metabolism.
The levels of valine and histidine only increased in the AP groups compared with the control group, which indicated an acceleration of protein catabolism.Considering the decreased ratio of aerobic respiration, we conjectured that the decomposition of proteins provided energy in the AP groups.Rheum produced an increase in the ratio of aerobic respiration and prohibited protein catabolism, as indicated by the absence of a significant difference between the control and medication groups.
Increased adenine and pyrimidine in the AP groups indicated accelerated nucleic acid metabolism.The level of formic acid also decreased in the AP groups, which proved that the consumption of formic acid was lifted following the improvement in the synthesis of purine nucleotides.In the medication and control groups, adenine, pyrimidine, and formic acid were present at the same levels, indicating that an equilibrium of nucleic acids occurred, and rheum stabilized nucleic acid metabolism.

Effect on metabolism of digestion and excretion
Creatinine and urea nitrogen levels are standard indicators for the evaluation of renal function.The creatinine content decreased in the control-medication group, and increased in the AP and AP-medication groups.According to some reports, emodin promotes creatine excretion (Yu et al., 2002), which is in-line with the results of the control-medication group.This was corroborated by a report that AP could cause pathological changes in kidneys and induces urinary abnormalities (oliguria, proteinuria, hematuria, or renal failure) with an occurrence rate of 50% (Xie et al., 2003).The increases in the levels of creatinine and urea nitrogen in the AP modeling groups confirmed the phenomenon.Furthermore, there was no significant difference in the level of urea nitrogen in the medication groups compared with the control group, which possibly indicates rheum as a prompt therapy for renal abnormality in AP model rats.
Bilirubin is one of the main components in the hepato-enteric circulation.In our test, the purgation effect of rheum (Yamamoto et al., 2000) possibly led to a decrease in the level of bilirubin.The level of bilirubin decreased significantly in the control-medication and medication groups, which corroborated the data from related reports on the purgation effect.

Other metabolites
Compared with the control group, the levels of cholestanone and testosterone increased in the controlmedication group.Rheum possibly promotes growth and development.However, it did not have a significant effect in the AP rat models, possibly owing to the abnormality of metabolism.TMAO is the degradation product of choline, and an increase of TMAO always indicates rupture of the cell membrane (Grifin et al., 2001).Therefore, the increase of TMAO in the serum samples from the AP and medication groups was induced by the digestion of pancreatic tissue.Hippuric acid and PAG are usually produced by intestinal bacteria (Delaney et al., 2004;Phipps et al., 1998), and PAG is the end product of phenylalanine in rodents (James et al., 1972).The increase in the levels of hippuric acid and PAG in the AP groups indicated intestinal bacteria entering the blood, whereas the remaining hippuric acid in the medication groups indicated inhibition by rheum.The decreased 5-HT level in the AP groups might have been caused by the rupture of islets of Langerhans in the AP rats, because 5-HT is principally secreted by the islets of Langerhans (Mann, Brent, Arango, 2001).
In general, the metabolism of fat, protein, and nucleic acids was accelerated, and the aerobic respiration ratio was reduced in the AP rats.During AP, the cell membrane ruptures, intestinal bacteria enter the blood, and circulatory, respiratory, urinary, and digestive system abnormalities occur.The results were consistent with previous reports that systemic inflammatory response syndrome (SIRS) and multiple organ failure (MOF) are caused by AP (Schlosser et al., 2002;Isenmann, Rau, Beger, 2001;Buter et al., 2002).When rheum was administered to the rats, multiple metabolic abnormalities such as the occurrence of amines were not observed, even though several metabolic pathways were impacted.This indicates that rheum might improve energization in organisms.Previous studies have shown that the therapeutic effect of rheum is associated with three underlying mechanisms: (1) rheum increases the level of testosterone in an organism, and leads to accelerated regeneration of erythrocytes (Istvan, 2005); (2) rheum is a traditional Chinese medicine with a therapeutic effect on acute pulmonary injury (Surbatovic et al., 2005), and can therefore improve the efficiency of lung ventilation; and (3) rheum exerts anti-inflammatory and antibiotic effects (Zhou et al., 2006;Wu et al., 2005), both of which decrease vascular permeability and increase relative blood volume, thereby improving oxygen supply in the blood capillaries.Moreover, the elimination of metabolic products via the purgation effect of rheum is also a critical mechanism.

CONCLUSION
Analysis of serum samples using 1 H NMR and PLS-DA revealed that 7.5% rheum exerted therapeutic action on acute pancreatitis rat models, which was consistent with the pathology evaluation.By profiling endogenous metabolites, we discovered that the therapeutic effects could be caused by an improvement in energization and the elimination of metabolic products in the organism.The significant differences in the levels of metabolites among the test groups also supported the potential use of NMR-based metabonomic analysis for the investigation of traditional Chinese drugs.

FIGURE 3 -
FIGURE 3 -Hematoxylin and eosin (HE) staining (200×).A, B: Normal pancreatic tissue with no pathological damage.C: Edema (indicated by the yellow arrow), inflammation infiltrate (indicated by the blue arrow), and mild hemorrhage (indicated by the blue oval) are evident.D: Inflammation infiltrate (indicated by the blue arrow) is less pronounced than in C. E: Infiltrate (indicated by the blue arrow) and severe hemorrhage (indicated by the blue oval) are clear.F: There is mild infiltrate (indicated by the blue oval).G: Severe acinar necrosis (indicated by the green oval).H: Less pronounced acinar necrosis compared with G (indicated by the green oval).

FIGURE 2 -
FIGURE 2 -Schmidt scores for the acute pancreatitis (AP) and medication groups.*p < 0.05 for comparison between the corresponding groups (6-h AP group vs 6-h medication group; 12-h AP group vs 12-h medication group; 24-h AP group vs 24-h medication group).

FIGURE 5 -
FIGURE 5 -Loading plot of the proton nuclear magnetic resonance ( 1 H NMR) spectra of serum samples from the control and control-medication groups.

FIGURE 4 -
FIGURE 4 -Score plot of the first two values of the proton nuclear magnetic resonance ( 1 H NMR) spectra of serum samples from the control and control-medication groups.

FIGURE 7 -
FIGURE 7 -Loading plot of the proton nuclear magnetic resonance ( 1 H NMR) spectra of serum samples from the control and acute pancreatitis (AP) groups (6-h, 12-h, and 24-h).

FIGURE 6 -
FIGURE 6 -Score plot of the first two values of the proton nuclear magnetic resonance ( 1 H NMR) spectra of serum samples from the control and acute pancreatitis (AP) groups (6-h, 12-h, and 24-h).

FIGURE 8 -
FIGURE 8 -Score plot of the first two values of the proton nuclear magnetic resonance ( 1 H NMR) spectra of serum samples from the control and medication groups (6-h, 12-h, and 24-h).

TABLE I -
Serum levels of major metabolites in the control, control-medication, and acute pancreatitis (AP) groups at 6 h, 12 h, and 24 h; and in the medication groups at 6 h, 12 h, and 24 h *p < 0.05 for comparison between the corresponding groups (6-h AP group vs 6-h medication group; 12-h AP group vs 12-h medication group; 24-h AP group vs 24-h medication group).† p < 0.05 vs control group.

TABLE II -
Summary of the variations in metabolites detected in the serum samples from the control-medication group, the acute pancreatitis (AP) groups (6-h, 12-h, and 24-h), and the medication groups (6-h, 12-h, and 24 h) compared with the control group

TABLE III -
Summary of the variations in metabolites detected in the serum samples from the medication groups compared with the acute pancreatitis (AP) groups Note: ↑, increases; ↓, decreases; -, no change