Zheng is the basic theory and essence of traditional Chinese medicine (TCM) in diagnosing diseases. However, there are no biological evidences to support TCM Zheng differentiation. In this study we elucidated the biological alteration of cirrhosis with TCM “Liver-Kidney Yin Deficiency (YX)” or “Dampness-Heat Internal Smoldering (SR)” Zheng and the potential of urine metabonomics in TCM Zheng differentiation. Differential metabolites contributing to the intergroup variation between healthy controls and liver cirrhosis patients were investigated, respectively, and mainly participated in energy metabolism, gut microbiota metabolism, oxidative stress, and bile acid metabolism. Three metabolites, aconitate, citrate, and 2-pentendioate, altered significantly in YX Zheng only, representing the abnormal energy metabolism. Contrarily, hippurate and 4-pyridinecarboxylate altered significantly in SR Zheng only, representing the abnormalities of gut microbiota metabolism. Moreover, there were significant differences between two TCM Zhengs in three metabolites, glycoursodeoxycholate, cortolone-3-glucuronide, and L-aspartyl-4-phosphate, among all differential metabolites. Metabonomic profiling, as a powerful approach, provides support to the understanding of biological mechanisms of TCM Zheng stratification. The altered urinary metabolites constitute a panel of reliable biological evidence for TCM Zheng differentiation in patients with posthepatitis B cirrhosis and may be used for the potential biomarkers of TCM Zheng stratification.
Cirrhosis is scarring of the liver and also is the final stage of many chronic liver diseases, leading to portal hypertension and end-stage liver disease [
Zheng (Chinese character transliteration) is a temporary state at one time and is also known as a traditional Chinese medicine (TCM) syndrome [
In this study, we conduct a urinary metabonomic study on a group of liver cirrhosis patients (
We used a multicenter, multistage sampling method to obtain a cohort of representative samples of male patients in the general liver cirrhosis population. Patients were eligible to enter the study if they were clinically diagnosed liver cirrhosis due to chronic hepatitis B infection according to the “Guideline on prevention and treatment of chronic hepatitis B in China (2005).” The guideline (2005 version) was jointly revised in 2007 by Chinese Society of Hepatology, Chinese Medical Association, and Chinese Society of Infectious Diseases, Chinese Medical Association [
Exclusion criteria in the study were patients having a history of hepatitis A, C infection, alcohol or drug abuse, liver cancer, neoplastic liver diseases, hepatotoxic medication, and autoimmune liver disease in the past 6 months before recruiting into the study and other conditions likely to interfere with the study, such as overt hepatic encephalopathy (West Heaven Criteria grade II through IV), spontaneous bacterial peritonitis, upper gastrointestinal haemorrhage, and hepatorenal syndrome. Those with a history of severe primary heart, brain, lung, spleen, kidney, endocrine diseases, hematological disorder, and psychosis were also excluded from the study. Meanwhile, those not matching TCM YX Zheng or SR Zheng diagnosis criteria of cirrhosis were ruled out.
A total of 63 patients, aged between 33 and 58, were enrolled in the study from Shuguang Hospital, Longhua Hospital, and Putuo District Center Hospital affiliated to Shanghai University of Traditional Chinese Medicine (Shanghai, China) and Shanghai Public Health Center affiliated to Fudan University (Shanghai, China) between January 1, 2007, and December 31, 2008. All patients were clinically stable at the time of assessment. Patients were spontaneously divided into YX Zheng subgroup (
A cohort of 31 male participants was recruited as healthy controls from the Physical Examination Center of Shuguang Hospital. There was no significant difference in age, height, body weight, and BMI between healthy controls and liver cirrhosis patients or between TCM YX Zheng and SR Zheng of cirrhosis patients (Table
Clinical information and characteristics of human subjects.
Variable | Control | Liver cirrhosis | TCM Zhengs of liver cirrhosis | |
---|---|---|---|---|
YX subgroup | SR subgroup | |||
Patients ( |
31 | 63 | 33 | 30 |
Age (y) |
|
|
|
|
Body height (cm) |
|
|
|
|
Body weight (kg) |
|
|
|
|
BMI |
|
|
|
|
RBC (1012/L) |
|
|
||
WBC (109/L) |
|
|
||
HB (g/L) |
|
|
||
NEUT# (109/L) |
|
|
||
LYM# (109/L) |
|
|
||
PLT (109/L) |
|
|
||
Alb (g/L) |
|
|
||
Glb (g/L) |
|
|
||
A/G (100%) |
|
|
||
ALT (IU/L) |
|
|
||
AST (IU/L) |
|
|
||
GGT (IU/L) |
|
|
||
ALP (IU/L) |
|
|
||
CHE (IU/L) |
|
|
||
TBiL ( |
|
|
||
DBiL ( |
|
|
||
PT (sec) |
|
|
||
INR (%) |
|
|
||
BUN (mmol/L) |
|
|
||
Cr ( |
|
|
||
TCH (mmol/L) |
|
|
||
TG (mmol/L) |
|
|
||
APOA-1 (g/L) |
|
|
||
FPG (mmol/L) |
|
|
||
AFP (ng/mL) |
|
|
Note: The results are presented as mean ± SD and were compared by
BMI: body-mass index; RBC: red blood cell; WBC: white blood cell; HB: haemoglobin; NEUT#: absolute neutrophil Count; LYM#: absolute lymphocyte count; PLT: platelet; Alb: albumin; Glb: globulin; A/G: albumin/globulin; ALT: alanine aminotransferase; AST: aspartate aminotransferase; GGT: gamma glutamyl transferase; ALP: alkaline phosphatase; CHE: cholinesterase; TBiL: total bilirubin; DBiL: direct bilirubin; PT: prothrombin time; INR: international normalized ratio; BUN: blood urea nitrogen; Cr: creatinine; TCH: total cholesterol; TG: triglycerides; APOA-1: apolipoprotein A-1; FPG: fasting plasma glucose; AFP: alpha-fetoprotein.
Ethical approval for these studies was obtained from the ethics committees of the four hospitals mentioned above. The study was carried out in compliance with the Declaration of Helsinki (55th World Medical Association General Assembly, Tokyo, 2004). All participants have written the informed consent prior to the study.
Before the study, all study investigators, including medical students, trained general practitioners, and nurses, had completed a training program for methods and requirements of samples collection and obtained a manual of detailed procedure that guided how to manage the questionnaires, anthropometric measurements, and biological samples (urine and serum). All participants completed a questionnaire documenting their anthropometric measurements (e.g., weight and height), sociodemographic status (e.g., age, sex, education, and career), personal and family health history (e.g., hypertension, diabetes, liver disease, and surgery), lifestyle (e.g., smoking and alcohol consumption), and TCM Zheng scale (published in
Serum biochemical assay was performed with an automatic biochemistry analyzer for the analysis of blood routine, liver, and renal function markers. The data were offered by clinical laboratory of each hospital participating in the study. All questionnaires and serum biochemical indices were stored and analyzed by Key Laboratory of Liver and Kidney Diseases (Ministry of Education), Institute of Liver Diseases, Shuguang Hospital, Shanghai, China.
GC-MS profiling and data analysis of urine samples completed by Center for Chinese Medical Therapy and Systems Biology, Shanghai University of Traditional Chinese Medicine, Shanghai, China. The urine sample preparation for GC-MS analysis was performed according to our previously published method with minor modification [
Raw GC-MS data were converted into AIA format (NetCDF) files by Agilent GC-MS 5975 Data Analysis software, and subsequently the data information was extracted by the XCMS toolbox using the parameters as previously described. The XCMS output (TSV file) was introduced to Matlab software version 7.0 (The MathWorks, Inc.), where internal standard (IS) peaks and impurity peaks from column bleeds and derivatization procedure were excluded. The remaining ion features with high correlation of abundance within the same retention time group were combined into a single compound so as to obtain the total numbers of compounds and simplify data matrix for multivariate statistical analysis. The intensities of ion features (area) were further normalized to the total area for each sample to eliminate the variations caused by the different volume of individual urine sample and arranged on a three-dimensional matrix consisting of arbitrary peak index (RT-
UPLC-QTOFMS profiling and data analysis of urine samples were completed by Key Laboratory of Bioorganic Phosphorus Chemistry and Chemical Biology (Ministry of Education), Department of Chemistry, Tsinghua University, Beijing, China. The urine sample preparation for UPLC-QTOFMS was performed according to our previous works [
A Waters ACQUITY UPLC system coupled with an orthogonal acceleration time-of-flight mass spectrometry equipped with an electrospray interface (Waters Corp., Milford, USA) was used for metabonomic profiling. Chromatographic separation was performed on a Waters ACQUITY BEH C18 column (100 × 2.1 mm, 1.7
The peak picking, peak alignment, and peak filtering of the raw UPLC-QTOFMS data were carried out with the MarkerLynx Application Manager Version 4.1 (Waters, Manchester, UK). The parameters used were retention time range 0–18 min, mass range 50–1000
Compound annotation from UPLC-QTOFMS data was performed by comparing the accurate mass (
The interpretation for data analysis derived from GC-MS and UPLC-QTOFMS was performed by Laboratory of Systems Biomedicine (Ministry of Education), Shanghai Center for Systems Biomedicine, Shanghai Jiao Tong University, Shanghai, China. The verification of data was performed by Cancer Center, University of Hawaii, Honolulu, USA, which ensured that data were complete, accurate, and verifiable from source data. The trial profile is shown in Figure
Scheme of the research design.
For GC-MS, the resulting three-dimensional matrix data was imported to SIMCA-P 11.0 software (Umetrics, Umea, Sweden). Principle component analysis (PCA) was performed on the mean-centered and UV-scaled data to visualize general clustering, trends, and outliers among all samples on the scores plot. Partial least squares-discriminant analysis (PLS-DA) was used to maximize the variation. These differential metabolites selected from the PLS-DA model with VIP value (VIP > 1) are validated at a univariate level with Wilcoxon-Mann-Whitney test with a critical
For UPLC-QTOFMS, the resulting three-dimensional matrix, including assigned peak index (retention time-
GraphPad Prism software (version 6.0) was used for data entry and management. All reported
The clinical characteristics of cirrhosis patients including TCM YX group and SR group were summarized in Table
A total of 165 ion features were obtained and 11 identified urine metabolites were differentially expressed in cirrhotic patients compared to those in healthy controls. Peak intensity comparison of the differentially expressed metabolite levels in liver cirrhosis patients compared to those in healthy controls was summarized in Table
List of urinary differential metabolites in cirrhosis patients and among TCM YX, SR Zheng subgroup relative to controls.
Compounds | Liver cirrhosis versus control | YX versus control | SR versus control | SR versus YX | |||||
---|---|---|---|---|---|---|---|---|---|
VIPa | FCb |
|
FCd |
|
FCe |
|
FCf |
| |
GC-MS | |||||||||
4-Pyridinecarboxylate | 1.855 | 0.46 |
|
0.47 |
|
0.38 |
|
0.81 |
|
Threonine* | 1.498 | 0.64 |
|
0.63 |
|
0.60 |
|
0.96 |
|
Proline* | 1.474 | 1.30 |
|
1.57 |
|
1.79 |
|
1.14 |
|
Citrate* | 1.33 | 1.50 |
|
1.60 |
|
1.37 |
|
0.85 |
|
Aconitate* | 1.393 | 1.36 |
|
1.46 |
|
1.27 |
|
0.87 |
|
2-Pentendioate | 1.734 | 1.58 |
|
2.14 |
|
1.52 |
|
0.71 |
|
Hippurate* | 1.905 | 0.47 |
|
0.57 |
|
0.36 |
|
0.64 |
|
2-Aminobutyrate* | 1.954 | 0.38 |
|
0.34 |
|
0.30 |
|
0.88 |
|
Acetyl citrate | 1.517 | 2.75 |
|
3.26 |
|
3.24 |
|
1.00 |
|
3,4-Dihydroxyphenylacetate* | 2.121 | 1.88 |
|
2.19 | <0.0001 | 1.74 |
|
0.80 |
|
4-Hydroxy-benzenepropanedioate | 1.723 | 4.29 |
|
4.41 |
|
5.51 |
|
1.25 |
|
|
|||||||||
UPLC-QTOF-MS | |||||||||
cis-Aconitate* | 2.0 | 0.75 |
|
0.74 | <0.0001 | 0.76 | <0.0001 | 1.03 |
|
Pyroglutamate* | 2.1 | 0.69 |
|
0.65 | <0.0001 | 0.75 | <0.0001 | 1.15 |
|
O-Phosphotyrosine | 2.0 | 0.70 |
|
0.72 | <0.0001 | 0.70 | <0.0001 | 0.97 |
|
3-Methoxy-4-hydroxyphenylglycol sulfate | 2.1 | 1.70 |
|
1.73 | <0.0001 | 1.72 | <0.0001 | 0.99 |
|
Alpha-hydroxyisobutyrate* | 2.4 | 0.42 |
|
0.46 | <0.0001 | 0.35 | <0.0001 | 0.75 |
|
3-Hydroxyisovalerate* | 2.4 | 0.55 |
|
0.55 | <0.0001 | 0.54 | <0.0001 | 0.99 |
|
Dopaxanthin | 1.8 | 0.23 |
|
0.30 |
|
0.14 | <0.0001 | 0.46 |
|
Alpha-hydroxyhippurate* | 2.1 | 0.35 |
|
0.42 | <0.0001 | 0.24 | <0.0001 | 0.57 |
|
Canavaninosuccinate | 3.1 | 25.57 |
|
25.94 | <0.0001 | 25.79 | <0.0001 | 0.99 |
|
L-Aspartyl-4-phosphate | 1.6 | 0.71 |
|
0.74 |
|
0.60 | <0.0001 | 0.82 |
|
Isoxanthopterin | 1.6 | 0.74 |
|
0.77 |
|
0.71 |
|
0.92 |
|
Tyrosine-betaxanthin | 2.6 | 0.38 |
|
0.41 | <0.0001 | 0.34 | <0.0001 | 0.83 |
|
Estrone* | 1.4 | 0.78 |
|
0.79 |
|
0.78 |
|
0.98 |
|
Glycocholic acid 3-glucuronide | 1.7 | 5.18 |
|
5.37 |
|
5.30 |
|
0.99 |
|
Taurohyocholate* | 1.5 | 119.52 |
|
119.94 |
|
150.74 |
|
1.26 |
|
Cortolone-3-glucuronide | 2.5 | 0.36 |
|
0.44 | <0.0001 | 0.24 | <0.0001 | 0.54 |
|
Tetrahydroaldosterone-3-glucuronide | 2.6 | 0.31 |
|
0.37 | <0.0001 | 0.22 | <0.0001 | 0.58 |
|
11-Beta-hydroxyandrosterone-3-glucuronide | 2.4 | 0.38 |
|
0.44 | <0.0001 | 0.31 | <0.0001 | 0.70 |
|
N-Acetyl-leukotriene E4 | 2.6 | 0.12 |
|
0.13 | <0.0001 | 0.08 | <0.0001 | 0.60 |
|
11-Oxo-androsterone glucuronide | 2.3 | 0.25 |
|
0.29 | <0.0001 | 0.19 | <0.0001 | 0.67 |
|
Glycocholate* | 1.9 | 12.72 |
|
12.57 |
|
16.45 | <0.0001 | 1.31 |
|
Dehydroepiandrosterone 3-glucuronide | 2.4 | 0.29 |
|
0.33 | <0.0001 | 0.20 | <0.0001 | 0.60 |
|
Androsterone sulfate | 2.5 | 0.01 |
|
0.00 | <0.0001 | 0.02 | <0.0001 | 234.20 |
|
Testosterone sulfate | 2.3 | 0.21 |
|
0.21 | <0.0001 | 0.19 | <0.0001 | 0.90 |
|
Glycoursodeoxycholate* | 1.3 | 16.41 |
|
9.19 |
|
23.03 |
|
2.51 |
|
Androsterone glucuronide | 3.1 | 0.27 |
|
0.31 | <0.0001 | 0.19 | <0.0001 | 0.62 |
|
17-hydroxyandrostane-3-glucuronide | 2.9 | 0.27 |
|
0.31 | <0.0001 | 0.20 | <0.0001 | 0.67 |
|
Glycolithocholate 3-sulfate | 2.9 | 0.04 |
|
0.06 | <0.0001 | 0.01 | <0.0001 | 0.12 |
|
Note: *Metabolites were verified by reference standards; avariable importance in the projection (VIP) was obtained from PLS-DA model with a threshold of 1.0; bfold change (FC) was obtained by comparing those metabolites in liver cirrhosis patients to controls; c
PCA and PLS-DA analysis were performed to distinguish healthy subjects from cirrhosis patients in TCM YX subgroup and SR subgroup. With the 165 features determined by GC-MS, a PCA scores plot (figure not shown) using 4 components (R2X = 0.486) and a cross-validated PLS-DA model using 1 predictive component and 2 orthogonal components (R2Xcum = 0.502, R2Ycum = 0.77, and Q2Ycum = 0.476) were constructed (Figure
PLS-DA scores plot of urinary metabolites from healthy controls and TCM YX and SR Zheng patients with posthepatitis B cirrhosis using GC-MS spectral data (a) and UPLC-QTOFMS spectral data (b).
A PCA scores plot using 5 components (R2Xcum = 0.472, Q2cum = 0.058) and a cross-validated PLS-DA model using one predictive component and three orthogonal components (R2Xcum = 0.121, R2Ycum = 0.742, and Q2Ycum = 0.237) were constructed with 8,163 ion features detected on the UPLC-QTOFMS spectra. Clear separation among healthy controls, TCM YX subgroup, and SR subgroup of cirrhosis patients (Figure
A total of 28 characteristic urinary metabolites were identified from UPLC-QTOFMS negative ion mode for liver cirrhosis, as summarized in Table
Methods identifying the liver condition have been established in cirrhosis study, including the histological observations characterized by scar tissue, fibrous septa and regenerative nodules [
The liver is the most important metabolic organ in the human body, responsible for metabolism of a large array of substrates, such as sugar, protein, fat, and phytochemical compound [
Two panels of markers, 11 and 28 urinary metabolites identified by GC-MS and UPLC-QTOFMS, respectively, were significantly altered in cirrhosis participants (Table
On the contrary, the abnormal alteration of hippurate and 4-pyridinecarboxylate was only found in TCM SR subgroup patients, but not in TCM YX subgroup. Decreased hippurate in TCM SR subgroup patients and
Bar charts of eight representative metabolite markers (mean ± SEM.) that are differentially expressed in healthy controls and TCM YX and SR Zheng patients with posthepatitis B cirrhosis.
The liver’s blood supply mainly comes from the intestine through the portal vein. The liver is vulnerable to exposure of bacterial products translocated from the gut lumen via the portal vein. Disruption of the intestinal barrier in developing liver cirrhosis results in the leaky gut, which contributes to bacterial translocation [
There are some limitations in our study. First, the included participants were restricted to men only because women are easier to suffer from hormone interference, such as menses and early pregnancy, and difficult to judge the pathological or physiological alteration of hormone associated with liver cirrhosis. Second, metabolites of great statistical significance warrant further validation and screening as potential biomarkers for posthepatitis B cirrhosis diagnosis and TCM Zheng stratification in larger group of participants with both genders and other TCM Zhengs.
In conclusion, our results suggest that a panel of unique urinary metabolite markers is of clinical potential for the disease diagnosis and patient stratification for liver cirrhosis. These metabolite markers reflect the essence of the patients with posthepatitis B cirrhosis characterized by the disorders of the TCA cycle, amino acid, bile acids, hormones, dopamine, intestinal microbial metabolism, and oxidative stress. Moreover, the energy metabolism disorder is special prominent in cirrhotic patients with TCM YX Zheng, while the abnormality of the dopamine, intestinal microbiota metabolism, and oxidative stress is more serious in those with TCM SR Zheng. Specific urinary metabolites may be used as the biomarkers for TCM Zheng stratification. One of the most remarkable things about this study is that metabonomic profiling, as a powerful approach, partly interprets the biological reasons inducing TCM Zheng manifestation. Cirrhosis with TCM SR Zheng manifests more serious changes in the physiopathology of disease. Enclosing the same cohort of participants, we have obtained more information by integrating two medicine systems, indicating that urinary metabolite variation not only is associated with the pathological progression of cirrhosis but also acted as evidence of TCM Zheng stratification, contributing to the personalized diagnosis or treatment.
Xiaoning Wang and Guoxiang Xie are the co-first authors.
The authors declare that there is no conflict of interests.
This study was financially supported in part by Program for Outstanding Medical Academic Leader of Shanghai Municipal Health Bureau, China (LJ06023), Budgeted Scientific Research Project of Shanghai Municipal Education Commission, China (I3YZ044, 2010JW027), Natural Science Foundation of Shanghai, China (14ZR1441400), and E-institutes of Shanghai Municipal Education Commission, China (E03008, 085ZY1205).