Metabonomics-Based Study of Clinical Urine Samples in Suboptimal Health with Different Syndromes

Objective. To explore the urinary biochemistry features of syndromes of traditional Chinese medicine (TCM) such as syndrome of stagnation of liver Qi, spleen deficiency, liver Qi stagnation, and spleen deficiency (LSSDS) in sub-optimal health status (SHS). Methods. 12 cases for each syndrome group in SHS were selected, 12 subjects were used as a normal control group, and 1H NMR detection was, respectively, carried out, and the data was corrected by the orthogonal signal correction (OSC) and then adopted a partial least squares (PLS) method for discriminate analysis. Results. The OSC-PLS (ctr) analysis results of the nuclear overhauser enhancement spectroscopy (NOESY) detection indicated that the syndromes in SHS could be differentiated, and there were significant differences in the levels of metabolites of the urine samples of the four groups; the biomarkers of LSSDS in SHS were found out. The contents of citric acid (2.54 and 2.66), trimethylamineoxide (3.26), and hippuric acid (3.98, 7.54, 7.58, 7.62, 7.66, 7.82, and 7.86) in the urine samples of LSSDS group were lower than that of the normal control group. Conclusion. There are differences in the 1H-NMR metabolic spectrum of the urine samples of the four groups, and the specific metabolic products of the LSSDS in SHS can be identified from metabonomics analysis.


Introduction
Since 1984, when the WHO gave the de�nition to health as paying attention to "the state of complete physical, mental and social well-being" in its "Charter, " the research on the "third state" has been carried out widely at home and abroad. In 1996, scholar Wang from China named this state as "SHS" [1]; thereaer, the study of Suboptimal health domestic started to be involved in many aspects. "Preventive treatment of disease, " namely, the prevention of diseases and maintaining health is the original characteristics and advantages of traditional Chinese medicine. e state of having no "disease" but having the "syndromes" of Suboptimal health has made the current treatment mainly based on the approach of Traditional Chinese Medicine.
It was found by the clinical epidemiological survey that 60%-70% of people in China are in SHS; the incidences of syndrome of stagnation of liver �i and spleen de�ciency among the population of SHS are high and can be up to 17.25 [2], and the incidence of the syndrome in those surveyed on the population of SHS in General Hospitals can be even up to 45.85% [3]. In recent years, metabonomics technology has been widely used in the diagnosis and research of hypertension [4] and other cardiovascular diseases, neurological diseases [5], urinary system diseases [6], diabetes [7], congenital metabolism disorders [8], and cancer [9], and so forth, and it has also been widely used in Chinese herbal compound research [10], safety evaluation of traditional Chinese medicine [11], screening of toxic markers [12], and so on.
Currently combination of syndromes classi�cation and biomedical diagnosis becomes a common model in diagnostics in TCM clinical practice. Clinical treatments of a patient rely on the successful differentiation of a speci�c TCM syndrome. Our previous studies have shown that [13], through the analysis of the 1 H NMR spectrum of the plasma of model rats with chronic restraint stress-induced syndrome of stagnation of liver �i and spleen de�ciency, the changes in the spectral peak shape of acetic acid, lactic acid, tyrosine acid, low-density lipoprotein, and the unknown compound at 3.44 ppm were very signi�cant. As a result, the metabonomics method was applied in this study in discussing the syndrome of stagnation of liver Qi and spleen de�ciency and assisted in revealing its biological changes. e objective of this study is using metabonomics method to differentiate syndromes with clinical urine samples in Suboptimal health status.

Materials and Methods
2.1. Source and Grouping of the Subjects. Among the personnel from the Chinese Academy of Sciences who had physical examinations in the Medical Center of Beijing Guanghua Hospital from December 2009 to April 2010, 12 cases of syndrome of stagnation of liver Qi and spleen de�ciency, syndrome of stagnation of liver Qi, and syndrome of spleen de�ciency in SHS who met the inclusion criteria were, respectively, selected, and 12 cases for the healthy group of normal control were selected. 48 subjects were typical permanent residents in Beijing, and it was shown that the four groups were well balanced on age and sex. Both the research protocol and written informed consent were reviewed and approved by an ethics committee of Beijing University of Chinese Medicine prior to the study initiation. In addition, a TCM syndrome questionnaire for measuring SHS was also used in the study [14], and the SHS was also evaluated as our previously reported [15].

Diagnostic Criteria for Suboptimal Health Population.
(1) Recurrent physical discomfort and decreased efficiency continued for more than three months due to persistent or excessive fatigue. (2) Have no signi�cant organic diseases or psychological disorders. Only those who meet both the above two criteria could be diagnosed as Suboptimal health population. Reagents are deutoxide (D2O), produced by Cambridge Isotope Laboratories Inc., USA, 99.9%; 2,2,3,3,-d(4)-3-Trimethylsilyl) propionic acid sodium (TSP), purchased from Merck, Germany. [16,17]. About 5 mL of middle-segment fasting morning urine sample of the subjects were collected. en the samples were centrifuged at 3500 rpm for 10 min aer collection. en 1 mL of the supernatant was taken and placed in the refrigerator with temperature at −20 ∘ C for cryopreservation. 30 L of TSP/D 2 O (1 mg/mL) was added into the 5 mm NMR tube. Aer the urine samples were thawed at room temperature, 350 L of phosphate buffer solution and 350 L of the urine were added into the EP tube, aer being well mixed by shaking, centrifuged at 13,000 rpm for 10 min, and 600 L of the supernatant was added into the NMR tube mentioned above and mixed and was ready for use. Under the conditions of 27 ∘ C, Carr-Purcell-Meiboom-Gill (CPMG) and longitudinal eddy delay (LED) were, respectively, adopted for the experiment on the VARIAN UNITYI-NOVA 600 MHz Superconducting Fourier Transform NMR spectrometer, the presaturated method was used to inhibit the water peak, the saturation time was 2 s, spectral width was 8000 Hz, the sampling point was 32 k, and it accumulated 64 times. e presaturation frequency and central frequency were both at the position of water peak. e free induction decay (FID) signal was converted to NMR spectrum through the Fourier transform at the 128 k point. e peak of the chemical shi of TSP was used as a reference position and set to 0 ppm. e obtained NMR data was converted to a spectrum by the Fourier transform, aer phase and baseline correction, the spectrum within a certain range was piece-wisely integrated at a width of 0.04 ppm for each segment, and the segments with solvent peaks and urea peaks (4.6 to 6.2) were excluded at the same time. e integral range of the NOESY data was 0.4 ppm ∼ 9.4 ppm. e integral was normalized by the total integral intensity of each spectrum. e obtained data was output and converted to an Excel �le and saved.

NMR Data Analysis.
All the integral data was output to Excel �les and saved and then entered into the SIMCA-P � soware (V10.04, Umetrics, Umeå, Sweden) for analysis. Orthogonal signal corrections (OSC) were performed �rst, and PLS analysis was then conducted. e analysis results were expressed as scores plot and loadings plot. e degree of separation for each group of data was determined from the scores plot, and the metabolites that contributed to the composition were obtained from the factor loadings plot. Figure 1).

Results of the Overall Comparative Analysis of the Four
Groups. e OSC-PLS analysis results of 1 H-NMR spectrums of the four groups-syndrome of stagnation of liver Qi and spleen de�ciency (group A), syndrome of stagnation of liver Qi (group B), syndrome of spleen de�ciency (group C), and the normal control (group D)-are shown in Figure 2. Figure 2(a) is a one-dimensional scores plot obtained by PLS analysis aer OSC denoising processing. We can see that the four groups of samples are distinguishable from each other, and there are no overlap and duplication among the four groups. e results indicate that there are differences in the metabolic phenotypes of the four groups, and the intergroup differences are signi�cant. e group of syndrome of stagnation of liver Qi and spleen de�ciency (Group A) shows obvious separation trends from the group of normal control (group D) and group of syndrome of spleen de�ciency (Group C) along the 1 axis; it can be said that the group of syndrome of stagnation of liver Qi (Group B) and the group of syndrome of spleen de�ciency (Group C) show rough separation trends along the 1 axis. e abscissa of Figure 2(b) corresponds to the chemical shi, wherein the positive values on the vertical axis represent that, within the chemical shi range, the corresponding metabolite concentrations in most samples of the syndrome of stagnation of liver Qi and spleen de�ciency are larger; the negative values represent that, within the chemical shi interval, as compared with the group of normal control and the group of the syndrome of spleen de�ciency, the metabolite concentrations in the group of syndrome of stagnation of liver Qi and spleen de�ciency are lower. e speci�c metabolites should be analy�ed in combination with the intergroup pairwise comparison results.
As the diet and lifestyle of the subjects in this study were not strictly restricted, in order to improve the classi�cation results, intergroup pairwise comparisons were using the PLS-DA analysis which was carried out aer OSC denoising processing.  Figure 3. It can be seen from the scores plot (Figure 3(a)) that these two groups of samples can be well separated along the 1 axis, with no overlap and duplication. is indicates that the two groups have differences in the metabolic phenotypes, and the intergroup difference is signi�cant. We can see from the loadings plot ( Figure 3(b)) that most points are concentrated near the origin, and only a few points are away from the origin, which represent the compounds that resulted in the differences between the two groups. We can �nd out the chemical shis of the metabolites NMR peak values with their differences by combining the loadings plot with the primary spectrum, and the metabolites with differences of these two groups of urine samples are shown in Table 1.  Figure 4. It can be seen from the scores plot (Figure 4(a)) that these two groups can be roughly separated by the green line as indicated, the vertical to the green line representing the intergroup differences, and the parallel to the green line representing the intra-group differences. We can see from the loadings plot (Figure 4(b)) that most points are concentrated near the origin, and only a few points are away from the origin, which represent the compounds that resulted in the differences between the two groups. We can �nd out the chemical shis of the metabolites NMR peak values with their differences by combining the loadings plot with the primary spectrum, and the metabolites with differences of these two groups of urine samples are shown in Table 1.      of spleen de�ciency (�roup C) are shown in Figure 5. It can be seen from the scores plot ( Figure 5(a)) that these two groups of samples can be well separated along the 1 axis, with no overlap and duplication, indicating that the two groups have differences in the metabolic phenotypes, and the intergroup difference is signi�cant. We can see from the loadings plot ( Figure 5(b)) that most points are concentrated near the origin, and only a few points are located away from the origin, which represent the compounds that resulted in the differences between the two groups. We can �nd out the chemical shis of the NMR peak values of the metabolites with its differences by combining the loadings plot with the primary spectrum, and the metabolites with its differences of these two groups of urine samples are shown in Table 1.

Summary of the Analysis Results of the Urine Samples from the Four Groups.
Based on the results above, there are certain differences among the samples from the four groups. Further analysis of the loadings plots of the principal component by intergroup comparison of the groups showed that the following substances (Table 1) contributed greatly to the classi�cation, namely, the potential markers.

Discussion
In this study, the PCA analysis and the PLS analysis which were carried out aer OSC denoising processing were comprehensively used, and the results were expressed by scores plots, and loadings plots. Wherein, the scores plots were plotted using the principal component as the axis, re�ecting the differences between the classes, to access classi�cation information of the samples. e loadings plots re�ected the variables and their contributions to classi�cation. By analy�ing the scores plots, loadings plots and the primary spectrums of the intergroup pairwise comparison, some biomarkers were obtained. e potential biomarkers of syndrome of  Syndrome of stagnation of liver Qi and spleen de�ciency is one of the common syndromes of SHS and mental disorders in TCM clinic [18], our recently studies shown: the Xiaoyaosan (XYS) decoction reversed chronic stress induced decreases in brain-derived neurotrophic factor (BDNF) and increases in tyroxine hydroxylase (TrkB), and neurotrophin 3 (NT-3) in the frontal cortex and the hippocampal CA1 subregion [19]. e xiaoyaosan decoction can signi�cantly downregulate the contents of leptin and leptin receptor (ob-R) at the arcuate nucleus (ARC) in the basal of hypothalamus in the hypothalamus of chronic stressed rats [20]. Meanwhile, the XYS-decoction-containing serum is signi�cant in improving mitochondrial membrane potential and apoptotic rate of hippocampus neuron induced by oxidative-stress [21]. Excessive pressure from work or a poor emotion state will lead to the dysfunction of "liver controlling conveyance and dispersion, " and the stagnation of liver Qi will result in spleen and stomach disorders. As a result, fatigue and other symptoms will occur.
Creatinine is a metabolite produced during metabolism of muscle tissue and is also a compound related to the renal function. erefore, the relative increase in creatinine levels suggests the decrease in the metabolic function of syndrome of stagnation of liver Qi and spleen de�ciency in SHS. Trimethylamine oxide (TMAO) has a role in promoting the growth of muscle tissue, and the decrease in the content of trimethylamine oxide suggests the decline of the "spleen governing muscles" function of the syndrome of stagnation of liver Qi and spleen de�ciency in SHS. Hippuric acid is a small molecule urine toxin, and there is a study indicating that hippuric acid can accurately re�ect the energy metabolism of the liver and any reason affecting the oxidative and phosphorylation ability of mitochondria of liver cells; thus, resulting in hepatic energy metabolism disorders can affect the synthesis ability of hippuric acid [22]. erefore, the changes in hippuric acid also demonstrate that the energy metabolism in the liver of syndrome of stagnation of liver Qi and spleen de�ciency in SHS has been damaged to a certain extent.  In this study, we analyzed the metabolic phenotype spectrums of the urine samples of the syndrome of stagnation of liver Qi and spleen de�ciency, syndrome of stagnation of liver Qi, syndrome of spleen de�ciency in SHS, and the group of normal control from the perspective of metabonomics, and there were signi�cant differences in the metabolites for the syndromes in SHS. ese differences are based on the changes in the metabolism of different substances or the metabolic network. e metabonomics technology may be used to identify the speci�c metabolites for the biological basis of the syndromes of TCM, and the methods of bioinformatics can also be used to analyze the functions of the biomarkers, and to determine the "Metabolic spectrum related to syndrome of TCM" [13].
We also describe in detail the metabolic consequences of syndromes in SHS. In particular, we have shown that the speci�c metabolic products of the syndrome of stagnation of liver Qi and spleen de�ciency in SHS can be identi�ed from metabonomics analysis: creatinine (3.06 and 4.06), trimethylamine oxide (3.26), and hippuric acid (3.98, 7.54, 7.58, 7.62, 7.66, 7.82, and 7.86). We consider the results of this study to add considerably to our understanding of the biochemistry of TCM syndromes.