Journal of Ethnopharmacology 244 (2019) 112139

Contents lists available at ScienceDirect

Journal of Ethnopharmacology
journal homepage: www.elsevier.com/locate/jethpharm

Fecal metabonomics combined with 16S rRNA gene sequencing to analyze T

the changes of gut microbiota in rats with kidney-yang deficiency syndrome
and the intervention effect of You-gui pill
Ruiqun Chena,∗,1, Jia Wanga,1, Runhua Zhanc, Lei Zhangb, Xiufeng Wangb,∗∗
a
School of Life Sciences and Biopharmaceutics, Guangdong Pharmaceutical University, Guangzhou, 510006, PR China
b
College of Medical Information Engineering, Guangdong Pharmaceutical University, Guangzhou, 510006, PR China
c
Shool of Pharmacy, Guangdong Pharmaceutical University, Guangzhou, 510006, PR China

ARTICLE INFO ABSTRACT

Keywords: Ethnopharmacological relevance: A myriad of evidence have shown that kidney-yang deficiency syndrome (KYDS)
Fecal metabonomics is associated with metabolic disorders of the intestinal microbiota, while TCMs can treat KYDS by regulating gut
16S rRNA gene sequencing microbiota metabolism. However, the specific interplay between KYDS and intestinal microbiota, and the in-
You-gui pill trinsic regulation mechanism of You-gui pill (YGP) on KYDS’ gut microbiota remains largely unknown so far.
Kidney-yang deficiency syndrome
Materials and methods: In the present study, fecal metabonomics combined with 16S rRNA gene sequencing analysis
Gut microbiota
were used to explore the mutual effect between KYDS and intestinal flora, and the intrinsic regulation mechanism of
YGP on KYDS's gut microbiota. Rats' feces from control (CON) group, KYDS group and YGP group were collected, and
metabolomic analysis was performed using 1H NMR technique combined with multivariate statistical analysis to
obtain differential metabolites. Simultaneously, 16S rRNA gene sequencing analysis based on the Illumina HiSeq
sequencing platform and ANOVA analysis were used to analyze the composition of the intestinal microbiota in the
stool samples and to screen for the significant altered microbiota at the genus level. After that, MetaboAnalyst
database and PICRUSt software were apply to conduct metabolic pathway analysis and functional prediction analysis
of the screened differential metabolites and intestinal microbiota, respectively. What's more, Pearson correlation
analysis was performed on these differential metabolites and gut microbiota.
Results: Using fecal metabonomics, KYDS was found to be associated with 21 differential metabolites and seven
potential metabolic pathways. These metabolites and metabolic pathways were mainly involved in amino acid
metabolism, energy metabolism, methylamine metabolism, bile acid metabolism and urea cycle, and short-chain
fatty acid metabolism. Through 16S rRNA gene sequencing analysis, we found that KYDS was related to eleven
different intestinal microbiotas. These gut microbiota were mostly involved in amino acid metabolism, energy me-
tabolism, nervous, endocrine, immune and digestive system, lipid metabolism, and carbohydrate metabolism.
Combined fecal metabonomics and 16S rRNA gene sequencing analysis, we further discovered that KYDS was pri-
marily linked to three gut microbiotas (i.e. Bacteroides, Desulfovibrio and [Eubacterium]_coprostanoligenes_group) and
eleven related metabolites (i.e. deoxycholate, n-butyrate, valine, isoleucine, acetate, taurine, glycine, α-gluconse, β-
glucose, glycerol and tryptophan) mediated various metabolic disorders (amino acid metabolism, energy metabolism,
especially methylamine metabolism, bile acid metabolism and urea cycle, short-chain fatty acid metabolism. nervous,
endocrine, immune and digestive system, lipid metabolism, and carbohydrate metabolism). YGP, however, had the
ability to mediate four kinds of microbes (i.e. Ruminiclostridium_9, Ruminococcaceae_UCG-007, Ruminococcaceae_UCG-

Corresponding author.
∗

Corresponding author.
∗∗

E-mail addresses: 975201721@qq.com (R. Chen), 849525858@qq.com (J. Wang), 853870052@qq.com (R. Zhan), zhanglei7965@126.com (L. Zhang),
wangxiufeng@gdpu.edu.cn (X. Wang).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.jep.2019.112139
Received 11 April 2019; Received in revised form 20 July 2019; Accepted 4 August 2019
Available online 08 August 2019
0378-8741/ © 2019 Elsevier B.V. All rights reserved.
R. Chen, et al. Journal of Ethnopharmacology 244 (2019) 112139

010, and uncultured_bacterium_f_Bacteroidales_S24-7_group) and ten related metabolites (i.e. deoxycholate, valine,
isoleucine, alanine, citrulline, acetate, DMA, TMA, phenylalanine and tryptophan) mediated amino acid metabolism,
especially methylamine metabolism, bile acid metabolism and urea cycle, short-chain fatty acid metabolism, en-
docrine, immune and digestive system, and lipid metabolism, thereby exerting a therapeutic effect on KYDS rats.
Conclusion: Overall, our findings have preliminary confirmed that KYDS is closely related to metabolic and
microbial dysbiosis, whereas YGP can improve the metabolic disorder of KYDS by acting on intestinal micro-
biota. Meanwhile, this will lay the foundation for the further KYDS's metagenomic research and the use of
intestinal microbiotas as drug targets to treat KYDS.

Abbreviations ACTH adrenocorticotropic hormone

CORT cortisol
KYDS kidney-yang deficiency syndrome 17-OHCS 17-hydorxycorticosteroids
YGP You-gui pill HPA hypothalamic-pituitary-adrenal axis
TCM Traditional Chinese Medicine; TSP 3-(trimethylsilyl)propionic 2,2,3,3-d4 acid
SDH Radix Rehmanniae Praeparata NaN3 sodium azide;
FZ Radix Aconiti Lateralis Prreparata PBS phosphate buffer solution
RG Cinnamomi cortex PCA principle component analysis
SY Rhizoma Dioscoreae PCoA principal coordinate analysis
SZY Fructus Corni OPLS-DA orthogonal partial least squares discriminant analysis
TSZ Semen Cuscutae PLS-DA partial least square discriminant analysis
GQ Fructus Lycii VIP variable importance in the projection
LJJ Cervi cornus colla CV-ANOVA cross-validated residuals
DZ Eucommiae cortex UPGMA unweighted pair group method with arithmetic mean
DG Radix Angelicae Sinensis DMA dimethylamine;
TSH thyroid stimulating hormone TMA trimethylamine;
T3 triiodothyronine; HMDB The Human Metabolome Database
T4 tetraiodothyronine; BCAAs branched-chain amino acids
LH luteinizing hormone AAAs aromatic amino acids
FSH follicle stimulating hormone SCFAs short-chain fatty acids
T testosterone CNS central nervous system

1. Introduction
The human body is a superorganism composed of its own genome
and metagenome of its symbiotic microorganism (Nicholson et al.,
2005; Nicholson, 2006; Lederberg, 2000). Gut microbiota is recognized
as an indispensable “metabolic organ” that plays crucial roles in
maintaining human health or initiating diseases, including digestion,
nutrient absorption, energy supply, immune regulation, disease re-
sistance, etc (Lin et al., 2018a; Kelly et al., 2005; Xu et al., 2003). Ac-
cumulating evidence indicates that perturbations of the gut microbiome
and its influence on metabolic and physiological functions may exert an
important role in the development of human diseases, such as: gastro-
intestinal diseases (i.e. colon cancer, non-inflammatory bowel disease)
and metabolic diseases (i.e. obesity, diabetes), etc (Lu et al., 2014;
Seksik et al., 2003; Ley et al., 2006; He et al., 2015). In contrast, an
increasing number of evidence suggests that traditional Chinese medi-
cines (TCMs)/TCM formulas may be used as prebiotics to modulate gut
microbiota structure and metabolic phenotype of the host, and further
as a new source for drug leads in gut microbiota-targeted disease
management (Yu et al., 2017; Chang et al., 2017; Xu et al., 2015). You-
gui pill (YGP), a classic Chinese medicine formula consisted of ten
TCMs, including Radix Rehmanniae Praeparata (SDH), Radix Aconiti
Lateralis Prreparata (FZ), Cinnamomi cortex (RG), Rhizoma Dioscoreae
(SY), Fructus Corni (SZY), Semen Cuscutae (TSZ), Fructus Lycii (GQ), Cervi
cornus colla (LJJ), Eucommiae cortex (DZ), and Radix Angelicae Sinensis
(DG), has been widely used to recuperate kidney-yang deficiency syn-
drome (KYDS) clinically for 400 years in China (Wang et al., 2015; Liu
et al., 2017). The composition and full details of YGP were shown in
Table S1 (Wang et al., 2015; Zhang et al., 2017).
Modern studies have indicated that the pathogenesis mechanism of
KYDS is mainly in the functional disorder with different degree of hy-
pothalamic-pituitary-target gland axis (adrenal, thyroid and gonad),
and it is characterized by diarrhea, cold limbs, decreased mobility, slow
response, decreased appetite, cowered and polyuria, etc (Lu et al., 2011;
Gao et al., 2009; Zou et al., 2012). Meanwhile, the early colonization
period of microbiota coincides with activation of the hypothalamic-
pituitary-adrenal (HPA) axis, which has an impact on the enteric ner-
vous system that innervates the gastrointestinal (GI) tract (Nicholson
et al., 2012). Notably, a myriad of evidences shows that KYDS is as-
sociated with metabolic disorders of the intestinal microbiota, while
TCMs can treat KYDS by regulating gut microbiota metabolism (Lu
et al., 2011; Zou et al., 2012, 2013; Gong et al., 2012; Zhao et al.,
2013a; Huang et al., 2013). Clinical studies have also found that the
intestinal microbiota of patients with KYDS has been dysregulated
(Wang et al., 2017; Li et al., 2011; Huang et al., 2017). Likewise, our
previous research shows that KYDS is related to gut metabolism, and
YGP can regulate the intestinal microbiota disorder of KYDS (Chen
et al., 2017, 2018a). However, the specific interplay between KYDS and
intestinal microbiota, and the intrinsic regulation mechanism of YGP on
KYDS′ gut microbiota remains largely unknown so far. Therefore, in the
present study, we further comprehensively explore intestinal microbial
changes of KYDS rats and specific intervention of YGP on KYDS's in-
testinal microbiota via combining fecal metabonomics and 16S rRNA
gene sequencing analysis from the perspective of intestinal microbiota.
2. Materials and methods
2.1. Chemicals and reagents
Deuterium oxide (D2O, containing 0.05% TSP: 3-(trimethylsilyl)
propionic 2,2,3,3-d4 acid) and 5 mm NMR tubes were purchased form
Qingdao Tenglong Weibo Technology Co., Ltd (Qingdao, China).
500 MHz Bruker AVANCE III NMR instrument (Bruker, Swiss) was
provided by Center Laboratory of Guangdong Pharmaceutical
University. All biochemical indicators in this study such as thyroid
stimulating hormone (TSH), triiodothyronine (T3), tetraiodothyronine

2
R. Chen, et al.
(T4), luteinizing hormone (LH), follicle stimulating hormone (FSH),
testosterone (T), adrenocorticotropic hormone (ACTH) and cortisol
(CORT) were obtained by entrusting the Beitu Dongya Biotechnology
Research Institute (Beijing, China), while 17-hydorxycorticosteroids
(17-OHCS) was detected by entrusting the Nanjing Jiancheng
Bioengineering Institute (Nanjing, China). 96-well PCR instrument was
bought from AB SCIEX (AB SCIEX, USA); Micro centrifuge and deio-
nized water were got from TIANGEN Biotech CO., LTD (Beijing, China);
MinElute® PCR Purification Kit was supplied by QIAGEN Co., Ltd
(Shanghai, China); 100 bp DNA Ladder Marker was provided by Takara
Biotechnology Co., Ltd (Tokyo, Japan); Anhydrous alcohol was ob-
tained from Beijing Chemical Works (Beijing, China); Phosphate buffer
solution (1X, PBS) was bought from Dailian Meilun Biotechnology Co.,
Ltd (Dailian, China). All other used chemicals were of analytical grade.
2.2. Extraction of YGP solution
The YGP solution was extracted according to the “Jing-Yue
Complete Works” of Zhongjing Zhang at Ming dynasty and some pub-
lished literatures (Wang et al., 2015; Yang et al., 2009; Ji et al., 2016a;
Chen et al., 2018a). SDH (24 g), FZ (6 g), RG (6 g), SY (12 g), SZY (9 g),
TSZ (12 g), GQ (12 g), DZ (12 g), and DG (9 g) were weighed according
to the original proportions recorded in ancient books. In addition to
LJJ, others nine traditional Chinese medicines were first soaked in
distilled water for 0.5 h, then decocted for 20 min and further filtered.
The filtrate was extracted, and filter residue was continued for the same
decoction and filtration procedure. The twice obtained filtrates were
blended, LJJ (12 g) was then added to it in the hot state and mixed with
a glass rod. The mixture was further concentrated to 111 mL with a
rotary evaporator, and a concentration (crude herb) of 1.0 g mL-1 YGP
solution was obtained eventually.
2.3. Rats and treatments
A total of 30 male Sprague-Dawley (SD) rats (180 ± 20 g, animal
licence No. SCXK-(Yue) 2013-0034, SPF grade) were commercially
purchased from Experimental Animal Center of Guangzhou University
of Chinese Medicine. All rats were housed under control conditions
(temperature of 24 ± 2 °C, relative humidity of 55 ± 5%, and a 12 h
light/dark cycle) with regular chow and water freely available. After
one week of adaptation, 30 rats were randomly divided into three
groups (n = 10): Control group (CON), KYDS group (KYDS) and YGP
group (YGP). Besides, all rats were evenly mixed and co-housing after
grouping and marking. The methods of modeling and administration
were conducted according to references described previously (Chen
et al., 2005, 2018a; Gao et al., 2009; Zou et al., 2012). In brief, rats in
CON group were given an intramuscular injection of 0.3 mL of 0.9%
saline into a hind limb for 22 days in succession. 0.5 mL/100 g hy-
drocortisone was injected intramuscularly into a hind limb of KYDS
group rats for 22 days in succession. Correspondingly, YGP group was
consistent with the modeling method of KYDS group, but YGP solution
(1.0 mL/100 g, 15 days in succession) was simultaneously administered
from the 8th day of modeling. Rats in YGP group were intragastric
administration with a dosage of 1.0 mL/100 g one time every day. The
dosage for rats was calculated according to the dosage for adults, and
the specific formula was: dosage for rats/d = dosage for
adults × 0.018 × 5/(kg⋅d) (Wang et al., 2013a, 2013b). However, rats
in CON group and KYDS group were treated with the same solvent (i.e.
distilled water), and the administration method and dosage as YGP
group. Our study was approved by the Animal Ethics Committee of
Guangdong Pharmaceutical University. All experimental procedures
were strictly performed in accordance with the National Institutes of
Health Guide for the Care and Use of Laboratory Animals.
3
Journal of Ethnopharmacology 244 (2019) 112139
2.4. Sample collections and preparation
At least 2 pellets of feces from each rat were collected at the end of
the 23rd day, divided into two parts in sterile EP tube with 10 μL of 1%
sodium azide (NaN3) and stored at -80 °C refrigerator for subsequent
fecal metabonomics and 16S rRNA gene sequencing analysis, respec-
tively.
Three groups of rats were sacrificed in parallel. Blood samples were
collected and centrifuged (4500 rpm for 10 min, 4 °C) to obtain super-
natants for biochemical analysis (TSH, T3, T4, LH, FSH, T, ACTH and
CORT). Urine samples were collected from metabolic cages and cen-
trifuged at 4 °C for 10 min at 5000 rpm for 17-OHCS test. Among these,
TSH, T3 and T4 were indicators that reflected thyroid function; LH, FSH
and T were indicators that reflected testicular function; ACTH and
CORT were indicators of adrenal function; while 17-OHCS was an im-
portant indicator for clinical diagnosis of KYDS. The tissues of patho-
logical axis (hypothalamus, pituitary, adrenal gland and thyroid) and
testicular tissues were collected and fixed in 10% formalin and Bouin's
fluid overnight for histopathological analysis, respectively. Meanwhile,
all tissues were embedded in paraffin wax, cut into 3 mm tissue sections
with a microtome, stained with hematoxylin and eosin (HE), and
photographed with an optical microscope.
2.5. Fecal metabonomics
250 mg of each sample stool was weighed and thawed at room
temperature before NMR analysis. 1 mL PBS (0.2 mol/L, Na2HPO4/
NaH2PO4, pH = 7.4) was mixed with 250 mg feces and centrifuged at
14,000 rpm at 4 °C for 10 min. Thereafter, 400 μL of the supernatant
and 100 μL of D2O (containing 0.05% TSP) were mixed, transferred into
5 mm nuclear magnetic tube, and stored in a refrigerator at 4 °C (sto-
rage time < 10 h) for further analysis.
All fecal samples’ spectra were processed at 298 K using a Bruker
spectrometer (Avance III 500 MHz) with a one-dimensional water pre-
saturated standard NOESYPR1D pulse sequence (recycle delay-90-t1-
90-tm-90-acquisition). A total number of 128 transients were collected
into 32 k data points using a spectral width of 10 kHz with a relaxation
delay of 3 s and the total echo time (2nτ) of 100 m s. An exponential
function corresponding to a line broadening factor of 0.3 Hz was ap-
plied to all acquired free induction decays (FIDs) before Fourier
transformation (Chen et al., 2018b).
The 1H NMR spectra of fecal samples were manually phased and
baseline using the TopSpin software (Version 2.1, Bruker Biospin,
Germany), and automatically integrated by using AMIX software
(V3.7.3, Bruker Biospin, Germany). The integral values were nor-
malized by total area. Moreover, the integral interval was δ 0.5 ~ 9.0
with an integral spacing of 0.004 ppm, and the integral value of δ 4.7
~ 5.2 were removed in order to eliminate the influence caused by the
water peak. All normalized values were introduced as raw data into
SIMCA-P 13.0 software (Umetrics, Sweden) for multivariate statis-
tical analysis, such as the principle component analysis (PCA) and
orthogonal partial least squares discriminant analysis (OPLS-DA). A
permutation test (200 permutations) was performed to validate
partial least square discriminant analysis (PLS-DA) model, and OPLS-
DA was validated by both a 7-fold cross-validation and ANOVA of the
cross-validated residuals (CV-ANOVA) methods. What's more, S-plots
were used to screen fecal biomarkers of KYDS treated by YGP.
Endogenous metabolites were assigned according to the HMDB da-
tabase (http://www.hmdb.ca/) and some published literatures (Le
Gall et al., 2011; Chen et al., 2018b; Zhao et al., 2013b; Saric et al.,
2008; Zhang et al., 2011; Jacobs et al., 2008). Taken together,
variables with P < 0.05 and VIP > 1 obtained from OPLS-DA ana-
lysis were considered as fecal biomarkers associated with KYDS.
R. Chen, et al.
2.6. 16S rRNA gene sequencing analysis
Microbial genomic DNA was extracted from fecal pellets using the
PowerSoil DNA Isolation Kit (MOBIO Laboratories) according to the
manufacture's protocols, and was stored at -80 °C refrigerator for fur-
ther testing. DNA quality and quantity were assessed by the ratios of
260 nm/280 nm and 260 nm/230 nm, respectively. The V3 + V4 se-
quenced regions of 16S rRNA gene were amplified by PCR with the
universal primers (forward: 5′- ACTCCTACGGGAGGCAGCA-3′ and re-
verse: 5′-GGACTACHVGGGTWTCTAAT-3′). The PCR amplified product
was purified, quantified, and normalized to form a sequencing library.
The constructed library was first subjected to library quality inspection,
Fig. 1. Histopathological analysis of hypothalamus, pituitary, adrenal gland, thyroid and testis.
4
Journal of Ethnopharmacology 244 (2019) 112139
and the quality-qualified library was sequenced using Illumina HiSeq
2500 (Illumina, San Diego, USA) according to the standard protocols of
Biotree Bio-technology Co., Ltd (Shanghai, China). Afterwards, the raw
image data files obtained by high-throughput sequencing were con-
verted into the original sequenced reads by Base Calling Analysis, and
the results were stored in the FASTQ (abbreviated as fq) file format
which contained sequence information of reads and its corresponding
sequencing quality information. Data preprocessing: according to the
overlap relationship among paired-end (PE) reads, the paired-end se-
quence data obtained by HiSeq platform was merged into a sequence of
Tags. Meanwhile, the quality of Reads and the effect of merge were
quality-controlled filtered through the following three steps: (1) PE
R. Chen, et al. Journal of Ethnopharmacology 244 (2019) 112139

reads splicing: Splicing the reads of each sample through overlap using
FLASH v1.2.7 software, the obtained splicing sequence was the raw tags
data; (2) Tags filtering: the raw tags obtained by splicing were filtered
by Trimmomatic v0.33 software to get high quality tags data (i.e. clean
tags); (3) Removal of chimeras: The chimeric sequences were identified
and removed using UCHIME v4.2 software to obtain the effective tags.
Clustering of Tags at 97% sequence similarity level using UCLUST
(Edgar, 2010) in QIIME (Caporaso et al., 2010) (version 1.8.0) software
to obtain operational taxonomic unit (OUT), and the OTU was tax-
onomically annotated based on the Silva (bacteria, http://www.arb-
silva.de) and UNITE (fungi, http://unite.ut.ee/index.php) taxonomic
databases. 0.005% of all sequence numbers as a threshold was used to
filter OUT (Bokulich et al., 2013). The OTU representative sequence
was compared with the microbial reference database to obtain the
species classification information corresponding to each OTU, and then
the composition of each sample was counted at each level (i.e. phylum,
class, order, family, genus, species). On this basis, the QIIME software
was used to generate the species abundance table at different classifi-
cation levels, and R package was used to map the community structure
plot under the taxonomic level of the sample. Species annotation was
performed using the RDP Classifier (version 2.2, http://sourceforge.
net/projects/rdpclassifier/) with a confidence threshold of 0.8 (Wang
et al., 2007).
Alpha diversity of samples (Rarefaction curve, Shannon index, Rank
abundance curve, Chao1 richness estimator, Ace richness estimator,
Simpson diversity index, etc) was analyzed using Mothur (version
v.1.30) software. Beta diversity analysis (PCA, PCoA, UPGMA) was
performed using Bray-Curtis algorithm of QIIME and R software to
compare the similarity of species diversity in different samples. Finally,
ANOVA analysis was performed to screen species with significant dif-
ferences at genus level (P < 0.05).
2.7. Pathway analysis
For metabolites, differential metabolites screened from fecal meta-
bonomics were imported into the MetaboAnalyst 4.0 database (https://
www.metaboanalyst.ca/) for pathway analysis. MetaboAnalyst is an
online database with statistical, functional and integrative analysis of
metabolomics data. It performs in-house mapping of common com-
pound names to a wide variety of database identifiers including KEGG,
HMDB, ChEBI, METLIN and PubChem prior to performing any func-
tional analysis (Chong et al., 2018). Metabolic pathways with P < 0.05
were considered as potential pathways that YGP affect the occurrence
of KYDS (Lin et al., 2018a; Chen et al., 2018a). For microbes, the
functional composition of samples was extrapolated by comparing the
species composition information obtained from the 16S sequencing
data using the PICRUSt software, thereby analyzing the functional
differences between different samples or groups (Parks et al., 2014).
Specific steps were as follows: First, the generated OTU-table needed to
be standardized as the different species contain different 16S copy
numbers. Then, through analyzing the GreenGene ID corresponding to
each OTU, the KEGG family information of the OTU can be obtained,
thereby calculating the abundance of the KEGG, and obtaining the re-
lated pathway from the information of the KEGG database. Finally, a
significant difference test between samples was performed at the genus
level for species abundance between different samples using the G-TEST
and Fisher test methods in the STAMP software (t-test, P-value < 0.05
indicates significant). In general, differences and changes in the meta-
bolic pathways of functional genes of microbial communities between
samples among different groups can be predicted by differential ana-
lysis of KEGG metabolic pathways.

Fig. 2. 1H NMR spectra of fecal samples from CON group, KYDS group and YGP group. “ × 32” and “ × 8” represent the amplification factor of the spectra. (1)
deoxycholate; (2) n-butyrate; (3) valine; (4) leucine; (5) isoleucine; (6) propionate; (7) lactate; (8) alanine; (9) citrulline; (10) lysine; (11) cadaverine; (12) acetate;
(13) glutamate; (14) N-acetyl-5-aminosalicylate; (15) succinate; (16) dimethylamine (DMA); (17) trimethylamine (TMA); (18) creatinine; (19) taurine; (20) glycine;
(21) α-glucose/β-glucose; (22) glycerol; (23) fumarate; (24) tyrosine; (25) phenylalanine; (26) phenylacetate; (27) tryptophan; (28) histidine; (29) formate; (30) 3-
hydrobutyrate.

5
R. Chen, et al. Journal of Ethnopharmacology 244 (2019) 112139

2.8. Correlation profiling between the metabolites and intestinal microbiotas
Pearson correlation analysis was performed between altered meta-
bolites screened from fecal metabonomics and perturbed gut microbes
genus screened from 16S rRNA gene sequencing analysis. The criteria
for screening were correlation coefficient |r| > 0.70, P < 0.05.
Eventually, significant related metabolites and gut microbes genus were
obtained and displayed as heat maps (⁎P < 0.05, ⁎⁎P < 0.01). Red
indicates a positive correlation while blue indicates a negative corre-
lation.
2.9. Statistical analysis
One-way ANOVA with Bonferroni's multiple comparisons test and
Student's t-test were performed using SPSS 17.0 software (IBM, USA).
The significance threshold was set at P < 0.05.
3. Results
3.1. Histopathological and biochemical analysis
These suggested that the target gland axis was inhibited and KYDS
symptoms appeared in rats. Histopathological analysis (Fig. 1) also
showed that the hypothalamic-pituitary-target gland axis (adrenal
gland, thyroid and testis) of KYDS group's rats had undergone patho-
logical changes to a certain extent. The hypothalamic neuron cells
atrophied and quantity decreased; Pituitary basophils became less and
vacuolar degeneration; Adrenal cortex thinned and the cells atrophied.
Thyroid follicular lumen decreased, and follicular cells appeared
atrophy. What's more, inflammation and tissue proliferation also oc-
curred; Testicular stromal cells decreased, spermatogenic cell hierarchy
and sperm count has also decreased to varying degrees. In addition, rats
in KYDS group also showed behavioral expression similar to KYDS (Lu
et al., 2011; Gao et al., 2009; Zou et al., 2012), which further demon-
strated the success of KYDS modeling in this study. In YGP group,
however, these biochemical indicators all increased significantly. At the
same time, pathophysiology of the target gland axis has also been im-
proved to a degree. Therefore, it was indicated that YGP can have a
therapeutic effect on KYDS. These were basically consistent with our
previous research results and reported results (Chen et al., 2017, 2018a;
Zhou et al., 2016; Liu et al., 2016; Zhang et al., 2016; Nan et al., 2016).

It can be seen from Fig. S1 that in KYDS group, levels of biochemical 3.2. Representative 1H NMR spectra of feces
indicators related to the target gland axis (TSH, T3, T4, LH, FSH, T,
ACTH and CORT) and urine (17-OHCS) were significantly decreased. Fig. 2 shows representative 1H NMR spectra of fecal samples from

Fig. 3. Results of multivariate analysis among three groups. A1, A2: PCA and 3D PCA plots of CON, KYDS and YGP group (R2X = 0.808, Q2 = 0.729). B1, C1: OPLS-
DA plots of CON vs KYDS (R2X = 0.660, R2Y = 0.720, Q2 = 0.592, P value of CV-ANOVA = 0.0309), and KYDS vs YGP group (R2X = 0.763, R2Y = 0.917,
Q2 = 0.743, P value of CV-ANOVA = 0.0249). B2, C2: S-plots of CON, KYDS and YGP group. Variables labeled with red triangles are recognized as differential
metabolites (VIP > 1, P < 0.05) and the labels next to red triangles correspond to the numbers of metabolites in Table S2. (For interpretation of the references to
color in this figure legend, the reader is referred to the Web version of this article.)

6
R. Chen, et al.
CON group, KYDS group and YGP group. Thirty endogenous metabo-
lites in fecal samples were ultimately assigned based on published lit-
eratures and databases including: (1) deoxycholate; (2) n-butyrate; (3)
valine; (4) leucine; (5) isoleucine; (6) propionate; (7) lactate; (8) ala-
nine; (9) citrulline; (10) lysine; (11) cadaverine; (12) acetate; (13)
glutamate; (14) N-acetyl-5-aminosalicylate; (15) succinate; (16) di-
methylamine (DMA); (17) trimethylamine (TMA); (18) creatinine; (19)
taurine; (20) glycine; (21) α-glucose/β-glucose; (22) glycerol; (23) fu-
marate; (24) tyrosine; (25) phenylalanine; (26) phenylacetate; (27)
tryptophan; (28) histidine; (29) formate; (30) 3-hydrobutyrate.
3.3. Fecal metabolic profiling of YGP in treating KYDS rats
Fecal metabolic profiling (PCA and 3D PCA) shows that CON group,
KYDS group and YGP group can be clearly separated, and YGP group
was closer to CON group than KYDS group (Fig. 3A1-A2). In addition,
further OPLS-DA analysis also displays that the three groups were ob-
viously separated (Fig. 3B1, C1). Meanwhile, CV-ANOVA of OPLS-DA
and permutations test of PLS-DA (200 permutations, Fig. S2) also
manifests that the model was reliable. S-plots in Fig. 3B2 and C2 an-
notate and display the variable with VIP > 1 and P < 0.05 (red tri-
angle). Besides, all the points of VIP < 1 in the S-plots were removed,
leaving only the variable with VIP > 1.
21 metabolic markers with significant changes were eventually
found (VIP > 1, P < 0.05) via the multivariate statistical analysis of
feces obtained from KYDS rats treated by YGP (Table S2). These me-
tabolic markers were mainly involved in amino acid metabolism (va-
line, leucine, isoleucine, alanine, lysine, glutamate, taurine, glycine,
phenylalanine and tryptophan); energy metabolism (creatinine, α-glu-
cose, β-glucose, glycerol and phenylacetate); methylamine metabolism
(DMA, TMA); bile acid metabolism and urea cycle (deoxycholate,
Fig. 4. Summary of fecal samples' pathway analysis using MetaboAnalyst (A), and KEGG functional prediction analysis of intestinal microbiota (B). The key pathways
(P < 0.05) include: (1) aminoacyl-tRNA biosynthesis; (2) valine, leucine and isoleucine biosynthesis; (3) nitrogen metabolism; (4) phenylalanine metabolism; (5)
valine, leucine and isoleucine degradation; (6) butanoate metabolism; (7) alanine, aspartate and glutamate metabolism. The pathways annotated in red in (B)
represent potential pathways associated with changes in the intestinal microbiota of KYDS. (For interpretation of the references to color in this figure legend, the
reader is referred to the Web version of this article.)
7
Journal of Ethnopharmacology 244 (2019) 112139
citrulline); and short-chain fatty acid metabolism (n-butyrate, acetate).
Among them, studies have shown that metabolites associated with in-
testinal microbiota metabolism included deoxycholate, n-butyrate, ly-
sine, acetate, DMA, TMA, creatinine, α-glucose, β-glucose, phenylala-
nine, phenylacetate and tryptophan (Nicholson et al., 2012).
3.4. Key metabolic pathway analysis for different metabolites from feces
The results of pathway analysis revealed that there were seven
primary disturbed pathways (P < 0.05) of feces in KYDS rats treated
by YGP, involving: (1) aminoacyl-tRNA biosynthesis; (2) valine, leucine
and isoleucine biosynthesis; (3) nitrogen metabolism; (4) phenylalanine
metabolism; (5) valine, leucine and isoleucine degradation; (6) bu-
tanoate metabolism; (7) alanine, aspartate and glutamate metabolism
(Fig. 4A, Table S3). The color and size of each circle in (A) was based on
the P values and pathway impact values, respectively.
3.5. Sequencing data quality assessment and alpha diversity analysis
A total of 1,520,099 pairs of reads were obtained by sequencing of
19 fecal samples, and 1,228,541 clean tags were generated after paired-
ended reads were merged and filtered. Each sample yielded at least
62,698 clean tags, and an average of 64,660 clean tags. Eventually, 392
OTUs were obtained at a similar level of 97%.
To verify whether the sequencing amount of this study was large
enough to reflect the diversity of the original microorganisms, Alpha
diversity analysis (e.g. Chao1, Ace, Shannon, Simpson and Coverage
indices) was conducted on the results of 16S rRNA sequencing based on
97% similarity level. The Chao1 and Ace indices were used to measure
the abundance of species (i.e. the number of species). The Shannon and
Simpson indices were applied to measure species diversity. Coverage
R. Chen, et al. Journal of Ethnopharmacology 244 (2019) 112139

(caption on next page)

8
R. Chen, et al.
Fig. 5. The histogram of species distribution at the phylum (A) and genus (B) levels in three groups revealed by 16S rRNA sequencing (different colors represent
different bacteria at phylum or genus levels). PCoA analysis among CON, KYDS and YGP groups (C). Hierarchical clustering analysis of UPGMA in three groups (D).
The ANOVA analysis of gut bacteria at the genus level (E). ⁎ and ⁎⁎ denote compared with CON group, *P < 0.05, ⁎⁎P < 0.01, respectively. # and ## represent
compared with KYDS group, #P < 0.05, ##P < 0.01, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the
Web version of this article.)
was used for measuring OTU coverage and reflecting whether the se-
quencing results represented the true state of the microbes in the
sample. It can be found from Fig. S4 that, within a certain range, the
rarefaction curve tended to be flat with the increase of the number of
sequence sampled, indicating that the sequencing amount of each
sample in this study was sufficient. Similarly, as the number of se-
quence sampled increased, the Shannon curve also tended to be flat,
revealing that the vast majority of microbial species information has
been covered in the samples of this study (Fig. S5). Collectively, sta-
tistical analysis of Alpha diversity index showed that the diversity of
KYDS and YGP group was smaller than that of CON group (Table S4).
What's more, the coverage of all sample libraries was > 99.9%, in-
dicating that the library size of this study was large enough to represent
the majority of bacteria.
3.6. Species annotation analysis
Fig. 5. A and B show the results of species annotation analysis at the
phylum and genus levels in three groups revealed by 16S rRNA se-
quencing. The figure manifested only bacteria of the top ten abundance
at phylum and genus level, and merged remaining species into
“Others”, while unclassified represented species that were not tax-
onomically annotated. At the phylum level, Firmicutes and Bacteroidetes
were the most predominant phylum in gut bacteria of rats and account
for the mainly relative abundance in all samples, as shown in Fig. 5A.
Compared with CON group, Firmicutes in KYDS group decreased, while
Bacteroidetes increased. In contrast, compared to KYDS group, Firmicutes
increased in YGP group, whereas Bacteroidetes decreased. Studies have
shown that the imbalance between the Firmicutes and Bacteroidetes ra-
tios was largely related to energy metabolism (Lin et al., 2018b;
Turnbaugh et al., 2006, 2009). In accordance with previous reports,
KYDS was also closely associated with energy metabolism disorders
(Zhao et al., 2013a; Chen et al., 2017, 2018a; Zou et al., 2012; Lu et al.,
2011).
3.7. Beta diversity analysis
It can be seen that the gut microbiome composition profiles among
CON, KYDS and YGP groups were separated obviously, as shown by the
principal coordinate analysis (PCoA) plot in Fig. 5C. Of note, it was
quite clear that the corresponding 3D PCA score plot based on OTU
(97% similarity) could also be well separated (Fig. S3). In agreement
with PCoA and 3D PCA analysis, hierarchical clustering analysis of
unweighted pair-group method with arithmetic mean (UPGMA) mani-
fested that CON, KYDS and YGP group clustered clearly in their own
groups (Fig. 5D). Based on the distance matrix (Bray-Curtis algorithm)
obtained from beta diversity analysis, UPGMA was used to conduct
hierarchical clustering of samples through the R language tool to judge
the similarity of species composition between samples. The closer the
samples were, the shorter the branches were, indicating that the species
composition of the two samples was more similar.
3.8. Screening of significantly altered microbiota
The results of ANOVA analysis showed that eleven microbes at the
genus level occurred significantly in KYDS group and YGP group
(Fig. 5E). These microbes included: Bacteroides; Clostridium_sensu_s-
tricto_1; Desulfovibrio; Lactobacillus; Prevotellaceae_NK3B31_group; Rumi-
niclostridium_9; Ruminococcaceae_UCG-007; Ruminococcaceae_UCG-010;
9
Journal of Ethnopharmacology 244 (2019) 112139
Streptococcus; [Eubacterium]_coprostanoligenes_group; unculture-
d_bacterium_f_Bacteroidales_S24-7_group. In KYDS group, four bacteria
showed a significant increased (Bacteroides, Desulfovibrio, Ruminiclos-
tridium_9 and Streptococcus), and seven bacteria significantly decreased
(Clostridium_sensu_stricto_1, Prevotellaceae_NK3B31_group, Ruminococca-
ceae_UCG-007, Lactobacillus, Ruminococcaceae_UCG-010, [Eubacterium]
_coprostanoligenes_group and uncultured_bacterium_f_Bacteroidales_S24-
7_group). In YGP group, however, there were three significant declined
(Bacteroides, Desulfovibrio and Streptococcus), one significant increased
(uncultured_bacterium_f_Bacteroidales_S24-7_group), and the remaining
seven changes were not significant.
3.9. Function prediction of differential microbe
KEGG functional prediction analysis was performed on eleven dif-
ferent microbes at the genus levels. It was showed that these differential
microbes were chiefly related to amino acid metabolism, energy me-
tabolism, immune system, endocrine system, nervous system, carbo-
hydrate metabolism, lipid metabolism, and digestive system, etc
(Fig. 4B). The pathways annotated with red in Fig. 4B represented
potential pathways associated with changes in the intestinal microbiota
of KYDS.
3.10. Correlation analysis for differential metabolites and microbes
Correlation heatmap was applied to represent the covariation be-
tween perturbed gut microbes genus and altered fecal metabolites, as
shown in Fig. 6A. In CON group, the intestinal microbiota (B1, B4, B5,
B6, B7) were significantly associated with metabolites (M2, M3, M4,
M5, M7, M8, M10, M12, M14, M17, M18). In KYDS group, there were
significant associations between the intestinal microbiota (B1, B3, B10)
and metabolites (M1, M2, M3, M5, M9, M14, M15, M16, M17, M18,
M21). In YGP group, there were significant correlations between the
intestinal microbiota (B6, B7, B8, B11) and metabolites (M1, M3, M5,
M6, M7, M9, M11, M12, M19, M21). Overall, the correlation and
changes between metabolites and gut microbiota in YGP group were
closer to CON group than KYDS group. The correlation and changes
between the metabolites and gut microbiota (e.g. B1, B2, B3, B7, B9,
B10) in KYDS group tend to be opposite to those in CON group.
In order to further study the intrinsic relationship between these
metabolites and intestinal microbiota, the significantly correlated me-
tabolites and microbes in each group were extracted, and combined
with the above pathway analysis (metabolic pathway analysis and 16S
function prediction) to construct the interaction network diagram of the
metabolites-microbes-pathways. It can be found from Fig. 6B that
compared with the CON group from the perspective of fecal metabo-
nomics, KYDS group was mainly related to amino acid metabolism,
energy metabolism, methylamine metabolism, bile acid metabolism
and urea cycle, short-chain fatty acid metabolism, especially methyla-
mine metabolism. From the perspective of the intestinal microbiota, it
can also be seen that the occurrence of KYDS was mainly related to
three kinds of bacteria (Bacteroides, Desulfovibrio and [Eubacterium]
_coprostanoligenes_group), and these three kinds of bacteria mainly in-
volved in amino acid metabolism, energy metabolism, nervous, endo-
crine, immune and digestive system, lipid metabolism, and carbohy-
drate metabolism. These results were mutually validated with the
results of above fecal metabonomics, and were consistent with the ex-
isting findings about KYDS (Huang et al., 2013; Lu et al., 2011; Zhang
et al., 2014). After YGP treatment, however, levels of metabolites
R. Chen, et al. Journal of Ethnopharmacology 244 (2019) 112139

Fig. 6. Correlation heatmap is applied to represent the correlation values between perturbed gut microbes genus and altered fecal metabolites (A). M1-M21 cor-
responds to 21 differential metabolites in Table S2. B1–B11 represents eleven different bacteria, including Bacteroides; Clostridium_sensu_stricto_1; Desulfovibrio;
Lactobacillus; Prevotellaceae_NK3B31_group; Ruminiclostridium_9; Ruminococcaceae_UCG-007; Ruminococcaceae_UCG-010; Streptococcus; [Eubacterium]_coprostanoligen-
es_group; uncultured_bacterium_f_Bacteroidales_S24-7_group, respectively. “⁎” (white) refer to the significance level of the corresponding correlation coefficient in each
group. ⁎P < 0.05, ⁎⁎P < 0.01. Interaction network diagram of the metabolites-microbes-pathways based on correlation analysis (B). Metabolites (microbes) labeled
with magenta and green indicate an increase or decrease in levels. Pathways labeled with indigo mean the same metabolism between KYDS and YGP groups. The red
(blue) line represents a positive correlation (negative correlation) between the metabolite and microbes. ⁎P < 0.05 and ⁎⁎P < 0.01 compared with CON group,
while #P < 0.05 and ##P < 0.01 compared with KYDS group. The statistical results of metabolites and microbes were shown in Table S2 and Fig. 5E. (For
interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

related to amino acid metabolism, methylamine metabolism, bile acid
metabolism and urea cycle, short-chain fatty acid metabolism increased
from the perspective of fecal metabonomics, especially methylamine
metabolism (DMA and TMA), but energy metabolism was not statisti-
cally correlated. From the perspective of gut microbiota, it was found
that YGP mostly regulated endocrine, immune and digestive system,
lipid metabolism and carbohydrate metabolism via acting on four mi-
crobes, including Ruminiclostridium_9, Ruminococcaceae_UCG-007, un-
cultured_bacterium_f_Bacteroidales_S24-7_group, Ruminococcaceae_UCG-
010, thereby exerting a therapeutic effect. Liu et al. (2016) showed that

10
R. Chen, et al.
AS1350 (a TCM formulae with ingredients were similar to YGP) can
improve KYDS by acting on fatty acid metabolism and amino acid
metabolism. Zhao et al. (2013a) also showed that two herbal extracts
from GQ and RG, etc can improve KYDS through energy metabolism,
lipid metabolism, and gut microbiota metabolism. These further de-
monstrated that YGP can improve KYDS through intestinal microbiota.
Of note, it can be found that the intestinal microbiota significantly
correlated with KYDS is totally different from that of YGP regulated
KYDS, but the metabolic pathways and metabolites of the action are
basically consistent. The potential mechanism may be related to the
interaction between traditional Chinese medicine and intestinal mi-
crobiota, which is worthy of further study.
4. Discussion
Numerous studies have shown that KYDS is closely related to the
neuroendocrine immune system (NEIS) of modern medicine, and
mostly manifests as functional disorders of the hypothalamic-pituitary-
target gland axis (adrenal, thyroid and gonad) (Liu, 2009; Chen et al.,
2017; Zhang et al., 2016). Clinical studies have also shown that Yang-
deficiency patients are susceptible to digestive disorders (e.g. peptic
ulcer, diarrhea, irritable bowel syndrome), and endocrine and meta-
bolic diseases (e.g. hypothyroidism) (Huang et al., 2017; McKenna
et al., 2008; Dethlefsen et al., 2008; Turnbaugh et al., 2009). The 16S
rRNA sequencing analysis of KYDS rats’ feces in this experiment also
show that the occurrence of KYDS is closely related to gut microbiota-
mediated amino acid metabolism, energy metabolism, nervous, endo-
crine, immune and digestive system, lipid metabolism, and carbohy-
drate metabolism. These also further prove the reliability of the results
of this study.
4.1. Amino acid metabolism
A variety of amino acids (e.g. valine, isoleucine, alanine, trypto-
phan, glutamate and phenylalanine) can be further converted into
glucose by gluconeogenesis, thereby replenishing energy for the body.
From Table S2, eight amino acids (i.e. valine, leucine, isoleucine, ala-
nine, lysine, glutamate, glycine, phenylalanine and tryptohan) were
found to have a significantly decreased in KYDS group, whereas after
the treatment of YGP, six amino acids (i.e. leucine, lysine, glutamate,
taurine, glycine and phenylalanine) were found to increase sig-
nificantly. The results of metabolic pathway analysis and 16S functional
prediction analysis also showed that KYDS was significantly correlated
with aminoacyl-tRNA biosynthesis and amino acid metabolism, re-
spectively. This suggests that when KYDS occurs, amino acid metabo-
lism is disrupted, while YGP can improve amino acid metabolism. Va-
line, leucine and isoleucine are three kinds of branched-chain amino
acids (BCAAs) that participate in valine, leucine and isoleucine bio-
synthesis/degradation. It has been reported to that the gut microbiome
can affect circulating levels of several metabolites such as BCAAs, thus
potentially causing insulin resistance (Zierer et al., 2018). Correlation
analysis also showed that valine was significantly negatively correlated
with Desulfovibrio, and isoleucine was positively correlated with Bac-
teroides in KYDS group. Besides, both Desulfovibrio and Bacteroides are
markedly elevated, and Bacteroides is a probiotics while Desulfovibrio is
a common pathogen in the gut (Li, 2009). This indicates that the valine,
leucine and isoleucine biosynthesis/degradation mediated by Desulfo-
vibrio and Bacteroides in KYDS rats is disturbed. In YGP group, however,
both valine and isoleucine were significantly positively linked to un-
cultured_bacterium_f_Bacteroidales_S24-7_group, which was also sig-
nificantly increased (Figs. 5–6). Bacteroides and unculture-
d_bacterium_f_Bacteroidales_S24-7_group belong to Bacteroidetes phylum.
Studies have shown that Bacteroidetes has the functions of carbohydrate
fermentation, polysaccharide metabolism, bile acid and steroid meta-
bolism, as well as maintaining the normal physiological function of the
intestinal tract (Salyers, 1984). Ormerod et al. have declared that the
11
Journal of Ethnopharmacology 244 (2019) 112139
uncultured Bacteroidales family S24-7 with the core metabolism of fer-
mentation are highly localized to the gastrointestinal tracts of home-
othermic animals and increasingly being recognized as a numerically
predominant member of the gut microbiota (Ormerod et al., 2016). This
suggests that YGP can regulate valine, leucine and isoleucine bio-
synthesis/degradation through acting on unculture-
d_bacterium_f_Bacteroidales_S24-7_group.
In KYDS group, taurine was positively related to Bacteroides, glycine
was negatively correlated with Desulfovibrio, and tryptophan was ne-
gatively linked to [Eubacterium]_coprostanoligenes_group. In YGP group,
alanine was determined to correlate positively with
uncultured_bacterium_f_Bacteroidales_S24-7_group. Phenylalanine and
tryptohan were negatively related to Ruminococcaceae_UCG-010. Both
Ruminococcaceae_UCG-007 and Ruminococcaceae_UCG-010 are patho-
gens, and studies have shown that Ruminococcus are low in Yang-defi-
ciency patients (Wang et al., 2017). This also suggests that YGP can
improve KYDS by acting on both bacteria. Previous studies have
highlighted that an increase in bacterial biosynthetic potential for fatty
acids, sugars and various amino acids, including BCAAs (valine, leucine
and isoleucine) and aromatic amino acids (AAAs: tryptophan, tyrosine
and phenylalanine), associated with steatosis and insulin resistance
(Hoyles et al., 2018a). Guan et al. indicates that the TCM clinical status
of a person with kidney deficiency may change, thereby leading to
metabolic syndrome and a complex diagnosis including Xiao-ke (dia-
betes) and chest pain, etc (Guan, 2000). Huang et al. manifests that the
glucocorticoid (e.g. hydrocortisone) excess can cause insulin resistance
(Huang et al., 2013). Chen et al. also deems that kidney deficiency
syndrome may serve as a good precursor or a starting point for the
study of the complex conditions related to metabolic disorders, espe-
cially at their early stages (e.g. type 2 diabetes) (Chen et al., 2005).
Metabolic pathway analysis showed that KYDS was significantly asso-
ciated with phenylalanine metabolism as well. These all indicate that
KYDS is associated with BCAAs and AAAs-mediated dysbiosis of the
intestinal microbiota.
Moreover, study has displayed that severe disturbance of trypto-
phan metabolism occurred in the state of KYDS (Zhou et al., 2016). The
metabolites involved in tryptophan metabolism, such as 5-hydro-
xytryptophane, N-acetylserotonin, indole-3-acetate (I3A) and indole-3-
propionic acid (IPA), reflected a level of central nervous activity (Zhang
et al., 2016). The gut microbiota disorder-related metabolism of tryp-
tophan was reported to be involved in gut microbiome-derived neuro-
developmental diseases, thereby exerting the effect of inhibiting central
nervous system (CNS) inflammation (Rothhammer et al., 2018), and
modulating inflammatory response in hepatocytes and macrophages
(Krishnan et al., 2018). Furthermore, Zhao et al. also indicated that IPA,
a deamination product of tryptophan formed by symbiotic bacteria in
the gastrointestinal tract of mammals, decreased significantly in rats
exposed to hydrocortisone and could be regarded as the characteristic
biomarkers of KDS-Yang associated with the impaired gut microflora
(Zhao et al., 2013a). This may be because IPA can alleviate the pa-
thogenesis of diarrhea-related bowel diseases such as the manifestation
of watery diarrhea in KYDS rats and chronic diarrhea of KDS-Yang
patients via protecting against oxidative damage (Zhao et al., 2013a;
Sengul et al., 2011). Consistent with these, 16S function prediction
analysis also discovered that KYDS was significantly related to the
nervous system and digestive system. In addition, although 16S func-
tion prediction did not find nervous system-related pathways in YGP
group, dozens of literatures published have shown that YGP could im-
prove the nervous system and promote nerve regeneration (Ji et al.,
2016b; Kou et al., 2014). Consequently, it can be considered that the
nervous system and digestive system mediated by [Eubacterium]_co-
prostanoligenes_group and Ruminococcaceae_UCG-010 are disordered
when KYDS occurs, whereas YGP can alleviate KYDS by adjusting this
disorder to some extent.
R. Chen, et al.
4.2. Energy metabolism
From Table S2, five metabolites (creatinine, α-glucose, β-glucose,
glycerol and phenylacetate) were found to decrease in KYDS group,
especially creatinine. According to previous studies, when KYDS oc-
curred, mitochondrial damage, and TCA cycle and glycolysis were in-
hibited, resulting in energy metabolism disordered (Gao et al., 2009;
Chen et al., 2017, 2018a). Li et al. showed that the level of creatinine in
urine of Yang-deficiency patients significantly decreased, and there was
imbalance between aerobic metabolism and anaerobic glycolysis, re-
sulting in disturbance of energy metabolism (Li et al., 2011). 16S
function prediction analysis also showed that KYDS was significantly
associated with energy metabolism and carbohydrate metabolism.
Likewise, the correlation analysis results revealed that α-glucose, β-
glucose and glycerol displayed strong negative correlations with De-
sulfovibrio, and Desulfovibrio increased significantly in KYDS group.
These also confirm that KYDS is closely linked to energy metabolic
disorders. Conversely, the levels of these five metabolites in YGP group
increased significantly, while Desulfovibrio decreased significantly. Our
previous study has also shown that YGP can improve KYDS by reg-
ulating energy metabolism (Chen et al., 2018a). Nonetheless, in YGP
group, metabolites and metabolic pathways significantly associated
with energy metabolism were not found by correlation analysis and 16S
functional prediction. Presumably, YGP exerts therapeutic effects by
regulating metabolites or microbes indirectly related to energy meta-
bolism, and its mechanism of action requires more indepth studies.
As discussed above, alanine and glutamate are all glucogenic amino
acids that can be converted into glucose to provide energy for the body
through gluconeogenesis and glycolysis pathways. Both metabolites
decreased significantly in KYDS group, while glutamate was found to be
significantly elevated in YGP group. This illustrates the metabolic dis-
orders associated with alanine and glutamate. Consistently, the meta-
bolic pathway analysis of this study and our previous studies also
showed a significant correlation between alanine, aspartate and gluta-
mate metabolism and KYDS (Chen et al., 2017, 2018a).
4.3. Methylamine metabolism
Metabolites associated with methylamine metabolism (e.g. DMA,
TMA) decreased in KYDS group, but significantly increased in YGP
group. Our previous study also found that KYDS was determined to
correlate with TMAO and betaine-mediated intestinal microbiota dis-
orders, while YGP significantly enhanced TMAO and betaine to improve
KYDS (Chen et al., 2017, 2018a). Li et al. (2011) also showed that
changes of dimethylamine in urine of people with Yang-deficiency in-
dicated an imbalance in the intestinal microflora. TMA and DMA are
two important kinds of methylamine metabolites yielded via the de-
gradation of choline under the action of intestinal microbiota (Chen
et al., 2018b; Messenger et al., 2013). Correlation analysis revealed that
there were no DMA and TMA-associated microbes in KYDS group, while
there was correlation between the two metabolites and microbes in
CON group and YGP group (Fig. 6B). In CON group, TMA had a sig-
nificantly positive correlation with Prevotellaceae_NK3B31_group, and
Prevotellaceae_NK3B31_group was significantly decreased in KYDS
group. Prevotellaceae_NK3B31_group belongs to Bacteroidales with func-
tion of anti-infammation (Wei et al., 2018), and studies have shown
that TMAO was correlated with Bacteroidales in the gut microbiota
(Wang et al., 2016). For YGP group, DMA and TMA have statistically
negative related to Ruminococcaceae_UCG-010 and Ruminiclostridium_9,
respectively. Both Ruminococcaceae_UCG-010 and Ruminiclostridium_9
belong to Clostridiales containing the TMAO metabolic gene CutC/D
(Falony et al., 2015), and have also been shown to be associated with
TMAO-mediated intestinal microbiota (Wang et al., 2016). Previous
studies have demonstrated that microbiome-associated factors involve,
methylamines, such as TMA and TMA N-oxide (TMAO), playing a part
in the development of insulin resistance and atherosclerosis (Hoyles
12
Journal of Ethnopharmacology 244 (2019) 112139
et al., 2018a, 2018b; Dumas et al., 2017) These all suggest that KYDS is
associated with Prevotellaceae_NK3B31_group-mediated intestinal mi-
crobiota dysbiosis, and YGP can improve KYDS by regulating Rumino-
coccaceae_UCG-010 and Ruminiclostridium_9-mediated methylamine
metabolism.
Additionally, our previous studies also found that KYDS may be
related to disruption of cell memberane (Chen et al., 2018a). Choline is
an important component of biofilm formation which can yield TMA
under the action of intestinal microbiota to participate in methylamine
metabolism. 16S function prediction was also found to be associated
with membrane transport in both KYDS and YGP groups (Fig. 4B).
Therefore, this also indicates that KYDS is related to methylamine
metabolism.
4.4. Bile acid metabolism and urea cycle
It can be seen from Table S2 that deoxycholate was significantly
decreased in KYDS group, and correlation analysis also showed a sig-
nificant negative correlation between deoxycholate and [Eubacterium]
_coprostanoligenes_group in KYDS group. Deoxycholate is one of the
secondary bile acids from the metabolism of gut bacteria (Chen et al.,
2018b). In YGP group, although deoxycholate was not notably in-
creased, it was positively correlated with the unculture-
d_bacterium_f_Bacteroidales_S24-7_group, which was also significantly
increased. Therefore, it can be considered that KYDS is related to dys-
biosis of bile acid metabolism mediated by [Eubacterium]_coprostanoli-
genes_group, while YGP can ameliorate the bile acid metabolism of KYDS
by acting on uncultured_handed um_f_bacteroidales_s24-7_group.
Citrulline displayed a significant decrease in KYDS group and in-
creased in YGP group. It is involved in the urea cycle in liver's mi-
tochondria. Studies have shown that L-citrulline supplementation can
play an anti-inflammatory effect on kidney via promoting nitric oxide
production (Chen et al., 2018b; Persson et al., 2014). Our previous
studies have also found that KYDS is closely related to mitochondrial
damage, inflammatory response, and liver and kidney damage (Chen
et al., 2018a). Zou et al. revealed that the ammonia transport function
was blocked in the state of KYDS, and the alanine-glucose cycle was
disturbed, which affected the urea cycle (Zou et al., 2012). The changes
of alanine and glucose in KYDS group and YGP group also confirm this
viewpoint. Besides, correlation analysis also manifested a significantly
negative correlation between citrulline and Pre-
votellaceae_NK3B31_group in CON group. Prevotellaceae_NK3B31_group is
a probiotic with a significant decrease in KYDS group. Meanwhile, ci-
trulline displayed strong negative correlations with Ruminococca-
ceae_UCG-007 in YGP group. Given Ruminococcaceae_UCG-007 did not
show significant changes in YGP group, a significant decrease could be
observed in KYDS group. Wang et al. also showed that Ruminococcus
decreased in Yang-deficiency patients (Wang et al., 2017). These sug-
gest that KYDS may be related to the urea cycle mediated by Pre-
votellaceae_NK3B31_group and Ruminococcaceae_UCG-007.
4.5. Short-chain fatty acid metabolism
N-butyrate and acetate, two end products of the intestinal microbial
fermentation of complex carbohydrates and amino acids, are short-
chain fatty acids (SCFAs) with potential anticarcinogenic activity of
microbial metabolites present in feces, and are important for colonic
health (Chen et al., 2018b; Saric et al., 2008; Macfarlane and
Macfarlane, 2012). They can be sensed by the G protein-coupled re-
ceptors, GPR41 and GPR43, and expressed by enteroendocrine cells
(Samuel et al., 2008). Nicholson et al. have shown that n-butyrate or
specific species of butyrate-producing gut bacteria may be a new target
for restoring host immune function and barrier integrity, and for reg-
ulating energy metabolism (Nicholson et al., 2012). Both n-butyrate
and acetate showed significant decrease in KYDS group, illustrating that
the short-chain fatty acid metabolism and energy metabolism were
R. Chen, et al.
inhibited. Of particular interest, n-butyrate and acetate displayed a
significant negative correlation with Desulfovibrio in KYDS group, while
Desulfovibrio significantly increased. Desulfovibrio is a kind of sulfate
reducing bacteria and pathogenic bacteria commonly found in the gut.
On one hand, it reduces sulfate to sulfide which has a toxic effect on
intestinal epithelial cells and induces abnormal proliferation and me-
tabolism of epithelial cells (Loubinoux et al., 2002). On the other hand,
it also destroys the intestinal barrier function by inhibiting the oxida-
tion of n-butyrate (Pitcher and Cummings, 1996; Roediger et al., 1997).
Resultantly, when KYDS occurs, Desulfovibrio increases significantly and
induces abnormal proliferation and metabolism of epithelial cells,
leading to a decline in n-butyrate. Simultaneously, Desulfovibrio also
destroys the intestinal barrier function by inhibiting the oxidation of n-
butyrate to yield ketone bodies which provide energy for intestinal
epithelial cells. Prevotellaceae_NK3B31_group is related to the synthesis
of SCFAs (Shen et al., 2017), which is significantly correlated with n-
butyrate in CON group of Fig. 6B. Wei et al. have indicated that Pre-
votellaceae_NK3B31_group correlated with SCFAs-production and anti-
inflammation decreased in T2DM rats, while increased after the treat-
ment of TCM formula (Wei et al., 2018). What's more, metabolic
pathway analysis manifested that butanoate metabolism was sig-
nificantly associated with KYDS as well (Fig. 4A). These results all
further demonstrates that n-butyrate plays an important role in the
development of KYDS.
After YGP treatment, however, both acetate and
uncultured_bacterium_f_Bacteroidales_S24-7_group increased significantly.
Coherently, the correlation analysis also showed that there was a sig-
nificant positive correlation between the two. As mentioned earlier,
uncultured_bacterium_f_Bacteroidales_S24-7_group belonged to
Bacteroidetes could maintain the normal physiological function of the
intestinal tract with the functions of carbohydrate fermentation, poly-
saccharide metabolism, bile acid and steroid metabolism (Salyers,
1984). Therefore, this indicates that YGP can improve the intestinal
microbiota disturbance of KYDS rats by regulating different microbes or
metabolites associated with short-chain fatty acid metabolism, thereby
exerting a therapeutic effect.
Although there were no significant correlation between metabolites
and three micrboes (i.e. Clostridium_sensu_stricto_1, Lactobacillus and
Streptococcus) in KYDS and YGP group from correlation analysis, they
showed significant changes between these two groups. Therefore, we
think that they should also be related to KYDS. Lysine decreased sig-
nificantly in KYDS group and significantly increased in YGP group. It is
one of the essential amino acids in the human body, which enhances
immunity and improves the function of central nervous tissues. This
also indicates that KYDS is involved in the metabolic disorder of the
neuroendocrine immune system. Correlation analysis also showed that
lysine was significantly negatively correlated with Lactobacillus in CON
group, but no correlation was found in KYDS and YGP group. In addi-
tion, Lactobacillus also decreased significantly in KYDS group. It is one
of the most widely used probiotics, which helps digestion, prevents the
adhesion of harmful bacteria in the intestinal epithelium, and enhances
host immunity (Ljungh and Wadström, 2006; Sanz et al., 2007). The
research also showed that Lactobacillus was significantly decreased in
patients with kidney-Yang deficiency, reflecting the basic character-
istics of spleen and stomach function impairment in patients with
kidney-Yang deficiency (Ding et al., 2007). In addition, a large number
of studies have also found that the intestinal microbiota of patients with
intestinal-brain disease changes significantly, and the intestinal mi-
crobiota may also cause central nervous system diseases (Seksik et al.,
2003; Bajaj et al., 2012). In contrast, YGP have been found to improve
the nervous system and promote nerve regeneration. Therefore, this
suggests that KYDS is associated with lysine and Lactobacillus-mediated
metabolic disorders, and YGP can improve this disorder to some extent.
Streptococcu is a conditional pathogen that causes suppurative in-
flammation and hypersensitivity diseases. It significantly increased in
KYDS group and significantly decreased in YGP group. Studies have
13
Journal of Ethnopharmacology 244 (2019) 112139
shown that Streptococcus was elevated in patients with irritable bowel
syndrome (IBS) compared with normal subjects (Kassinen et al., 2007).
People with yang deficiency are often susceptible to diseases such as
diarrhea and stool, which are typical manifestations of IBS. Thus, this
suggests that KYDS is associated with Streptococcu-mediated gastro-
intestinal disorders, while YGP can treat KYDS by regulating Strepto-
coccu. Clostridium_sensu_stricto_1 belongs to Clostridia, and most Clos-
tridia are butyrate-producing bacteria. Butyrate is the final metabolite
of polysaccharides that cannot be absorbed by the host in the fermented
food of the intestinal microbiota, and plays an important role in the
energy supply of the host and the development of intestinal epithelial
cells (Pryde et al., 2002). Clostridium_sensu_stricto_1 decreased sig-
nificantly in KYDS group, indicating that the production of butyrate
was reduced, therefore resulting in inhibition of energy metabolism.
Although TCM and microecology have great differences in form, both
are based on multi-pathway and multi-target to explore the wholeness
law between the organism and environment. TCM formula can cure
diseases by fostering probiotics to repel pathogenic bacteria. Conse-
quently, the combination of TCM and microecology is of great sig-
nificance for the treatment of complex diseases.
In conclusion, given this study initially confirmed that KYDS is as-
sociated with intestinal microbiota disorders, and YGP can play a role in
the treatment of KYDS by regulating intestinal microbiota. However, if
more comprehensive and reliable results want to be obtained, we be-
lieve that the following aspects of works need to be done. First of all, a
metagenomics study was conducted to further analyze the function of
these gut microbiotas. Meanwhile, studies of bioactivity and fecal
transplantation experiments were performed on the differential in-
testinal microbiota. Next, expand the sample size and combine the
clinical data for more in-depth research. Last but not least, it is neces-
sary to elucidate the active ingredient and optimal dosage of YGP in the
treatment of KYDS's gut microbiota dysbiosis in combination with
pharmacology.
5. Conclusions
In the present work, fecal metabonomics combined with 16S rRNA
gene sequencing analysis were used to study the specific interplay between
KYDS and intestinal microbiota, and the intrinsic regulation mechanism of
YGP on KYDS's gut microbiota. Using fecal metabonomics, KYDS was
found to be associated with 21 differential metabolites and seven potential
metabolic pathways. These metabolites and metabolic pathways were
mainly involved in amino acid metabolism, energy metabolism, methy-
lamine metabolism, bile acid metabolism and urea cycle, and short-chain
fatty acid metabolism. Through 16S rRNA gene sequencing analysis, we
found that KYDS was related to eleven different intestinal microbiotas at
the genus level. These gut microbiotas were mostly involved in amino acid
metabolism, energy metabolism, nervous, endocrine, immune and diges-
tive system, lipid metabolism, and carbohydrate metabolism. Combined
fecal metabonomics and 16S rRNA gene sequencing analysis, we further
discovered that KYDS was primarily linked to three gut microbiotas (i.e.
Bacteroides, Desulfovibrio and [Eubacterium]_coprostanoligenes_group) and
eleven related metabolites (i.e. deoxycholate, n-butyrate, valine, iso-
leucine, acetate, taurine, glycine, α-gluconse, β-glucose, glycerol and
tryptophan) mediated various metabolic disorders (amino acid metabo-
lism, energy metabolism, especially methylamine metabolism, bile acid
metabolism and urea cycle, short-chain fatty acid metabolism, nervous,
endocrine, immune and digestive system, lipid metabolism, and carbo-
hydrate metabolism). YGP, however, has the ability to mediate four kinds
of microbes (i.e. Ruminiclostridium_9, Ruminococcaceae_UCG-007,
Ruminococcaceae_UCG-010, and uncultured_bacterium_f_Bacteroidales_S24-
7_group) and ten related metabolites (i.e. deoxycholate, valine, isoleucine,
alanine, citrulline, acetate, DMA, TMA, phenylalanine and tryptophan)
mediated amino acid metabolism, especially methylamine metabolism,
bile acid metabolism and urea cycle, short-chain fatty acid metabolism,
endocrine, immune and digestive system, and lipid metabolism, thereby
R. Chen, et al.
exerting a therapeutic effect on KYDS rats. Overall, our findings have
preliminary confirmed that KYDS is closely related to metabolic and mi-
crobial dysbiosis, and indicates that YGP can improve the metabolic dis-
order of KYDS by acting on intestinal microbiota. Meanwhile, this will lay
the foundation for the further KYDS's metagenomic research and the use of
intestinal microbial as drug targets to treat KYDS.
Conflicts of interest
The authors declare no competing financial interest.
Author contributions
Xiufeng Wang obtained funding and helped reviewing; Ruiqun Chen
performed the experiments, analyzed the data and wrote the manu-
script; Runhua Zhan were responsible for helping the collection of ex-
perimental samples; Jia Wang and Lei Zhang helped with data analysis.
All authors have reviewed the manuscript and given approval to the
final version of the manuscript.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China of China (No. 81403153), and Science and
Technology Planning Project of Guangdong Province, China (No.
2014A020212603), and Scientific Research Projects of Guangdong
Traditional Chinese Medicine Bureau (20191195).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.jep.2019.112139.
References
Bajaj, J.S., Ridlon, J.M., Hylemon, P.B., Thacker, L.R., Heuman, D.M., Smith, S.,
Sikaroodi, M., Gillevet, P.M., 2012. Linkage of gut microbiome with cognition in
hepatic encephalopathy. Am. J. Physiol. Gastrointest. Liver Physiol. 302 (1),
G168–G175.
Bokulich, N.A., Subramanian, S., Faith, J.J., Gevers, D., Gordon, J.I., Knight, R., Mills,
D.A., Caporaso, J.G., 2013. Quality-filtering vastly improves diversity estimates from
Illumina amplicon sequencing. Nat. Methods 10 (1), 57–59.
Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K.,
Fierer, N., Peña, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T.,
Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D.,
Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J.,
Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of high-
throughput community sequencing data. Nat. Methods 7 (5), 335–336.
Chang, C.J., Lin, C.S., Lu, C.C., Martel, J., Ko, Y.F., Ojcius, D.M., Tseng, S.F., Wu, T.R.,
Chen, Y.M., Young, J.D., Lai, H.C., 2017. Corrigendum: ganoderma lucidum reduces
obesity in mice by modulating the composition of the gut microbiota. Nat. Commun.
8 (6), 16130.
Chen, M., Zhao, L., Jia, W., 2005. Metabonomic study on the biochemical profiles of a
hydrocortisone-induced animal model. J. Proteome Res. 4 (6), 2391–2396.
Chen, R., Wang, J., Liao, C., Ma, N., Zhang, L., Wang, X., 2017. 1H NMR studies on serum
metabonomic changes over time in a kidney-Yang deficiency syndrome model. RSC
Adv. 7 (54), 34251–34261.
Chen, R., Wang, J., Liao, C., Zhang, L., Guo, Q., Wang, X., 2018a. Exploring the bio-
markers and therapeutic mechanism of kidney-yang deficiency syndrome treated by
You-gui pill using systems pharmacology and serum metabonomics. RSC Adv. 8 (2),
1098–1115.
Chen, R., Liao, C., Guo, Q., Wu, L., Zhang, L., Wang, X., 2018b. Combined systems
pharmacology and fecal metabonomics to study the biomarkers and therapeutic
mechanism of type 2 diabetic nephropathy treated with Astragalus and Leech. RSC
Adv. 8 (48), 27448–27463.
Chong, J., Soufan, O., Li, C., Caraus, I., Li, S., Bourque, G., Wishart, D.S., Xia, J., 2018.
MetaboAnalyst 4.0: towards more transparent and integrative metabolomics analysis.
Nucleic Acids Res. 46 (W1), W486–W494.
Dethlefsen, L., Huse, S., Sogin, M.L., Relman, D.A., 2008. The pervasive effects of an
antibiotic on the human gut microbiota, as revealed by deep 16S rRNA sequencing.
PLoS Biol. 6 (11), e280.
Ding, W.J., Gao, F., Yang, J., Yang, H.Y., Zhou, B.J., 2007. Clinical study on intestinal
flora imbalance in patients with kidney-yang deficiency syndrome. J.N. Chin. Med.
39 (1), 9–10.
Dumas, M.E., Rothwell, A.R., Hoyles, L., Aranias, T., Chilloux, J., Calderari, S., Noll, E.M.,
14
Journal of Ethnopharmacology 244 (2019) 112139
Pean, N., Boulange, C.L., Blancher, C., Barton, R.H., Gu, Q., Fearnside, J.F., Deshayes,
C., Hue, C., Scott, J., Nicholson, J.K., Gauguier, D., 2017. Microbial-host co-meta-
bolites are prodromal markers predicting phenotypic heterogeneity in behavior,
obesity, and impaired glucose tolerance. Cell Rep. 20 (1), 136–148.
Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST.
Bioinformatics 26 (19), 2460–2461.
Falony, G., Vieira-Silva, S., Raes, J., 2015. Microbiology meets big data: the case of gut
microbiota-derived trimethylamine. Annu. Rev. Microbiol. 69, 305–321.
Yang, G., Gao, G., Lou, Z., 2009. Metabonomic study on urine of rats with hydrocortisone-
induced kidney deficiency syndrome. Acad. J. Second Mil. Med. Univ. 29 (5),
565–568.
Gong, M., Ye, W., Xie, Y., Wang, S., Liang, S., Zou, Z., 2012. Metabonomic study of in-
tervention effects of Morinda officinalis on 'kidney-yang deficiency syndrome'. China
J. Chin. Mater. Med. 37 (11), 1682–1685.
Guan, Z.A., 2000. Modern Diabetes Research. Tianjin Science and Technology Press,
Tianjin, pp. pp247.
He, C., Shan, Y., Song, W., 2015. Targeting gut microbiota as a possible therapy for
diabetes. Nutr. Res. (N.Y.) 35 (5), 361–367.
Hoyles, L., Fernandez-Real, J.M., Federici, M., Serino, M., Abbott, J., Charpentier, J.,
Heymes, C., Luque, J.L., Anthony, E., Barton, R.H., Chilloux, J., Myridakis, A.,
Martinez-Gili, L., Moreno-Navarrete, J.M., Benhamed, F., Azalbert, V., Blasco-Baque,
V., Puig, J., Xifra, G., Ricart, W., Tomlinson, C., Woodbridge, M., Cardellini, M.,
Davato, F., Cardolini, I., Porzio, O., Gentileschi, P., Lopez, F., Foufelle, F., Butcher,
S.A., Holmes, E., Nicholson, J.K., Postic, C., Burcelin, R., Dumas, M.E., 2018a.
Molecular phenomics and metagenomics of hepatic steatosis in non-diabetic obese
women. Nat. Med. 24 (7), 1070–1080.
Hoyles, L., Jimenez-Pranteda, M.L., Chilloux, J., Brial, F., Myridakis, A., Aranias, T.,
Magnan, C., Gibson, G.R., Sanderson, J.D., Nicholson, J.K., Gauguier, D., McCartney,
A.L., Dumas, M.E., 2018b. Metabolic retroconversion of trimethylamine N-oxide and
the gut microbiota. Microbiome 6 (1), 73.
Huang, D., Yang, J., Lu, X., Deng, Y., Xiong, Z., Li, F., 2013. An integrated plasma and
urinary metabonomic study using UHPLC-MS: intervention effects of Epimedium
koreanum on 'Kidney-Yang Deficiency syndrome' rats. J. Pharm. Biomed. Anal. 76,
200–206.
Huang, T., Li, Q., Yin, Y., Wang, J., Sun, S., Yang, F., Li, L., Wang, J., Wang, Q., Li, Y.,
2017. Composition of intestinal microflora associated with Yang-deficiency. J. Tradit.
Chin. Med. Sci. 4 (3), 254–260.
Jacobs, D.M., Deltimple, N., van Velzen, E., van Dorsten, F.A., Bingham, M., Vaughan,
E.E., van Duynhoven, J., 2008. (1)H NMR metabolite profiling of feces as a tool to
assess the impact of nutrition on the human microbiome. NMR Biomed. 21 (6),
615–626.
Ji, Y.Z., Geng, L., Zhou, H.B., Wei, H.C., Chen, H.D., 2016a. Chinese herbal medicine
Yougui Pill reduces exogenous glucocorticoid-induced apoptosis in anterior pituitary
cells. Neural. Regen. Res. 11 (12), 1962–1968.
Ji, X., Liu, H., An, C., Wang, Y., Zhao, H., Zhang, Q., Li, M., Qi, F., Chen, Z., Wang, X.,
Wang, L., 2016b. You-Gui pills promote nerve regeneration by regulating netrin1,
DCC and Rho family GTPases RhoA, Racl, Cdc42 in C57BL/6 mice with experimental
autoimmune encephalomyelitis. J. Ethnopharmacol. 187, 123–133.
Kassinen, A., Krogius-Kurikka, L., Mäkivuokko, H., Rinttilä, T., Paulin, L., Corander, J.,
Malinen, E., Apajalahti, J., Palva, A., 2007. The fecal microbiota of irritable bowel
syndrome patients differs significantly from that of healthy subjects.
Gastroenterology 133 (1), 24–33.
Kelly, D., Conway, S., Aminov, R., 2005. Commensal gut bacteria: mechanisms of immune
modulation. Trends Immunol. 26 (6), 326–333.
Kou, S., Zheng, Q., Wang, Y., Zhao, H., Zhang, Q., Li, M., Qi, F., Fang, L., Liu, L., Ouyang,
J., Zhao, H., Wang, L., 2014. Zuo-Gui and You-Gui pills, two traditional Chinese
herbal formulas, downregulated the expression of NogoA, NgR, and RhoA in rats with
experimental autoimmune encephalomyelitis. J. Ethnopharmacol. 158, 102–112
2014.
Krishnan, S., Ding, Y., Saedi, N., Choi, M., Sridharan, G.V., Sherr, D.H., Yarmush, M.L.,
Alaniz, R.C., Jayaraman, A., Lee, K., 2018. Gut microbiota-derived tryptophan me-
tabolites modulate inflammatory response in hepatocytes and macrophages. Cell Rep.
23 (4), 1099–1111.
Le Gall, G., Noor, S.O., Ridgway, K., Scovell, L., Jamieson, C., Johnson, I.T., Colquhoun,
I.J., Kemsley, E.K., Narbad, A., 2011. Metabolomics of fecal extracts detects altered
metabolic activity of gut microbiota in ulcerative colitis and irritable bowel syn-
drome. J. Proteome Res. 10 (9), 4208–4218.
Lederberg, J., 2000. Infect. Hist. Sci. 288 (5464), 287–293.
Ley, R.E., Turnbaugh, P.J., Klein, S., Gordon, J.I., 2006. Microbial ecology: human gut
microbes associated with obesity. Nature 444 (7122), 1022–1023.
Li, M., 2009. Co-variation Analysis of Human Gut Microbial Structure and Host Global
Metabolism. Shanghai Jiao Tong University Doctoral dissertation.
Li, Y.S., Wang, Q., Yuan, Z.J., 2011. NMR-based metabonomics studies on serum and
urine of Yang-deficiency constitution. Chem. J. Chin. Univ. 32 (11), 2521–2527.
Lin, Z., Ye, W., Zu, X., Xie, H., Li, H., Li, Y., Zhang, W., 2018a. Integrative metabolic and
microbial profiling on patients with spleen-yang-deficiency syndrome. Sci. Rep. 8 (1),
6619.
Lin, Z., Ye, W., Zu, X., Xie, H., Li, H., Li, Y., Zhang, W., 2018b. Integrative metabolic and
microbial profiling on patients with spleen-yang-deficiency syndrome. Sci. Rep. 8 (1),
6619.
Liu, Y., 2009. Progress and prospects of the molecular mechanism of kidney-YANG de-
ficiency syndrome. J. Tradit. Chin. Med. Univ. Hunan 29 (10), 66–67.
Liu, Q., Zhang, A., Wang, L., Yan, G., Zhao, H., Sun, H., Zou, S., Han, J., Ma, C.W., Kong,
L., Zhou, X., Nan, Y., Wang, X., 2016. High-throughput chinmedomics-based pre-
diction of effective components and targets from herbal medicine AS1350. Sci. Rep.
6, 38437.
R. Chen, et al.
Liu, H., Qiu, F., Bian, B., Yang, X., Zou, H., Zhao, H., Chen, Z., Wang, Q., Zhao, H., Wang,
L., 2017. Integrating qualitative and quantitative assessments of Yougui pill, an ef-
fective traditional Chinese medicine, by HPLC-LTQ-Orbitrap-MSn and UPLC-QqQ-
MS/MS. Anal. Methods 9 (23), 3485–3496.
Ljungh, A., Wadström, T., 2006. Lactic acid bacteria as probiotics. Curr. Issues Intest.
Microbiol. 7 (2), 73–89.
Loubinoux, J., Bronowicki, J.P., Pereira, I.A., Mougenel, J.L., Faou, A.E., 2002. Sulfate-
reducing bacteria in human feces and their association with inflammatory bowel
diseases. FEMS Microbiol. Ecol. 40 (2), 107–112.
Lu, X., Xiong, Z., Li, J., Zheng, S., Huo, T., Li, F., 2011. Metabonomic study on 'Kidney-
Yang Deficiency syndrome' and intervention effects of Rhizoma Drynariae extracts in
rats using ultra performance liquid chromatography coupled with mass spectrometry.
Talanta 83 (3), 700–708.
Lu, K., Abo, R.P., Schlieper, K.A., Graffam, M.E., Levine, S., Wishnok, J.S., Swenberg, J.A.,
Tannenbaum, S.R., Fox, J.G., 2014. Arsenic exposure perturbs the gut microbiome
and its metabolic profile in mice: an integrated metagenomics and metabolomics
analysis. Environ. Health Perspect. 122 (3), 284–291.
Macfarlane, G.T., Macfarlane, S., 2012. Bacteria, colonic fermentation, and gastro-
intestinal health. J. AOAC Int. 95 (1), 50.
McKenna, P., Hoffmann, C., Minkah, N., Aye, P.P., Lackner, A., Liu, Z., Lozupone, C.A.,
Hamady, M., Knight, R., Bushman, F.D., 2008. The macaque gut microbiome in
health, lentiviral infection, and chronic enterocolitis. PLoS Pathog. 4 (2), e20.
Messenger, J., Clark, S., Massick, S., Bechtel, M., 2013. A review of trimethylaminuria:
(fish odor syndrome). J. Clin. Aesthet. Dermatol. 6 (11), 45–48.
Nan, Y., Zhou, X., Liu, Q., Zhang, A., Guan, Y., Lin, S., Kong, L., Han, Y., Wang, X., 2016.
Serum metabolomics strategy for understanding pharmacological effects of ShenQi
pill acting on kidney yang deficiency syndrome. J. Chromatogr. B Analyt. Technol.
Biomed. Life Sci. 1026, 217–226.
Nicholson, J.K., 2006. Global systems biology, personalized medicine and molecular
epidemiology. Mol. Syst. Biol. 2 (52), 1–6.
Nicholson, J.K., Holmes, E., Wilson, I.D., 2005. Gut microorganisms, mammalian meta-
bolism and personalized health care. Nat. Rev. Microbiol. 3 (5), 431–438.
Nicholson, J.K., Holmes, E., Kinross, J., Burcelin, R., Gibson, G., Jia, W., Pettersson, S.,
2012. Host-gut microbiota metabolic interactions. Science 336 (6086), 1262–1267.
Ormerod, K.L., Wood, D.L.A., Lachner, N., Gellatly, S.L., Daly, J.N., Parsons, J.D.,
Dal'Molin, C.G.O., Palfreyman, R.W., Nielsen, L.K., Cooper, M.A., 2016. Genomic
characterization of the uncultured Bacteroidales family S24-7 inhabiting the guts of
homeothermic animals. Microbiome 4 (1), 36.
Parks, D.H., Tyson, G.W., Hugenholtz, P., Beiko, R.G., 2014. STAMP: statistical analysis of
taxonomic and functional profiles. Bioinformatics 30 (21), 3123–3124.
Persson, P., Fasching, A., Teerlink, T., Hansell, P., Palm, F., 2014. L-Citrulline, but not L-
arginine, prevents diabetes mellitus-induced glomerular hyperfiltration and protei-
nuria in rat. Hypertension 64 (2), 323–329.
Pitcher, M.C., Cummings, J.H., 1996. Hydrogen sulphide: a bacterial toxin in ulcerative
colitis? Gut 39 (1), 1–4.
Pryde, S.E., Duncan, S.H., Hold, G.L., Stewart, C.S., Flint, H.J., 2002. The microbiology of
butyrate formation in the human colon. FEMS Microbiol. Lett. 217 (2), 133–139.
Roediger, W.E., Moore, J., Babidge, W., 1997. Colonic sulfide in pathogenesis and
treatment of ulcerative colitis, Dig. Dis. Sci. 42 (8), 1571–1579.
Rothhammer, V., Borucki, D.M., Tjon, E.C., Takenaka, M.C., Chao, C.C., Ardura-Fabregat,
A., de Lima, K.A., Gutiérrez-Vázquez, C., Hewson, P., Staszewski, O., Blain, M., Healy,
L., Neziraj, T., Borio, M., Wheeler, M., Dragin, L.L., Laplaud, D.A., Antel, J., Alvarez,
J.I., Prinz, M., Quintana, F.J., 2018. Microglial control of astrocytes in response to
microbial metabolites. Nature 557 (7707), 724–728.
Salyers, A.A., 1984. Bacteroides of the human lower intestinal tract. Annu. Rev.
Microbiol. 38, 293–313.
Samuel, B.S., Shaito, A., Motoike, T., Rey, F.E., Backhed, F., Manchester, J.K., Hammer,
R.E., Williams, S.C., Crowley, J., Yanagisawa, M., Gordon, J.I., 2008. Effects of the
gut microbiota on host adiposity are modulated by the short-chain fatty-acid binding
G protein-coupled receptor, Gpr41. Proc. Natl. Acad. Sci. U.S.A. 105 (43),
16767–16772.
Sanz, Y., Nadal, I., Sánchez, E., 2007. Probiotics as drugs against human gastrointestinal
infections. Recent Pat. Anti-Infect. Drug Discov. 2 (2), 148–156.
Saric, J., Li, J.V., Wang, Y.L., Keiser, J., Bundy, J.G., Holmes, E., Utzinger, J., 2008.
Metabolic profiling of an Echinostoma caproni infection in the mouse for biomarker
discovery. PLoS Neglected Trop. Dis. 2 (7), e254.
Seksik, P., Rigottier-Gois, L., Gramet, G., Sutren, M., Pochart, P., Marteau, P., Jian, R.,
Dore, J., 2003. Alterations of the dominant faecal bacterial groups in patients with
Crohn's disease of the colon. Gut 52 (2), 237–242.
Sengul, N., Isik, S., Aslim, B., Ucar, G., Demirbag, A.E., 2011. The effect of exopoly-
saccharide-producing probiotic strains on gut oxidative damage in experimental co-
litis. Dig. Dis. Sci. 56 (3), 707–714.
Shen, F., Zheng, R., Sun, X., Ding, W., Wang, X., Fan, J., 2017. Gut microbiota dysbiosis in
patients with non-alcoholic fatty liver disease. Hepatobiliary Pancreat. Dis. Int.:
HBPD INT 16 (4), 375–381.
15
Journal of Ethnopharmacology 244 (2019) 112139
Turnbaugh, P.J., Ley, R.E., Mahowald, M.A., Magrini, V., Mardis, E.R., Gordon, J.I., 2006.
An obesity-associated gut microbiome with increased capacity for energy harvest.
Nature 444 (7122), 1027–1031.
Turnbaugh, P.J., Hamady, M., Yatsunenko, T., Cantarel, B.L., Duncan, A., Ley, R.E., Sogin,
M.L., Jones, W.J., Roe, B.A., Affourtit, J.P., Egholm, M., Henrissat, B., Heath, A.C.,
Knight, R., Gordon, J.I., 2009. A core gut microbiome in obese and lean twins. Nature
457 (7228), 480.
Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., 2007. Naive Bayesian classifier for rapid
assignment of rRNA sequences into the new bacterial taxonomy. Appl. Environ.
Microbiol. 73 (16), 5261–5267.
Wang, X.F., Yi, L.N., Zhang, L., Li, J., Luo, L.C., 2013a. Study on the law of compatibility
in three categorized formulas for tonifying Shen Yang: shenqi Pill, Yougui Pill, and
Yougui Drink based on rough set. Chin. J. Integr. Tradit. West. Med. 33 (1), 114–118.
Wang, X.F., Zhang, L., Tang, X.Y., Deng, S.Y., Luo, L.C., 2013b. Study on the law of
compatibility in yougui pill based on decision rules model and uniform design. Chin.
J. Exp. Tradit. Med. Formulae 19 (3), 321–324.
Wang, L., Cao, A.L., Chi, Y.F., Ju, Z.C., Yin, P.H., Zhang, X.M., Peng, W., 2015. You-gui
Pill ameliorates renal tubulointerstitial fibrosis via inhibition of TGF-beta/Smad
signaling pathway. J. Ethnopharmacol. 169, 229–238 2015.
Wang, S., Xia, G.H., He, Y., Liao, S.X., Yin, J., Sheng, H.F., Zhou, H.W., 2016. Distribution
characteristics of trimethylamine N-oxide and its association with gut microbiota. J.
South. Med. Univ. 36 (4), 455–460.
Wang, J., Li, Y., Luo, B., Yue, Z., 2017. NMR-based metabonomics study on faeces of
subjects with yang-deficiency constitution. J. Beijing Univ. Tradit. Chin. Med. 40 (1),
65–69.
Wei, X., Tao, J., Xiao, S., Jiang, S., Shang, E., Zhu, Z., Qian, D., Duan, J., 2018. Xiexin
Tang improves the symptom of type 2 diabetic rats by modulation of the gut mi-
crobiota. Sci. Rep. 8 (1), 3685.
Xu, J., Bjursell, M.K., Himrod, J., Deng, S., Carmichael, L.K., Chiang, H.C., Hooper, L.V.,
Gordon, J.I., 2003. A genomic view of the human-Bacteroides thetaiotaomicron
symbiosis. Science 299 (5615), 2074–2076.
Xu, J., Lian, F., Zhao, L., Zhao, Y., Chen, X., Zhang, X., Guo, Y., Zhang, C., Zhou, Q., Xue,
Z., Pang, X., Zhao, L., Tong, X., 2015. Structural modulation of gut microbiota during
alleviation of type 2 diabetes with a Chinese herbal formula. ISME J. 9 (3), 552–562.
Yang, Y., Li, Z., Sun, J., 2009. Gene chip study on cerebral gene of effect of Jinkui
Shenqiwan and Youguiwan on mouse model of kidney-yang asthenia with syndrome
disproved according to therapeutic efficacy of drugs used. China J. Chin. Mater. Med.
34 (9), 1124–1128.
Yu, M., Jia, H.M., Zhou, C., Yang, Y., Sun, L.L., Zou, Z.M., 2017. Urinary and fecal me-
tabonomics study of the protective effect of Chaihu-Shu-Gan-San on antibiotic-in-
duced gut microbiota dysbiosis in rats. Sci. Rep. 7 (1), 46551.
Zhang, L., Ye, Y., An, Y., Tian, Y., Wang, Y., Tang, H., 2011. Systems responses of rats to
aflatoxin B1 exposure revealed with metabonomic changes in multiple biological
matrices. J. Proteome Res. 10 (2), 614–623.
Zhang, L., Han, X., Li, Z., Liu, R., Xu, W., Tang, C., Wang, X., Xiao, H., 2014.
Metabolomics research on time-selected combination of Liuwei Dihuang and Jinkui
Shenqi pills in treating kidney deficiency and aging by chemometric methods.
Chemometr. Intell. Lab. Syst. 130, 50–57.
Zhang, A., Liu, Q., Zhao, H., Zhou, X., Sun, H., Nan, Y., Zou, S., Ma, C.W., Wang, X., 2016.
Phenotypic characterization of nanshi oral liquid alters metabolic signatures during
disease prevention. Sci. Rep. 6, 19333.
Zhang, L., Wang, P.E., Ying, J., Jin, X., Luo, C., Xu, T., Xu, S., Dong, R., Xiao, L., Tong, P.,
Jin, H., 2017. Yougui Pills attenuate cartilage degeneration via activation of TGF-
beta/Smad signaling in chondrocyte of osteoarthritic mouse model. Front.
Pharmacol. 8, 611.
Zhao, L., Wu, H., Qiu, M., Sun, W., Wei, R., Zheng, X., Yang, Y., Xin, X., Zou, H., Chen, T.,
Liu, J., Lu, L., Su, J., Ma, C., Zhao, A., Jia, W., 2013a. Metabolic signatures of kidney
yang deficiency syndrome and protective effects of two herbal extracts in rats using
GC/TOF MS, Evid. Based Complement. Altern. Med. 2013 (11), 540957.
Zhao, Y., Wu, J., Li, J.V., Zhou, N., Tang, H., Wang, Y., 2013b. Gut microbiota compo-
sition modifies fecal metabolic profiles in mice. J. Proteome Res. 12 (6), 2987–2999.
Zhou, X.H., Zhang, A.H., Wang, L., Tan, Y.L., Guan, Y., Han, Y., Sun, H., Wang, X.J., 2016.
Novel chinmedomics strategy for discovering effective constituents from ShenQiWan
acting on ShenYangXu syndrome. Chin. J. Nat. Med. 14 (8), 561–581.
Zierer, J., Jackson, M.A., Kastenmuller, G., Mangino, M., Long, T., Telenti, A., Mohney,
R.P., Small, K.S., Bell, J.T., Steves, C.J., Valdes, A.M., Spector, T.D., Menni, C., 2018.
The fecal metabolome as a functional readout of the gut microbiome. Nat. Genet. 50
(6), 790–795.
Zou, Z.J., Gong, M.J., Xie, Y.Y., Wang, S.M., Liang, S.W., 2012. Urinary metabonomic
study of kidney-yang deficiency syndrome induced by hydrocortisone. Chin. J. Exp.
Tradit. Med. Formulae 18 (8), 133–136.
Zou, Z.J., Xie, Y.Y., Gong, M.J., Han, B., Wang, S.M., Liang, S.W., 2013. Urine metabo-
nomic study of intervention effects of Morinda officinalis How. on ‘kidney-yang de-
ficiency syndrome’. Acta Pharm. Sin. 48 (11), 1733–1737.