2023.58 云南中医药大学 赵婷宣传画册-在线电子书制作-云展网在线书城 (2023)


Clinical efficacy of Yiqi Yangxue

formula on knee osteoarthritis and

unraveling therapeutic

mechanism through plasma

metabolites in rats

Ting Zhao1†

, Shiqi Wang1†

, Wenbin Liu2†

, Jiayan Shen1


Youwu Dai 3

, Mingqin Shi 3

, Xiaoyi Huang3

, Yuanyuan Wei 3

, Tao Li 4


Xiaoyu Zhang1

, Zhaohu Xie 3

, Na Wang5

*, Dongdong Qin3

* and

Zhaofu Li 3



The First School of Clinical Medicine, Yunnan University of Chinese Medicine, Kunming, China, 2


Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Enshi, China, 3

School of Basic Medical

Sciences, Yunnan University of Chinese Medicine, Kunming, China, 4

Qujing Hospital Affiliated to Yunnan

University of Traditional Chinese Medicine, Qujing, China, 5

Institutes of Integrative Medicine, Fudan

University, Shanghai, China

Objective: To observe the clinical efficacy and safety of Yiqi Yangxue formula

(YQYXF) on knee osteoarthritis (KOA), and to explore the underlying therapeutic

mechanism of YQYXF through endogenous differential metabolites and their

related metabolic pathways.

Methods: A total of 61 KOA patients were recruited and divided into the treatment

group (YQYXF, 30 cases) and the control group (celecoxib, Cxb, 31 cases). Effects

of these two drugs on joint pain, swelling, erythrocyte sedimentation rate (ESR)

and c-reactive protein (CRP) were observed, and their safety and adverse reactions

were investigated. In animal experiments, 63 SD rats were randomly divided into

normal control (NC) group, sham operation (sham) group, model (KOA) group,

Cxb group, as well as low-dose (YL), medium-dose (YM), and high-dose groups of

YQYXF (YH). The KOA rat model was established using a modified Hulth method.

Ultra-high-performance liquid chromatography/Q Exactive HF-X Hybrid

Quadrupole-Orbitrap Mass (UHPLC-QE-MS)-based metabolomics technology

was used to analyze the changes of metabolites in plasma samples of rats.

Comprehensive (VIP) >1 and t-test p < 0.05 conditions were used to screen

the disease biomarkers of KOA, and the underlying mechanisms of YQYXF were

explored through metabolic pathway enrichment analysis. The related markers of

YQYXF were further verified by ELISA (enzyme-linked immunosorbent assay).

Results: YQYXF can improve joint pain, swelling, range of motion, joint function,

Michel Lequesen index of severity for osteoarthritis (ISOA) score, Western Ontario

and McMaster Universities Osteoarthritis Index (WOMAC) score, ESR, and CRP. No

apparent adverse reactions were reported. In addition, YQYXF can improve

cartilage damage in KOA rats, reverse the abnormal changes of 16 different

metabolites, and exert an anti-KOA effect mainly through five metabolic

pathways. The levels of reactive oxygen species (ROS) and glutathione (GSH)

were significantly decreased after the treatment of YQYXF.



Xiao-Ling Xu,

Zhejiang Shuren University, China


Venketesh Sivaramakrishnan,

Sri Sathya Sai Institute of Higher Learning

(SSSIHL), India

Heather Walker,

The University of Sheffield,

United Kingdom

Elizabeth R. Lusczek,

University of Minnesota Twin Cities,

United States


Zhaofu Li,


Na Wang,


Dongdong Qin,


These authors have contributed equally

to this work.


This article was submitted

to Human and Medical Genomics, a

section of the journal

Frontiers in Genetics

RECEIVED 12 November 2022

ACCEPTED 23 March 2023

PUBLISHED 05 April 2023


Zhao T, Wang S, Liu W, Shen J, Dai Y,

Shi M, Huang X, Wei Y, Li T, Zhang X, Xie Z,

Wang N, Qin D and Li Z (2023), Clinical

efficacy of Yiqi Yangxue formula on knee

osteoarthritis and unraveling therapeutic

mechanism through plasma metabolites

in rats.

Front. Genet. 14:1096616.

doi: 10.3389/fgene.2023.1096616


© 2023 Zhao, Wang, Liu, Shen, Dai, Shi,

Huang, Wei, Li, Zhang, Xie, Wang, Qin and

Li. This is an open-access article

distributed under the terms of the

Creative Commons Attribution License

(CC BY). The use, distribution or

reproduction in other forums is

permitted, provided the original author(s)

and the copyright owner(s) are credited

and that the original publication in this

journal is cited, in accordance with

accepted academic practice. No use,

distribution or reproduction is permitted

which does not comply with these terms.

Frontiers in Genetics 01 frontiersin.org

TYPE Original Research

PUBLISHED 05 April 2023

DOI 10.3389/fgene.2023.1096616


Conclusion: YQYXF can significantly improve the clinical symptoms of KOA

patients without obvious adverse reactions. It mainly improved KOA through

modulating lipid metabolism-related biomarkers, reducing lipid peroxidation and

oxidative stress.


knee osteoarthritis, YQYXF, clinical efficacy, metabolome, biomarkers


Knee osteoarthritis (KOA) is the most prevalent form of arthritis

characterized by a degeneration of articular cartilage resulting in the

development of osteophytes, or bone spurs (Horecka et al., 2022).

The global prevalence of KOA is around 3.8% (Kao et al., 2022).

KOA is closely associated with age, as radiographic evidence of KOA

occurs in most people over the age of 65 years (Lespasio et al., 2017).

Furthermore, the incidence of KOA increases with a higher average

weight of the population, particularly in obese women (Zhang and

Jordan, 2010). Strenuous physical activity, especially activities

requiring kneeling, knee-bending, squatting, and prolonged

standing, as well as knee trauma and injury have also been

linked to a high prevalence of KOA (Heidari, 2011).

Although the mechanisms of degenerative changes are betterunderstood thanks to numerous biochemical and genetic studies,

drugs that can stop the degenerative cascade remain unknown. So

far, arthritis have been managed pharmacologically and nonpharmacologically, including common pharmacotherapies,

surgery, and lifestyle changes. All available forms of KOA

therapy are based on symptomatic treatment, such as pain relief

and joint function improvement (Nowaczyk et al., 2022). Pain

medications, including the most popular non-steroidal antiinflammatory drugs (NSAIDs), are the first-line treatment

(Gnylorybov et al., 2020). Surgery should be considered only in

the case of no improvement and the presence of advanced lesions

visible in imaging tests. Currently, an increasing number of studies

are being published suggesting that traditional Chinese medicine

may be as effective or even more effective than NSAIDs and result in

fewer systemic adverse effects (Glyn-Jones et al., 2015; Wang et al.,

2020). Yiqi Yangxue formula (YQYXF) is a prescribed Chinese

herbal formula for treating KOA based on the traditional Chinese

medicine theory. The YQYXF consists of astragalus, codonopsis,

tangerine peel, cohosh, bupleurum, angelica, atractylodes, cassia

twig, white peony, licorice, divaricate saposhniovia root, ligusticum

wallichii, rhizoma drynariae, epimedium, a total of 14 herbs. In our

previous study, YQYXF could inhibit the levels of matrix

metalloproteases 1 (MMP-1) and MMP-13, promoting

chondrocyte proliferation (Duan R. et al., 2020). However, the

potential mechanism of YQYXF in treating KOA is still unclear.

In the present study, we aimed to evaluate the clinical efficacy

and safety of YQYXF in patients with KOA. We observed the effects

of YQYXF and celecoxib (Cxb) on visual analogue scale (VAS) score,

swelling, range of motion (ROM) and joint function, Michel

Lequesen index of severity for osteoarthritis (ISOA) score, the

Western Ontario and McMaster Universities Osteoarthritis Index

(WOMAC), Kellgren-Lawrence score, erythrocyte sedimentation

rate (ESR) and c-reactive protein (CRP) index of KOA patients

(Kellgren and Lawrence, 1957; Brosseau et al., 2003; Zhang et al.,

2009; Anil et al., 2021). The safety and adverse reactions were

investigated by blood cell analysis (white blood cell, red blood

cell, hemoglobin, and platelet), liver function (alanine

transaminase, aspartate aminotransferase), and kidney function

(blood urea nitrogen, creatinine) (Wu et al., 2021). In addition,

we used the KOA rat model to further analyze the changes of

metabolites in rat plasma samples using metabonomics technologies

based on ultra-high-performance liquid chromatography/Q

Exactive HF-X Hybrid Quadrupole-Orbitrap Mass (UHPLC-QEMS) (Xiao et al., 2012; Wang et al., 2014). The potential mechanism

of YQYXF on endogenous differential metabolites and related

metabolic pathways was also discussed, providing evidence for

the treatment of KOA.

Materials and methods

Clinical study design

Sample source and grouping

Sixty-one KOA patients (Kellgren-Lawrence score I-III) were

recruited from Yunnan Provincial Hospital of Traditional

Chinese Medicine. The patients were divided into a treatment

group with 30 patients (YQYXF) and a control group with

31 patients (celecoxib, Cxb). All patients fulfilled the

American College of Rheumatology criteria for primary KOA,

and the subjects agreed to sign the informed clinical consent. The

age was between 38 and 70 years old—patients who discontinued

NSAIDs for 7 days or more. The exclusion criteria included

patients with other rheumatic diseases, such as rheumatoid

arthritis, Sjögren’s syndrome, and gouty arthritis. In addition,

patients with allergies or severe other systemic diseases will not

participate in this study. The design scheme of this project has

been approved by the Medical Ethics Committee of Yunnan

University of Traditional Chinese Medicine (ethics number:


Experimental drugs and treatments

YQYXF granules, provided by Jiangyin Tianjiang

Pharmaceutical Co., Ltd. YQYXF granules consist of astragalus

30 g, codonopsis 15 g, tangerine peel 10 g, cohosh 10 g,

bupleurum 10 g, angelica 20 g, cassia twig 15 g, white peony 15 g,

atractylodes 15 g, licorice 5 g, divaricate saposhniovia root 15 g,

ligusticum wallichii 15 g, rhizoma drynariae 15 g, epimedium

15 g. The Cxb capsules, provided by Ruihui Pharmaceutical Co.,

Ltd., are approved by Chinese medicine J20080059. In this study, the

treatment group was given YQYXF, one bag/time, three times a day.

The control group was given 200 mg/day of Cxb capsules once times

a day.

Frontiers in Genetics 02 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


Observation indicators and methods

All patients were evaluated using the VAS score and ISOA

score, WOMAC, Kellgren-Lawrence score, range of motion

(ROM), and joint function grades before the treatment and at

4 weeks after treatment (Kellgren and Lawrence, 1957; Brosseau

et al., 2003; Zhang et al., 2009; Anil et al., 2021). ESR and CRP

were detected before treatment and on the fourth weekend of

treatment, respectively. ESR was detected by the ESR analyzer.

C-reactive protein (CRP) was detected by immunoturbidimetry.

The Kellgren-Lawrence grading system for KOA is the most

widely used method and has become a widely accepted

method for the diagnosis of KOA, which is a grading method

for the severity of KOA. According to X-ray findings of the knee

joint, grading of the severity of KOA can be divided into grade 0

(normal knee joint), grade I, grade II, grade III, and grade IV (the

most severe KOA) (Kellgren and Lawrence, 1957; Zeng et al.,

2017; Di Martino et al., 2022). Grading of joint function can be

divided into four grades. Grade I refers to various activities that

can be done. Grade II refers to moderate limitation. Although one

or more joints are uncomfortable or have limited movement, they

can still engage in normal activities. Grade III refers to limited

actions and can only take care of themselves but cannot engage in

general activities. Grade IV refers to lying in bed or sitting in bed

and you cannot take care of yourself (Zheng, 2002). Joint pain

(VAS of patient and physician) was scored using a 10-cm visual

analog scale (VAS), and the patients were instructed to mark the

corresponding position on the VAS that represented their pain

(Abbott et al., 2013; Su et al., 2018). 0 cm: no pain; 1–3 cm: mild

pain, but still able to engage in normal activities; 4–6 cm:

moderate pain, affecting work, but able to take care of

themselves; 7–9 cm: severe pain, unable to take care of

themselves; 10 cm: extreme pain. The efficacy evaluation, joint

function, swelling, and range of motion in this study are

determined according to the guiding principles for clinical

research of new drugs of traditional Chinese medicine (Zheng,

2002). Highly effective: pain and swelling disappear, joint activity

is normal, and the score decreases by ≥ 95%. Moderately effective:

pain and swelling disappear, joint activity is not limited, and

score decreases ≥70% and <95%. Lowly effective: pain and

swelling symptoms are basically eliminated, joint activity is

slightly limited, and the score decreased by ≥ 30% and <70%.

Ineffective: pain, swelling, and joint range of motion did not

improve significantly, and the score decreased by < 30%.

Animal experimental design

Preparation of experimental drugs

The 14 herbs in YQYXF are provided by the Chinese Pharmacy

of the First Affiliated Hospital of Yunnan University of Traditional

Chinese Medicine, and their composition and dosage are the same as

those of YQYXF granules. The high dose group of YQYXF was

administered by gavage with an aqueous solution containing 18.4 g

of crude drug/kg, the medium dose group was administered with

9.2 g of crude drug/kg, and the low dose group was administered

with 4.6 g of crude drug/kg of rats (respectively equivalent to 0.25,

0.5, and 1 time of the human clinical equivalent dose according to

the body surface area dose conversion method of humans and rats

Meeh-Rubner formula). In the positive drug group, 18 mg Cxb/kg

aqueous solution was administered by gavage. Cxb capsules,

provided by Ruihui Pharmaceutical Co., Ltd., approved by

Chinese medicine J20080059.

Identification of the compounds in YQYXF by


Compounds in YQYXF were analyzed using a UHPLC system

(Vanquish, Thermo Fisher Scientific) equipped with a waters

UPLC BEH C18 column (1.7 μm 2.1 100 mm), and the flow rate

was set to 0.5 mL/min, and an injection volume was set to 5 μL.

Mobile phase A consisted of 0.1% formic acid solution, and

mobile phase B was 0.1% formic acid in acetonitrile. The

multi-step linear elution gradient program was as follows:

0–11 min, 85%–25% A; 11–12 min, 25%–2% A; 12–14 min,

2%–2% A; 14–14.1 min, 2%–85% A. During each acquisition

cycle, the mass range was from 100 to 1,500, the top four of

every process were screened, and the corresponding MS/MS data

were further acquired. Sheath gas flow rate: 35 Arb, Aux gas flow

rate: 15 Arb, Ion Transfer Tube Temp: 350°

C, Vaporizer Temp:


C, Full ms resolution: 60,000, MS/MS resolution: 15,000,

Collision energy: 16/32/48 in NCE mode, Spray Voltage: 5.5 kV

(positive) or −4 kV (negative). An Orbitrap Exploris 120 mass

spectrometer coupled with Xcalibur software was employed to

obtain the MS and MS/MS data information of YQYXF based on

the IDA acquisition mode. The raw data of mass spectra were

imported into XCMS software for processing, such as retention

time correction, peak identification, peak extraction, peak

integration, and peak alignment. The peak information of

compounds was searched through the in-house secondary

mass spectrometry database provided by Shanghai BIOTREE

Biotech Co., Ltd.

Administration of the KOA rat model

Sixty-three SPF-grade female SD rats (180 ± 20 g) were

purchased from Hunan Slike Jingda Laboratory Animal Co.,

Ltd., license number: SCXK (Xiang) 2019–0004. The

experiments were conducted under full authorization from the

Ethics Committee of Yunnan University of Chinese Medicine

(ethical code no. R-062019065). The rats were randomly divided

into seven groups, including normal control (NC), shamoperated (sham), model (KOA), Cxb (18 mg/kg), YQYXF low-,

middle-, and high-dose groups (18.4, 9.2, and 4.6 g of crude

drug/kg, respectively). Each group had nine rats. The KOA

models were established by the modified Hulth method

(Rogart et al., 1999). The anterior ligament was severed, and

the medial meniscus was removed with the tibial joint reduction

(Gu et al., 2021). The knee in the sham group was only treated

with joint capsule opening and suturing. The operation was

conducted in an aseptic environment. After the successful

modeling, the NC group and the sham group were provided

with normal drinking water and diet. The rest were given the

corresponding drug suspension by gavage once a day for eight

consecutive weeks. The plasma was collected by centrifugation,

and the right knee joint was taken. The knee joints were fixed

with 10% neutral formaldehyde solution and decalcified with

10% EDTA for 56 days. The tissue was embedded in paraffin and

sliced with a thickness of 5 μm.

Frontiers in Genetics 03 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


Hematoxylin-eosin (HE) staining

Routine dewaxing, rinsed with tap water, stained with

hematoxylin solution for 5 min, dehydrated in acid water and

ammonia, rinsed with tap water, dehydrated and stained with

eosin for 3 min. Dehydrated from low to high concentrations of

alcohol, clear, sealed with neutral glue.

Metabolites extraction and UHPLC-QE-MS analysis

Add 400 μL of extract (methanol: acetonitrile = 1:1 (V/V),

containing isotope-labeled internal standard mixture) to 100 μL

of plasma and vortex to mix for 30 s. Sonicate for 10 min and let

stand at −40°

C for 1 h, centrifuge at 12,000 rpm for 15 min at 4°


and collect the supernatant for assay. The quality control (QC)

sample was prepared by mixing an equal aliquot of the supernatants

from all plasma samples. UHPLC-QE-MS analyses were performed

using a UHPLC system (Vanquish, Thermo Fisher Scientific) with a

UPLC BEH Amide column (2.1 mm × 100 mm, 1.7 μm) coupled to

Q Exactive HFX mass spectrometer (Orbitrap MS, Thermo). The

mobile phase consisted of 25 mmol/L ammonium acetate and

25 mmol/L ammonia hydroxide in water (A) and acetonitrile (B).

The auto-sampler temperature was 4°

C, and the injection volume

was 2 μL. The QE HFX mass spectrometer was used for its ability to

acquire MS/MS spectra in information-dependent acquisition mode

in the control of the acquisition software (Xcalibur, Thermo). In this

mode, the acquisition software continuously evaluated the full scan

MS spectrum. The ESI source conditions were set as follows: sheath

gas flow rate as 30 Arb, Aux gas flow rate as 25 Arb, capillary

temperature as 350°

C, full MS resolution as 60,000, MS/MS

resolution as 7,500, collision energy as 10/30/60 in NCE mode,

spray Voltage as 3.6 kV (positive) or −3.2 kV (negative),

respectively. The raw data were converted to the mzXML format

using ProteoWizard and processed with an in-house program,

which was developed using R and based on XCMS, for peak

detection, extraction, alignment, and integration. Then an inhouse MS2 database (BiotreeDB) was applied in metabolite

annotation. The cutoff for annotation was set at 0.3. The Thermo

Q Exactive HFX mass spectrometer is capable of primary and

secondary mass spectral data acquisition under the control of the

acquisition software (Xcalibur, Thermo).

Principal components analysis (PCA) and orthogonal

correction partial least squares discriminant analysis (OPLSDA) were conducted using the SIMCA16.0.2 software package

(Sartorius Stedim Data Analytics AB, Umea, Sweden). PCA, an

unsupervised analysis that reduces the dimension of the data, was

carried out to visualize the distribution and the grouping of the

samples. A 95% confidence interval in the PCA score plot was

used as the threshold to identify potential outliers in the dataset.

In order to visualize group separation and find significantly

changed metabolites, supervised orthogonal projections to

latent structures-discriminate analysis (OPLS-DA) were

applied. Then, 7-fold cross-validation was performed to

examine the quality of the model. Permutation tests were used

to test the validity of the model. The first principal component of

variable importance in the projection (VIP) and Student’s t-test

were obtained to refine the analysis. Suppose VIP>1 and p < 0.05,

the variable was defined as a significantly different metabolite

between the two groups. The significantly different metabolites

were used for plotting hierarchical clustering based on the

Euclidean distance formula and drawn heat maps using the

Pheatmap package in R studio. The volcano plots were used to

filter the metabolites of interest based on Log 2 (fold change)

and–Log 10 (p-value). The Kyoto Encyclopedia of Genes and

Genomes (KEGG, http://www.genome.jp/kegg/) and

MetaboAnalyst (http://www.metaboanalyst.ca/) databases were

used for pathway enrichment analysis.

The determination of the content of lipid

peroxidation-related indexes

The plasma of the rats was collected, and the levels of lipid

peroxidation-related indexes, such as reactive oxygen species (ROS)

and glutathione (GSH), were tested by enzyme-linked

immunosorbent assay (ELISA). The ELISA kit (Jiangsu Jingmei

Biological Technology Co., Ltd., China) was used according to the

manufacturer’s instructions.

Statistical method

The data were processed and analyzed by using SPSS

26.0 software. If the data followed the normal distribution, they

were presented as mean ± SD (standard deviation). Data were

compared for differences using two independent samples t-tests

or one-way ANOVA. If the data were not normally distributed, they

were presented as median (IQR, interquartile range), and a nonparametric test was adopted, and p < 0.05 was considered

statistically significant.


Clinical experiments

Participant’s characteristics

As illustrated in Table 1, the ages of the treatment group and

control group were 57.30 ± 7.82 years and 56.94 ± 7.95 years,

respectively. There was no significant difference in age between

the two groups (p = 0.97). Among the treatment group, 4 males

(13.33%) and 26 females (86.67%). While in the control group,

5 were males (16.13%) and 26 were females (83.87%). There was

no significant difference in gender between the two groups (p =

0.76). The disease duration of the treatment and control groups

were 46.33 ± 21.34 months and 45.29 ± 20.92 months,

respectively. There was no significant difference in disease

duration between the two groups (p = 0.84). In grading of

severity of KOA assessed by Kellgren-Lawrence score grading,

6 cases were grade I, 20 were grade II, and 4 were grade III among

the treatment group. While, in the control group, 7 cases were

grade I, 19 were grade II, and 5 were grade III. There was no

significant difference in severity of KOA between the two groups

(p = 0.99). The joint function grades in the treatment group

included 10 cases of grade I, 18 cases of grade II, and 2 cases of

grade III. In comparison, the control group had 11 cases of grade

I, 18 cases of grade II, and 2 cases of grade III. There was also no

significant difference in joint function between the two groups

(p = 0.87). Therefore, the age, gender, disease duration, severity of

arthritis, and joint function of the two groups of patients are

comparable, and follow-up research can be carried out.

Frontiers in Genetics 04 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


Efficacy evaluation

The clinical observation results showed that 3 (10%) cases were

highly effective, 17 (56.67%) cases were moderately effective, 9

(30%) cases were lowly effective, and 1 (3.33%) case was

ineffective, with a total effective rate of 96.67% in the treatment

group. In the control group, 2 (6.45%) cases were highly effective, 15

(48.39%) cases were moderately effective, 12 (38.71%) cases were

lowly effective, and 2 (6.45%) cases were ineffective, and the total

effective rate was 93.55% (all p > 0.05) (Figure 1A). The VAS score of

patients in the treatment group was 4.23 ± 1.19 before

administration and 1.46 ± 0.93 after 4 weeks of administration

(p = 2.81 × 10−14). The VAS score of patients in the control

group was 4.35 ± 1.20 before administration and 1.86 ±

0.83 after 4 weeks of administration (p = 1.14 × 10−13). In the

treatment group, the VAS score of physician was 4.20 ±

0.92 before administration and 1.73 ± 0.79 after 4 weeks of

administration (Figure 1B, p = 5.44 × 10−16). In the control

group, the VAS score of physician was 4.10 ± 1.25 before

administration, and the VAS score of physician was 1.97 ±

0.66 after 4 weeks of administration (Figure 1B, p = 1.07 × 10−11).

In the treatment group, the medium of swelling score was

2.00 before administration and 0.00 after 4 weeks of

administration (Figure 1C, p = 1.80 × 10−5

). The swelling score

of the control group was 2.00 before administration and 0 after

4 weeks of administration (Figure 1C, p = 1.10 × 10−5

). In the

treatment group, the joint range of motion (ROM) was 2.00 before

administration and 0.00 after 4 weeks of administration (Figure 1D,

p = 9.40 × 10−7

). The ROM in the control group was 2.00 before

administration and 0.00 after 4 weeks of administration (Figure 1D,

p = 0.002). In the treatment group, ISOA score was 9.00 ±

1.80 before administration and 3.07 ± 1.36 after 4 weeks of

administration (Figure 1E, p = 8.49 × 10−21). The ISOA in the

control group was 8.74 ± 1.63 before administration and 4.10 ±

1.33 after 4 weeks of administration (Figure 1E, p = 4.78 × 10−18). In

the treatment group, the WOMAC score was 52.27 ± 6.74 before

administration and 16.13 ± 3.85 after 4 weeks of administration

(Figure 1F, p = 3.24 × 10−33). The WOMAC score of the control

group was 53.13 ± 6.38 before administration and 19.52 ± 3.71 after

4 weeks of administration (Figure 1F, p = 9.10 × 10−34). In the

treatment group, the ESR was 20.67 ± 8.21 (mm/h) before

administration and 10.03 ± 4.80 (mm/h) after 4 weeks of

administration (Figure 1G, p = 7.21 × 10−18). The ESR of the

control group was 20.90 ± 8.94 (mm/h) before administration

and 10.61 ± 3.07 (mm/h) after 4 weeks of administration

(Figure 1G, p = 1.90 × 10−7

). In the treatment group, the CRP

was 8.56 ± 3.05 (mg/L) before administration and 2.06 ± 1.15 (mg/L)

after 4 weeks of administration (Figure 1H, p = 3.12 × 10−15). The

CRP of the control group was 8.80 ± 2.81 (mg/L) before

administration and 2.09 ± 1.06 (mg/L) after 4 weeks of

administration (Figure 1H, p = 2.71 × 10−17). After drug

intervention, the joint function grades in the treatment group

included 22 cases of grade I, 8 cases of grade II, and 0 cases of

grade III, while the control group had 19 cases of grade I, 12 cases of

grade II, and 0 cases of grade III and no significant difference was

observed (p = 0.97).

Safety evaluation

Compared with before treatment, there was no significant

difference in safety indicators such as blood cell analysis (white

blood cell, red blood cell, hemoglobin, platelet), liver function

(alanine transaminase, aspartate aminotransferase), renal function

(blood urea nitrogen, creatinine), and electrocardiogram after

4 weeks of treatment (all p > 0.05) (Table 2). Our research has

proved that YQYXF is safe for KOA patients.

Animal experiment

Screening active components of YQYXF with


The compounds in YQYXF were identified by the UHPLC-QEMS method. A total of 447 compounds were characterized in

TABLE 1 Comparison of baseline characteristics between treatment group and control group.

Treatment group (n = 30) Control group (n = 31) p-value

Age, yrs, mean ± SD 57.30 ± 7.82 56.94 ± 7.95 0.97

Female sex, n (%) 26 (86.67) 26 (83.87) 0.76

Disease duration, months, mean ± SD 46.33 ± 21.34 45.29 ± 20.92 0.84

Grading of severity of KOA

Ⅰ, n (%) 6 (20.00) 7 (22.58) 0.99

Ⅱ, n (%) 20 (66.67) 19 (61.29)

Ⅲ, n (%) 4 (13.33) 5 (16.13)

Grading of joint function

Ⅰ, n (%) 10 (33.33) 11 (35.48) 0.87

Ⅱ, n (%) 18 (60.00) 18 (58.06)

Ⅲ, n (%) 2 (0.07) 2 (6.45)

Note: there were no significant differences in baseline characteristics between therapy group and control group.

Frontiers in Genetics 05 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


YQYXF (Figure 2A: positive ion modes; Figure 2B, negative ion

modes). Among these, 17 kinds of components were detected,

including flavonoids, phenols, terpenoids, amino acid derivatives,

phenylpropanoids, alkaloids, aromaticity aliphatic acyl, xanthones,

jasmonic acid, organic acids and derivatives, fatty acids, prenol

lipids, lipoic acids and derivatives, carboxylic acids and

derivatives, carbohydrates and derivatives, alkaloids, and

quinones. The peak area represents the relative abundance of the

substance, and ppm is the deviation between the measured mz value

of the substance and the theoretical mz value. ppm = (measured mz

value - theoretical mz value) × 1,000,000 ÷ theoretical mz value. The

leading substances were defined as the top 10 substances identified

by UHPLC-QE-MS analysis. As in Table 3, the top ten substances

were ranked from highest to lowest according to their respective

peak area. The larger the peak area, the higher the ranking.

Effects of YQYXF in KOA rats

The time flow chart of the experiment is shown in Figure 3A.

The body weight of the rats in each group showed a steadily

increasing trend (Figure 3B). The overall growth trend of the

body weight of the KOA group was lower than that of the NC

group. The weight gain trend of Cxb and YQYXF groups with


Clinical efficacy assessment. (A) Disease effectiveness comparison. (B) Joint pain VAS score. (C) Joint swelling score. (D) Joint range of motion

(ROM) score. (E) Michel Lequesen index of severity for osteoarthritis (ISOA) score. (F) WOMAC score. (G) Erythrocyte sedimentation rate (ESR). (H)

C-reactive protein (CRP). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Frontiers in Genetics 06 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


different doses was higher than that of KOA group, but there was no

significant difference in the body weight (p > 0.05). The point

indicated by the arrow was the cartilage surface (Figure 3C). The

NC and sham operation groups’ cartilage surfaces were smooth and

flat, without cracks and defects. Chondrocytes were orderly

arranged, clearly stratified, and evenly distributed, without

obvious cell clusters. In the model group, the cartilage surface

ulcer became thinner, and the local cartilage calcification layer

was ruptured and disappeared. Chondrocytes were disordered.

The stratification was not easy to recognize, and regional cell

clusters were apparent. The Cxb group’s cartilage surface was

smooth, without obvious cracks and defects, and the

chondrocytes were arranged in order with a regular hierarchical

structure. The cartilage surface was rough in the YQYXF low-dose

(YL) group, and some cartilage tissues were defective. The cartilage

surface was not smooth in the YQYXF medium (YM) dose

group. Some cartilages were defective or cracked, and the

chondrocytes were complex and disordered. The hierarchical

structure was obvious, and local cell clusters appeared. The

staining matrix was uniform. The cartilage surface in the YQYXF

high-dose (YH) group was smooth, without apparent defects

and tear.

Effects of YQYXF on potential biomarkers in KOA


To further explore the effect of YQYXF on the endogenous

differential metabolites and their related metabolic pathways of

KOA, we used the plasma of rats in the NC group, KOA group,

and YH group for metabolomic analysis. We included QC samples

throughout the experimental process to ensure the stability and

reliability of the data and the system. The QC samples (yellow dots)

were closely clustered together, indicating that the UHPLC-QE-MS

system had good stability and was reliable for metabolomic analysis

of the samples. LC-MS data obtained from the plasma samples were

analyzed using PCA for metabolic changes between the NC, KOA,

YH, and QC samples (Figure 4A: positive ion modes; Figure 4B,

negative ion modes). In both the positive and negative ion modes, we

found a large deviation in 1 rat in the NC group, and no meaningful

results could be drawn. Therefore, we excluded it from further

analysis. The contribution ratio of principal component 1 (PC1) was

39.4%, and that of PC2 was 15.4% (Figure 4A: positive ion modes).

The contribution ratio of principal component 1 (PC1) was 22.2%,

and that of PC2 was 11.2% (Figure 4B, negative ion modes). In the

positive ion mode, the separation of KOA and NC samples was

insignificant. It is worth noting that the KOA samples were

significantly separated from the NC samples in the negative

mode plot, indicating significant metabolic differences between

the two groups. Meanwhile, YH and OA samples were separated,

but the separations were not significant in both positive and negative

ion modes. Due to the complex multidimensional characteristics of

metabolic data, unsupervised PCA model analysis alone could not

well distinguish group differences among samples. The OPLS-DA

model was employed to characterize the differential metabolites

among NC, KOA, and YH groups to further identify the differences

in the composition of the metabolites. R2 indicated how well the

variation of a variable was explained, and Q2 meant how well a

variable could be predicted. The replacement test established the

corresponding OPLS-DA model to obtain the R2 and Q2 values of the

random model by randomly changing the order of the classification

variable Y and repeating it for several times (n = 200), which played

TABLE 2 Safety index evaluation.

Project Group Before treatment After 4 weeks of treatment p-value

WBC (×109

/L) Treatment group 6.03 ± 1.17 5.58 ± 1.04 0.12

Control group 5.10 ± 0.63 5.14 ± 0.51 0.77

RBC (×109

/L) Treatment group 4.65 ± 0.28 4.70 ± 0.43 0.59

Control group 4.63 ± 0.52 4.84 ± 0.51 0.88

Hb (g/L) Treatment group 141.17 ± 7.30 141.47 ± 8.51 0.88

Control group 139.23 ± 10.70 143.94 ± 7.75 0.12

PLT (×109

/L) Treatment group 225.00 ± 41.55 217.40 ± 57.96 0.56

Control group 216.06 ± 43.89 228.45 ± 54.87 0.34

ALT (U/L) Treatment group 18.80 ± 7.83 16.87 ± 6.97 0.32

Control group 16.71 ± 7.92 19.26 ± 6.40 0.17

AST (U/L) Treatment group 21.57 ± 5.34 19.17 ± 5.36 0.09

Control group 18.71 ± 5.81 19.97 ± 4.18 0.33

BUN (mmol/L) Treatment group 4.80 ± 1.46 4.72 ± 1.22 0.82

Control group 5.00 ± 1.54 4.47 ± 1.02 0.12

Cr (umol/L) Treatment group 67.20 ± 11.74 69.10 ± 14.08 0.57

Control group 67.00 ± 12.84 67.58 ± 11.10 0.85

WBC, white blood cell; RBC, red blood cell; Hb, hemoglobin; PLT, platelet; ALT, alanine transaminase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; Cr, creatinine, all p > 0.05.

Frontiers in Genetics 07 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616



The total ion chromatograms (TIC) of YQYXF that were obtained in positive ion mode and negative ion mode. (A) UHPLC-QE-MS analysis base peak

intensity chromatograms of YQYXF in positive ion mode. (B) UHPLC-QE-MS analysis base peak intensity chromatograms of YQYXF in negative ion mode.

TABLE 3 Identification of components of YQYXF by UHPLC-QE-MS analysis (top 10).

Name Composite score Rtmed (s) Mzmed ppm Formula Peak area

3,5,7,8-tetramethoxy-2-(3,4,5-trimethoxyphenyl)chromen-4-one 0.78 399.66 433.15 1.04 C22H24O9 3.17×109

2″-O-beta-L-galactopyranosylorientin 0.63 88.34 609.15 2.69 C27H30O16 1.66×109

Herbacetin 1.00 132.90 303.05 1.25 C15H10O7 1.40×109

Isoliquiritigenin 0.99 101.62 257.08 3.29 C15H12O4 1.32×109

Biochanin-7-O-glucoside 0.90 98.83 447.13 0.00 C22H22O10 1.25×109

Liquiritin 0.72 102.92 417.12 3.15 C21H22O9 1.06×109

4-Methoxysalicylic acid 0.93 63.18 167.04 0.40 C8H8O4 9.81×108

Formononetin-7-O-glucoside 0.87 190.81 431.13 0.96 C22H22O9 9.71×108

Licoricesaponin H2 0.90 352.46 823.41 2.05 C42H62O16 8.85×108

Naringin 0.61 130.20 581.18 4.13 C27H32O14 8.51×108

Frontiers in Genetics 08 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


an essential role in avoiding the over-fitting of the test model and

evaluating the robustness of the model.

The OPLS-DA score plots showed an obvious separation

between the KOA and NC group in the positive and negative ion

mode from the metabolic profiles (Figure 5, A1: positive ion, A2:

negative ion). The permutation test results R2

Y and Q2 between the

KOA and NC groups were respectively 0.98 and 0.40 in the positive

ion mode and 0.99 and 0.58 in negative ion mode (Figure 5, B1:

positive ion, B2: negative ion). The YH and KOA groups OPLS-DA

scores plots were also well separated in the positive and negative ion

mode (Figure 5, C1: positive ion, C2: negative ion); R2

Y and Q2 were

respectively, 0.99 and 0.72 in the positive ion mode and 0.99 and

0.68 in the negative ion mode (Figure 5, D1: positive ion, D2:

negative ion). All R2

Y were very close to 1, indicating that the

established model conformed to the real situation of the sample data.

The intercept between the regression line of Q2 and the longitudinal

axis was less than zero. Meanwhile, with the gradual reduction of

displacement retention, the proportion of the Y variable of

displacement increased, and the Q2 of random model gradually

decreased. It showed that the model in this study had good

robustness, and there was no over-fitting phenomenon.

Therefore, the OPLS-DA results further confirmed the successful

establishment of our OA rat model and that YH administration

could regulate the metabolic profile of the rat.


Effects of YQYXF in OA rats. (A) Time flow chart. (B) Effects of YQYXF on body weight. (C) Cartilage tissue HE staining. NC: normal control group,

sham: sham operation group, OA: model group, Cxb: celecoxib, YL: low-dose groups of YQYXF, YM: medium-dose groups of YQYXF, YH: high-dose

groups of YQYXF.


PCA score plots of the QC, NC, KOA, and YH groups. (A) Positive ion modes. (B) Negative ion modes. NC group (n = 5), KOA group (n = 6), YH group

(n = 6).

Frontiers in Genetics 09 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


To investigate the contribution of potential biomarkers between

the two groups, a volcano plot was drawn, followed by Student’s

t-test. Volcano plot of comparison groups: KOA vs. NC (Figure 6,

A1: positive ion; A2: negative ion); YH vs. KOA (Figure 6, B1:

positive ion; B2: negative ion). It summarized the contribution of

each variable to the model. The metabolites with VIP>1 and p <

0.05 were considered as significantly changed metabolites. In total,

66 metabolites were identified in the heatmap of hierarchical

clustering analysis between the KOA and NC groups, including

29 and 37 metabolites in the positive and negative ion modes,

respectively. Between the YH and KOA groups, 81 metabolites

were identified, including 57 and 24 metabolites in the positive

and negative ion modes, respectively. Heatmap of comparison

groups: KOA vs. NC (Figure 6, C1: positive ion, C2: negative

ion); YH vs. KOA (Figure 6, D1: positive ion, D2: negative ion).

Taking the intersection of differential metabolites between the KOA

vs. NC group and the YH vs. KOA group, there were 16 potential

biomarkers, including 8 lipids and lipid-like molecules, 2 organic

acids and derivatives, 2 phenylpropanoids and polyketides,

2 oganoheterocyclic compounds, 2 oganic acids and derivatives

(Table 4).

In order to more intuitively express the intervention effect of

YQYXF components on the screened metabolites, PCA analysis was

performed on the 16 screened potential biomarkers (Figure 7A). The

intra-group aggregation and inter-group dispersion were obvious in

the NC and the KOA group, indicating that the plasma metabolites

in the KOA rats were distinguished from the NC. The KOA group

was significantly separated from the NC group and YH group, while

the NC group and YH group were significantly aggregated,

indicating that YH can regulate KOA-related metabolites.

Further, we mapped the differential metabolites to authoritative

metabolite databases. After obtaining the matching information of

the differential metabolites, we searched and analyzed the metabolic

pathway of the pathway database of Rattus norvegicus (rat). The

metabolic pathway enrichment analysis of differential metabolites

showed that YQYXF mainly exerted its anti-inflammatory effect by

regulating five pathways: linoleic acid metabolism, α-linolenic acid

metabolism, sphingolipid metabolism, arachidonic acid metabolism,


OPLS-DA analysis of serum of mice. OPLS-DA scores plots: KOA vs. NC (A1: positive ion, A2: negative ion), YH vs. KOA (C1: positive ion, C2: negative

ion). Permutation test of OPLS-DA model: KOA vs. NC (B1: positive ion, B2: negative ion); YH vs. KOA (D1: positive ion, D2: negative ion).

Frontiers in Genetics 10 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


and glycerophospholipid metabolism (Figure 7B). In addition, we

performed correlation analysis on 16 biomarkers. Internal

interaction and crosstalk networks are established between

phenylpropanoic acids and organoheterocyclic compounds, fatty

acyls and organic sulfonic acids, carboxylic acids and diazinanes,

fatty acyls and organonitrogen compounds (Figure 7C).

YQYXF modulates oxidative stress and lipid

peroxidation capacity in KOA rats

The results of metabonomics experiments showed that YQYXF

mainly regulated the signal pathway related to lipid metabolism.

Therefore, it may play the role of anti-KOA cartilage destruction by

repairing lipid metabolism disorder. We further explored the

changes of oxidative stress and lipid peroxidation-related factors

in KOA rats. The GSH of NC, sham, OA, YL, YM, YH, and Cxb

group was 235.69 ± 33.23, 250.43 ± 24.85, 467.97 ± 29.25, 397.37 ±

22.89, 374.49 ± 29.29, 319.31 ± 18.88, 284.02 ± 28.56 mmol/L,

respectively. The ROS of NC, sham, OA, YL, YM, YH, and Cxb

group was 290.41 ± 50.77, 318.59 ± 31.99, 681.62 ± 25.38, 540.76 ±

48.31, 495.07 ± 47.84, 433.13 ± 51.34, 359.10 ± 30.95 pg/mL,

respectively. Compared with the NC and sham group, the

expression of GSH (p = 1.59 × 10−19 for the KOA group vs. the

NC group, p = 1.67 × 10−18 for the KOA group vs. the sham group)

and ROS (p = 9.23 × 10−21 for the KOA vs. the NC group, p = 1.43 ×


Multivariate statistical analysis of metabolite profiles in plasma. Volcano plot of comparison groups: KOA vs. NC (A1: positive ion, A2: negative ion);

YH vs. KOA (C1: positive ion, C2: negative ion). Heatmap of comparison groups: KOA vs. NC (C1: positive ion, C2: negative ion); YH vs. KOA (D1: positive

ion, D2: negative ion). Screening of differential metabolites by metabolomic analysis. Significantly upregulated metabolites are shown in red, significantly

downregulated metabolites in blue, and non-significantly different metabolites in grey.

Frontiers in Genetics 11 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


TABLE 4 Identified Potential Biomarkers Regulated by Yiqi Yangxue formula (YQYXF).

Name MS2 score Rt




vs. NC)



vs. NC)




vs. NC)







vs. KOA




vs. KOA)



vs. NC)










vs. NC




1,3-Dihydro-(2H)-indol-2-one 0.99 28.61 134.06 1.61 0.002 0.32 1.30 0.03 2.16 0.83 0.86 63.53 + ↓ ↑

Oleamide 0.98 82.95 282.28 2.38 0.01 3.60 1.72 0.02 0.50 1.00 0.92 71.07 + ↑ ↓

Sphingosine 0.97 82.95 300.29 2.40 0.01 3.71 1.71 0.02 0.48 1.00 0.92 73.62 + ↑ ↓

DL-2-Aminooctanoic acid 0.96 243.66 160.13 2.39 0.001 3.20 2.17 0.01 0.32 1.00 1.00 66.90 + ↑ ↓

LysoPE (18:1 (9Z)/0:0) 0.93 220.69 480.31 1.74 0.03 0.67 1.53 0.03 1.41 0.75 0.92 29.06 + ↓ ↑

7-Ketocholesterol 0.90 31.98 401.34 2.44 0.03 0.19 2.23 0.01 4.62 1.00 1.00 88.44 + ↓ ↑




0.78 152.05 856.58 1.91 0.01 0.65 1.52 0.04 1.33 0.89 0.83 49.78 + ↓ ↑

Frontiers in Genetics 1

Zhao et al.



PC(22:4 (7Z,10Z,13Z,16Z)/14:0) 0.68 159.91 782.57 1.74 0.04 0.76 1.4 0.04 1.23 0.92 0.86 58.83 + ↓ ↑

L-Cyclo (alanylglycyl) 0.62 379.50 129.07 2.02 0.004 0.50 1.76 0.01 2.03 0.97 0.89 60.36 + ↓ ↑

PC(16:0/14:0) 0.61 169.58 706.54 1.86 0.02 0.64 1.59 0.02 1.37 0.89 0.89 27.85 + ↓ ↑




0.56 148.69 878.57 2.07 0.01 0.41 1.63 0.04 1.62 0.94 0.81 48.93 + ↓ ↑

2-Hydroxyethanesulfonate 0.98 160.46 124.99 1.82 0.05 2.29 1.74 0.04 0.40 0.81 0.92 64.01 - ↑ ↓

Phenyllactic acid 0.97 125.42 165.05 2.40 0.03 0.13 2.27 0.003 3.57 0.86 1.00 95.37 - ↓ ↑

Citraconic acid 0.96 453.31 129.02 1.79 0.03 0.70 1.54 0.04 1.41 0.81 0.83 28.08 - ↓ ↑

(9xi,10xi,12xi)-9,10-Dihydroxy12-octadecenoic acid 0.93 178.74 313.24 1.87 0.02 0.39 1.87 0.04 7.68 0.78 0.94 125.36 - ↓ ↑

4-Hydroxycinnamic acid 0.68 58.39 163.04 2.00 0.01 0.66 1.87 0.005 1.41 0.89 0.92 22.82 - ↓ ↑

frontiersin.org 2




Metabolic profiling of 16 potential biomarkers. (A) Principal components analysis score plot of 16 differential metabolites. (B) Metabolic pathway

bubble plot of 16 differential metabolites. (C) Heatmap of correlation analysis. The horizontal and vertical coordinates in the figure represent the

contrasting differential metabolites. Red represents a positive correlation, blue represents a negative correlation, and the darker the color, the stronger

the correlation. Significant correlations are marked with an asterisk (*).


The effect of YQYXF on anti-oxidative stress and lipid peroxidation in OA rats. (A) Levels of ROS in plasma. (B) Levels of GSH in plasma. All data were

expressed as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Frontiers in Genetics 13 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


10−19 for the KOA group vs. the sham group) were significantly

increased (Figures 8A, B). The GSH of drug-treated groups was

lower than that of the KOA group (p = 1.60 × 10−5 for the lowYQYXF-dose group, p = 8.6826 × 10−8 for the middle-YQYXF-dose

group, p = 5.00 × 10−13 for the high-YQYXF-dose group, and p =

5.50 × 10−16 for the Cxb group) and the ROS of drug-treated groups

was lower than that of the KOA group (p = 1.76 × 10−7 for the lowYQYXF-dose group, p = 2.34 × 10−10 for the middle-YQYXF-dose

group, p = 5.74 × 10−14 for the high-YQYXF-dose group, and p =

9.69 × 10−18 for the Cxb group), suggesting that the YQYXF could

regulate the plasma lipid metabolism disorder in the KOA rats.

Hence, YQYXF may exert anti-KOA cartilage degeneration by

regulating lipid peroxidation.


Clinical research results showed that YQYXF could improve the

clinical symptoms of KOA patients without noticeable adverse

reactions. In animal experiments, the 66 potential disease

biomarkers of KOA and 81 related metabolites of YQYXF were

screened out. After comparative analysis, there were 16 potential

biomarkers in the intervention of YQYXF in KOA rats, and exert

anti-KOA through five metabolic pathways, including sphingolipid

metabolism, glycerophospholipid metabolism, linoleic acid

metabolism, α-linolenic acid metabolism, and arachidonic acid


Sphingolipids display important functions in various

pathologies such as obesity, diabetes, and OA (El Jamal et al.,

2020). Sphingosine 1-phosphate (S1P) is a metabolite of cell

membrane sphingolipids enriched in circulating fluid. It binds to

G protein-coupled S1P receptors to regulate embryonic

development and organ function. S1P binding triggers multiple

cellular and physiological events, including the localization of

immune cells to sites of inflammation and regulation of T-cell

differentiation (Th17 and Treg cells) (Robinson et al., 2022). The

balance between the levels of S1P and sphingosine has been

considered as a switch that could determine whether a cell

proliferates or dies (El Jamal et al., 2020). Masuko found that

S1P may play a unique role in the pathophysiology of KOA by

regulating VEGF expression in chondrocytes (Masuko et al., 2012).

A study found that the activity of sphingosine kinase 1 increased

with osteoclast differentiation, and its expression was enhanced in

the subchondral bone of mice with KOA (Cherifi et al., 2021). The

lipid mediator S1P was identified as a therapeutic target for KOA

(Stradner et al., 2013; Ustyol et al., 2017). In this study, potential

biomarkers were significantly enriched in the sphingolipid

metabolism pathway, suggesting that YQYXF may reduce KOA

cartilage damage and improve lipid metabolism mainly by

regulating sphingolipid metabolism changes.

Glycerophospholipids form the essential lipid bilayer of all

biological membranes and are intimately involved in signal

transduction, regulation of membrane trafficking, and many

other membrane-related phenomena (Farooqui et al., 2000). The

alterations in phospholipid composition and concentrations are

associated with the development of KOA (Kosinska et al., 2014;

Zhang W. et al., 2014). A study found activation of

glycerophospholipid metabolism and oxidative stress pathways in

synovial fluid metabolism in patients with KOA (Carlson et al.,

2019). In this study, YQYXF may improve lipid metabolism by

affecting the level of glycerophospholipids, alleviating the

progression of KOA.

In addition, linoleic acid, α-linolenic acid, and arachidonic acid

all belong to polyunsaturated fatty acids (PUFAs). While all PUFAs

reduced markers of oxidative stress, omega-3 PUFAs additionally

decreased prostaglandin production (Loef et al., 2019). The omega-3

PUFAs have been shown to decrease markers of inflammation and

cartilage degradation. Oxidative stress can be directly assessed by

measuring ROS. Known ROS include superoxide, hydrogen

peroxide, peroxyl radicals, and reactive nitrogen species

(including nitric oxide and peroxynitrite derived from the nitric

oxide) (Duan L. et al., 2020). Previous studies have found that

excessive ROS generated by lipid metabolism disorders can induce

chondrocyte apoptosis (Poulet and Staines, 2016). Excessive

accumulation of ROS can cause chondrocyte damage and

cartilage matrix degradation, promoting the occurrence of KOA

(Bolduc et al., 2019). ROS can also destroy proteoglycans and type II

collagen in the cartilage matrix by activating matrix

metalloproteinases, inhibiting matrix synthesis, and leading to

loss of cartilage integrity (Mehana et al., 2019). GSH can exert a

destructive effect on ROS through an enzymatic mechanism,

reducing the level of ROS or inhibiting its activity. Imbalanced

ROS/GSH may result from a direct increase of ROS, consumption of

GSH, intracellular oxidoreductase interference, or thioredoxin

activity reduction (Liu et al., 2022). GSH and ROS in the model

group were significantly increased, suggesting an imbalance between

ROS production and elimination. Furthermore, the intervention of

different doses of YQYXF may activate the feedback regulation

mechanism, promote the reduction of ROS level, and then lead to

the corresponding decrease of GSH, which can alleviate the

imbalance. Thus, YQYXF may reduce ROS production by

balancing lipid metabolism disorders and inhibiting KOA

TABLE 5 The relevance of metabolic pathways to symptoms and pathology.

Pathway Symptoms and pathology

Linoleic acid metabolism Regulate inflammatory reactions and pain, and maintain the stability of blood glucose and blood fat levels

Alpha-Linolenic acid metabolism Regulate lipid metabolism and inflammatory reactions

Sphingolipid metabolism Regulate inflammatory reactions and pain

Arachidonic acid metabolism Participate in immune and inflammatory reactions

Glycerophospholipid metabolism Regulate lipid metabolism

Frontiers in Genetics 14 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


cartilage destruction. However, further experimental verification is

still needed.

A study used liquid chromatography/mass spectrometry (LC/

MS)-based metabolomics to explore the serum metabolomics in rats

with KOA. The six biomarkers were identified, which were

metabolized through tryptophan metabolism, glutamate

metabolism, nitrogen metabolism, spermidine metabolism, and

fatty acid metabolism pathways (Zhao et al., 2021). Another

study used ultra-high performance liquid chromatography-triple

quadrupole mass spectrometry (UPLC-TQ-MS), followed by

multivariate statistical analysis, to determine the serum amino

acid profiles of KOA patients and healthy controls. The

metabolic pathways with the most significant effects were

involved in the metabolism of alanine, aspartate, glutamate,

arginine, and proline (Chen et al., 2018). In our study, the

plasma metabolites of KOA and control group participated in

alanine, aspartate and glucose metabolism, pyrimidine

metabolism, and biosynthesis of unsaturated fatty acids. The

results are consistent with the findings of the above studies. In

addition, the fasting serum of KOA patients and healthy controls

was assessed by metabolomic analysis (Tootsi et al., 2020). The

changes in the serum levels of amino acids, sphingomyelins,

phoshatidylcholines and lysophosphatidylcholines of the KOA

patients compared with healthy controls suggest systemic

inflammation in severe KOA patients. In our study, YQYXF is

mainly involved in lipid metabolism pathways, such as sphingolipids

metabolism and glycerol phospholipids metabolism, which indicates

that YQYXF may play a role in treating systemic inflammation

in KOA.

For note, the flavocoxid is a medical food consisting of plantderived flavonoids which have anti-inflammatory activity and are

used to treat chronic KOA. Studies have shown that flavocoxid was

as effective as naproxen in managing the signs and symptoms of

KOA (Levy et al., 2010). In our study, 99 flavonoids in YQYXF were

identified, including naringin, baicalin, icariin, and quercetin.

Naringin, a natural flavanone found in citrus fruits, and its

aglycone have been demonstrated to ameliorate obesity and

dyslipidemia. The principal mechanisms by which these

flavonoids exert their action involve upregulation of peroxisome

proliferator activated receptor α and adenosine 5′-monophosphateactivated protein kinase, and the downregulation of genes involved

in lipid metabolism (Massaro et al., 2022). Naringin is an effective

therapeutic drug for the treatment of KOA and KOA-related

symptoms, which can support the recovery of hind-limb weightbearing (Xu et al., 2017). Naringin can prevent cartilage destruction

in KOA by inhibiting the nuclear factor kappa-B (NF-κB) signaling

pathway, which reduces Tumor necrosis factor-α (TNF-α)-mediated

chondrocyte inflammation and cartilage matrix degradation (Zhao

et al., 2016). Baicalein can ameliorate inflammatory-related

apoptotic and catabolic phenotypes in human chondrocytes

(Zhang X. et al., 2014). In addition, the anti-inflammatory and

anti-apoptotic effects of baicalein are mediated by inhibiting the

translocation of phosphorylated p65 to the nucleus (Li et al., 2017).

Icariin has been shown to stimulate osteogenic differentiation and

bone formation and to increase the synthesis of the cartilage

extracellular matrix (Zhao et al., 2019). A study has

demonstrated that the IKBKB, NFKBIA, MAPK8, MAPK9, and

MAPK10 may be the hub genes affected by icariin when providing

its beneficial effects on KOA. In addition, icariin can alleviate KOA

by inhibiting NOD-like receptor thermal protein domain associated

protein 3-mediated pyroptosis (Zu et al., 2019). Quercetin may be

related to the inhibition of interleukin-1β (IL-1β) and TNF-α

production via the Toll-like receptor 4/NF-κB pathway in KOA

rats (Zhang et al., 2020). The use of quercetin partially abrogated

intestinal flora disorder and reversed fecal metabolite abnormalities

(Lan et al., 2021). In addition, 32 phenolic compounds were found in

YQYXF, such as paeonol, 4-methylatechol, and thymol. Paeonol, as

an essential component in traditional Chinese medicine, has antiinflammatory activity and can offer therapy for a multitude of

inflammatory-related diseases. Studies have shown that applying

paeonol can attenuate the secretion of cartilage extracellular matrix

and cartilage degrading enzymes induced by IL-1β in chondrocytes

(Liu et al., 2017). Besides, paeonol can also alleviate destabilization of

the medial meniscus-induced articular cartilage degeneration in vivo

(Liu et al., 2017). There are few studies on phenolic compounds in

OA, which need further research and verification.

For note, the metabolic pathways likely contribute to symptoms

and pathology in KOA. Some studies demonstrated that metabolites

in the synovial fluid and blood could be used as biomarkers for KOA

incidence, prognosis, and response to therapy (Rockel and Kapoor,

2018). Various metabolites can directly influence the perception of

pain. Secreted phospholipase A2 catalyzes the conversion of

phosphatidylcholine (PC) analogues to lysoPC analogues.

Subsequent metabolism of lysoPCs via autotaxin generates

lysophosphatidic acid, an inflammatory and pain-producing

signal. Endocannabinoids, endogenously produced lipid-derived

compounds, could be beneficial for individuals with KOA to

reduce pain symptoms. In turn, the endocannabinoids may result

in the production of lysoPCs, which could promote joint pain due to

metabolism of lysophosphatidic acid (Muccioli, 2010). In addition,

the oxidized linoleic acid metabolite and partial TRPV1 agonist 9-

hydroxyoctadecandienoic acid was shown to be involved in chronic

inflammatory pain (Wedel et al., 2022). The linoleic acid can

attenuate inflammatory responses and reduce LPS-induced

phosphorylation of proteins associated with NF-κB signaling

(Kim et al., 2020). In addition, the dysregulation of sphingolipid

metabolism contributes to neuropathic pain (Stockstill et al., 2018)

(Table 5).

The limitation of this study is that the components of traditional

Chinese medicine are complex, with the characteristics of multiple

targets and pathways. Although the components of YQYXF were

identified, the analysis of members entering the blood and the study

of pharmacokinetics still need to be clarified.

In conclusion, YQYXF could be an effective and promising

agent for treating KOA, which might exert its action by regulating

multiple lipid metabolism-related pathways. Our study provides

new insights into studying the underlying molecular mechanism of

YQYXF against oxidative stress in the KOA model. However,

further research exploration is needed.

Data availability statement

The original contributions presented in the study are publicly

available. This data can be found here: https://www.ebi.ac.uk/

metabolights/. Accession number: MTBLS5667.

Frontiers in Genetics 15 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


Ethics statement

The studies involving human participants were reviewed and

approved by the Medical Ethics Committee of Yunnan University of

Traditional Chinese Medicine (ethics number: 2019YXLL005).

Written informed consent to participate in this study was

provided by the participants’legal guardian/next of kin. The

animal study was reviewed and approved by the Ethics

Committee of Yunnan University of Chinese Medicine (ethical

code no. R-062019065). Written informed consent was obtained

from the individual(s) for the publication of any potentially

identifiable images or data included in this article.

Author contributions

All authors listed have made a substantial, direct, and intellectual

contribution to the work and approved it for publication.


National Natural Science Foundation of China grant Nos.

(81960870, 31960178, 82160923, 81560781, and 81760822);

Yunnan Provincial Ten Thousands Program Famous Doctor

Special; Yunnan Province Qingguo Wang Expert Workstation

Construction Project (202005AF150017); Yunnan Applied Basic

Research Projects-Yunnan University of Chinese Medicine Union

Foundation (202101AZ070001-247); Yunnan Applied Basic

Research Projects-Union Foundation [2019FF002(-031)]; Applied

Basic Research Programs of Science and Technology Commission

Foundation of Yunnan Province (2019FA007); Key Laboratory of

Traditional Chinese Medicine for Prevention and Treatment of

Neuropsychiatric Diseases, Yunnan Provincial Department of

Education; Scientific Research Projects for High-level Talents of

Yunnan University of Chinese Medicine (2019YZG01); Young TopNotch Talent in 10,000 Talent Program of Yunnan Province

(YNWR-QNBJ-2019-235); National Science and Technology

Innovation 2030 Major Program (2021ZD0200900); Yunnan Key

Research and Development Program (202103AC100005); Yunnan

Province Fabao Gao Expert Workstation Construction Project

(202105AF150037); Yunnan Engineering Research Center of

Drug Development for Bone Diseases. Scientific Research Fund

Project of Yunnan Provincial Department of Education

(2021Y461 and 2022Y348).


We would like to thank Shanghai Biotree Biomedical

Biotechnology Co., Ltd., for assistance in UHPLC-QE-MS analysis.

Conflict of interest

The authors declare that the research was conducted in the

absence of any commercial or financial relationships that could be

construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors

and do not necessarily represent those of their affiliated

organizations, or those of the publisher, the editors and the

reviewers. Any product that may be evaluated in this article, or

claim that may be made by its manufacturer, is not guaranteed or

endorsed by the publisher.


Abbott, J. H., Robertson, M. C., Chapple, C., Pinto, D., Wright, A. A., Leon de la

Barra, S., et al. (2013). Manual therapy, exercise therapy, or both, in addition to

usual care, for osteoarthritis of the hip or knee: A randomized controlled trial. 1:

Clinical effectiveness. Osteoarthr. Cartil. 21, 525–534. doi:10.1016/j.joca.2012.


Anil, U., Markus, D. H., Hurley, E. T., Manjunath, A. K., Alaia, M. J., Campbell, K. A.,

et al. (2021). The efficacy of intra-articular injections in the treatment of knee

osteoarthritis: A network meta-analysis of randomized controlled trials. Knee 32,

173–182. doi:10.1016/j.knee.2021.08.008

Bolduc, J. A., Collins, J. A., and Loeser, R. F. (2019). Reactive oxygen species, aging and

articular cartilage homeostasis. Free Radic. Biol. Med. 132, 73–82. doi:10.1016/j.


Brosseau, L., Yonge, K. A., Robinson, V., Marchand, S., Judd, M., Wells, G., et al.

(2003). Thermotherapy for treatment of osteoarthritis. Cochrane Database Syst. Rev.

2003, Cd004522. doi:10.1002/14651858.Cd004522

Carlson, A. K., Rawle, R. A., Wallace, C. W., Brooks, E. G., Adams, E., Greenwood, M.

C., et al. (2019). Characterization of synovial fluid metabolomic phenotypes of cartilage

morphological changes associated with osteoarthritis. Osteoarthr. Cartil. 27, 1174–1184.


Chen, R., Han, S., Liu, X., Wang, K., Zhou, Y., Yang, C., et al. (2018). Perturbations in

amino acids and metabolic pathways in osteoarthritis patients determined by targeted

metabolomics analysis. J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1085, 54–62.


Cherifi, C., Latourte, A., Vettorazzi, S., Tuckermann, J., Provot, S., Ea, H. K., et al.

(2021). Inhibition of sphingosine 1-phosphate protects mice against chondrocyte

catabolism and osteoarthritis. Osteoarthr. Cartil. 29, 1335–1345. doi:10.1016/j.joca.


Di Martino, A., Boffa, A., Andriolo, L., Romandini, I., Altamura, S. A., Cenacchi, A.,

et al. (2022). Leukocyte-rich versus leukocyte-poor platelet-rich plasma for the

treatment of knee osteoarthritis: A double-blind randomized trial. Am. J. Sports

Med. 50, 609–617. doi:10.1177/03635465211064303

Duan, L., Duan, H., Wan, C. P., Xie, Z. H., Lv, X. M., and Li, Z. F. (2020b). Effects of

Yiqi Yangxue Decoction to protein level of β-catenin, WISP-1 in IL-1β induced

Osteoarthritis chondrocyte model. Lishizhen Med. Materia Medica Reseach 31,

2576–2579. doi:10.3969/j.issn.1008-0805.2020.11.005

Duan, R., Xie, H., and Liu, Z. Z. (2020a). The role of autophagy in osteoarthritis.

Front. Cell. Dev. Biol. 8, 608388. doi:10.3389/fcell.2020.608388

El Jamal, A., Bougault, C., Mebarek, S., Magne, D., Cuvillier, O., and Brizuela, L.

(2020). The role of sphingosine 1-phosphate metabolism in bone and joint pathologies

and ectopic calcification. Bone 130, 115087. doi:10.1016/j.bone.2019.115087

Farooqui, A. A., Horrocks, L. A., and Farooqui, T. (2000). Glycerophospholipids in brain:

Their metabolism, incorporation into membranes, functions, and involvement in

neurological disorders. Chem. Phys. Lipids 106, 1–29. doi:10.1016/s0009-3084(00)00128-6

Glyn-Jones, S., Palmer, A. J., Agricola, R., Price, A. J., Vincent, T. L., Weinans, H., et al.

(2015). Osteoarthritis. Lancet 386, 376–387. doi:10.1016/s0140-6736(14)60802-3

Gnylorybov, A. M., Ter-Vartanian, S. K., Golovach, I. Y., Vyrva, O. E., Burianov, O.

A., Yesirkepova, G. S., et al. (2020). Expert opinion on the extensive use of prescription

crystalline glucosamine sulfate in the multimodal treatment of osteoarthritis in Ukraine,

Kazakhstan, Uzbekistan, and Armenia. Clin. Med. Insights Arthritis Musculoskelet.

Disord. 13, 1179544120946743. doi:10.1177/1179544120946743

Gu, W., Shi, Z., Song, G., and Zhang, H. (2021). MicroRNA-199-3p up-regulation

enhances chondrocyte proliferation and inhibits apoptosis in knee osteoarthritis via

DNMT3A repression. Inflamm. Res. 70, 171–182. doi:10.1007/s00011-020-01430-1

Frontiers in Genetics 16 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616


Heidari, B. (2011). Knee osteoarthritis prevalence, risk factors, pathogenesis and

features: Part I. Casp. J. Intern Med. 2, 205–212. PMID: 24024017.

Horecka, A., Hordyjewska, A., Blicharski, T., and Kurzepa, J. (2022). Osteoarthritis of

the knee - biochemical aspect of applied therapies: A review. Bosn. J. Basic Med. Sci. 22,

488–498. doi:10.17305/bjbms.2021.6489

Kao, W. C., Chen, J. C., Liu, P. C., Lu, C. C., Lin, S. Y., Chuang, S. C., et al. (2022). The

role of autophagy in osteoarthritic cartilage. Biomolecules 12, 1357. doi:10.3390/


Kellgren, J. H., and Lawrence, J. S. (1957). Radiological assessment of osteo-arthrosis.

Ann. Rheum. Dis. 16, 494–502. doi:10.1136/ard.16.4.494

Kim, S. M., Park, E. J., Kim, J. Y., Choi, J., and Lee, H. J. (2020). Anti-inflammatory

effects of fermented Lotus root and linoleic acid in lipopolysaccharide-induced RAW

264.7 cells. Life (Basel) 10, 293. doi:10.3390/life10110293

Kosinska, M. K., Liebisch, G., Lochnit, G., Wilhelm, J., Klein, H., Kaesser, U., et al.

(2014). Sphingolipids in human synovial fluid--a lipidomic study. PLoS One 9, e91769.


Lan, H., Hong, W., Qian, D., Peng, F., Li, H., Liang, C., et al. (2021). Quercetin

modulates the gut microbiota as well as the metabolome in a rat model of osteoarthritis.

Bioengineered 12, 6240–6250. doi:10.1080/21655979.2021.1969194

Lespasio, M. J., Piuzzi, N. S., Husni, M. E., Muschler, G. F., Guarino, A., and Mont, M.

A. (2017). Knee osteoarthritis: A primer. Perm. J. 21, 16–183. doi:10.7812/tpp/16-183

Levy, R., Khokhlov, A., Kopenkin, S., Bart, B., Ermolova, T., Kantemirova, R., et al.

(2010). Efficacy and safety of flavocoxid compared with naproxen in subjects with

osteoarthritis of the knee-a subset analysis. Adv. Ther. 27, 953–962. doi:10.1007/s12325-


Li, Y., Wang, J., Song, X., Bai, H., Ma, T., Zhang, Z., et al. (2017). Effects of baicalein on

IL-1β-induced inflammation and apoptosis in rat articular chondrocytes. Oncotarget 8,

90781–90795. doi:10.18632/oncotarget.21796

Liu, M., Zhong, S., Kong, R., Shao, H., Wang, C., Piao, H., et al. (2017). Paeonol

alleviates interleukin-1β-induced inflammatory responses in chondrocytes during

osteoarthritis. Biomed. Pharmacother. 95, 914–921. doi:10.1016/j.biopha.2017.09.011

Liu, T., Sun, L., Zhang, Y., Wang, Y., and Zheng, J. (2022). Imbalanced GSH/ROS and

sequential cell death. J. Biochem. Mol. Toxicol. 36, e22942. doi:10.1002/jbt.22942

Loef, M., Schoones, J. W., Kloppenburg, M., and Ioan-Facsinay, A. (2019). Fatty acids

and osteoarthritis: Different types, different effects. Jt. Bone Spine 86, 451–458. doi:10.


Massaro, L., Raguzzini, A., Aiello, P., and Valencia, D. V. (2022). The potential role

of naringin and naringenin as nutraceuticals against metabolic syndrome. Endocr.

Metab. Immune Disord. Drug Targets 23, 428–445. doi:10.2174/


Masuko, K., Murata, M., Beppu, M., Nakamura, H., Kato, T., and Yudoh, K. (2012).

Sphingosine-1-phosphate modulates expression of vascular endothelial growth factor in

human articular chondrocytes: A possible new role in arthritis. Int. J. Rheum. Dis. 15,

366–373. doi:10.1111/j.1756-185X.2012.01756.x

Mehana, E. E., Khafaga, A. F., and El-Blehi, S. S. (2019). The role of matrix

metalloproteinases in osteoarthritis pathogenesis: An updated review. Life Sci. 234,

116786. doi:10.1016/j.lfs.2019.116786

Muccioli, G. G. (2010). Endocannabinoid biosynthesis and inactivation, from

simple to complex. Drug Discov. Today 15, 474–483. doi:10.1016/j.drudis.2010.


Nowaczyk, A., Szwedowski, D., Dallo, I., and Nowaczyk, J. (2022). Overview of firstline and second-line pharmacotherapies for osteoarthritis with special focus on intraarticular treatment. Int. J. Mol. Sci. 23, 1566. doi:10.3390/ijms23031566

Poulet, B., and Staines, K. A. (2016). New developments in osteoarthritis and cartilage

biology. Curr. Opin. Pharmacol. 28, 8–13. doi:10.1016/j.coph.2016.02.009

Robinson, G., Pineda-Torra, I., Ciurtin, C., and Jury, E. C. (2022). Lipid metabolism in

autoimmune rheumatic disease: Implications for modern and conventional therapies.

J. Clin. Invest. 132, e148552. doi:10.1172/jci148552

Rockel, J. S., and Kapoor, M. (2018). The metabolome and osteoarthritis: Possible

contributions to symptoms and pathology. Metabolites 8, 92. doi:10.3390/


Rogart, J. N., Barrach, H. J., and Chichester, C. O. (1999). Articular collagen

degradation in the Hulth-Telhag model of osteoarthritis. Osteoarthr. Cartil. 7,

539–547. doi:10.1053/joca.1999.0258

Stockstill, K., Little, J. W., Braden, K., Lauro, F., Giancotti, L. A., Harada, C. M.,

et al. (2018). Dysregulation of sphingolipid metabolism contributes to bortezomibinduced neuropathic pain. J. Exp. Med. 215, 1301–1313. doi:10.1084/jem.20170584

Stradner, M. H., Gruber, G., Angerer, H., Huber, V., Setznagl, D., Kremser, M. L., et al.

(2013). Sphingosine 1-phosphate counteracts the effects of interleukin-1β in human

chondrocytes. Arthritis Rheum. 65, 2113–2122. doi:10.1002/art.37989

Su, K., Bai, Y., Wang, J., Zhang, H., Liu, H., and Ma, S. (2018). Comparison of

hyaluronic acid and PRP intra-articular injection with combined intra-articular and

intraosseous PRP injections to treat patients with knee osteoarthritis. Clin. Rheumatol.

37, 1341–1350. doi:10.1007/s10067-018-3985-6

Tootsi, K., Vilba, K., Märtson, A., Kals, J., Paapstel, K., and Zilmer, M. (2020).

Metabolomic signature of amino acids, biogenic amines and lipids in blood serum

of patients with severe osteoarthritis. Metabolites 10, 323. doi:10.3390/


Ustyol, E. A., Karaarslan, F., Bekpinar, S., Ozkuk, K., and Erdogan, N. (2017). Effects

of balneotherapy on serum levels of shingosine-1-phosphate in patients with

osteoarthritis. Altern. Ther. Health Med. 23, 10–14. PMID: 28646803.

Wang, J., Christison, T. T., Misuno, K., Lopez, L., Huhmer, A. F., Huang, Y., et al.

(2014). Metabolomic profiling of anionic metabolites in head and neck cancer cells by

capillary ion chromatography with Orbitrap mass spectrometry. Anal. Chem. 86,

5116–5124. doi:10.1021/ac500951v

Wang, M., Liu, L., Zhang, C. S., Liao, Z., Jing, X., Fishers, M., et al. (2020). Mechanism

of traditional Chinese medicine in treating knee osteoarthritis. J. Pain Res. 13,

1421–1429. doi:10.2147/jpr.S247827

Wedel, S., Osthues, T., Zimmer, B., Angioni, C., Geisslinger, G., and Sisignano, M.

(2022). Oxidized linoleic acid metabolites maintain mechanical and thermal

hypersensitivity during sub-chronic inflammatory pain. Biochem. Pharmacol. 198,

114953. doi:10.1016/j.bcp.2022.114953

Wu, J., Li, W., Ye, B., and Yao, Y. (2021). The efficacy and safety of xianling

gubao capsules in the treatment of knee osteoarthritis: A protocol for a

randomized, double-blind, controlled trial. Med. Baltim. 100, e27086. doi:10.


Xiao, J. F., Zhou, B., and Ressom, H. W. (2012). Metabolite identification and

quantitation in LC-MS/MS-based metabolomics. Trends Anal. Chem. 32, 1–14.


Xu, Q., Zhang, Z. F., and Sun, W. X. (2017). Effect of naringin on monosodium

iodoacetate-induced osteoarthritis pain in rats. Med. Sci. Monit. 23, 3746–3751. doi:10.


Zeng, X., Ma, L., Lin, Z., Huang, W., Huang, Z., Zhang, Y., et al. (2017). Relationship

between Kellgren-Lawrence score and 3D kinematic gait analysis of patients with medial

knee osteoarthritis using a new gait system. Sci. Rep. 7, 4080. doi:10.1038/s41598-017-


Zhang, J., Yin, J., Zhao, D., Wang, C., Zhang, Y., Wang, Y., et al. (2020). Therapeutic

effect and mechanism of action of quercetin in a rat model of osteoarthritis. J. Int. Med.

Res. 48, 300060519873461. doi:10.1177/0300060519873461

Zhang, W., Doherty, M., Leeb, B. F., Alekseeva, L., Arden, N. K., Bijlsma, J. W., et al.

(2009). EULAR evidence-based recommendations for the diagnosis of hand

osteoarthritis: Report of a task force of ESCISIT. Ann. Rheum. Dis. 68, 8–17. doi:10.


Zhang, W., Likhodii, S., Zhang, Y., Aref-Eshghi, E., Harper, P. E., Randell, E., et al.

(2014a). Classification of osteoarthritis phenotypes by metabolomics analysis.

BMJ Open 4, e006286. doi:10.1136/bmjopen-2014-006286

Zhang, X., Zhu, Y., Chen, X., Zhang, Y., Zhang, Y., Jia, Y., et al. (2014b). Baicalein

ameliorates inflammatory-related apoptotic and catabolic phenotypes in human

chondrocytes. Immunopharmacol 21, 301–308. doi:10.1016/j.intimp.2014.05.006

Zhang, Y., and Jordan, J. M. (2010). Epidemiology of osteoarthritis. Clin. Geriatr.

Med. 26, 355–369. doi:10.1016/j.cger.2010.03.001

Zhao, C. F., Li, Z. H., Li, S. J., Li, J. A., Hou, T. T., and Wang, Y. (2019). PLGA scaffold

carrying icariin to inhibit the progression of osteoarthritis in rabbits. R. Soc. Open Sci. 6,

181877. doi:10.1098/rsos.181877

Zhao, J., Liu, M., Shi, T., Gao, M., Lv, Y., Zhao, Y., et al. (2021). Analysis of serum

metabolomics in rats with osteoarthritis by mass spectrometry. Molecules 26, 7181.


Zhao, Y., Li, Z., Wang, W., Zhang, H., Chen, J., Su, P., et al. (2016). Naringin protects

against cartilage destruction in osteoarthritis through repression of NF-κB signaling

pathway. Inflammation 39, 385–392. doi:10.1007/s10753-015-0260-8

Zheng, X. Y. (2002). Guiding principles for clinical research of new drugs of traditional

Chinese medicine. Beijing: China Pharmaceutical Science and Technology Press.

Zu, Y., Mu, Y., Li, Q., Zhang, S. T., and Yan, H. J. (2019). Icariin alleviates

osteoarthritis by inhibiting NLRP3-mediated pyroptosis. J. Orthop. Surg. Res. 14,

307. doi:10.1186/s13018-019-1307-6

Frontiers in Genetics 17 frontiersin.org

Zhao et al. 10.3389/fgene.2023.1096616

Top Articles
Latest Posts
Article information

Author: Van Hayes

Last Updated: 09/23/2023

Views: 6488

Rating: 4.6 / 5 (46 voted)

Reviews: 93% of readers found this page helpful

Author information

Name: Van Hayes

Birthday: 1994-06-07

Address: 2004 Kling Rapid, New Destiny, MT 64658-2367

Phone: +512425013758

Job: National Farming Director

Hobby: Reading, Polo, Genealogy, amateur radio, Scouting, Stand-up comedy, Cryptography

Introduction: My name is Van Hayes, I am a thankful, friendly, smiling, calm, powerful, fine, enthusiastic person who loves writing and wants to share my knowledge and understanding with you.