Acupuncture Regulates Serum Differentially Expressed Proteins in Patients with Chronic Atrophic Gastritis: A Quantitative iTRAQ Proteomics Study

Objective To identify differentially expressed proteins (DEPs) in sera of patients with chronic atrophic gastritis (CAG) using isobaric tags for relative and absolute quantitation (iTRAQ) and to explore acupuncture's mechanism in CAG. Methods Peripheral sera from 8 healthy volunteers (HC), 8 chronic nonatrophic gastritis (NAG) patients, 8 CAG patients, and 8 CAG patients who underwent acupuncture treatment (CAG + ACU) were collected followed by labeling with iTRAQ reagent for protein identification and quantification using two-dimensional liquid chromatography-tandem mass spectrometry (2D-LC-MS/MS). Representative DEPs were selected through bioinformatics, and proteins were verified by enzyme-linked immunosorbent assay (ELISA). Results A total of 4,448 unique peptides were identified, corresponding to 816 nonredundant proteins. A 1.4-fold difference was used as the threshold. Compared with the HC group, 75 and 106 DEPs were identified from CAG and NAG groups, respectively. Compared with the CAG group, 110 and 66 DEPs were identified from the NAG and CAG + ACU groups, respectively. The DEPs were mainly involved in protein binding and the Notch signaling pathway-related proteins, and the upregulated proteins included actin-binding proteins (thymosin beta-4, tropomyosin-4, profilin-1, transgelin-2), while the downregulated proteins included Notch2 and Notch3. After acupuncture, the expression of these proteins in CAG patients was less differentiated from that in healthy people. The level of the above 6 proteins were verified by ELISA, and the results were similar to the results of iTRAQ analysis. Conclusions Actin-binding proteins and Notch signaling pathway-related proteins were correlated with the development and progression of CAG and thus are potential diagnostic markers for CAG. Acupuncture may play a role in regulating actin-binding proteins and Notch signaling pathway-related proteins to play a therapeutic role in CAG.


Introduction
Chronic gastritis is characterized as either chronic atrophic gastritis (CAG) or nonatrophic gastritis (NAG), of which CAG includes gland atrophy and intestinal metaplasia and is recognized as a precancerous lesion [1,2]. Correa's cascade summarizes the evolution of chronic gastritis as NAG ⟶ CAG ⟶ gastric intestinal metaplasia of the gastric mucosa ⟶ gastric mucosal dysplasia ⟶ gastric cancer [3]. Global Cancer Statistics in 2018 showed that gastric cancer was the fifth most common malignant tumor and the third leading cause of cancer death globally [4]. Hence, as CAG is a precancerous lesion, its early diagnosis and treatment are particularly important to prevent the occurrence of gastric cancer.
In clinical practice, most CAG patients have no obvious symptoms. When symptoms are present, they manifest primarily as dyspepsia, epigastric pain, abdominal distension, and other nonspecific symptoms. e diagnosis of CAG mainly relies on gastroscopy and gastric mucosal biopsy [1]. However, for various reasons, gastroscopy is not a popular procedure for CAG diagnosis in many countries. In recent years, some researchers have proposed a combined analysis of serum gastrin-1, pepsinogen I, pepsinogen II, and their ratios for CAG diagnosis [5], but studies have shown that serum gastrin-17 levels are not significantly correlated with the degree of antral atrophy and cannot be used as the gold standard of CAG diagnosis and treatment [6]. In addition, another therapeutic regimen for Helicobacter pylori (HP)positive CAG involves using triple therapy (i.e., proton pump inhibitor or bismuth plus 2 antibiotics) to eradicate HP in order to improve the clinical symptoms of CAG; in addition, gastric mucosal-protective agents, digestive enzyme inhibitors, and gastrointestinal motility drugs can be used according to the symptoms.1 However, some studies have shown that long-term application of proton pump inhibitors in HP-infected subjects causes an increased risk of gastric cancer and is related to drug dosage and time factors [7]. A growing number of studies have shown that acupuncture treatment holistically improves the clinical symptoms of CAG patients with low cost and no toxic side effects [8][9][10]. In recent years, research on the mechanism of acupuncture and moxibustion in CAG treatment has mainly focused on immune function, mucosal protection, gastrointestinal hormones, and metabolic changes [11][12][13]. However, the effect of acupuncture on differentially expressed proteins (DEPs) in the serum of CAG patients remains unknown.
Isobaric tags for relative and absolute quantitation (iTRAQ) is a high-throughput screening technology that has been extensively used in quantitative proteomics in recent years. Its characteristics include high sensitivity, strong separation ability, and high throughput. Furthermore, iTRAQ has often been used to screen for and identify biomarkers [14]. Recent studies have shown that iTRAQ can identify DEPs in gastric cancer, such as new molecular biomarkers (e.g., glutaminase 1, c-glutamyl cyclotransferase, calcium-binding proteins, and granulin) [15][16][17]. Nevertheless, there are few studies on the pathogenesis of CAG that examine DEPs. us, screening for serum markers of CAG may provide vital knowledge about CAG pathogenesis and the prevention of gastric cancer.
Our previous clinical studies have shown that acupuncture, one of the therapies in traditional Chinese medicine, significantly improves clinical symptoms in CAG patients, such as stomachache, abdominal distension, anxiety, and depression, with a total effective rate of 84.38%, improving their quality of life [18,19]. On this basis, we chose 8 effective patients in acupuncture treatments using iTRAQ and liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) to conduct relative quantitative analysis of proteins to determine the targets of acupuncture treatment in CAG. is study also used functional annotation, pathway enrichment, and other bioinformatic analysis methods to explore the pathogenesis of CAG and the related mechanism of acupuncture treatment.

Research Subjects.
is study included 24 research subjects, consisting of healthy volunteers (n � 8, A group, labeled 114, 18-70 years old), NAG patients (n � 8, B group, labeled 115, 18-70 years old), CAG patients (n � 8, C group, labeled 116, 18-70 years old), and CAG patients who underwent acupuncture treatment (n � 8, D group, labeled 117, 18-70 years old). Healthy volunteers were those with normal gastric mucosa on gastroscopy, while the other patients had mild and moderate CAG or NAG and were treated at the medical outpatient clinic of the Shanghai Research Institute of Acupuncture and Meridian, Shanghai, China, from May 2016 to June 2019. e diagnostic criteria were in accordance with the Consensus on Chronic Gastritis in China issued by the Chinese Society of Gastroenterology [20]. e mucosa specimens of gastric were taken under gastroscopy, and the degree of atrophy/intestinal metaplasia were evaluated by the OLGA (Operative Link for Gastritis Assessment) staging system [21].
Inclusion criteria were (1) age between 18 and 70 years, no gender limitation; (2) the pathological diagnosis was consistent with chronic gastritis on gastroscopy; and (3) understand and agree to participate in this study and sign the informed consent. Exclusion criteria were (1) individuals that fell outside the age range; (2) patients diagnosed with schizophrenia, major depression, and other psychiatric disorders; (3) patients previous diagnosis of gastric cancer, massive gastrointestinal bleeding and gastric perforation; (4) patients with hypertension, diabetes, heart disease, renal dysfunction, liver disease, or malignant diseases; (5) pregnant women, breastfeeding women, and women planning to be pregnant; (6) patients who had participated in other clinical trials within 1 month; and (7) the selected acupoint area has skin diseases, limb deformity, and other conditions that cannot be treated by acupuncture.
is study was approved by the Ethics Committees of the Shanghai Yueyang Integrated Traditional Chinese Medicine and Medicine Hospital affiliated to the Shanghai University of Traditional Chinese Medicine (Approval number: 2016-028) and has been registered in the Chinese Clinical Trial Registration Center (http://www.chictr.org.cn) (registration number: ChiCTR1900026044). All participants signed the informed consent and were diagnosed via gastroscopy and histopathological examinations of gastric mucosa biopsy specimens. Basic clinical parameters of healthy volunteers, NAG patients, and CAG patients before treatment are displayed in Table 1. 8 CAG patients were treated with acupuncture at the selected acupoints: Zhongwan (RN12), Neiguan (PC6, bilateral), and Zusanli (ST36, bilateral), according to the standards for acupoint selection established by the World Health Organization, the Global Standard Acupuncture Point Locations [22]. e needles used in this study were 0.3 × 40 mm disposable sterile acupuncture needles (Hwato; Suzhou Medical Appliance Factory, Jiangsu Province, China). e acupuncture depth was 15-20 mm, and the acupuncture technique was straight, flattening, and replenishing. e needles remained inserted in the patients for 20 min after deqi (feeling the sensation of soreness, numbness, distension, and heaviness). Acupuncture was performed 3 times per week, 10 times per treatment course, for a total of 6 courses by the same acupuncturist. Clinical features of CAG patients before and after acupuncture treatment along with clinical outcomes are displayed in Table 2.

Specimen Collection.
Morning fast peripheral blood (10 mL) from each group was collected after enrolling, and 10 mL blood from the D group was collected after 6 courses' acupuncture treatment. e samples were kept at room temperature for 30 min before centrifugation at 3,000 rpm for 15 min in a 4°C centrifuge (Fresco17; ermo Fisher Scientific, Waltham, MA, USA) to collect supernatants that were aliquoted in cryopreservation tubes, labeled, and stored in a −80°C freezer for later experiments.

Removing High-Abundance Proteins.
After thawing the frozen serum samples on ice, 50 μL of each serum sample was removed, and samples from the same group were pooled before centrifugation at 13,200 rpm for 10 min to collect and transfer the middle layer of serum into a new tube. e samples were sorted and labeled independently as A (HC), B (NAG), C (CAG), or D (CAG + ACU) group according to the sample grouping information. First, the samples were subjected to the Agilent-Human 14 multiple affinity removal system (MARS human-14; Agilent technologies, Santa Clara, CA, USA) to remove 14 high-abundance serum proteins, including albumin, immunoglobulin (Ig)G, antitrypsin, IgA, transferrin, haptoglobin, fibrinogen, α2-macroglobulin, α1acid glycoprotein, IgM, apolipoprotein AI, apolipoprotein AII, complement protein C3, and trans-thyroglobulin, followed by quantifying the protein concentrations in each group using a bicinchoninic acid (BCA) assay kit (Beyotime, Shanghai, China) and subsequently adjusting the concentration to 2 mg/mL in each group.

Filter-Aided Sample Preparation (FASP) for Whole
Protein Cleavage. A protein sample (100 µL) from each group was added with dithiothreitol and incubated in a 56°C water bath for 1 h followed by adding 100 µL iodoacetamide for further incubation at room temperature in the dark for 40 min. After reductive alkylation, the samples were transferred to 10 K ultrafiltration tubes and centrifuged at 12,000 rpm for 10 min to discard the bottom solution of the collection tube. Each supernatant sample was added to 100 µL lysis buffer (triethylammonium bicarbonate) in an ultrafiltration tube. Each sample was washed 4 times followed by adding 40 µL trypsin to the ultrafiltration tube for enzymatic hydrolysis in a 37°C incubator overnight. On the next day, the collection tubes were renewed and centrifuged at 12,000 rpm for 10 min. Each sample was diluted with lysis buffer to 100 µL. Enzymatic hydrolysates (5 µL for each sample) were collected for LC-MS analysis to calculate the number of missed pancreatin sites.

iTRAQ Labeling.
e iTRAQ Reagent-Multiplex Buffer Kit (ABSCIEX, Darmstadt, Germany) was equilibrated at room temperature for 20 min followed by centrifuging the reagents at 12,000 rpm for 10 min. Next, 50 µL isopropanol was added and mixed well using a vortex mixer. e iTRAQ reagent was added into a 50 µL sample from each group in a new centrifuge tube and labeled. e reaction was stopped after 3 h of incubation at room temperature (22 ± 2°C). e labeled samples were mixed and freeze-dried.

Offline High-Performance Liquid Chromatography (HPLC) Preclassification.
Each freeze-dried sample was reconstituted with 30 μL of 20 mM ammonium formate, vortexed, and centrifuged at 12,000 rpm for 10 min. Subsequently, 25 μL of supernatant was collected and subjected to high-pH reversed-phase chromatography for preclassification in the HPLC system (EASY-nLC 1000; ermo Fisher Scientific). A total of 10 tubes of sample solution were obtained after screening and were centrifuged and freeze-dried.

LC-MS/MS Analysis.
e separated and lyophilized peptide samples were redissolved in Buffer A (acetonitrile: water:formic acid � 2 : 98 : 0.1), dissolved on a vortex mixer, moved to a sample bottle, and subsequently subjected to MS analysis. Online Nano-RPLC was performed via the Easy-nLC 1000 system ( ermo Fisher Scientific). e dissolved sample was loaded onto a trap column (PepMap100, C18 3 μm 75 μm × 20 mm NanoViper, ermo Fisher Dionex ion chromatography; ermo Fisher Scientific) at a flow rate of 2 µL/min followed by rinse desalination for 10 min. e analytical column was a C18 reverse-phase chromatography column (PepMap100, C18 2 μm 75 μm × 150 mm Nano-Viper, ermo Fisher Dionex ion chromatography). e gradient used in the experiment increased the mobile phase B from 5% to 35% within 70 min. e Q-Exactive MS system ( ermo Fisher Scientific) combined with a nanospray ion source ( ermo Fisher Scientific) was used. e spray voltage was 1.6 kV, and the temperature of the capillary tube was 250°C. e MS scanning mode was set as the data-dependent acquisition mode (Data Dependent Analysis, DDA). Twenty fragment maps were collected after each full scan. e full-scan resolution was 70,000, the MS/MS resolution was 17,500, the precursor ion scan range was 300-1,800 m/z, and the collision energy was 27% higher energy C-trap dissociation.

Data Processing.
e original data were converted into mgf files using the Proteome Discoverer (version 1.3; ermo Fisher Scientific) software, and the data retrieval and analysis were performed in the UniProt human protein reference database (148,986 sequences). Subsequently, the target-decoy database was used to evaluate the results. For protein identification, a mass tolerance of 20 ppm was permitted for intact peptide masses and 0.05 Da for fragmented ions. One missed cleavage was allowed in the trypsin digests. e variable modifications were oxidation (M) and iTRAQ 4plex (Y); the fixed modifications were carbamidomethyl (C), iTRAQ 4plex (N-term), and iTRAQ 4plex (K). e charge state of peptides was set to +2 and +4. e protein identification was grouped according to the peptide tags in all samples. e protein ratio was calculated according to the specific peptides, and the protein identification confidence was set to 95% (i.e., Prot Score > 1.3) to search and remove the false-positive proteins in the decoy database (FDR < 1%). e threshold was protein difference fold, which reached 1.4-fold (ratio ≥ 1.4 or ≤ 0.714). In addition, Gene Ontology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis, protein interaction network analysis, clustered heap map analysis, expression trend analysis, and similarity and difference analysis of DEPs between groups were conducted to discover the functional properties of these DEPs and their relevance to the research goal.
is analysis was used to identify the representative proteins for subsequent ELISA detection.

Validation of the Candidate Proteins.
Six representative serum DEPs, including thymosin beta-4, profilin-1, tropomyosin-4, transgelin-2, Notch2, and Notch3, were selected according to the results of bioinformatics analysis. e DEPs were quantified using ELISA kits purchased from Shanghai Lengton Bioscience (Shanghai, China). A total of 32 samples were applied to determine the concentration of these proteins in each serum of the validation set from 4 groups (8 HC, 8 NAG, 8 CAG, 8 CAG + ACU). Each sample was performed in duplicate according to the ELISA manufacturer's instructions. e enzymatic reaction substrate was added to produce a chromogenic reaction. e chromogenic differences between the experimental group and the control group were compared to calculate the content of proteins.

Statistical
Analysis. SPSS software (version 19; IBM SPSS Statistics, USA) was used for statistical analysis. Measurement data with normal distribution were shown as mean ± standard deviation, and one-way ANOVA was used.
e data with non-normal distributions were showed as median (P25, P75), and nonparametric tests were used. Fisher's exact test was used to classify the classification variables. e significance level adopted was 0.05.

Protein Identification from Plasma Samples.
After LC-MS/MS analysis, a total of 99,509 spectra were matched to 6,628 peptides (including 4,448 unique  ese identified peptides were matched to 816 proteins. Using a protein difference ratio of 1.4-fold as the threshold (ratio ≥ 1.4 or ratio ≤ 0.714) and using healthy volunteers as the HC group, a total of 178 DEPs were identified in the NAG, CAG, and CAG + ACU groups (Supplementary File 1). Table 3 and Figure 1 summarize the DEP statistics.

GO Enrichment Analysis.
e 4 groups of DEPs were classified according to the molecular function (MF), cellular component (CC), and biological process (BP) of each protein via the GO database followed by using Fisher's exact test (p < 0.05) to obtain the protein function classifications that were significantly related to CAG (Figure 2). In MF, DEPs were mainly involved in protein binding in NAG, CAG, and CAG + ACU groups. In CC, DEPs were mainly involved in extracellular exosome, and in BP, DEPs were mainly involved in platelet degranulation and innate immune response in three groups. ese results suggested that protein binding and immune system response were related to CAG.

KEGG Pathway Enrichment Analysis.
Based on the KEGG database (http://www.genome.jp/kegg/), the DEPs were enriched and analyzed. e enriched pathways are shown in Figure 3; results showed that the occurrence and progression of NAG were mainly involved in the complement system, platelet activation, and chemokine signaling pathway. In addition to the above signaling pathways, CAG also participated in actin cytoskeleton regulation and Notch signaling pathways. Acupuncture may play a therapeutic role by regulating the actin cytoskeleton and Notch signaling pathway.
us, acupuncture may exert its therapeutic role on CAG by regulating these pathways.

Clustered Heatmap Analysis.
e expression levels of DEPs in different groups were displayed in different colors in the form of Log 2 expression. As shown in Figure 5, most of the proteins in the 3 groups (NAG, CAG, and CAG + ACU) were different. In the clustered heatmap, protein expression in the NAG group is mostly shown in black and demonstrates no significant differences from the HC group. Most proteins showed upregulated expression in the CAG group (depicted in red), showing significant differences from the HC group. e expression levels of most proteins in the CAG + ACU group tended to be similar to those in the NAG group. Among them, thymosin beta-4, profilin-1, tropomyosin-4, transgelin-2, Notch2, and Notch3 showed a strong trend in three groups.
3.6. Trend Analysis. According to the above results, a trend analysis of upregulated and downregulated DEPs was performed. A total of 50 model profiles were selected to summarize the expression patterns of proteins to further study the changing DEP trends across the different groups ( Figure 6). Among the upregulated DEPs, 4 protein trends were significant (Figure 6(a)), of which the most prominent pattern was profile 39, which contained 27 proteins in total (Figure 7(b)). Among the downregulated DEPs, 4 protein trends were significant (Figure 6(b)), and the most prominent pattern was profile 46, which contained 15 proteins in total (Figure 7(a)).

ELISA Verification.
e results of bioinformatics analysis showed that actin-binding proteins (ABPs) and the Notch signaling pathway-related proteins were closely related to the occurrence and progression of CAG. Six related proteins were selected for ELISA verification in this study, including thymosin beta-4, profilin-1, tropomyosin-4, transgelin-2, Notch2, and Notch3. e trends of the 6 DEPs  Evidence-Based Complementary and Alternative Medicine verified by ELISA were largely similar to those identified via iTRAQ analysis (Table 4). e serum thymosin beta-4, profilin-1, tropomyosin-4, and transgelin-2 protein levels in the CAG group were significantly higher (p < 0.01), while the Notch2 and Notch3 protein levels were significantly lower (p < 0.05) than in the HC and NAG groups. After acupuncture treatment, the protein contents of thymosin beta-4, profilin-1, tropomyosin-4, and transgelin-2 of the CAG + ACU group were significantly reduced (p < 0.05), while the protein contents of Notch2 and Notch3 were significantly increased (p < 0.05), compared with the HC and NAG groups (Figure 8).

Discussion
CAG is the most critical precancerous precursor to gastric cancer. Its pathogenesis is not yet fully elucidated. CAG may be closely related to HP infection, eating habits, bile reflux, and age, genetic, immune, and psychological factors [1]. Our previous studies showed that acupuncture could improve         the histopathological changes of gastric mucosa, anxiety, and depression in CAG patients and enhance their living quality [18,19]. e specific mechanism of CAG's transformation into gastric cancer is not yet understood. Proteomics can be used to study the characteristics of cells, tissues, or biological proteins, such as protein expression levels and post-translational modifications. Understanding the molecular changes in the early stages of the transition from CAG to gastric cancer and is important for the prevention and treatment of gastric cancer.
is study used iTRAQ combined with 2D-LC-MS/MS technology to identify a total of 178 DEPs from the 4 groups.
e results of GO enrichment analysis showed that in CAG, the MF mainly involved protein binding, the CC mainly involved cell exosomes, and the BP mainly involved platelet degranulation. In KEGG analysis, the complement system, platelets, chemokines, and other related signaling pathways were activated in NAG, suggesting that NAG involved inflammation and mucosal trauma. In addition to the above signaling pathways, CAG also showed actin cytoskeleton regulation and Notch signaling pathway-related proteins, suggesting that CAG may involve abnormal cytoskeletal changes. PPI analysis also showed that the protein interaction network of CAG involves ABPs, which further confirmed that CAG may involve abnormal cytoskeletal changes and indicated that acupuncture exerted a therapeutic effect on CAG by regulating cytoskeleton-related proteins. Heatmap and trend analyses showed that the DEPs in the NAG group were not significant, and the DEPs in the CAG group were mostly upregulated. After acupuncture treatment, most of the proteins tended to resume normal expression, suggesting that acupuncture simultaneously regulated multiple targets to treat CAG. e trends of the 6 DEPs verified by ELISA were similar to the trends identified by iTRAQ analysis.  H7C3R1  P01860  J3KPV0  J3QS23  G1AUC5  P01880  B7Z795  Q6GMW4  Q9UL82  Q5NV90  Q8N589  Q6P089  S6BGD6  L8E853  P01859  B7ZLM6  Q7Z443  B7Z539  P04196  Q04721  Q1HP67  M0R3C9  E9PH42  H7C3N9  J3KQY4  H0Y5V5  O60747  K7EQ47  Q71MF9  H7C226  P00748  P05155  Q68CK4  Q04756  B4DW08  Q6MZQ6  P01781  S6B2A1  A2KBB9  Q0KKI6  P01605  H0YID2  E5RGE1  Q15849  M0R2W8  H0YNC7  H0YAP2  A2J1M4  P10599  A7E2F7  A2NB46  M0R1V7  Q6P2N0  L8EAW6  P61626  P01833  Q9P273  B7Z4X6  Q86XU5  B7Z8A9  H0YIN9  P10720  B4E3J7  910124  E9PNQ9  P40197  P05090  B2RBZ5  F6KPG5  H7C229  B7Z7M2  Q5TCU6  C9JL85  B4E3I0  G3V1A4  K7EIW7  Q71V99  G3V226  P07737  Q14019  B7ZKS9  P02679  D6REL8  P07996  Q86YW5  P02776  A6NML8  Q5HYB6  Q86TQ3  K7EL66  B4E190  Q5T123  P37802  P67936  A8KA84  P02671  P02647  H6VRF8  P35527  P13645  Q8WTQ7  P62328  P01009  A8K2W3  A8K7I4  Q6ZW64  Q7Z7Q0   NAG  CAG  ACU + CAG   2   1 0 -1 -2 Figure 5: Clustered heatmap analysis of the differential proteins in NAG, CAG, and CAG + ACU groups. Each column represents the protein information of one group of samples, and each row represents the relative expression level of each protein. Red represents significantly upregulated proteins, green represents significantly downregulated proteins, and black represents no significant difference in proteins.
is study showed that most of the DEPs upregulated in the sera of CAG patients were related to ABPs, such as thymosin beta-4, profilin-1, tropomyosin-4, and transgelin-2. Actin is the main component of the microfilament system and one of the main components of the cytoskeleton. It plays a vital role in many cell activities, such as muscle contraction, maintenance of cell morphology, movement, division, material transport, and signaling transduction [23]. In addition to actin, there are many other proteins involved in the structure of microfilaments and that regulate microfilament polymerization. ese proteins are collectively called ABPs [24,25]. e malignant transformation of normal epithelial cells is characterized by abnormal changes in the cytoskeleton. ABPs are important targets for regulating cytoskeletal changes [26]. Many studies have found that ABPs are involved in the formation and metastasis of a variety of tumors [27,28]. However, further studies are needed to verify whether the high levels of ABPs in the serum of CAG patients are related to gastric cancer. ymosin beta-4 is one of the main actin regulatory molecules in the human body. It plays an important role in regulating actin microfilament assembly, cell movement and migration, inflammation, damage repair, and other biological activities [29]. Its roles in cornea repair, skin healing, and tumorigenesis have aroused more attention [30][31][32][33]. Studies have shown that thymosin beta-4 promotes tumor metastasis and is highly expressed in pancreatic cancer, esophageal cancer, colon cancer, and gastric cancer [34][35][36][37]. A study by Sahin et al. [38] also showed that thymosin beta-4 was overexpressed in the gastric mucosa of patients with gastrointestinal stromal tumors (GISTs). Overexpression of thymosin beta-4 in the gastric mucosa was related to tumor progression and poor prognosis in the GIST patients. erefore, thymosin beta-4 may be a potential diagnostic marker and prognostic indicator of GIST. Profilin-1 is a kind of low-molecular-weight ABPs that play an important role in regulating the polymerization and depolymerization of actin to ensure the steadiness of the microfilament assembly process [39]. Profilin-1 and thymosin beta-4 strengthen or weaken the polymerization of actin, thereby ensuring smooth microfilament assembly [40]. Profilin-1 imbalance has been found to be related to the occurrence and progression of amyotrophic lateral sclerosis, atherosclerosis, hypertension, and tumors [41][42][43][44][45]. Abnormal expression of profilin-1 is related to gastric cancer tumor invasion, lymph node metastasis, and tumor node metastasis (TNM) staging. Silencing profilin-1 inhibits the proliferation and metastasis of gastric cancer cells by inducing cell cycle arrest [45]. Tropomyosin-4 is an important regulatory protein in cell activity and muscle contraction [46]. Tropomyosin combines with F-actin and ABPs to mediate the interaction between F-actin and ABPs.
us, it participates in cytoskeleton regulation, as well as cell migration and invasion [47]. Studies have shown that abnormal expression of tropomyosin leads to changes in the cytoskeletal structure, along with related incidences of muscle diseases and malignant tumors [48]. Tropomyosin-4 is abnormally expressed in different tumors and plays different roles in promoting or suppressing cancer [49][50][51]. As a smallmolecule ABP, transgelin-2 is widely expressed in smooth muscle cells as well as various tissues and organs in the body and is significantly expressed in organs, such as the stomach,   pancreas, and bone marrow [52]. Transgelin-2 regulates the processes involved in cell proliferation, differentiation, migration, and signal transduction by binding to actin [53] and regulates the activation, differentiation, and phagocytosis of lymphocytes and macrophages by regulating the cytoskeleton of actin, thereby regulating the balance of immunity in the body [54]. Transgelin-2 is abnormally and highly expressed in colorectal cancer, ovarian cancer, kidney cancer, gastric cancer, and other malignant tumors [55][56][57]. In addition, studies have shown that trasgenlin-2 regulates the expression levels of metastasis-related factors and the nuclear factor-kappa B signaling pathway to inhibit cell viability, thereby inhibiting the migration and invasion of cervical cancer cells [58]. At present, no relevant literature has been found on the relationship between acupuncture and ABPs. In this study, compared with the HC group, serum thymosin beta-4, profilin-1, tropomyosin-4, and transgelin-2 were significantly upregulated in CAG patients. However, no significant difference in the levels of those proteins was found between the NAG and the HC groups, suggesting that CAG may involve abnormal changes in the cytoskeleton. e combination of thymosin beta-4, profilin-1, tropomyosin-4, and transgelin-2 expression may be a potential diagnostic marker for CAG. After acupuncture treatment, the expression of above 4 proteins was significantly reduced, suggesting that acupuncture may play a therapeutic role in CAG by reducing ABPs and inhibiting the transformation of CAG into gastric cancer.
is study also showed that Notch signaling pathwayrelated proteins may be involved in the occurrence and progression of CAG. e Notch family is a highly conserved gene group that involved in cell differentiation, apoptosis, proliferation, and tissue formation [59]. Studies have shown that the Notch signaling pathway plays an important regulatory role in tumor cells and tumor microenvironment, participating in tumor angiogenesis, metastasis, and epithelial-mesenchymal transition in tumor cells, as well as the regulation of drug resistance in tumors [60][61][62][63]. Studies have shown that the Notch gene is abnormally activated in many malignant tumors [60][61][62]. However, some studies indicated that whether the Notch signaling pathway plays an oncogenic or tumor-suppressive role depends on the environment [63,64]. For example, a study by Hassan et al. [65] showed that Notch3 was overexpressed in non-small-cell lung cancer. However, the expression of Notch3 in small-cell lung cancer was lower than that in the normal lung tissues. e Notch signaling pathway also plays an important role in the tumorigenesis and progression of gastric cancer. Studies have shown that activation of Notch1 and Notch2 in gastric cancer tissues aggravates gastric cancer progression [66,67]. Inhibition of the Notch signaling pathway enhances the sensitivity of gastric cancer cells to chemotherapy [68]. Expression of Notch1 and Notch2 decreases after gastric cancer resection [69]. In contrast, Bauer et al. [70] showed that patients with high Notch1 expression in early gastric cancer had relatively high survival rates, and thus, Notch1 might play a role in tumor suppression in the early stage of gastric cancer. Previous studies have not examined the Notch signaling pathway-related proteins and CAG. e results of this study indicated that Notch2 and Notch3 protein contents were significantly downregulated in CAG patients. However, no Notch1 abnormalities were found. After acupuncture treatment, the protein contents of Notch2 and Notch3 increased, suggesting that the decrease in serum Notch2 and Notch3 in CAG patients may be related to CAG progression, and acupuncture may treat CAG by increasing Notch2 and Notch3 protein expression.
Collectively, this study utilized iTRAQ technology to conduct exploratory research on serum DEPs in CAG patients vs. healthy controls, as well as in CAG patients after acupuncture treatment. ABPs and Notch signaling pathwayrelated proteins may play an important role in the occurrence and progression of CAG and may be potential diagnostic markers for CAG. Moreover, results suggest that acupuncture regulated ABPs and proteins related to the Notch signaling pathways in CAG patients. is effect may be an important therapeutic mechanism by which acupuncture treats CAG. Further studies are needed to verify whether these DEPs can be used as diagnostic markers for CAG and whether they are targets of CAG acupuncture treatment.

Conclusion
Our data presented a valuable resource for diagnosis and treatment for CAG. Acupuncture can modulate multiple abnormal DEPs, such as ABPs and proteins related to the Notch signaling pathways in CAG patients.

Data Availability
All data are available upon reasonable request to the corresponding authors.

Disclosure
Feng Li and Bai Yang are the co-first authors and Huangan Wu and Xiaomei Wang are the co-corresponding authors.