Proteomics analysis of differentially-expressed proteins in uterus of primary dysmenorrhea mice following administration of nuangong zhitong

Purpose: To use label-free proteomic method to investigate the mechanism of action of nuanggong zhitong decoction (NZD) on primary dysmenorrhea (PD). Methods: A mouse model of PD was established through oxytocin administration. The mice were divided into control group (normal mice), model group (PD mice administered normal saline), and treatment group (mice given NZD). The serum levels of PGE2 and PGF2α in the mice were measured by ELISA. The differentially expressed proteins (DEPs) among the three groups were revealed by identifying the proteins that were up-regulated (or down-regulated) in model group and down-regulated (or up-regulated) in the treatment group. The DEPs in the three groups were identified using NanoHPLC-MS/MS, and their functions were investigated using bioinformatics analyses. The accuracy of proteomics was verified with western blot analysis. Results: Thirty-eight up-regulated and 66 down-regulated DEPs were identified. Bioinformatics analysis revealed that the DEPs were related to immune response, signal conduction, protein binding, and metabolism. STRING analysis indicated a total of 53 DEPs have direct or indirect functional links. Western blot results revealed that levels of Stat1, Rock1, vinculin and vaveolin-1 were consistent with the results of proteomic analysis. Conclusion: These findings provide further insights into the mechanism underlying the protective effects of NZD.


INTRODUCTION
Primary dysmenorrhea (PD) refers to recurrent menstrual cramps that are not due to other diseases. It occurs in approximately 50% of menstruating females. The pain associated with PD is extremely severe in 15% of patients, and results in psychological distress such as anxiety and depression [1]. Moreover, PD pain may be accompanied with nausea-and-vomiting, fatigue, and diarrhea [3]. Currently, the principal pharmacological therapies for PD include oral contraceptives and non-steroidal antiinflammatory drugs (NSAIDs). [4]. However, these drugs are associated with adverse side effects [5]. Thus, in their place, Chinese herbal medicine is used for treating PD due to its fewer adverse effects and lower degree of PD recurrence [6,7].
Nuangong zhitong decoction (NZD) has been used for clinical treatment of PD for many years in China. It was developed from the traditional Chinese prescription of Wen jing decoction which has been used clinically treating dysmenorrhea for decades. [8]. Nuangong zhitong decoction (NZD) is composed of Cinnamomi ramulus, evodiamine, asarum, Radix linderae and rhizoma corydalis at the ratio of 1:2:1:2:2. Cinnamaldehyde and cinnamic acid are the two main constituents of Cinnamomi ramulus. They have been reported to suppress oxytocininduced uterine contractions [9,10]. Evodiamine, asarum, Radix linderae and rhizoma corydalis warm the meridians so as to dissipate cold and relieve pain. However, the molecular mechanisms that underlie the analgesic effect of NZD are poorly understood. Therefore, there is need for more studies in this area.
Label-free quantitative proteomics has been employed to explore the mechanisms of medicine, including traditional Chinese Medicine (TCM) [11][12][13][14]. It emerged as a powerful approach for large-scale protein analysis with quantifying peptides and proteins with the use of a peptide's response as a quantitative measure [15,16]. In this study, proteomic alterations in PD mice in response to NZD treatment were investigated using Nano-HPLC-MS/MS technology.

EXPERIMENTAL Animal model of PD
Animal experiments in this study were approved by Ethics Committee of Taicang TCM Hospital Affiliated to Nanjing University of Traditional Chinese Medicine. Female KM mice (mean weight = 25 ± 5 g, 6 -8 weeks of age) were purchased from Cavens Laboratory Animal Co. Ltd (Changzhou, China). The mice were intragastrically administered a decoction made from gypsum, gentiana, Phellodendron chinense and Rhizoma anemarrhena, mixed in a ratio 2:1.2:1:1.5, at a dose of 4 g/mL for 14 days to establish a mouse model of cold-type asthenia. The mice were then subcutaneously injected with estradiol benzoate injection (2 mg/kg) daily for 12 days to improve the sensitivity of the mice uterine tissues to oxytocin. On the 12 th day, oxytocin (20 U/kg) was injected intraperitoneally to the mice to induce severe uterine contraction.
Seven days after establishment of ACT-PD model, the mice were randomly divided into 2 groups (10 mice/group) administered normal saline (model group) or NZD at a dose of 30.00 g/kg body weight (bwt, treatment group) for another 7 days. Ten (10) healthy Balb/c mice which were intragastrically administered normal saline for 13 days (10 mL/kg bwt) served as control group. On the 13 th day, writhing reaction was induced through intraperitoneal injection of oxytocin (33 U/kg).

Writhing test
The mice were placed in a box and intraperitoneally injected with oxytocin. The number of writhes in 30 min was counted. Analgesia (A) was calculated according to Eq 1. (1) where W PD and Wt are the no. of writhes in PD and treatment groups, respectively

Enzyme-linked immunosorbent assay (ELISA)
Blood was collected from the retroorbital plexus of mice after administration of NZD or its bioactive components for 40 min. The serum levels of PGE2 and PGF2α were measured with ELISA kits according to the kit protocol.

Sample preparation and protein digestion
There were three samples of mouse uterus tissue in each group. After cutting them into smaller pieces, RIPA lysis buffer was added, and the tissues were mechanically homogenized using a tissue homogenizer thrice, each for 3 sec. After an incubation of 15 min on ice, the samples were centrifuged at 12,000 g for 15 min at 4℃, and the supernatants were separately transferred into new Eppendorf tubes. BCA assay was used for the detection of the protein concentration of the supernatant. Proteins were diluted with 8 M urea solution followed by a further incubation of 1 h at 37℃. Thereafter, the mixture was transferred into 10 K Microcon centrifugal filter unit (Millipore, Billerica, MA). The samples were centrifuged to remove urea. The proteins were then alkylated by iodoacetamide at room temperature for 20 min (in the dark) and digested with sequence-grade modified trypsin (Promega) and lyophilized.

LCMS/MS analysis
Solvent A (0.1% formic acid, 30 μL) was used for resuspending peptides. Separations were performed with an EASY-nano-LC 1200 system (Thermo Fisher Scientific). Peptide sample (6 μL) was loaded into a trap column (C18, 75 μm x 2 cm, flow rate: 300 nL/min) , and subsequently separated and loaded onto an analytical column (C18, 75 μm x 50 cm) using a linear gradient of 5 -38% B (0.1% formic acid in ACN) for 120 min. A 2 kV electrospray voltage between the sprayer and ion inlet of the mass spectrometer was utilized in the study.

Identification of DEPs
PEAKS Studio (version 8.5, Bioinformatics Solutions Inc., Waterloo, Canada) was used to analyze tandem mass spectra. PEAKS DB was used to search the UniProt-mouse database (ver.201711, 52194 entries). The search parameters were 0.05 and 7 ppm for the fragment and parent ions mass tolerances, respectively. The fixed modification was carbamidomethylation (C), while the variable modifications were deamidation (NQ), oxidation (M), and acetylation (Protein N-term). Peptides were filtered with 1 % FDR and 1 unique. The abundance of peptide and protein was calculated using ANOVA. The averaging the abundance of all peptides was normalization using medians. Protein with fold-changes over 1.5 and at least 2 unique peptides with significance over 13 (p < 0.05) was considered to be a Differentlyexpressed protein (DEP).

Bioinformatics analysis
The obtained DEPs were analyzed using three databases: Kyoto Encyclopedia of Genes and Genomes (KEGG), Gene Ontology (GO), and the Clusters of Orthologous Groups (KOGs) databases. The interaction network of DEPs was built with the STRING.

Western blot analysis
Uterus tissue from each mouse was used to extract protein using RIPA lysis buffer containing 1% PMSF and cocktail (Beyotime, Haimen, China). BCA assay was employed to estimate the protein concentration. The protein was separated by 8-10% SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. After blocking with non-fat milk (5 %), the membranes were then incubated overnight with anti-vinculin, anti-caveolin, anti-stat1, anti-rock1 and anti-GAPDH at 4 °C. The membranes were washed thrice with TBST buffer and then incubated with secondary antibodies at room temperature for 1 h. Enhanced chemiluminescence (ECL) reagent (Thermo Fisher Scientific, Inc.) was used for the detection of interest protein bands. Image J v1.48u software (National Institutes of Health, Bethesda, MD, USA) was employed to analyze the relative optical densities of interest bands.

Statistical analysis
Statistical analysis was performed with SPSS 19.0 software (IBM Corp., NY). All data are expressed as mean ± standard deviation (SD). The differences between two groups were analyzed by Student's t-test. One-way ANOVA was used for multiple-group comparisons. Statistical significance was assumed at p < 0.05.

Effect of NZD on writhing in primary dysmenorrhea
As shown in Table 1, compared with the control mice, a remarkable increased number of writhes was observed in model mice, indicating the successful establishment of PD model. Treatment with NZD significantly reduced the number of writhes (p < 0.01), and the percentage analgesia was 61.82. Model mice showed significant increases in the levels of serum PGE2 and PGF2α. However, administration of NZD induced a remarkable decrease in the levels PGE2 and PGF2α (p < 0.01). These findings suggest that NZD can significantly relieved PD.

Identification of DEPs in uterus of PD mice administered NZD
Nano-HPLC-MS/MS was applied to identify DEPs in uterine tissues in the three groups. As shown in Table 2, a total of 556 DEPs were identified between control group and model group, out of which 245 were up-regulated, while 311 were down-regulated. Four hundred and four (404) DEPs were identified between the model group and treatment group, 238 of which were up-regulated, while 166 were down-regulated. There were 471 DEPs between the control group and the treatment group, with 267 up-regulated and 204 down-regulated. The DEPs among the three groups were then further analyzed via identification of the up-regulated or downregulated proteins in model and treatment groups. Sixty-six proteins were up-regulated in model group and down-regulated in treatment group, while 38 proteins which were downregulated in model group were up-regulated in treatment group. These results are displayed in Table 3.

GO analysis
To extract information relevant to involved pathways of DEPs, the protein data obtained were analyzed using DAVID network analysis tool. Moreover, GO analysis was carried out on cellular components (CC), molecular functions (MF), and biological processes (BPs) associated with the DEPs. In the BP analysis, majority of DEPs were associated with immune response, immune system process, regulation of localization and single-organism transport ( Figure 1). The CC analysis showed that most of DEPs were present in the cytoplasm (Figure 1). Molecular functional classification of DEPs showed that DEPs were mainly involved in protein binding, catalytic activity and hydrolase activity (Figure 1).

KEGG pathway analysis
The results of KEGG analysis revealed that the DEPs were significantly associated with tuberculosis, Staphylococcus aureus infection, leukocyte transendothelial migration and phagosome. The results also indicated that the DEPs were associated with Parkinson's disease, Fc gamma R-mediated phagocytosis and chemokine signaling pathway (Figure 2).

KOGs analysis
The results of KOGs analysis showed that the functions of the DEPs were mostly in information storage and processing, cellular processes and signaling, and metabolism ( Figure 3). The results indicated that these processes may be involved in the therapeutic effect of NZD.

PPI analysis
For further exploring the mechanisms involved in the protective effect of NZD, STRING database was used to construct the PPI network of the DEPs. As shown in Figures 4, a total of 53 DEPs in the map have direct or indirect links.

Validation by western blot on DEPs
As shown in Fig.5, the expression levels of Stat1 and Rock1 were significantly elevated in model group and down-regulated in treatment group, while vinculin and caveolin-1 showed significant decreases in model group and increases in treatment group. These results were consistent with the observations in proteomics analysis.

DISCUSSION
Although NZD has been used in China to treat PD in clinics for many years, the underlying mechanism remains largely unknown. To the best of the authors' knowledge, the present study is the first to use label-free proteomic based method to investigate the mechanism of NZD on a PD model. The results showed that NZD reduced oxytocin-induced writhing response after oxytocin injection in PD mice. The serum levels of PGE2 and PGF2α, which are regarded as the most critical pain factors in PD [17], were significantly decreased after NZD administration. These findings suggest that NZD exerts a significant analgesic effect in PD mice. Label-free quantitative proteomics is useful in searching for disease-associated factors and has been used to investigate the mechanism of TCM in recent years [11][12][13][14]. The present study identified 38 up-regulated DEPs and 66 downregulated DEPs after NZD treatment. The GO and KEGG analyses revealed significant alteration of functions and signaling pathways in PD mice after NZD administration. These changes affected protein binding, immune response, catalytic activity and chemokine signaling pathway. These GOs and pathways may play important roles in the analgesic action of NZD. The results of KOGs analysis revealed that the functions of the DEPs mainly involved in metabolism, cellular processes and signaling, information storage and processing.
The accuracy of proteomics was verified using western blotting with respect to the expressions of vinculin, caveolin-1, Stat1 and Rock1 in the three groups. The levels of Stat1 and Rock1 in uteruses were elevated in model group and down-regulated in treatment group. Vinculin and caveolin-1 showed decreases in model group and increases in treatment group. These results confirmed the credibility of proteomic analysis.

CONCLUSION
The proteomic studies in the present investigation have revealed a number of DEPs involved in the response to NZD administration in PD mice. It is hoped that these findings will provide a database resource for further investigations on the mechanisms involved in the protective effect of NZD against PD.

Acknowledgement
This work was supported by the Project of Taicang Science and Technology (no. TC2017YYJC03).