iTRAQ‐based quantitative proteomics analysis of immune thrombocytopenia patients before and after Qishunbaolier treatment

Rationale Treatment of immune thrombocytopenia (ITP) usually involves long‐term use of immunosuppressive corticosteroids and splenectomy. However, these treatments often have side effects in patients. The Mongolian medicine Qishunbaolier (QSBLE) has a high curative effect, reduces the chances of relapse, and has no obvious side effects. This study was designed to identify potential therapeutic targets of QSBLE for treating ITP. Methods To reveal differences in protein expression between ITP patients (ITPs) before and after QSBLE treatment, comparative proteomics studies were performed using isobaric tags for relative and absolute quantification (iTRAQ). The analysis used nanospray liquid chromatography/tandem mass spectrometry (nano‐LC/MS/MS) in positive ion electrospray ionization mode. Key proteins relevant to ITP were revealed by the Kyoto Encyclopedia of Genes and Genomes (KEGG) and other bioinformatics tools. Real‐time polymerase chain reaction (RT‐PCR) analysis was carried out for confirmation of differentially expressed proteins. Results A total of 982 differentially expressed proteins were identified in ITPs compared with the controls. Compared with the pre‐QSBLE treatment group, 61 differentially expressed proteins were identified in the post‐QSBLE treatment group, with 48 proteins being significantly upregulated and 13 downregulated. Twenty‐nine pathways were significantly enriched. Q6N030 and other proteins were the key players in the protein‐pathway network. Twenty proteins that may play important roles in the treatment of ITP were further filtered. RT‐PCR and Western blot analyses further confirmed that MIF, PGK1 and IGHM were upregulated in ITPs after QSBLE treatment, in accordance with the proteomics data. Conclusions It is believed that the identified proteins and the results of bioinformatics analysis will provide a potential therapeutic target site for QSBLE for ITP therapy and biomarkers.


| INTRODUCTION
Immune thrombocytopenia (ITP) is an autoimmune disease with an increased tendency of hemorrhage, characterized by increased cell and antibody-mediated platelet destruction, purpura, ecchymosis, and mucosal bleeding. 1 ITP has a disease group in people of different ages, genders, and races. The age of onset is polarized, with higher incidence in children under 14 years of age and older people over 60 years of age. 2,3 ITP can be asymptomatic or can present with petechiae, ecchymoses, mucocutaneous bleeding, nasal bleeding, or serious bleeding into the intestine or brain. 4 Intracranial hemorrhage occurs in <1% of ITP patients (ITPs) but it is a life-threatening complication of the disease. 5 Although an enhanced tendency of bleeding is a typical clinical feature of ITP, this disease is paradoxically associated with thromboembolic events. 6 Recently, new drugs for thrombopoietin receptor antagonists have been used to treat chronic ITP. However, these methods often have adverse reactions during treatment, including infection, bone marrow suppression or fibrosis, and recurrence. 7,8 In addition, corticosteroids usually have a therapeutic effect in only 70% of patients, while the alternative splenectomy surgical treatment increases the risk of developing internal bleeding. 8,9 Based on these side effects, corticosteroids and splenectomy should only be used carefully. Searching for a highly effective, low-rebound, and safer drug is an active direction for the treatment of this disease.
Mongolians have long used natural products, including herbs, animals, and minerals, to fight diseases in their nomadic and hunting life. Since the 1960s, Mongolian medicines have been used to treat a large number of diseases, including ITP, aplastic anemia (AA), allergic purpura (AP), and other bleeding disorders such as myelodysplastic syndrome (MDS). 10 Qishunbaolier (QSBLE) is a Mongolian preparation containing gardenia, bezoar, cornu bubali, cattle gall powder, saffron, lithospermum, and rubia. 10 Toxicity studies have shown that QSBLE has no toxic side effects on peripheral blood mononuclear cells or on the liver, heart, and kidney in mice; no obvious symptoms were found after high-dose administration. 10 Previously, we profiled miRNA expression in ITPs before and after QSBLE treatment and identified 44 miRNAs that were differentially expressed. 11 Of these 44 miRNAs, 25 were downregulated in ITPs. 11 These 25 miRNAs may be closely related to the pathogenesis of ITP. However, the impact of changes in protein expression before and after QSBLE treatment needs further study. Proteomics is an effective tool for the characterization of protein expression profiles, and has been widely used to study protein-related diseases. 12,13 Therefore, further identification and exploitation of potential molecular biomarkers will contribute to the prevention and treatment of ITP.
Our study aimed to use isobaric tags for relative and absolute quantification (iTRAQ) methods to determine the presence of differentially expressed proteins and biological pathways between ITPs before and after QSBLE treatment. Our results provide a comprehensive list of proteins associated with the progression of ITP; these may prove to be useful as potential markers or therapeutic targets.

| Sample collection and processing
In order to quantify protein sample preparation for mass spectrometry analysis, serum was depleted of most abundant proteins using a Human 14 multiple affinity removal system column (Agilent Technologies, Santa Clara, CA, USA) following the manufacturer's protocol. A 10 kDa ultrafiltration tube (Sartorius, Göttingen, Germany) was used for desalination and concentration of low-abundance components. One volume of SDT (2% sodium dodecyl sulfate (SDS), 100 mM dithiothreitol (DTT), 100 mM Tris-HCl, pH 7.6) buffer was added, and the solution was then boiled for 15 min and centrifuged at 14000 g for 20 min. The supernatant was quantified using a BCA Protein Assay Kit (Bio-Rad, Hercules, CA, USA). The samples were stored at −80 C.

| Western blot analysis
Western blot analysis was performed as described previously. 16

| Data analysis
The raw mass spectrometry data were acquired in a RAW file, and subjected to library identification and quantitative analysis using to be upregulated or downregulated, respectively. A protein-pathway network analysis was carried out using Cytoscape, as described previously. 17 The protein-pathway network was prepared based on the differential protein before and after treatment of the ITPs. All the pathways with P <0.05 in the KEGG and the proteins involved in the pathway were visualized with Cytoscape software (www. cytoscape.org).

| Statistical analysis
The ITP data pre-and post-QSBLE treatments were compared using the Mann-Whitney T-test in GraphPad (San Diego, CA, USA) Prism 5.0 software. When the P-value was <0.05, the difference was considered to be significant.  showed an increase in the platelet count from 46.9 × 10 9 /L to 115.2 × 10 9 /L, which effectively treated ITP (Table 1). Further observations upon closely following these QSBLE patients for up to 12 weeks did not reveal any side effects usually associated with glucocorticoid therapy, including loss of appetite, vomiting, increased heart rate, nausea, increased sweating, difficulty in sleeping or any other symptoms. We did not observe any significant recurrence after QSBLE treatment (data not shown).

| Differentially expressed proteins in ITPs before and after QSBLE treatment
In order to study and compare the differentially expressed proteins in the blood of ITPs and controls, protein samples were labeled with iTRAQ reagents and analyzed by nano-LC/MS/MS (Figure 1).
In our study, a total of 5717 unique peptide groups belonging to 982 proteins were identified (

| Gene ontology analysis of identified proteins
Proteins that were identified only in ITP patients before or after QSBLE treatment were found to differ in their characteristics. Gene ontology (GO) enrichment indicated that these differentially expressed proteins were mainly involved in biological processes such as biological regulation, biological adhesion, immune system process, signaling, locomotion, and developmental process ( Figure 2). In terms of molecular function annotation, these differentially expressed proteins were mainly involved in signal transducer activity, molecular transducer activity, transporter activity, molecular function regulator, binding, structural molecular activity and chemorepellent activity  that these proteins were related to immune response which may make them potential biomarkers for ITP diagnosis.

| KEGG pathway analysis
The KEGG pathway is a utility database resource for understanding advanced functions and biological systems (such as cells, organisms, and ecosystems), genome sequencing from molecular-level information, especially large molecular data sets, and other highthroughput experimental techniques. Pathway enrichment analysis was also performed by searching against the KEGG database.
According to the results of the KEGG enrichment analyses, 29 pathways were included in the differentially expressed proteins ( Figure 3A). ( Figure 3A). In order to further analyze the potential target proteins in ITP therapy in the KEGG signaling pathway, we constructed a protein-pathway network ( Figure 3B). The results revealed that these differentially expressed proteins were highly associated with immune diseases, and immune system development and function, such as the Fc epsilon RI signaling pathway and immune network for IgA production are potential therapeutic target proteins for ITP.

| MS analysis in ITP pathogenesis and gene expression verification
It is unclear whether protein dysregulation is associated with the pathogenesis of ITP. We conclude that ITP-associated proteins are abnormally expressed in ITPs, but can be restored to control levels after QSBLE treatment. To further screen for functional proteins, we reanalyzed differentially expressed proteins and focused on proteins that are abnormally expressed in ITP which are also expressed in control and treatment. Finally, we identified 20 proteins that met these criteria ( Figure 4A, and Table S5, supporting information). Of these 20 proteins, P14174, P00558 and Q6N030, which are encoded by MIF, PGK1 and IGHM, respectively, were closely related to immunity. [18][19][20] The results of the RT-PCR and Western blot analyses showed that the expression of these three genes was upregulated in ITPs after QSBLE treatment, which is consistent with the proteomics results ( Figures 4C and 4D). When MIF is inhibited, it usually leads to a functional reversal from immunosuppressive MDSCs to immunostimulatory dendritic cells (DCs), some of which may be due to a decrease in MDSC prostaglandin E2 (PGE2). 18 In a recent study, PGK1 was used as an immune target in Kawasaki disease. 19 Genetic studies have shown that patients affected by ARA due to μ heavy chain defects are complex heterozygotes of the IGHM gene deletion. 20 The three genes were all upregulated in the ITPs after treatment compared with those before treatment ( Figure 4B, and  treatment in pediatric ITP are very rare, and only five cases have been reported to date. [21][22][23][24] Traditionally, macrophage and autoantibodymediated phagocytosis in the spleen play a major role in platelet destruction in the pathogenesis of ITP. 25,26 At present, the source of these antibodies is not clear, but recent studies have shown that when infected with hepatitis virus or HIV virus, it is easy to induce anti-autoreactivity, which in turn stimulates the production of antiplatelet antibodies. 27 Studies have also shown that T cells play an important role in the immune pathogenesis of ITP. Because T cells have a self-tolerance that tends to be self-reactive, dysregulation of T cells is crucial in the development of ITP. [28][29][30][31] There are many factors that can cause the pathogenesis of ITP, The goal of primary ITP therapy is to increase the patient's platelet count to a safe level and reduce the mortality rate, but excessive treatment should be avoided as much as possible.  Table 2). The data for the invalid group are also listed in Table 2; they were not directly related to the effect of QSBLE on the treatment of ITP, but they do have potential research value. After identifying significant differentially expressed proteins, we performed bioinformatics analysis to further extract information from the proteomics data. GO analysis provided insights into the molecular functions and biological processes of relevant ITP proteins, and pathway enrichment analysis was also performed by searching against the KEGG database.
According to the results of KEGG enrichment analyses, 29 pathways were significantly enriched. The KEGG enrichment analyses revealed that these differentially expressed proteins were highly associated with immune diseases, and immune system development and function, such as the Fc epsilon RI signaling pathway and immune network for IgA production. We further screened out 20 proteins which are expressed in normal people as well as in pre-and post-QSBLE treatments in ITPs. Of these 20 proteins, MIF, PGK1, and IGHM are closely related to immunity. [18][19][20] As ITP is an autoimmune disorder characterized by autoantibody production against platelets and increased platelet destruction, platelets can both supply and respond to signals at the early stages of immune control. 41 The results of GO analysis also showed that the differentially expressed proteins are involved in the biological process of adhesion, response to stimulus and immune system processes and molecular function of molecular transducer activity, binding and catalytic activity, cellular component of cell junction, and macromolecular complex ( Figure 2).
The above-mentioned results imply that MIF, IGHM, and PGK1 may affect the function of platelets which are highly related to ITP.
Overexpression of the miR-98-5p can inhibit the PI3K/Akt signaling pathway and impair the therapeutic effect of MSCs in ITP mice. 42 Furthermore, the PPAR and NF-κB signaling pathways are involved in regulating ITPs. 43 The PI3K/Akt, PPAR, and NF-κB signaling pathways related to ITP are also those that are highly enriched by the 20 differentially expressed proteins described here (Figure 3).
Although the pathways and biological processes of these differentially expressed proteins have been suggested to be enriched and involved, the relationship between the proteins and the miRNAs that we previously reported still needs to be further studied. 11 ITP is a rare disease, our clinical screening standards were very strict, and the number of patients with effective treatment drugs will be further reduced. Thus, the number of patients ultimately available is very limited. The data from this study show differentially expressed proteins between ITP patients and normal people to a certain extent in Inner Mongolia. In addition, before putting these results into clinical use, it is necessary to further verify the stability and pathways of the