Phosphorylated protein marker for glioma angiogenesis and application of phosphorylated protein marker

Abstract

The invention relates to the technical field of biomedicine. The invention provides a phosphorylated protein marker for angiogenesis of glioma and application of the phosphorylated protein marker. Through a series of analysis such as iTRAQ quantitative phosphorylated proteomics, KEGG signal channel enrichment, a kinase spectrum, glioblastoma transcriptome high-throughput sequencing data and immunohistochemistry, it is determined that significantly different phosphorylated proteins and related signal channels can significantly affect neuroglioma angiogenesis; and the phosphorylated protein changes along with tumor progression, and is a phosphorylated protein marker and a potential therapeutic target for glioma angiogenesis.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a phosphorylated protein marker of glioma angiogenesis and an application thereof. Background Art

[0002] Glioma is the most common intracranial primary brain tumor, accounting for 81% of malignant brain tumors. The disease is difficult to treat, with severe mortality and morbidity, and a 5-year overall survival rate of less than 35%. Based on pathological characteristics, the World Health Organization (WHO) divides gliomas into different pathological stages (I-IV), among which stage IV is the most malignant and has the most abundant neovascular tissue in the tumor, such as glioblastoma multiforme (GBM). Although traditional therapies and immunotherapy have made certain progress, the improvement in the prognosis of patients with gliomas is limited, especially for patients with high-grade gliomas.

[0003] The formation of tumor vascular networks is called angiogenesis, which is a necessary condition for the continued survival and malignant transformation of tumor cells and is involved in tumor cell proliferation, invasion and treatment resistance. Pro-angiogenic factors released by tumor cells bind to membrane receptors of vascular endothelial cells (VECs), triggering a pro-angiogenic cascade in target cells, thereby inducing angiogenic behavior of VECs and forming endothelial tubular structures. Given the role of angiogenesis in tumor progression, targeting the tumor vasculature by inhibiting the signaling pathways required for angiogenesis has received widespread attention in the treatment of gliomas. However, targeted therapy for glioma angiogenesis still has certain limitations, partly because the heterogeneity of tumors makes it impossible for targeted therapy to benefit all patients.

[0004] Therefore, more unknown angiogenesis biomarkers need to be studied to expand the group that can benefit from targeted angiogenesis treatment of glioma. Summary of the invention

[0005] The purpose of the present invention is to provide phosphorylated protein markers related to glioma angiogenesis. Through a series of analyses such as iTRAQ quantitative phosphorylation proteomics, it is determined that significantly differentially phosphorylated proteins and related signal pathways will significantly affect glioma angiogenesis and change with tumor progression. They are phosphorylated protein markers of glioma angiogenesis and potential therapeutic targets.

[0006] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0007] The present invention provides phosphorylated protein markers related to glioma angiogenesis, including ADD1, ANK1, ANK2, AQP4, ATP2B2, CAMK2A, CAMK2D, CANX, CAV1, CRYAB, DKC1, DMTN, EPB41, FBN1, FGA, FLNA, GAP43, HNRNPA1, HSP90AB1, HSPB1, KNG1, L1CAM, MAP1B, MAP2, MAPT, MBP, MYH10, MY H11, MYL12A, MYLK, NCAM1, NEFM, NES, PACSIN1, PAK1, PLEC, PSMA3, RBM10, RBM25, RPLP0, RPLP2, SF1, SF3A1, SNT A1, SORBS1, SPTB, SPTBN1, STXA1, SYN1, SYP, TAGLN, TGFB1I1, TJP1, TLN1, TPPP, TUBA1C, VIM, ZYX, PDHA1 and PDK2.

[0008] The present invention also provides the use of a reagent for detecting the marker in preparing a glioma diagnosis kit.

[0009] The present invention also provides the use of a reagent for detecting the marker in preparing a glioma prognosis assessment kit.

[0010] The present invention also provides the use of the marker as a screening target in screening drugs for treating glioma.

[0011] By adopting the above technical solution, the present invention has the following beneficial effects:

[0012] 1. In the present invention, 195 significantly differentially phosphorylated proteins (DPPs) and 635 differentially phosphorylated sites (DPSs) were identified by iTRAQ quantitative phosphorylation proteomics analysis, of which 58 DPPs were identified as core hub molecules. Then, KEGG signaling pathway enrichment analysis was performed on the 195 DPPs, and it was found that phosphorylation significantly affected angiogenesis-related signaling pathways, including cell adhesion-related pathways and HIF-1 signaling pathways.

[0013] 2. In the present invention, 321 kinases were found to possibly regulate phosphorylation modification by kinase spectrum analysis, among which the protein expression levels of 14 kinases including pyruvate dehydrogenase kinase 2 (PDK2) changed in glioma neovascularization and were identified as differentially expressed kinases (DEKs) in glioma neovascularization tissue. Analysis of DPPs regulated by DEKs revealed that two DPPs (CAMK2D and PDHA1) enriched in the HIF-1 signaling pathway were regulated by DEKs. Further verification of targeted therapeutic drugs for key DPPs showed that the small molecule chemotherapy drug staurosporine (STS) affects the phosphorylation modification of the differential phosphorylation site serine 19 (MYL12A-Ser19) of myosin light chain 12A (MYL12A) in glioma vascular endothelial cells (VECs), thereby affecting glioma angiogenesis. Therefore, it is concluded that glioma angiogenesis may be related to certain DPPs.

[0014] 3. In the present invention, 23 core hub molecules were identified by combining the high-throughput sequencing data analysis of the glioblastoma (GBM) transcriptome in the Cancer Genome Atlas (TCGA) online database, including DPPs (MAPT, FLNA, CAMK2A, VIM, PAK1 and MAP1B) related to glioma disease, and 3 DPPs that can be used as prognostic molecules for the disease. Combined immunohistochemistry (IHC) analysis demonstrated that phosphorylation of pyruvate dehydrogenase E1 subunit alpha 1 (PDHA1), a key protein in the mitochondrial energy metabolism pathway in the HIF-1 signaling pathway, at serine 293 (PDHA1-Ser293) and serine 300 (PDHA1-Ser300) in glioma VECs and its upstream kinase pyruvate dehydrogenase kinase 2 (PDK2) were reduced, which was associated with the tumor pathological stage and changed with tumor progression. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1Statistical results of DPPTs, DPPs, DPSs and amino acid types identified in human glioma vascular tissues ( Figure 1 A represents DPPTs, B represents DPPs, C represents DPSs, and D represents amino acid type);

[0016] Figure 2 This is the GO enrichment clustering result of DPPs identified in human glioma vascular tissue;

[0017] Figure 3 Figure 2 shows the cell adhesion, HIF-1, and neurodegenerative disease-related signaling pathways of DPPs enriched in human glioma vascular tissue. Figure 3 A in the figure is the cell adhesion signaling pathway, B is the HIF-1 signaling pathway, and C is the neurodegenerative disease-related signaling pathway; red marks are proteins with upregulated phosphorylation, green marks are proteins with reduced phosphorylation, and yellow marks are proteins with upregulated phosphorylation at certain sites and reduced phosphorylation at other sites);

[0018] Figure 4 The PPI network constructed for the interaction between DPPs and the identified hub molecules ( Figure 4 The protein marked in green is the core hub molecule);

[0019] Figure 5 Differential phosphorylation profiles of central molecules associated with glioma progression ( Figure 5 A in the figure indicates MAPT, B indicates FLNA, C indicates HSP90AB1, D indicates PAK1, E indicates VIM, F indicates CAMK2A, G indicates CAVI, H indicates AQP4, I indicates GAP43, J indicates KNG1, K indicates MAP1B, and L indicates CANX; red marks indicate amino acids that are upregulated by phosphorylation, green marks indicate amino acids that are downregulated by phosphorylation, S indicates serine, T indicates threonine, and Y indicates tyrosine);

[0020] Figure 6 is the kinase spectrum of DPPs;

[0021] Figure 7 The protein interaction network diagram between DPPs and predicted kinases;

[0022] Figure 8 The network diagram of differentially expressed kinases in glioma angiogenesis and their protein interaction with DPPs ( Figure 8 T stands for tumor, N stands for normal);

[0023] Fig. 9The DPPs-kinase-small molecule chemical drug interaction network diagram is the hub Fig. 9 Medium and dark blue lines highlight the interaction between CAMK2D and its substrate DPPs; light blue lines highlight the interaction between MYLK and its substrate DPPs; orange lines highlight the interaction between CAMK2D and MYLK and their small molecule drugs; highlighted kinases are kinases with connectivity greater than 10 with DPPs; highlighted drugs are small molecule drugs targeting these kinases);

[0024] Fig.10 This is a diagram of the interaction network of DPPs-kinase-small molecule chemical drugs related to cell adhesion pathway ( Fig.10 The black lines highlight the interaction between MYLK and its substrate DPPs and small molecule drugs; the blue lines highlight the interaction between CAMK2D and its substrate DPPs and small molecule drugs; the pink lines highlight the interaction between PIM1 and its substrate DPPs and small molecule drugs; the dark green lines highlight the interaction between CSNK2A1 and its substrate DPPs and small molecule drugs; the light green lines highlight the interaction between CSNK2A2 and its substrate DPPs and small molecule drugs; the orange lines highlight the interaction between MAPK14 and its substrate DPPs and small molecule drugs);

[0025] Fig.11 The 10x microscope images of tubule formation in HUVEC cells and hCMEC / D3 cells after 6 hours of treatment with different concentrations of STS and the tubule statistics ( Fig.11 A in the figure represents the microscopic image of HUVEC cells, B represents the bifurcation point of HUVEC cells, C represents the number of branches of HUVEC cells, D represents the microscopic image of hCMEC / D3 cells, E represents the bifurcation point of hCMEC / D3 cells, and F represents the number of branches of hCMEC / D3 cells; ***P<0.001, **P<0.01, *P<0.05);

[0026] Fig.12 Western blotting was used to detect the changes in MYL12A-Ser19 phosphorylation in HUVEC and hCMEC / D3 cells treated with STS ( Fig.12 A in the figure indicates HUVEC cells, and B indicates hCMEC / D3 cells);

[0027] Fig.13 Effects of different concentrations of STS on glioma model mice ( Fig.13A in the figure indicates glioma model mice, B indicates glioma after dissection, C indicates the weight statistics of tumor tissue obtained after dissection, and D indicates the weight growth of model mice; ***P<0.001, **P<0.01, *P<0.05);

[0028] Fig.14 IHC determination of CD31 expression in different tumor tissues obtained by dissection and its statistical results ( Fig.14 A in the figure indicates the staining result, and B indicates the expression level of CD31);

[0029] Fig.15 This is the common molecular analysis diagram of DEGs and DPPs in glioblastoma;

[0030] Fig.16 The results of KEGG pathway enrichment analysis of common molecules of DEGs and DPPs in glioblastoma;

[0031] Fig.17 The analysis results of the core hub common molecules of the PPI interaction network diagram ( Fig.17 The red marks in the figure indicate the core hub common molecules);

[0032] Fig.18 The DPPs with prognostic value in univariate regression analysis;

[0033] Fig.19 The DPPs with prognostic value in multivariate regression analysis;

[0034] Fig. 20 iTRAQ quantitative phosphorylation proteomics analysis of changes in the phosphorylation levels of PDHA1-Ser293, PDHA1-Ser300, PDK1 and PDK2 in tumor blood vessels ( Fig. 20 A represents PDHA1-Ser300, B represents PDHA1-Ser293, C represents PDK2, and D represents PDK1; T represents tumor, C represents control, N represents negative, II represents tumor samples in stage II, III represents tumor samples in stage III, and IV represents tumor samples in stage IV);

[0035] Fig.21 IHC was used to detect the phosphorylation level of PDHA1-Ser300 in tumor samples and vascular endothelial cells of normal tissues. Fig.21 A in the figure represents the staining result, B represents the positive staining rate between N and L, H, and C represents the positive staining rate between L and H);

[0036] Fig. 22 IHC was used to detect the phosphorylation level of PDHA1-Ser293 in vascular endothelial cells of tumor samples and normal tissues. Fig. 22 A represents the staining result, B represents the positive staining rate between N and L, H, and C represents the positive staining rate between L and H; N represents negative, L represents low, and H represents high);

[0037] Fig.23 IHC was used to detect the expression level of PDK2 protein in tumor samples and vascular endothelial cells of normal tissues. Fig.23 A represents the staining result, B represents the positive staining rate between N and L, H, and C represents the positive staining rate between L and H; N represents negative, L represents low, and H represents high);

[0038] Fig.24 IHC was used to detect the expression level of PDHA1 protein in tumor samples and vascular endothelial cells of normal tissues. Fig.24 A represents the staining result, B represents the positive staining rate between N and L, H, and C represents the positive staining rate between L and H; N represents negative, L represents low, and H represents high);

[0039] Fig.25 The statistical results of clinical staging of tumors in different PDHA1-pSer300, PDHA1-Ser293, PDK2, and PDHA1 expression subgroups ( Fig.25 In the figure, A represents PDHA1-pSer300, B represents PDHA1-Ser293, C represents PDK2, and D represents PDHA1; L represents low, and H represents high). DETAILED DESCRIPTION

[0040] The present invention provides phosphorylated protein markers related to glioma angiogenesis, characterized by comprising adducin1 (ADD1), ankyrin 1 (ANK1), ankyrin 2 (ANK2), aquaporin 4 (AQP4), ATPase plasma membrane Ca2+ transporting 2 (ATP2B2), calcium / calmodulin dependent protein kinase II alpha (CAMK2A), calcium / calmodulin dependent protein kinaseII delta (CAMK2D), calnexin (CANX), caveolin 1 (CAV1), crystallin alpha B (CRYAB), dyskerin pseudouridine synthase 1 (DKC1), dematin actin binding protein (DMTN), erythrocyte membrane protein band 4.1 (EPB41), fibrillin 1 (FBN1), fibrinogen alpha chain (FGA), fibrinogen alpha chain (FLNA), growth-associated protein 43(GAP43), heterogeneous nuclear ribonucleoproteinA1(HNRNPA1), heat shockprotein90alpha family B member 1(HSP90AB1), heat shock protein family B(small) member1(HSPB1), kininogen 1(KNG1), L1 cell adhesion molecule(L1CAM), microtubuleassociated protein 1B(MAP1B), microtubule-associated protein 2(MAP2), microtubule associated protein tau(MAPT), myelinbasicprotein(MBP), myosin heavy chain 10(MYH10), myosin heavy chain 11(MYH11), myosin light chain 12A(MYL12A), myosinlight chain kinase (MYLK), neural cell adhesion molecule 1 (NCAM1), neurofilament medium chain (NEFM), nestin (NES), protein kinase C and caseinkinase substrate in neurons 1 (PACSIN1), P21 (RAC1)-activated kinase 1 (PAK1), plectin (PLEC), phosphorylated proteasome 20S subunit alpha 3 (PSMA3), RNA binding motif protein 10 (RBM10), RNA binding motif protein 25 (RBM25), ribosomal protein lateral stalk subunit P0 (RPLP0), ribosomal protein lateral stalk subunit P2 (RPLP2), splicing factor 1 (SF1), splicing factor 3a subunit 1 (SF3A1), syntrophinalpha 1 (SNTA1), sorbin and SH3 domain containing 1(SORBS1),spectrin beta, erythrocytic (SPTB), non-erythrocytic 1 (SPTBN1), syntaxin A (STXA1), synapsin I (SYN1), synaptophysin (SYP), transgelin (TAGLN), transforming growth factor beta1induced transcript 1 (TGFB1I1), tightjunction protein1 (TJP1), talin 1 (TLN1), tubulin polymerization promoting protein (TPPP), tubulin alpha 1c (TUBA1C), vimentin (VIM), zyxin (ZYX), PDHA1 and PDK2.

[0041] The present invention also provides the use of a reagent for detecting the marker in preparing a glioma diagnosis kit.

[0042] The present invention also provides the use of a reagent for detecting the marker in preparing a glioma prognosis assessment kit.

[0043] The present invention also provides the use of the marker as a screening target in screening drugs for treating glioma.

[0044] The technical solutions provided by the present invention are described in detail below in conjunction with the embodiments, but they should not be construed as limiting the protection scope of the present invention.

[0045] Example 1

[0046] 1.1 Collection and processing of clinical samples of glioma

[0047] Thirty-eight clinical glioma specimens (pathological stages II-IV) and three normal control neovascular tissue specimens were obtained from the Department of Neurosurgery, Xiangya Hospital, Central South University. Each sample was histopathologically identified. Laser capture microdissection (LCM) was used to obtain neovascular tissue in glioma tissues, which were mixed for downstream analysis.

[0048] (II) iTRAQ quantitative phosphoproteomics to study the differential phosphorylated protein profiles of glioma angiogenesis

[0049] 1. Preparation of Protein Samples

[0050] 100 mg of glioma neovascular tissue was treated with 1 mL of urea thermolysis solution (consisting of 9M urea, 20mM 2-hydroxyethyl HEPES, 2.5mM sodiumpyrophosphate, 1mM sodium orthovanadate and 1mMβ-glycerophosphate, pH 8.0) in an ice bath with ultrasound (100W, 10s / time, 10 times in total), and then centrifuged to obtain the supernatant as the extracted protein, and the protein was quantitatively analyzed using the Bradford method (Bradford Protein Quantification Kit, YEASEN, Cat#20202ES76).

[0051] 2. Enzymatic hydrolysis and peptide quantification

[0052] Protein samples extracted from glioma neovascular tissue and normal vascular tissue were mixed separately. 200 μg of neovascular tissue protein was treated with 30 μL solution I (consisting of 4% SDS, 100 mM dithiothreitol (DTT) and 150 mM Tris-HCl, pH = 8.0) and incubated at 37 ° C for 2.5 h, followed by repeated ultrafiltration to remove these reagents using solution II (consisting of 8 M urea and 150 mM Tris-HCl, pH = 8.0). The reduced cysteine ​​residues were blocked with 100 μL solution III (consisting of 50 mM iodoacetamide, 8 M urea, 150 mM Tris-HCl, pH = 8.0), incubated in the dark for 2.5 h, washed 3 times with 100 μL solution II, and washed 2 times with 100 μL dissolution buffer. 4 μg of trypsin was added to digest the protein mixture (37 ° C, 18 h). Tryptophan was desalted using a C18 powder box and concentrated by vacuum centrifugation. The tryptophan content was determined by ultraviolet spectroscopy at 280 nm.

[0053] 3. iTRAQ-labeled peptides

[0054] 100 μg of the lyophilized peptide mixture was taken from each sample and labeled with iTRAQ reagents according to the manufacturer's instructions (AB Sciex, Foster City, CA). The peptides were dissolved in 600 μL TEAB solution (0.5 M, pH = 8.5), and the labeling reagent (20 units per 2 mg of peptide) was added to 1.4 mL ethanol and incubated for 1 h. The reaction was then stopped with Tris-HCl (50 mM, pH = 7.5). Each sample was labeled with three different iTRAQ reagents.

[0055] 4. TiO2 Enrichment of Phosphopeptides

[0056] The 6 iTRAQ-labeled peptides were mixed evenly, vacuum concentrated, and then 500 μL of DHB buffer was added to resuspend the iTRAQ-labeled peptides. TiO2 microspheres were added to the peptide sample, stirred for 2 hours, and centrifuged at 5000g for 1 minute. The supernatant was taken to retain the microbeads and washed with 50 μL of wash buffer I (30% acetonitrile and 3% trifluoroacetic acid) and 50 μL of wash buffer II (80% acetonitrile and 0.3% trifluoroacetic acid). The phosphopeptides were eluted with 50 μL of elution buffer (40% acetonitrile and 15% NH3·H2O) and lyophilized.

[0057] 5. Identification of DPPs in Glioma Neovascularization

[0058] The TiO2-enriched phosphopeptides were analyzed by LC-MS / MS for 60 min on Easy nLC (Proxeon Biosystems, now Thermo Fisher Scientific) and Q accurate mass spectrometer (Thermo Fisher Scientific). The protein database was analyzed, identified and quantified using MaxQuant software. Differentially phosphorylated peptides (DPSs) and differentially phosphorylated peptides (DPPTs) in glioma neovascularization tissue were identified based on the screening threshold (FDR < 0.05, phosphorylation probability > 50) compared with the control group. Proteins containing DPPTs were named DPPs.

[0059] 6. Biometric Analysis of DPPs

[0060] Gene ontology (GO) enrichment analysis was performed on biological processes (BPs), cellular components (CCs), and molecular functions (MFs) involved in DPPs using the publicly available DAVID database (https: / / david.ncifcrf.gov / ), with identification thresholds of enrichment score>1.5 and P-value<0.05. GO clustering analysis was performed on BPs, CCs, and MFs in the DAVID database with the following parameters: classification stringency=custom, similarity word overlap=3, similarity threshold=0.50, final membership=2, multiple linkage threshold=0.50, enrichment score>1.5, and P-value<0.05. The Kyoto Encyclopedia of Genes and Genomes (KEGG) in the metscape database (https: / / metascape.org / gp / index.html) was used to enrich and analyze the DPPs-mediated signaling pathways (enrichment score>1.5 and P-value<0.05).

[0061] 7. Results

[0062] iTRAQ phosphoproteomics was used to study the phosphorylated proteome in human glioma neovascular tissue and control vascular tissue obtained by LCM. By setting the screening threshold (FDR < 0.05; PhosphositeProbabilities ≥ 50), we identified 468 peptides (202 of which were increased and 266 were decreased) and 635 amino acid sites (269 of which were increased and 366 were decreased) whose phosphorylation modifications changed significantly in glioma vascular tissue, and named them differentially phosphorylated peptides (DPPTs) and differentially phosphorylated sites (DPSs), respectively. These DPPTs and DPSs are located in 195 DPPs ( Figure 1 and Table 1).

[0063] Table 1 Changes of DPPTs, DPSs and DPPs in human glioma neovascularization identified by iTRAQ quantitative phosphoproteomics

[0064]

[0065] GO enrichment analysis was performed to analyze the cellular functions (including biological processes, cellular components, and molecular functions) involved in DPPs. Figure 2 It was shown that 195 DPPs were involved in 14 categories of cellular functions, including regulation of skeletal muscle and cardiomyocytes, actin cytoskeleton organization, neural regulation, protein cytoskeleton organization, heart rate regulation, ion transport, microtubule cytoskeleton, plasma membrane region regulation, cell junctions, intermediate filament cytoskeleton regulation, protein kinase C activity, dystrophin-related glycoprotein regulation, regulation of kinase activity, and telomere regulation.

[0066] Through KEGG pathway enrichment analysis (such as Figure 3 ), 195 DPPs were found to be involved in 20 signaling pathways, mainly including cell adhesion, proteoglycans in cancer, tight junctions of epithelial cells, actin cytoskeleton regulation, HIF-1 signaling pathway, and neurodegenerative disease-related signaling pathways. The relevant DPPs involved in cell adhesion, HIF-1 signaling pathway, and neurodegenerative disease signaling pathways are shown in Table 2.

[0067] Table 2 DPPs involved in cell adhesion, HIF-1 signaling pathway and neurodegenerative disease signaling pathway

[0068]

[0069]

[0070]

[0071] Example 2

[0072] (I) Construction of protein-protein interaction network and identification of hub molecules

[0073] The protein and protein interaction (PPI) network of DPPs was constructed using the STRING database (https: / / cn.string-db.org / ) and visualized using Cytoscape software. The hub molecules in the PPI network were determined by overlapping six groups of hub genes, which were identified by six algorithms in the Cytohubba program, including molecular complex detection (MCC; cutoff ≥ 10, connectivity ≥ 5), maximum neighborhood component (MNC; cutoff ≥ 4, connectivity ≥ 5), edge percolated component (EPC; cutoff ≥ 10, connectivity ≥ 5), degree of interaction (cutoff ≥ 10, connectivity ≥ 5), closeness (cutoff ≥ 50, connectivity ≥ 5), and authenticity (cutoff ≥ 6, connectivity ≥ 5). The hub genes identified by each algorithm were overlapped by Venn analysis to obtain the final hub molecules and marked in the PPI network using Cytoscape software.

[0074] To analyze the interactions among the 195 DPPs and identify the DPPs at the core hub position, a PPI network was constructed for these DPPs. The results showed that these DPPs had a tightly connected interaction network ( Figure 4 To identify the core hub molecules, the hub proteins obtained by the above six algorithms in the Cytohubba plug-in in the Cytoscape biological software were analyzed for common molecules, and 58 DPPs were finally identified as hub molecules in the PPI network (such as Figure 4as shown), specifically including: adducin 1 (ADD1), ankyrin 1 (ANK1), ankyrin 2 (ANK2), aquaporin 4 (AQP4), ATPase plasma membrane Ca2+ transporting 2 (ATP2B2), calcium / calmodulin dependent protein kinase II alpha (CAMK2A), calcium / calmodulin dependent protein kinase II delta (CAMK2D), calnexin (CANX), caveolin 1 (CAV1), crystallin alpha B (CRYAB), dyskerin pseudouridine synthase 1 (DKC1), dematin actin binding protein (DMTN), erythrocyte membrane protein band 4.1 (EPB41), fibrillin 1 (FBN1), fibrinogen alpha chain (FGA), fibrinogen alpha chain (FLNA), growth-associated protein 43 (GAP43), heterogeneous nuclear ribonucleoprotein A1 (HNRNPA1), heat shock protein 90 alpha family B member 1 (HSP90AB1), heat shock protein family B (small) member 1 (HSPB1), kininogen 1 (KNG1), L1 cell adhesion molecule (L1CAM), microtubule associated protein 1B (MAP1B), microtubule-associated protein 2 (MAP2), microtubule associated protein tau (MAPT), myelin basic protein (MBP), myosin heavy chain 10 (MYH10), myosin heavy chain 11 (MYH11), myosin light chain 12A (MYL12A), myosin light chainkinase(MYLK)、neural cell adhesion molecule 1(NCAM1)、neurofilament medium chain(NEFM)、nestin(NES)、protein kinase C andcaseinkinase substrate in neurons1(PACSIN1)、P21(RAC1)-activated kinase 1(PAK1)、plectin(PLEC)、phosphorylated proteasome 20Ssubunit alpha 3(PSMA3)、RNAbinding motifprotein 10(RBM10)、RNAbinding motifprotein 25(RBM25)、ribosomalprotein lateral stalk subunit P0(RPLP0)、ribosomal protein lateral stalksubunit P2(RPLP2)、splicing factor 1(SF1)、splicing factor 3a subunit 1(SF3A1)、syntrophin alpha 1(SNTA1)、sorbin and SH3domain containing 1(SORBS1)、spectrinbeta,erythrocytic(SPTB)、non-erythrocytic 1(SPTBN1)、syntaxin A(STXA1)、synapsin I(SYN1)、synaptophysin(SYP)、transgelin(TAGLN)、transforming growthfactorbeta1induced transcript 1(TGFB1I1)、tightjunction protein 1(TJP1)、talin1(TLN1)、tubulin polymerization promoting protein(TPPP)、tubulin alpha1c (TUBA1C), vimentin (VIM), zyxin (ZYX). Among the above 58 hub molecules, 12 molecules are involved in the regulation of glioma progression, including MAPT, FLNA, AQP4, GAP43, HSP90AB1, CAMK2A, CAV1, MAP1B, VIM, CANX, PAK1 and KNG1. The basic functions of these proteins and their roles in glioma progression are detailed in Table 3.

[0075] Table 3 Basic functions of 12 glioma progression-related DPPs and their roles in tumor progression

[0076]

[0077]

[0078] These DPPs identified by iTRAQ phosphoproteomics have different types of phosphorylation modifications (e.g. Figure 5 ), in which 8 different types of phosphorylated modified forms of MAPT protein were identified in glioma neovascularization; 2 different types of phosphorylated modified forms of HSP90AB1; 2 different types of phosphorylated modified forms of VIM; 5 different types of phosphorylated modified forms of CAMK2A; 2 different types of phosphorylated modified forms of AQP4; 3 different types of phosphorylated modified forms of GAP43; 20 different types of phosphorylated modified forms of MAP1B; 3 different types of phosphorylated modified forms of CANX; and 1 phosphorylated modified form of FLNA, PAK1, CAV1 and KNG1.

[0079] (II) Identification of DPPs kinase spectrum and analysis of corresponding chemotherapeutic drugs

[0080] 1. Construction of DPPs-kinase interaction network

[0081] Protein phosphorylation is usually regulated by kinases and phosphatases. Kinases add phosphate groups to amino acid residues of proteins, which is called phosphorylation; phosphatases remove phosphate groups from amino acid residues of proteins, which is called dephosphorylation.

[0082] iGPS software was used to predict the kinases that regulate DPSs in the above 195 DPPs (cutoff value>1). By deleting duplicate values, 321 kinases were finally identified as potential kinases of 195 DPPs in glioma angiogenesis. Among these kinases, 50 belonged to the AGC kinase family, 11 belonged to the atypical kinase family, 70 belonged to the CAMK kinase family, 12 belonged to the CK1 kinase family, 60 belonged to the CMGC kinase family, 46 belonged to the STE kinase family, 28 belonged to the TK kinase family, 30 belonged to the TKL kinase family, and 14 belonged to other families ( Figure 6 ).

[0083] The DPPs-kinase interaction network was constructed using the STRING online database and visualized using Cytoscape software. In the DPPs-kinase network, DPPs and kinases were labeled in different colors and forms. The interaction correlation between DPPs and their upstream kinases was analyzed using the PPI network. The results showed that the predicted kinases and DPPs had a close protein-protein interaction relationship ( Figure 7 ), further demonstrating that these kinases have potential regulatory effects on DPPs.

[0084] The predicted kinases were combined with the results of the above iTRAQ quantitative proteomics study on proteomic changes in glioma neovascularization tissues for analysis ( Figure 8), and found that the protein expression levels of 14 kinases were changed in glioma neovascularization, including calcium / calmodulin-dependent protein kinase Ib(KCC1A), misshapen like kinase 1(MINK1), protein kinase N1(PKN1), protein kinase N2(PKN2), solute carrier family 12member 5(KCC2B), serine / threonine kinase 10(STK10), CaM kinase like vesicle associated(CAMKV), kalirin RhoGEF kinase(KACRN), epidermal growth factor receptor(EGFR), interleukin 1receptoraassociated kinase 4(IRAK4), pyruvate dehydrogenase kinase 2(PDK2), eukaryotictranslation initiation factor 2alpha kinase 2(E2AK2), mitogen-activatedproteinkinase kinase kinase 10 (M3K10), Bruton tyrosine kinase (BTK). These kinases were named differentially expressed kinases (DEKs).

[0085] Further analysis of the DPPs regulated by DEKs revealed that eight DPPs (MYLK, MYL12A, FLNA, TJP1, CAV1, CAV2, MYH11, and CAMK2D) enriched in the cell adhesion signaling pathway were regulated by the above DEKs; two DPPs (CAMK2D and PDHA1) enriched in the HIF-1 signaling pathway were also regulated by the above DEKs ( Figure 8 ).

[0086] 2. Construction of DPPs-kinase-small molecule drug network

[0087] For potential targeted therapeutic drugs, the upstream kinases of the hub molecules were selected, and the MoA module of the connectivity map (cMAP, https: / / clue.io / command) online database was used to perform small molecule drug analysis on the predicted kinases. Due to the large number of kinases, drug prediction analysis was performed on 270 regulatory kinases of the core hub molecules in DPPs, and 280 small molecule drugs targeting 115 of these kinases were finally identified.

[0088] Based on the interaction between DPPs and kinases, and between kinases and small molecule drugs, the Cytoscape software was used to construct a hub DPPs-kinase-chemotherapeutic drug network to further screen the small molecule drugs with the most potential therapeutic value ( Fig. 9 ).

[0089] Since the DPPs regulated by DEKs are mainly involved in the cell adhesion signaling pathway, cell adhesion is considered to be the main signaling pathway affected by phosphorylation proteomics. Among the predicted kinases, 11 DPPs related to the cell adhesion signaling pathway regulated by 95 kinases were screened and obtained. The targeted small molecule drugs of these 95 kinases were predicted through the CMap biological online database, and finally 262 small molecule drugs targeting related kinases were determined. The cell adhesion-related DPPs-kinase-small molecule drug interaction network diagram was constructed to further screen the small molecule drugs with the most potential therapeutic value related to this signaling pathway ( Fig.10 ).

[0090] Through the DPPs-kinase-small molecule drug interaction network analysis, it was found that four kinases including casein kinase 2alpha2 (CSNK2A2), casein kinase 2alpha 1 (CSNK2A1), mitogen-activated protein kinase14 (MAPK14) and Pim-1proto-oncogene serine / threonine kinase (PIM1) were the most significant kinases regulating DPPs (connectivity with DPPs greater than 10). In addition, MYLK and CAMK2D were found to be upstream kinases regulating hub DPPs and phosphorylation modifications of cell adhesion pathway DPPs ( Fig.10 ).

[0091] Among the predicted small molecule drugs, 38 can target more than 10 protein kinases (connectivity with kinases is greater than 10), and are therefore considered to have the greatest potential for therapeutic value. The 38 small molecule drugs include staurosporine (STS), U-0126, PD-98059, SB-203580, 4,5,6,7-tetrabromobenzotriazole (TBB), LY-294002, bosutinib, SB-202190, KN-62, bisindolylmaleimide-ix, etc. Functional analysis revealed that these small molecule drugs are protein kinase inhibitors, including PKC inhibitors, MAP kinse-ERK kinase (MEK) inhibitors, microtubule-associated protein (MAP) kinase inhibitors, p38 MAPK inhibitors, casein kinase inhibitors, mTOR inhibitors, phosphatidylinositol 3-kinase (PI3K) inhibitors, DNA-dependent protein kinase inhibitors, phosphodiesterase inhibitors, Polo-like kinase (PLK) inhibitors, ABL tyrosine kinase inhibitors, BCRactivator ofRhoGEF and GTPase-Abl tyrosine kinase (BCR-ABL) kinase inhibitors, non-receptor tyrosine kinase inhibitors, calmodulin-dependent protein kinase inhibitors, purinergic receptor antagonists and cyclin-dependent kinase inhibitors (as shown in Table 4).

[0092] Table 4 Kinases and corresponding predicted small molecule chemotherapeutic drugs

[0093]

[0094]

[0095]

[0096]

[0097] 3. STS disrupts endothelial cell angiogenesis and glioma tumor progression

[0098] By integrating DEKs, kinases of cell adhesion signaling pathway-related DPPs, and targeted small molecule drugs of core hub DPPs kinases, STS was judged to be the phosphorylated proteome targeted drug with the most potential therapeutic value. Therefore, the function of STS in glioma angiogenesis was further studied.

[0099] The results of the cellular level study showed that STS treatment of human umbilical vein endothelial cells HUVEC (donated by Dr. Li Zhaoshui from Qingdao University) and immortalized human brain microvascular endothelial cells hCMEC / D3 cells (immortalized human brain microvascular endothelial cells) significantly inhibited the cell phenotype tubule formation ability in a dose-dependent manner ( Fig.11 ). Western blotting results further showed that STS treatment inhibited the phosphorylation of MYL12A serine 19 site (MYL12A-Ser19) in cells, and this site was identified as a highly phosphorylated state in glioma neovascularization ( Fig.12 ).

[0100] Mouse glioma cells GL261 were transplanted into the right axilla of nude mice, and STS was injected intraperitoneally after tumor formation for treatment analysis. BALB / c nude mice, 4-6 weeks old (female), were purchased from Nanjing Institute of Animal Model Biology. Animal experiments were approved by the Ethics Committee of Shandong First Medical University. All studies were performed in accordance with the regulations of the Association for Assessment and Accreditation of Laboratory Animal Care. 1×10 6 The mouse glioma cell line GL261 (gift from Professor Lu Ling of the School of Medicine, Ocean University of China) was subcutaneously injected into the right armpit of the mouse, and a tumor formed after 4 days. The mice were then randomly divided into three groups and the drugs were injected intraperitoneally according to the following scheme: the control group was injected with normal saline, and the other two groups of mice were injected with low-dose (0.3 mg / kg) and high-dose (0.6 mg / kg) STS, respectively, once a day for 7 consecutive days. The length (L) and width (W) of the tumor were measured and recorded every day, and the volume of the tumor was calculated according to the formula π×(L×W2) / 6. The results showed that both the high-dose and low-dose STS groups significantly inhibited the tumor growth rate ( Fig.13 ), however, there was no significant difference between the STS high-dose group and the low-dose group.

[0101] When the tumor tissue grows to a certain dimension, the mice are killed by resecting the neck, and the subcutaneous tumor tissue is obtained by dissection. After taking pictures and weighing, it is fixed with 4% paraformaldehyde, and the density of tumor neovascular tissue is detected by immunohistochemistry using the primary antibody of the new vascular tissue marker CD31.

[0102] The results showed that the tumor blood vessel density marked by CD31 was significantly reduced in the STS drug treatment group, and the tumor blood vessel density of the high-dose group mice was significantly lower than that of the low-dose group mice ( Fig.14 ). It can be seen that STS treatment significantly inhibits glioma tumor angiogenesis, thereby inhibiting tumor progression. This also shows that glioma angiogenesis is related to the phosphorylation of serine at certain DPPs sites.

[0103] Example 3 DPPs gene expression and prognostic analysis in glioma

[0104] Human GBM RNA sequencing and clinical data were obtained from The Cancer Genome Atlas (TCGA) online database (https: / / tcga-data.nci.nih.gov / tcga / ). The gene expression of 195 DPPs was analyzed by the RNA sequencing cohort of GBM in the TCGA database (TCGA-GBM), which included 144 tumor samples and 5 normal samples after removing duplicate samples, as well as samples without clinical data and patient clinical information, including survival, age, sex, postoperative, radiotherapy, tissue source site, and race. The annotation file and RNA sequencing file were integrated using the “tidyverse” R package. Differentially expressed genes (DEGs) were identified using the “Deseq 2” R package, with the identification threshold of fold change ≥ 2 and P-value < 0.05. For genes related to overall survival (OS) in GBM patients, the “Survival” R package was used to calculate the correlation between OS and the expression value of each gene in the tcga-glioma RNA sequencing cohort (P-value<0.05). For univariate and multivariate Cox regression analysis, clinical data including age, sex, and survival status were used to determine prognostic features.

[0105] From the TCGA-GBM cohort, 5223 DEGs were first identified, and the common molecular analysis was performed with 195 DPPs in the previous phosphorylation proteome. The results showed that the coding genes of 82 DPPs were differentially expressed in GBM tissues ( Fig.15 ). KEGG pathway enrichment analysis was performed on 82 common molecules. The results showed that the signaling pathways involved in these common molecules were the same as those involved in previous DPPs, mainly including cell adhesion, proteoglycans in cancer, and signaling pathways related to neurodegenerative diseases ( Fig.16 ).

[0106] The constructed PPI interaction network and core hub molecule analysis identified 23 core hub molecules, including glioma disease-related DPPs (MAPT, FLNA, CAMK2A, VIM, PAK1 and MAP1B) ( Fig.17Univariate regression analysis showed that the 23 core hub molecules included 15 prognostic factors, including SYN1, DMTN, HSPB1, proteinkinase cAMP-dependenttype II regulatory subunitbeta (PRKAR2B), STX1A, lymphocytespecificprotein 1 (LSP1), PAK1, L1CAM, MYL12A, TUBA1C, ANK1, IQ motifand Sec7domainArfGEF 1 (IQSEC1), LIM domain 4 (PDLIM4), reticulon4 (RTN4) and Sushi domaincontaining 5 (SUSD5). Fig.18 ). Multivariate regression analysis of 23 core hub molecules showed that three DPPs, ANK1, L1CAM, and LSP1, were identified as prognostic factors ( Fig.19 ).

[0107] Example 4

[0108] To verify the results of previous quantitative phosphoproteomic studies, clinical tissues of gliomas of different pathological stages (stages I-IV) were collected, and IHC was used to detect the differential phosphorylation change sites of PDHA1 in tumor VECs (phosphorylation levels of Ser293 and Ser300) as well as the expression levels of PDHA1 and upstream kinase PDK2 in quantitative proteomics.

[0109] Glioma tissues of pathological stage I to IV were obtained from the Department of Neurosurgery, Tumor Hospital Affiliated to Shandong First Medical University, and were pathologically diagnosed and identified. The tissues were fixed in order with 4% paraformaldehyde, dehydrated with different concentrations of alcohol and xylene (75% alcohol for 1.5h, 95% alcohol for 1h, alcohol for 1.5h, alcohol for 1h, xylene for 0.5h, xylene for 0.5h), and embedded in paraffin. Finally, the embedded tissues were cut into 5μm thickness and attached to slides. Before immunohistochemistry, the tissue sections were baked at 65℃ for 1h, and then gradually dewaxed and hydrated (xylene for 10min 3 times, alcohol for 5min, 90% alcohol for 5min, 80% alcohol for 5min, 70% alcohol for 5min, and distilled water for 3min 3 times). The tissue sections were repaired with citric acid buffer, 3% hydrogen peroxide inactivated endogenous enzyme activity, and blocked with 5% bovine serum albumin for 20min, and then incubated with specific primary antibodies at 4℃ overnight. On the second day, tissue sections were removed, placed at room temperature for 30 minutes, washed 3 times for 5 minutes with phosphate buffered saline containing 0.5% Tween-20, and then incubated at room temperature for 30 minutes with streptomyces antibiotic-peroxidase secondary antibody. After washing 3 times with PBST, incubated with diaminobenzidine and observed under a microscope. Finally, the stained sections were terminated with distilled water, retained with hematoxylin, ammonia cyanide, differentiated with hydrochloric acid and alcohol, and gradually dehydrated (70% alcohol for 3 minutes, 80% alcohol for 3 minutes, 90% alcohol for 5 minutes, 95% alcohol for 5 minutes, alcohol for 5 minutes, xylene for 5 minutes for a total of 2 times). Finally, the sections were dried and sealed with neutral resin. The protein expression level was detected after panoramic scanning with a pick scanner.

[0110] The results of iTRAQ quantitative phosphoproteomics analysis of the phosphorylation levels of PDHA1-Ser293, PDHA1-Ser300, and PDK2 in tumor blood vessels are shown in Figure 2. Fig. 20 shown.

[0111] The results of IHC studies showed that the positive staining rates of PDHA1-pSer293 and -pSer300 in glioma tumor VECs were 28.18% and 25.41%, respectively, while those in normal tissue VECs were 36.61% and 34.02%, respectively, indicating that PDHA1-pSer293 and -pSer300 were decreased in glioma tumor VECs ( Figure 21-22 The positive staining rate of protein kinase PDK2 in glioma tumor VECs was 35.76%, which was significantly lower than its positive staining rate of 43.97% in normal tissue VECs ( Fig.23 ). However, there was no significant difference in the positive staining rate of PDHA1 expression between tumors and normal VECs ( Fig.24 ).

[0112] The glioma samples were further divided into different subgroups based on the expression levels of PDHA1-pSer293, PDHA1-pSer300 and PDK2, and their clinical staging results were statistically analyzed to analyze the relationship between the protein phosphorylation modification and glioma tumor progression. The results showed that the PDHA1-pSer293, -pSer300 and PDK2 low expression subgroups had a higher proportion of high-grade glioma samples, while there was no significant difference in the proportion of high-grade glioma samples between the PDHA1 low expression subgroup and the high expression subgroup ( Fig.25 ).

[0113] In summary, it can be seen that the 195 significantly differentially phosphorylated proteins identified and the related signaling pathways significantly affect glioma angiogenesis and change with tumor progression. They can serve as phosphorylated protein markers and potential therapeutic targets for glioma angiogenesis.

[0114] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A phosphorylated protein marker associated with glioma angiogenesis, characterized in that: Including ADD1, ANK1, ANK2, AQP4, ATP2B2, CAMK2A, CAMK2D, CANX, CAV1, CRYAB, DKC1, DMTN, EPB41, FBN1, FGA, F LNA, GAP43, HNRNPA1, HSP90AB1, HSPB1, KNG1, L1CAM, MAP1B, MAP2, MAPT, MBP, MYH10, MYH11, MYL12A, MYL K, NCAM1, NEFM, NES, PACSIN1, PAK1, PLEC, PSMA3, RBM10, RBM25, RPLP0, RPLP2, SF1, SF3A1, SNTA1, SORB S1, SPTB, SPTBN1, STXA1, SYN1, SYP, TAGLN, TGFB1I1, TJP1, TLN1, TPPP, TUBA1C, VIM, ZYX, PDHA1, and PDK2.

2. Use of a reagent for detecting the marker according to claim 1 in preparing a glioma diagnostic kit.

3. Use of a reagent for detecting the marker according to claim 1 in preparing a kit for evaluating the prognosis of glioma.

4. Use of the marker according to claim 1 as a screening target in screening drugs for treating glioma.

Images