A label-free LC-MS combined with RPPA deep proteomics analysis method and its application

Abstract

The present invention provides a deep proteomics analysis method and application of label-free LC-MS combined with RPPA, belonging to the field of protein detection technology. The label-free LC-MS workflow provided by the present invention has no special requirements for LC-MS reagents and complex MS operation settings, and is combined in parallel with the RPPA proteomics technology to establish a streamlined workflow from sample preparation to data acquisition in tissue and cell line settings, expanding the coverage of protein targets discovered by tissue-based proteomics. The use of an integrated detection process improves sample utilization and optimizes the use of rare sample resources. Moreover, the two technologies complement each other to conduct more comprehensive and specific analysis and research on proteomics data, while also providing a more comprehensive chain of evidence for the mechanism pathways of TNBC tumors during their occurrence and development.

Description

Technical Field

[0001] The present invention relates to the technical field of protein detection, and in particular to a label-free LC-MS combined with RPPA deep proteomics analysis method and application. Background Art

[0002] Nowadays, quantitative proteomics, especially high-throughput proteomics technology, has become a cutting-edge analytical method for discovering new protein biomarkers in many disease settings. Proteomics takes the tumor proteome as the research object and analyzes the dynamic changes in the protein composition and change patterns in tumor cells from a holistic perspective. Proteomics is also one of the important methods for predicting tumor efficacy. People expect that clinical proteomics research will make a significant contribution to the development of better diagnostic and prognostic tools, the identification of new therapeutic targets, and ultimately the realization of personalized treatment for patients.

[0003] High-throughput proteomics technology has become a cutting-edge analytical method for discovering new protein biomarkers in many disease settings. However, common protein detection technologies, such as ELISA and western blot, cannot achieve high-throughput protein marker detection, nor can they mine high-throughput data to obtain more bioinformatics information such as mechanisms and pathways.

[0004] In addition, traditional detection only uses one detection platform, such as liquid phase tandem mass spectrometry (a biomarker and system for lung cancer detection CN115575636B), enzyme-linked immunosorbent assay ELISA (a system for diagnosing whether the subject has breast cancer and a biomarker CN115798712B), etc. to screen or detect protein biomarkers in cancer. In order to obtain a deeper and more comprehensive understanding of proteomics data, a single technology itself has its limitations, and it is not clear whether the technology is irreplaceable and comprehensive in data analysis results.

[0005] In addition, traditional testing is a technology corresponding to one test sample. It is difficult for the same sample to be used for multiple testing technologies at the same time, which is a waste of resources for many precious samples. Therefore, the diversity of samples also increases the difficulty and limitations of testing, and to a certain extent causes a waste of sample resources. If one sample can be used for joint proteomics testing of two technologies at the same time, the sample processing process can be optimized, the utilization rate of samples can be effectively improved, and the detection obstacles caused by sample scarcity can be reasonably avoided.

[0006] Mass spectrometry (MS)-based detection methods have become a key strategic choice for qualitative and quantitative detection of proteins in biological samples due to their fast analysis speed and high sensitivity. First, tandem mass spectrometry (MS / MS) has very high sensitivity because it can accumulate and capture precursor ions, and tandem mass spectrometry is usually less affected by chemical noise than MS scanning, making it easier to detect peptides than mass spectrometry. Second, due to the inherent complexity of proteomics analysis (i.e., a large number of peptide ions and charge states), tandem mass spectrometry can achieve higher detection specificity. In addition, the presence of multiple fragment ions has potential benefits for quantitative analysis, and more accurate detection results can be obtained (LC-MASS). At present, protein identification widely uses high-performance liquid chromatography (HPLC)-electrospray ionization (ESI)-tandem mass spectrometry (MS / MS) technology for peptide separation, and quantitative methods can be divided into label-based methods and label-free methods. Label-free methods have been widely used due to their low cost and the improvement of mass spectrometry resolution. However, mass spectrometry is affected by the mass accuracy of the substance. Reduced mass accuracy can lead to false positives, and the convolution spectrum caused by ion fragments of pseudo-isobaric peptides can easily affect the mass spectrometry accuracy.

[0007] Reverse Phase Protein Array (RPPA) technology is a high-throughput proteomics technology that combines planar high-precision large-scale sample protein antigen microarray printing and antibody detection. The RPPA technology platform is rigorously designed, the process is highly automated, and the quality control is extremely strict. RPPA can analyze up to 500 or more proteins of different abundance in 15 mg of tissue (the size of a grain of rice) at one time. It has ultra-high specificity and sensitivity that are unmatched by other high-throughput proteomics. It can perform systematic and in-depth detection of a large number of cell surface receptor proteins, key cell signaling proteins and protein modifications (phosphorylation, acetylation, methylation, etc.), proteases, transcription factors and other representative targets in different categories. It can also customize the target protein to be studied according to needs for analysis, including direct and indirect upstream and downstream protein network analysis. RPPA can be applied to different fields such as protein function and regulatory mechanism analysis, panoramic scanning of cell signal transduction, tumor markers and protein molecular typing, drug target discovery and drug mechanism analysis.

[0008] Although mass spectrometry-based proteomics has important prospects in biomarker analysis, no articles or inventions have been found to report related research on the combination of mass spectrometry and RPPA technology or technology comparison. In the field of cancer biology, reverse phase protein microarray (RPPA) has been listed as an excellent experimental method for cancer-related biomarker analysis. RPPA was originally designed to address changes in signaling pathways in cancer and is suitable for high-throughput targeted proteomics. Due to the highly sensitive antibody-based detection and amplification methods, RPPA focuses on proteins in key signaling pathways, transcriptional regulation and its modified protein forms. These proteins are low-protein expression bodies, but they have important functional determinants in pharmacodynamics and drug targets during carcinogenesis. Therefore, in addition to traditional mass spectrometry detection methods, RPPA technology is necessary to be included in the detection of cancer proteomics biomarkers. It can be used as a complementary technology to mass spectrometry detection, exploring the impact on proteomics and mechanism pathways from another technical dimension, and effectively increasing the coverage of protein targets. For data analysis, the use of a single proteomics detection method cannot explain a comprehensive and complete biological mechanism through GO and KEGG analysis.

[0009] In 2018, Gromov et al. published an article in MOLECULAR ONCOLOGY (A single lysis solution for the analysis of tissue samples by different proteomic technologies), introducing a single lysis solution for tissue sample analysis using different proteomics techniques (including tandem mass spectrometry and RPPA). However, this work used differential gel electrophoresis (DIGE) to separate proteins and matrix absorption laser dissociation to fragment proteins for mass spectrometry analysis, rather than liquid chromatography tandem mass spectrometry (LC-MS). In addition, the formulations and processing methods of the two lysis solutions are also different.

[0010] Patent CN 109725072 A introduces a targeted qualitative and quantitative metabolomics analysis method for screening cancer biomarkers based on LC-MS / MS technology. The invention only uses one technology, namely LC-MS / MS technology as the analysis method, and quantitatively analyzes the cancer-related core metabolites in plasma, and analyzes the possible secondary metabolites through KEGG.

[0011] Therefore, the present invention proposes a joint workflow using two commonly used high-throughput proteomics technologies, label-free mass spectrometry (label-free MS) and reverse phase protein array (RPPA), to enhance the coverage of protein targets discovered by tissue-based proteomics. It not only improves the utilization rate of samples, but also enables the two technologies to complement each other, conduct more comprehensive and specific analysis and research on proteomics data, and also provides a more comprehensive chain of evidence for the mechanism pathway information of TNBC tumors during the occurrence and development process. Summary of the invention

[0012] The purpose of the present invention is to provide a label-free LC-MS combined with RPPA deep proteomics analysis method and application.

[0013] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0014] The present invention provides a label-free LC-MS combined with RPPA deep proteomics analysis method, comprising the following steps:

[0015] The tissue to be tested is placed in a lysis buffer and fully ground and lysed, and the protein is quantified. Then, LC-MS / MS analysis and RPPA detection processes are performed respectively, and finally the test data are subjected to bioinformatics analysis.

[0016] Further, the composition of the lysis buffer is: 1-100 mM HEPES with pH 7.4, 10-800 mM NaCl, 0.1-10 mM EGTA, 1-100 mM sodium pyrophosphate, 10-1000 mM sodium fluoride, 0.1-10 mM MgCl 2 , 0.1-10% Triton X-100, 1-50% glycerol, 0.1-10 mM sodium orthocyanate, protease inhibitors, phosphatase inhibitors. The protease inhibitors and phosphatase inhibitors were tested at an addition concentration of 50 ml / template.

[0017] Furthermore, the mass volume ratio of the tissue to be tested to the lysis buffer is 1:5-1:50, the grinding and lysis parameters are: 1-4 cycles, each cycle is 15-60 seconds, the cooling time interval is 5-30 seconds, and the homogenized sample is centrifuged at 0-20°C and 10000-18000rpm for 10-20min.

[0018] Furthermore, the protein is quantified to a protein concentration higher than 1 mg / mL.

[0019] Furthermore, the mobile phase of the LC-MS / MS analysis is: 0.1% formic acid water for solution A and 80% acetonitrile for solution B, and the flow rate of the mobile phase is 300 nL / min; the data processing method of the data of the LC-MS / MS analysis is: using ProteomeDiscoverer 2.4 software to analyze the original files collected by the DDA method; using Spectronaut v17 software and human fasta files to analyze the original files collected by the DIA method.

[0020] Furthermore, the data processing method of the RPPA detection is: using MicroVigene software 5.6.0.8 to digitally transform the image; outputting text and image files using R package SuperCurve for SuperCurve fitting.

[0021] Furthermore, the bioinformatics analysis is as follows: all data generated by mass spectrometry and RPPA analysis are processed in R Version-4.1.0; all data cleaning, clustering, differential expression analysis, correlation analysis, and overlap analysis are mainly drawn using R packages: dplyr, ggplot2, pheatmap, clusterProfiler, limma, and VeenDiagram; correlation analysis is drawn using logarithmically transformed mass spectrometry data, and RPPA is drawn using normalized normlog2 data; for the differential expression of a single RPPA target, the Willcox rank sum test is used.

[0022] The present invention also provides an application of the above analysis method in cancer detection.

[0023] The label-free LC-MS workflow provided by the present invention has no special requirements for LC-MS reagents and complex MS operation settings, and is combined in parallel with RPPA proteomics technology to attempt to establish a streamlined workflow from sample preparation to data acquisition in tissue and cell line settings, expanding the coverage of protein targets discovered by tissue-based proteomics. The use of an integrated detection process improves sample utilization and optimizes the use of rare sample resources. Moreover, the two technologies complement each other to conduct more comprehensive and specific analysis and research on proteomics data, and also provide a more comprehensive chain of evidence for the mechanism pathways of TNBC tumors during their occurrence and development.

[0024] The present invention uses bioinformatics analysis processes such as volcano maps, heat maps, dot maps, Venn diagrams, GO maps, KEGG maps, etc. to explain in detail the differentially expressed proteins and downstream pathway activation patterns in mass spectrometry and RPPA, and explains that mass spectrometry data focuses on analyzing RNA splicing, translation regulation, and metabolic processes during tumor development, while RPPA focuses more on post-translational modifications and cell functions such as proliferation, adhesion, and apoptosis. KEGG pathway analysis also reveals the differences in their individual regulatory patterns. Mass spectrometry emphasizes neurodegeneration regulation, oxidation processes, and RNA / protein synthesis regulation, and RPPA is in a variety of cancers and key oncogenic signaling pathways such as EGFR, HER2, mTOR, FoxO, HIF-1, and PD-1 / DP-L1. Dynamic regulation.

[0025] In addition, the present invention also verifies the technical stability at the cellular level. For RPPA, since it focuses more on intracellular signal transduction, most of the top differentially expressed proteins are enriched in cell growth regulation, serine / threonine phosphorylation, TOR signal transduction and other cell differentiation and development processes. In general, these data further support the overall feasibility of our method in in vitro cell analysis in different states, and illustrate the technical robustness and complementarity of the two proteomics technologies in analyzing biological samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flowchart of deep proteomics analysis using label-free LC-MS combined with RPPA in Example 1;

[0027] Figure 2 The volcano plots of deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1, A is a volcano plot comparing differentially expressed targets of RPPA between TNBC and normal tissues, and B is a volcano plot comparing differentially expressed targets of MS between TNBC and normal tissues;

[0028] Figure 3 This is an unsupervised hierarchical clustering heat map of deep proteomic analysis of TNBC and paired normal tissues by label-free LC-MS combined with RPPA in Example 1. A is an unsupervised hierarchical clustering heat map of TNBC and paired normal tissues in MS; B is an unsupervised hierarchical clustering heat map of TNBC and paired normal tissues in RPPA;

[0029] Figure 4 This is a dot plot of the correlation ranking of matching targets of MS and RPPA between different samples by deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1, and the correlation ranking of matching targets of MS and RPPA between different samples;

[0030] Figure 5This is a correlation ranking dot plot of deep proteomics analysis of label-free LC-MS combined with RPPA for MS and RPPA matched differentially expressed targets in Example 1;

[0031] Figure 6 This is a Venn diagram of deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1, A is the number of differentially expressed genes enriched in MS and RPPA between TNBC and normal tissues; B is the number of total target genes enriched in MS and RPPA between TNBC and normal tissues; C is the number of genes in GO (MF) terms enriched in MS and RPPA between TNBC and normal tissues; D is the number of genes in the KEGG pathway enriched in MS and RPPA between TNBC and normal tissues;

[0032] Figure 7 The GOBiological Process of MS in the deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1 enriched differentially expressed targets;

[0033] Figure 8 The GOBiological Process of RPPA enriched differentially expressed targets in the deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1;

[0034] Fig. 9 KEGG pathway enrichment of differentially expressed targets of MS in deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1;

[0035] Fig.10 This is the KEGG pathway enrichment of differentially expressed targets of RPPA in the deep proteomics analysis of label-free LC-MS combined with RPPA in Example 1. DETAILED DESCRIPTION

[0036] The embodiments of the present invention are described in detail below, and the embodiments are intended to explain the present invention and should not be construed as limiting the present invention. If specific techniques or conditions are not specified in the embodiments, they are carried out according to the techniques or conditions described in the literature in this area or according to the product specifications. If the manufacturers of reagents or instruments are not specified, they are all conventional products that can be obtained commercially.

[0037] The following terms are used in this article: ACN, acetonitrile; AMBIC, ammonium bicarbonate; BCA, bicyclic acid assay; CF, correction factor; CPTAC, Clinical Proteomics Tumor Analysis Consortium; DDA, data-dependent acquisition; DIA, data-independent acquisition; DIGE, differential gel electrophoresis; EGFR, epidermal growth factor receptor; HEPES, 4-hydroxyethylpiperazineethanesulfonic acid; EGTA, ethylene glycol tetraacetic acid; ESI, electrospray ionization; FA, formic acid; FASP, filter-assisted sample preparation; FF, fresh frozen sample; GDC, common genomic data; GO, gene ontology database; HPLC, high performance liquid chromatography; KEGG, Kyoto Encyclopedia of Genes and Genomes; LC-MS, liquid chromatography-mass spectrometry; MALDI, matrix absorption laser dissociation ionization; MRM, multiple reaction monitoring; PTM, protein post-translational modification; RPPA, reversed-phase protein microarray; SDS, Sodium dodecyl sulfate; SP3, solid-phase enhanced sample preparation; TCGA, The Cancer Genome Atlas; TFA, trifluoroacetic acid; TMT, tandem mass tag; TNBC, triple-negative breast cancer.

[0038] Example 1

[0039] according to Figure 1 Perform protein detection according to the following steps:

[0040] 1) Sample source:

[0041] Samples were obtained from tumor tissue and paired normal tissue of surgical specimens. Paired normal tissue was obtained at least 2 cm away from the tumor edge.

[0042] 2) Preparation of lysis buffer:

[0043] The lysis buffer was prepared as follows: 50 mM HEPES (pH 7.4), 150 mM NaCl, 1 mM EGTA, 10 mM sodium pyrophosphate, 100 mM sodium fluoride, 1.5 mM MgCl2, 1% Triton X-100, 10% glycerol, 1 mM sodium orthocyanate, protease inhibitors (Roche 05056489001) and phosphatase inhibitors (Roche 04906837001). The lysis buffer was stored at -20°C and thawed on ice before use.

[0044] 3) Clinical sample preparation and protein quantification for dual proteomics analysis by LC-MS and RPPA

[0045] Fresh frozen TNBC tissue and paired normal tissue were removed from cryovials. Tissues were half-thawed and weighed before processing in grinding tubes. For each sample, cryolysis buffer was added at a ratio of 1:20 (w / v). The typical starting amount of tissue for RPPA analysis was 15 mg, and 70 mg per sample was used for LC-MS. Six ceramic beads were added to each sample grinding tube for tissue disruption and pre-cooled at -5°C. Tissue disruption was set to 2 cycles of 30 seconds each with a cooling interval of 10 seconds to prepare homogenates. The homogenized samples were centrifuged at 14,000 rpm for 15 min at 4°C and the supernatant was retained for protein quantification. Protein quantification was performed according to the standard BCA protocol (Pierce BCA kit 23225). Protein quantification was performed using a standard microplate reader (Biotek, Epoch2). Protein concentrations above 1 mg / ml were considered appropriate and continued for subsequent RPPA and LC-MS analysis.

[0046] ① Label-free liquid chromatography-mass spectrometry

[0047] Sample preparation: The protein extracts after solubilization were purified by acetone precipitation. Specifically, 80 μL of pre-cooled acetone was added to 50 μg of sample and placed at -20 °C overnight. The precipitated protein was precipitated by centrifugation at 16,000 × g for 10 min. The supernatant containing detergents and salts was removed. The protein was washed three times with pre-cooled acetone and dried at room temperature for 10 min. The sample was dissolved in a buffer consisting of 8 M urea and 50 mM ammonium bicarbonate (AMBIC), pH 7.8, and sonicated. Disulfide bonds were reduced with 5 mM dithiothreitol (DTT) at 30 °C for 1 h and alkylated with fresh 10 mM iodoacetamide at RT in the dark for 30 min. Urea was diluted in 4 × 50 mM AMBIC buffer and the ratio of mass spectrometry grade trypsin (Promega) to protein was 1:25 (w / w). After digestion at room temperature overnight, the sample was acidified with trifluoroacetic acid (TFA) and further purified with C18 stage tip. Each C18 tip was first equilibrated with 50 μL of 0.1% TFA, 80% acetonitrile (ACN) solvent, and then the sample was loaded into the C18 and washed twice with 55 μL of 0.1% TFA. Finally, the peptides were eluted from the C18 tip twice with 10 μL of 0.1% TFA, 80% ACN solvent and dried in a vacuum concentrator.

[0048] LC-MS / MS analysis

[0049] The purified samples were resuspended in 0.1% formic acid (FA) and loaded onto the autosampler of an Ultimate 3000 UPLC system (ThermoFisher Scientific). Each injection contained 250 ng of peptides and was separated on a homemade analytical column (30 cm, 100 µm ID, 1.9 µm C18). The LC mobile phase was set to 0.1% formic acid in water in solution A and 80% acetonitrile in solution B at a flow rate of 300 nL / min. For samples analyzed by data-dependent acquisition (DDA) method, the LC gradient and mass spectrometry parameters were set as follows: mobile phase B started at 4%, increased to 15% in 2 min, and gradually increased to 37.5% in 80 min. MS1 spectra were acquired in the mass spectrometer scan range of 300–1500 m / z, orbital resolution of 60K (200 m / z), maximum injection time of 30 ms, and normalized AGC target of 250%. MS2 spectra were acquired with a scan range of 200–1400 m / z, an isolation window of 1.6, an ion trap resolution of 15K (200 m / z), a normalized HCD energy of 27%, a maximum injection time of 22 ms, and a normalized AGC target of 100%. For data-independent acquisition (DIA), the parameters were set as follows: On the LC, mobile phase B started at 8% and increased to 37% over 120 min. Complete mass spectra were acquired in the Orbitrap with a scan of m / z 350–1150 at a resolution of 120K (200 m / z), an AGC target of 1E6, and a maximum injection time of 50 ms. For data-independent acquisition (DIA), the parameters were set as follows: On the LC, mobile B started at 8% and increased to 37% over 120 min. Full mass spectrometry was performed in the Orbitrap by scanning m / z 350–1150 at a resolution of 120K (200 m / z), an AGC target of 1E6, and a maximum injection time of 50 ms. A full mass spectrometric event was followed by 30 MS / MS windows in one cycle. The precursor was fragmented by HCD at a normalized collision energy of 32%. MS / MS was performed in the Orbitrap by scanning m / z 200–1600 at a resolution of 30K (m / z 200), an AGC target of 1E6, and a maximum injection time of 54 ms. To minimize the effects of LC-MS / MS system instability on the measurements, two technical runs were performed in random order for TNBC and control samples. Three biological replicates were analyzed for the cell samples.

[0050] Mass spectrometry data analysis

[0051] Raw files collected by the DDA method were analyzed using Proteome Discoverer 2.4 software (Thermo Fisher Scientific). Searches were performed by matching spectra against the UniProt Homo sapiens database. Parameters were set as follows: protease defined as trypsin, maximum missed cleavage of 2, minimum peptide length of 6, maximum peptide length of 44, precursor mass tolerance of 10 ppm, fragment mass tolerance of 0.02 Da, dynamic modifications as methionine oxidation and protein N-terminal acetylation, and static modifications as cysteine ​​carbamidomethylation. Percolator was used, and filtering parameters were set to a false discovery rate (FDR) of 1% at the peptide and protein level. Label-free quantification was based on extracted peptide peak areas, using a small feature mapper that allows matching features between runs. Unique peptides and razor peptides were used for quantification. Raw files collected by the DIA method were analyzed using Spectronaut v17 (Biognosys) software with human fasta files (Uniprot, UP000005640) with the following settings: enzyme was Trypsin / P. The false discovery rate (FDR) for both peptides and proteins was set to 1%. Cysteine ​​carbamidomethylation was a fixed modification. Protein N-terminal acetylation and methionine oxidation were defined as variable modifications. Normalization was based on total peptide amount. Protein abundance was calculated for the top 3 most abundant peptides. For the treatment of mass spectrometry missing data, we used the NAguideR package.

[0052] ② Reverse Phase Protein Array

[0053] Sample processing and RPPA workflow: Protein lysates were premixed with sample dilution buffer (50% glycerol, 4x SDS buffer plus 6 ml β-mercaptoethanol) to an equal final concentration of 1.5 mg / ml. The concentration-normalized samples were then diluted 2-fold in sample dilution buffer (lysis buffer, 50% glycerol, 4 x SDS buffer, 6 ml β-mercaptoethanol in a ratio of 3:4:1) and five serial dilution steps were performed at 1, 1 / 2, 1 / 4, 1 / 8, 1 / 16 on an automated liquid handling workstation (Tecan Fluent series). Samples were prepared in 384-well plates and then deposited on nitrocellulose-coated glass slides (Gracelab ONCYTEsuper NOVA) using a solid pin contact printer (Quanterix 2470 array). Slide controls of cell lines (with or without treatment) and a mixture of cell lines and tonsil tissue lysates were used as staining quality control (QC) and quantitative QC measures. Colorimetric signal quantification was then performed using a validated 305 antibody panel. Basically, slides were blocked with Re-Blot (Millipore) at room temperature, blocked with I-block (Fisher), and antigen retrieved with hydrogen peroxide (Fisher). Slides were then blocked sequentially with avidin, biotin, and protein block (DAKO) before incubation with the next antibody for 1 hour at RT. Rabbit or mouse secondary antibodies (DAKO) were then applied, followed by TSA (Akoya)-mediated signal amplification and DAB colorimetric visualization (DAKO). All staining procedures were automated on a DAKO Link 48 Autostainer (Agilent) with humidity controlled at 70%. Stained slides were scanned on a high-throughput slide scanner with a scanning resolution of 10um (Huron LE120), and images were used for downstream processing.

[0054] 4) RPPA data processing and analysis

[0055] Images were digitally transformed using MicroVigene software (version 5.6.0.8). The output text (txt) and image (tiff) files were fitted using the R package SuperCurve. This was done to generate expression data (rawlog2 files) and QC data for each slide. Correction factors (CF) were also calculated to assess sample outliers within and between experiments. For raw log2 data normalization, the median was subtracted from each column (antibody) in the column direction, and then the median was subtracted from each row (sample) in the row direction. This generates normalized log2 files that are further squared to generate linear data sets (Normlinear). All of these data sets were processed downstream for quantitative comparison and graphical visualization.

[0056] 10) Bioinformatics analysis

[0057] All data generated by mass spectrometry and RPPA analysis were processed in R (Version-4.1.0). All data cleaning, clustering, differential expression analysis, correlation analysis, and overlap analysis were mainly plotted using R packages (dplyr, ggplot2, pheatmap, clusterProfiler, limma, and VeenDiagram). External RPPA proteomics data (batch-normalized level 4 data) and corresponding metadata for TNBC were downloaded from Genomic Data Commons (GDC data portal: https: / / portal.gdc.cancer.gov / ). TNBC mass spectrometry data (qc by counts and normalized counts) and corresponding metadata were downloaded from Proteomic Data Commons (PDC data portal: https: / / proteomic.datacommons.cancer.gov / pdc / ). Correlation analysis was plotted using log-transformed mass spectrometry data, and RPPA was plotted using normalized log2 (normlog2) data. For differential expression of individual RPPA targets, the Willcox rank sum test was used. Unless otherwise stated, a p value of 0.05 was considered statistically significant. Figure 2 As shown, log2FC is set to 1.5 and 0.8 respectively. ), heat map (such as Figure 3 ), dot plot (as shown in Figure 4 , Figure 5 As shown in the figure, blue dots represent positively correlated targets, and black dots represent negatively correlated targets. ), Venn diagram (as shown in the figure Figure 6 ), Gene Ontology (GO) (as shown in Figure 7 , Figure 8 ), Kyoto Encyclopedia of Genes and Genomes (KEGG) (as shown in Fig. 9 , Fig.10 as shown).

[0058] Figures 2 to 7The bioinformatics analysis process detailed the differentially expressed proteins and downstream pathway activation patterns in mass spectrometry and RPPA, and explained that mass spectrometry data focuses on analyzing RNA splicing, translation regulation, and metabolic processes during tumor development, while RPPA focuses more on post-translational modifications and cell functions such as proliferation, adhesion, and apoptosis. KEGG pathway analysis also revealed differences in their individual regulatory patterns, with mass spectrometry emphasizing neurodegeneration regulation, oxidation processes, and RNA / protein synthesis regulation, and RPPA in a variety of cancers and key oncogenic signaling pathways such as EGFR, HER2, mTOR, FoxO, HIF-1, and PD-1 / DP-L1 dynamic regulation.

[0059] 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 label-free LC-MS combined with RPPA deep proteomics analysis method, characterized in that: The steps include: The tissue to be tested is placed in a lysis buffer for full grinding and lysis, protein quantification, and then LC-MS / MS analysis and RPPA detection processes are performed respectively. Finally, the test data are subjected to bioinformatics analysis; The mobile phase of the LC-MS / MS analysis is: 0.1% formic acid water in solution A and 80% acetonitrile in solution B, and the flow rate of the mobile phase is 300 nL / min; the data processing method of the data analyzed by the LC-MS / MS analysis is: using Proteome Discoverer 2.4 software to analyze the original files collected by the DDA method; using Spectronautv17 software and human fasta files to analyze the original files collected by the DIA method; The data processing method of the RPPA detection is as follows: using MicroVigene software version 5.6.0.8 to digitize the image; outputting text and image files to perform SuperCurve fitting using the R package SuperCurve; The bioinformatics analysis is as follows: all data generated by mass spectrometry and RPPA analysis are processed in RVersion-4.1.0; all data cleaning, clustering, differential expression analysis, correlation analysis, and overlap analysis are mainly drawn using R packages: dplyr, ggplot2, pheatmap, clusterProfiler, limma, and VeenDiagram; correlation analysis is drawn using logarithmically transformed mass spectrometry data, and RPPA is drawn using normalized normlog2 data; for differential expression of a single RPPA target, the Willcox rank sum test is used; The composition of the lysis buffer is: 1-100 mM HEPES with pH 7.4, 10-800 mM NaCl, 0.1-10 mM EGTA, 1-100 mM sodium pyrophosphate, 10-1000 mM sodium fluoride, 0.1-10 mM MgCl2, 0.1-10% Triton X-100, 1-50% glycerol, 0.1-10 mM sodium orthocyanate, protease inhibitors, phosphatase inhibitors, and the protease inhibitors and phosphatase inhibitors are tested according to the added concentration of 50 ml / template; The protease inhibitor is Roche 05056489001, and the phosphatase inhibitor is Roche 04906837001.

2. The analysis method according to claim 1, characterized in that: The mass volume ratio of the tissue to be tested to the lysis buffer is 1:5-1:50, the grinding and lysis parameters are: 1-4 cycles, each cycle is 15-60 seconds, the cooling time interval is 5-30 seconds, and the homogenized sample is centrifuged at 0-20°C and 10000-18000rpm for 10-20min.

3. The analytical method according to claim 1, characterized in that the protein is quantified to a protein concentration higher than 1 mg / mL.

4. Use of the analytical method according to any one of claims 1 to 3 in cancer detection.

Images