Labeled-free quantitative proteomic analysis of cervical squamous cell carcinoma identifies potential protein biomarkers

4D-label free quantitative proteomics technology

In recent years, the rapid development of 4D-label free quantitative proteomics technology has brought a new breakthrough for the study of biopharmaceutical mass spectrometry. The technique quantifies protein expression levels by analyzing ionized peptides or protein fragments in a protein sample. It is based on mass spectrometry and allows for highly sensitive Quantitative analysis, which uses high-resolution mass spectrometry instruments and advanced data processing algorithms to detect and quantify trace amounts of proteins in samples, providing more accurate and reliable quantitative results. 4D-label free technology combines efficient sample preparation method with automated mass spectrometry platform, which can analyze a large number of samples quickly and improve the analysis efficiency and sample processing ability.

Clinical samples

The specimens were collected from five pairs of cervical squamous cell carcinoma and corresponding paracervical normal tissues from Renmin Hospital of Wuhan University. All samples were stored at −80 °C and a preliminary histopathological diagnosis was made. Cancerous tissues and normal cervical tissues were used for proteomic analysis. This study was approved by the Ethics Committee of Wuhan University Renmin Hospital (WDRY2022-K173). We have obtained written informed consent from study participants.

Protein extraction

An appropriate amount of SDT lysate was added into the tissue or microbial body precipitation, which was then transferred to Lysing Matrix A tube and homogenized by MP homogenizer (24 × 2, 6.0 M/S, 30 s, twice). After ultrasound, it was placed in a boiling water bath for 10 min. A total of 14,000 g was placed in a centrifuge for 15 min, and the supernatant was filtered with a 0.22 μm centrifuge tube to collect filtrate. The protein was quantified by the BCA method. Samples were stored at −80 °C.

SDS-PAGE electrophoresis

A total of 20 μg protein from each sample was added to 6X loading buffer, placed in a boiling water bath for 5 min, 12% SDS-PAGE electrophoresis (constant pressure 250 V, 40 min), and Coomasil bright blue staining.

FASP enzymatic hydrolysis

A total of 100 ug protein solution was taken for each sample, and DTT was added to the final concentration of 100 mM, boil water for 5 min, and cool to room temperature. A total of 200 μL UA buffer was added and mixed well, then transfered to a 30 kD ultrafiltration centrifuge tube to be centrifuged 12,500 g for 15 min, then the filtrate was discarded (this step was repeated once). Then, a total of 100 μL IAA buffer (100 mm IAA in UA) was added and shaken at 600 rpm for 1 min, react at room temperature without light for 30 min, centrifuge 12,500 g for 15 min. A total of 100 μL UA buffer should be added and 12,500 g should be centrifuged for 15 min. This step was repeated twice. A total of 100 μL 50 mM NH4HCO3 solution was added 12,500 g was centrifuged for 15 min, and this step was repeated twice. The collection tube was replaced, 40 μL Trypsin buffer (4 μg Trypsin in 40 μL 50 mM NH4HCO3 solution) was added, and oscillated at 600 rpm for 1 min, and placed at 37 °C for 16–18 h. A total of 12,500 g was centrifuged for 15 min; then, 40 μL 50 mM NH4HCO3 solution was added, 12,500 g from the center for 15 min, and the filtrate was collected. The peptide was desorted using C18 Cartridge, lyophilized, redissolved in 40 μL 0.1% formic acid solution, and quantified (OD280).

Mass spectrum analysis

GO function annotation and KEGG path analysis (Ding et al., 2022)

Gene ontology (GO) function annotation analysis was performed on all identified proteins, and the GO function of cell components, biological processes, and molecular functions corresponding to all proteins were analyzed. see http://www.geneontology.org for more information. Genomic (KEGG) pathway analysis was performed on the selected DEPs using the online analysis software Omicsbean (http://www.omicsbean.cn/). The data is normalized based on Z-Score and then clustered. The cluster analysis was drawn using R script (Complex-Heatmap 2.4.3).

GSEA analysis (Li et al., 2024)

Gene set enrichment analysis (GSEA) was performed using GSEA v.4.2.3 (Broad Institute, Cambridge, MA, USA, https://www.gsea-msigdb.org/gsea/). We set parameters and run enrichment tests on a gene-set database to obtain permutation number (1,000) and permutation type (phenotype).

Immunohistochemistry and western blotting

We analysed the protein expression of three molecules (DSP, PPP1R13L and ANXA8) by immunohistochemistry and western blotting in normal and cervical cancer tissue.

Statistical analysis

Statistical analysis was performed with Statistical Program for Social Sciences (SPSS) (SPSS, version 22.0; SPSS Inc., Armonk, NY, USA). The significant difference was analyzed with a t-test between two groups, P-value < 0.05 was considered statistically significant. Proteins with P-value < 0.05 and fold changes (FC) >1.5 were considered as differentially expressed proteins.

GEPIA analysis

GEPIA is an online website for analyzing the RNA sequencing expression data of tumor and normal tissue data from the TCGA and the GTEx databases (http://gepia.cancer-pku.cn/) (Tang et al., 2017). We used the GEPIA database to validate the expression of DSP, PPP1R13L and ANXA8 in human cancers and normal cervix.

Differential expression of the proteins

In the analysis of significant differences in the quantitative results, we first filter the data within the sample group where at least half of the repeated experimental data are non-null values for differential comparative analysis. Proteins meeting the criteria of an expression difference multiple greater than 1.5 and a P-value less than 0.05 are considered differentially expressed proteins (DEPs). Compared to normal results, there are a total of 562 DEPs in the cervical cancer group, with 340 up-regulated and 222 down-regulated (Fig. 1A). The top 20 up-regulated and down-regulated DEPs are shown in Table 1. Cluster analysis found significant differences between the cervical cancer group and the normal group, and the data patterns of five samples in each group have high similarity, proving that some protein levels indeed differ between the two groups (Fig. 1B).

GO analysis of the differentially expressed proteins

To understand the function of different proteins and the signaling pathways they are involved in, we annotated 562 different proteins using GO and KEGG. GO analysis includes biological process (BP), cellular component (CC), and molecular function (MF) (Fig. 2A). The overall functional enrichment characteristics of all different proteins was revealed based on GO functional entries and displayed in the form of a bubble chart (Fig. 2B). To further understand the characteristics of these differentially expressed proteins, GO classification and enrichment analysis were conducted (Fig. 2C). In biological processes, the terms enriched the most are viral transcription, mRNA export from the nucleus, platelet degranulation, and extracellular matrix organization. In molecular function, proteins are mainly enriched in serine-type endopeptidase inhibitor activity, actin filament binding, structural constituents of the cytoskeleton, etc. In cellular components, most proteins are involved in focal adhesion, extracellular space, endoplasmic reticulum lumen, et al. To further understand the functions of the DEPs, we analyse the top 10 most enriched items in biological processes, cellular components, and molecular functions in each category for up-regulated or down-regulated differential protein enrichment analysis. The functions of these differential proteins may provide clues and theoretical foundations for further elucidating the pathogenesis and development of cervical cancer (Fig. 3).

KEGG analysis

The Kyoto Encyclopedia of Genes and Genomes (KEGG) is one of the databases commonly used for pathway research. It includes pathway information from various aspects such as metabolism, genetic information processing, environmental information processing, cellular processes, biological systems, human diseases, and drug development. Through KEGG pathway annotation for proteins with significant differential expression, it aids in understanding the metabolic or signal pathways these proteins may participate in. This reveals a series of changes from the cell surface to the cell nucleus, uncovering a series of biological events and acting factors involved in the process and indicating the potential biological consequences of the interruption or change in a particular process. Based on KEGG annotation results, the differential proteins in the pathways are sorted, displaying the TOP 20 KEGG pathway annotation results as shown in Fig. 4A. The KEGG pathway enrichment analysis method is similar to GO enrichment analysis. It uses the KEGG pathway as a unit, all qualitative proteins as a background, and analyzes and calculates the significant level of protein enrichment in each pathway through Fisher’s Exact Test, thereby determining the significantly affected metabolism and signal transduction pathways. The results are shown in bubble charts (Fig. 4B) and bar charts (Fig. 4C). KEGG enrichment analysis results suggest that significantly expressed proteins play important roles in complement and coagulation cascades, spliceosomes, coronavirus disease, nucleocytoplasmic transport, regulation of the actin cytoskeleton, cholesterol metabolism, etc. The functions of DEPs involved may provide clues and a theoretical basis for further clarifying the pathogenesis and development of cervical cancer.

GSEA analysis

The GSEA (Gene Set Enrichment Analysis) is not limited to differentially expressed proteins and directly uses the expression levels of all genes for analysis, effectively compensating for the deficiencies of traditional enrichment analysis. GSEA can also be used to determine whether a certain pathway is activated or inhibited in a set of samples. GSEA operates by sorting all genes based on the degree of differential expression between two sample groups and then calculating the enrichment degree and significance at the top or bottom of the pre-defined gene set in this sorted list. The Enrichment Score (ES) reflects the degree of gene set members enriched at both ends of the gene-sorted list. The Normalized Enrichment Score (NES) is obtained by normalizing according to the size of the gene set, and the false discovery rate (FDR) value is calculated to control the false-positive rate. GSEA analysis for the GO term is performed, and the ES plot of the GO function with the highest NES value in the experimental group is shown in Fig. 5-1. The bubble chart is used to visualize the GSEA-GO enrichment results, selecting the top 20 functions by NES value and calculating the Gene_ratio of the GO function, sorted from largest to smallest. The most relevant are mRNA end processing regulation and RNA catabolic regulation, as shown in Fig. 5A. The ES plot of the GO function with the highest NES value in the experimental group, namely GOCC CAJAL BODY, is shown in Fig. 5C. For KEGG Pathway GSEA analysis, a bubble chart is used to visualize the GSEA-KEGG enrichment results, selecting the top 20 pathways by NES value and calculating the Gene_ratio of the KEGG pathway, sorted from largest to smallest, as shown in Fig. 5B. The most relevant are the ribosome and spliceosome. The ES plot of the KEGG pathway with the highest NES value in the experimental group, namely KEGG RIBOSOME, is shown in Fig. 5D.

Structural domain, transcription factor analysis, and sub-cellular localization analysis

A protein domain represents a specific structure and independent functional region within a protein, serving as a fundamental unit for protein structure, function, and evolution. Compared to the vast number of proteins, the number of structural domains is limited, each encompassing a finite number of members. Exploring these domains is crucial for understanding the biological function and evolution of proteins. Typically ranging between 25 and 500 amino acids in length, structural domains are analyzed in this study using the InterPro database, a comprehensive resource that integrates protein families, domains, and functional sites. The functional structural domains of differential proteins are analyzed, and the annotated results are enriched. The top 20 enriched domains are visually represented through bubbles (Fig. 6A) and bar graphs (Fig. 6B). The domains with the highest Rich factor are: Guanine nucleotide exchange factor, anaphylatoxin, complement system domain, complement_C3_C4_C5, anaphylatoxin, complement system, serpin, conserved site RNA recognition motif domain, serpin family, serpin domain, serpin superfamily, RNA-binding domain superfamily, etc. Transcription Factors (TFs) are proteins capable of binding specifically to certain sequences upstream of the 5′ end of a gene, ensuring the gene’s expression at particular intensities, times, and spaces. Given the significance of TFs, in-depth analysis and annotation of these proteins are conducted. The Animal TFDB (Animal Transcription Factor Database) encompasses information about animal TFs and TF families, enabling the prediction of whether a protein under observation is a TF and to which TF family it belongs. The differential proteins are annotated as TFs, and the top 10 TF families are statistically analyzed, including Myb_DNA-bd, CSD, Fork_head, STAT_bind, MBD, and HMG_box families, as depicted in Fig. 6C.

Sub-cellular localization refers to the specific location within a cell where a protein or expressed product is found. The main Sub-cellular locations in eukaryotic cells include extracellular space, cytoplasm, cell nucleus, cell membrane, mitochondria, Golgi apparatus, endoplasmic reticulum, peroxisomes, cytoskeleton, nucleoplasm, nuclear matrix, and ribosomes, among others. Understanding the sub-cellular localization of proteins is pivotal for functional protein research. Proteins must be transported to the correct cellular organelle to participate in various life activities and effectively execute their functions. In this study, the software WoLF PSORT is employed for Sub-cellular localization analysis of differential proteins. The results are illustrated in Fig. 6D, where the top three Sub-cellular locations are cytosol (29.6%), nucleus (26.0%), and extracellular (17.6%).

Protein-protein interaction network analysis (PPI)

Proteins do not exist independently within biological organisms; their functional roles are exerted through the regulation and mediation of other proteins. This regulatory or mediating action first requires binding or interaction between proteins. Research on the interactions between proteins and the networks formed by these interactions is essential for revealing protein function. For instance, highly aggregated proteins may possess the same or similar functions, and proteins with high connectivity may be key points affecting the entire system’s metabolism or signal transduction pathways. In this project, we selected proteins with a P-value less than 0.05 and the most significant expression (50 up-regulated and 50 down-regulated) as target proteins for direct interaction network analysis (Fig. 7A). Furthermore, these 100 target proteins were used with all identified proteins in this project for indirect interaction network analysis (Fig. 7B).

Verification of results of three differential proteins

Upon analysis of this proteome, among the top 20 significantly up-regulated and top 20 significantly down-regulated differential proteins (as shown in Table 1), this study identified three potential up-regulated proteins that may be related to the occurrence and development of cervical cancer: desmoplakin(DSP), protein phosphatase 1, regulatory (inhibitor) subunit 13 like (PPP1R13L), and ANXA8. In the immunohistochemical staining of DSP, PPP1R13L and ANXA8 in paracancer and cervical cancer tissues, the staining intensity of three proteins in cancer tissues was found to be enhanced, primarily expressed in the cytoplasm and cell membrane compared with paracancer control (Fig. 8A). In order to further clarify the expression levels of these three proteins in cervical cancer tissue, then we tested the protein expression levels in cervical cancer and paracancer tissues. The results showed that the protein expression of all the three protein levels (DSP, PPP1R13L and ANXA8) increased in cervical cancer tissue compared with the paracancer control (Fig. 8B), and further statistical analysis showed that the expression levels of the three different proteins in cervical cancer were significantly increased (Fig. 8C, P < 0.01). To further verify the validity of the data, we used the GEPIA database to validate the expression of DSP, PPP1R13L and ANXA8 in human cancers and normal cervix. We found that the transcription levels of DSP, PPP1R13L and ANXA8 increased significantly (Fig. 8D). Therefore, the results suggest that these three proteins may play a significant role in the occurrence and development of cervical cancer.