Differentially expressed serum proteins in children with or without asthma as determined using isobaric tags for relative and absolute quantitation proteomics

Experimental design

The study proceeded according to the flowchart shown in Fig. 1.

Clinical information and serum sample collection

From September to October 2019, serum samples were collected from eight children who had received no treatment but who had been admitted to the Maternal and Child Health Hospital of Anhui Medical University Affiliated Hospital. No child included in the study had received a diagnosis of an immune disease, chronic kidney disease, or other disease affecting serum proteins. The included samples were collected from four children with asthma (experimental group, all samples were belongs to eosinophilic asthma, and obtained after the diagnosis immediately but before drug treatment) and four children without asthma (control group). All clinical diagnoses followed the 2019 Global Initiative for Asthma guidelines. The experiment was approved by the Medical Ethics Committee of Anhui Medical University (approval number 20200284), and the parents or guardians of all participants signed informed consent forms.

On the morning of the second day after the children were admitted to the hospital without drug treatment, blood samples (four mL) were collected, placed at room temperature (22–25 °C) in the dark for 1 h, and centrifuged at 3000 rpm at 4 °C for 15 min. The supernatant was transferred to a new tube with a pipette and stored at −80 °C until it was used in an experiment.

Protein extraction and quality control

A ProteoExtract albumin/IgG removal kit (Merck & Co.) was used to extract the serum samples. The total amount of protein extracted from each one serum sample was more than 400 µg. The protein bands were clear, complete and uniform. The protein was not degraded, and the total amount of protein in each sample was able to be used to do two or more experiments. The protein solution was prepared, and the Bradford Protein Assay working solution was added. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and Coomassie blue staining were performed to evaluate the sample quality after detecting the total amount of protein.

Labeling after enzymatic hydrolysis of proteins

We used iTRAQ techniques to label the peptide segments after enzymolysis (Sandberg et al., 2012). After protein quantification, 60 µg of the protein solution was placed in a centrifuge tube and 5 µL dithiothreitol solution was added. The mixture remained at 37 °C for 1 h. Next, 20 µL of iodoacetamide solution was added, and the mixture was placed at room temperature in the dark for 1 h. The samples were centrifuged, and the collected supernatant was discarded. The precipitated sediment was twice treated with 100 µL of uric acid buffer (8 M urea, 100 mM Tris–HCl; pH 8.0). The samples were washed with NH₄HCO₃ (50 mM, 100 µL) three times, and trypsin (the ratio of protein to enzyme 50:1) was added to the sample in an ultrafiltration tube. Enzymolysis was performed on the samples at 37 °C for 12–16 h. Finally, the samples were labeled and desalted using a C18 cartridge.

LC-MS/MS analysis

We identified differentially expressed proteins (DEPs) between the two groups by using an LC-MS/MS Spectrum system (Sandberg et al., 2012). After re-dissolving the labeled samples in 40 µL of 0.1% formic acid aqueous solution, we analyzed them using nano-LC-MS/MS. The mobile phases were phase A (2% acetonitrile/0.1% formic acid/98% water) and phase B (80% acetonitrile/0.08% formic acid/20% water). The column was equilibrated with 95% phase A liquid. The gradient from phase B was adjusted as follows: 0–80 min, linear increase from 0% to 40%; 80–80.1 min, increased to 95%; 80.1–85 min, maintained at 95%; 85–88 min, decreased to 6%. The separated samples were analyzed using mass spectrometry with a Q Exactive HF-X Mass Spectrometer (Thermo Fisher Scientific). Proteome Discover software, version 2.2.0.388, was used to search the Uniport Human database in FASTA format. Protein fold changes (FC) of at least 1.5 were obtained, and P < 0.05 was considered a statistically significant difference. A FC < 1.5 represented upregulated proteins and a FC < 0.667 represented downregulated proteins; a FC between 0.667 and 1.5 indicated no obvious change in protein expression between the two groups (Morrissey et al., 2013; Oliveira et al., 2020; Rawat et al., 2016).

Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Cluster of Orthologous Groups of Proteins (COG) Signaling Pathway Analyses

GO is a database that can be applied to various species to define and describe the functions of genes and proteins (Fang et al., 2019). The GO database is often used to clarify the roles of eukaryotic genes and proteins in cells. GO is useful for comprehensively describing the attributes of genes and gene products in organisms. GO consists of three domains (Liu et al., 2020; Wang et al., 2019a; Xing et al., 2020): (1) the cellular component domain contains the descriptions of proteins related to cell composition, which may be a subcellular structure (e.g., endoplasmic reticulum or nucleus) or a protein production component (e.g., ribosome, or proteasome); (2) the molecular function domain contains descriptions of all proteins related to molecular functions, such as biological activities and operations performed by specific gene products (i.e., molecules or complexes); (3) the biological process domain contains descriptions of all proteins related to biological processes, that is, a series of events in which molecular functions cooperate with one another. The GO function annotation result refers to the statistical number of DEPs detected between the two groups of serum samples in the three GO domains. The GO functional significance enrichment analysis provides the GO functional terms significantly enriched with the DEPs compared with all identified proteins, thus determining which biological functions the DEPs are significantly related to. The GO terms can explain the role of eukaryotic genes and proteins in cells, thus comprehensively describing the attributes of genes and gene products in organisms (Cai et al., 2015).

KEGG is a group of databases used to connect a series of genes in the genome with a molecular interaction network in cells to identify biological functions at the genomic and molecular levels. KEGG contains the signaling pathways of multiple cell processes, such as metabolism, membrane conversion, signal transduction, and the cell cycle (Kanehisa & Goto, 2000). The results of a KEGG analysis provides insights into the higher biological functions of cells. KEGG analysis can provide the most important biochemical metabolic pathways and signal transduction pathways involved in protein (Yang et al., 2018). COG annotates the functions of homologous proteins and includes both the COG database (clusters of homologous proteins from prokaryotes) and the KOG database (clusters of homologous proteins in eukaryotes) (Tatusov, Koonin & Lipman, 1997). COG database can provide the function of differential proteins (Wu et al., 2019). Briefly, using BLASTP (BLAST version 2.2.30, http://blast.ncbi.nlm.nih.gov/Blast.cgi), the sequence of protein set was compared with the COG database (the expected value of BLAST alignment parameter was set to 1 e-5), and the corresponding COG number of protein was obtained. The corresponding function description and function classification of protein were acquired.

Enzyme linked immunosorbent assay (ELISA)

Glucose-6-phosphate dehydrogenase (G6PD), the downregulated protein in the asthma group, was verified by ELISA kit. The experiment was performed according to the kit instructions. Blood samples were collected from children with or without asthma.

Statistical analysis

Two-tailed Mann–Whitney test was performed with SigmaPlot software. The chi-square test by Fisher’s exact test was used to compare the categorical variables (only gender data). Values are expressed as means ± SEM. A value of P < 0.05 was considered statistically significant.

Participants

In the participants of LC-MS/MS experiments, no significant difference in age (asthma: control, means ± SD, 4.0 ± 1.0 vs. 3.3 ± 0.6 years; n = 8, P = 0.5, gender (asthma: control, 3 males and 1 female vs. 2 males and 2 females, P = 0.5) or body mass index (asthma: control, means ± SD, 19.1 ± 0.6 vs. 17.9 ± 0.9 kg/m2; n = 8, P = 0.3) was detected between the experimental group (children with asthma) and the control group (children with trauma but no infection or asthma).

SDS-PAGE

Serum samples obtained from children with or without asthma were separated using SDS-PAGE. Total protein from eight samples was effectively separated without protein degradation within the molecular weight range of 15–220 kDa. The protein levels were sufficient to be used in subsequent experiments (Fig. 2).

LC-MS/MS analysis and identification of DEPs

LC-MS/MS is a powerful tool that enables identification of proteins in a mixed sample. In total, 103 proteins were identified in the serum of children with or without asthma, of which 46 DEPs were detected. As shown in Table 1, 12 DEPs were upregulated and 34 DEPs were downregulated. We plotted the magnitude of FC (log₂ FC) on the x-axis and the statistical significance of that change (−log₁₀ of the P value) on the y-axis to obtain a volcano plot (Fig. 3A). The cluster analysis for the expression of the DEPs clearly indicated that the expression patterns between children with or without asthma differed and that the protein expression in each group was clustered together (Fig. 3B). These results suggested that there was a significant difference in the levels of proteins expressed in the serum of children with or without asthma.

GO functional annotation and enrichment analysis

The GO functional annotation results of the DEPs between children with or without asthma are shown in Fig. 4. For the cellular component domain, 37 DEPs were mainly concentrated in the cellular anatomical term, among which the top 3 upregulated proteins were IGKV2-40, IGHV3-74, and IGKV1-27. For the molecular function domain, the role of 38 DEPs was mainly in binding, among which the top 3 upregulated proteins were IGKV2-40, IGHV3-74, and IGKV1-27. For the biological process domain, the DEPS were primarily involved in cellular processes (31 DEPs) and responding to stimuli (30 DEPs), with the top 3 upregulated proteins being IGKV2-40, IGHV3-74, and V1-19.

Significant enrichment analysis of the GO function refers to a rough understanding of the biological processes in which DEPs are enriched based on the simple annotation of genes, which increases the reliability of research focused on determining pathogenesis. By analyzing the GO functional significance enrichment map of the DEPs between children with or without asthma (Fig. 5), it was found that the detected DEPs play important roles as cellular anatomical entities and in binding, cellular processes, and responding to stimuli.

KEGG metabolic pathway analysis

In organisms, different proteins perform their biological functions in coordination with one another. Analyses based on metabolic pathways are helpful to further understand the biological functions of the DEPs. KEGG is the main public database used to analyze such pathways, and analyses using KEGG can determine the most important biochemical metabolic pathways and signal transduction pathways in which the DEGs participate. Here, KEGG functional annotation analysis was carried out on the DEPs identified in the serum samples from children with or without asthma. The results showed that KEGG pathways annotated with the DEPs included dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), phagosomes, regulation of the actin cytoskeleton, focal adhesions, and metabolic pathways (Fig. 6). The downregulated DEPs associated with DCM/HCM were ITGB1, ITGA2, ACTB, and CACNA2D1. The downregulated DEPs associated with phagosomes were ITGB1, ITGA2, ACTB, and MPO. The downregulated DEPs associated with regulation of the actin cytoskeleton were ITGB1, ITGA2, ACTB, and PFN1. The downregulated DEPs associated with focal adhesions were ITGB1, ITGA2, TNXB, and ACTB. The downregulated DEPs associated with metabolic pathways were AOC3, ALPL, PKM2, and G6PD. The pathway enrichment analysis is the same as the GO functional enrichment analysis, which uses the KEGG pathway as a unit and applies hypergeometric testing to determine the pathways significantly enriched with DEPs compared with all identified proteins. The most important biochemical metabolic pathways and signal transduction pathways of DEPs can be determined by pathway enrichment analysis (Fig. 7). The present analysis showed that most of the detected downregulated proteins were concentrated in metabolic pathways such as focal adhesion, MAPK signaling pathway, platelet activation and Rap1 signaling pathway.

COG protein functional analysis

COG is a database for orthologous classification of proteins. It is assumed that the group of proteins constituting each COG are all derived from the same ancestral protein and are divided into orthologs and paralogs. Orthologs refer to proteins evolved from vertical families (speciation) of different species and typically retain the same functions as the original proteins. Paralogs refer to proteins derived from gene replication in certain species that may evolve new functions related to the original functions. The DEPs between children with or without asthma were analyzed using the COG database (Fig. 8). We predicted the possible functions of these protein, and generated the functional classification statistics. The results showed that the functions of these DEPs were mainly concentrated in signal transduction mechanisms; posttranslational modification, protein turnover, and chaperones; amino acid transport and metabolism; cytoskeleton; and extracellular structures.

G6PD protein concentration in serum

To confirm the DEP in the LC-MS/MS experiment, we used an ELISA experiment and choose a key protein G6PD, which is decreased in asthma reported by studies from other groups (Hirasawa et al., 2003). Our data also showed a significant reduction in asthma group compared to control group (Fig. 9). Therefore, the result is consistent with the data from LC-MS/MS experiment.