TMT labeled comparative proteomic analysis reveals spleen active immune responses during Clostridium perfringens type C infected piglet diarrhea

Ethics statement

This study was carried out in accordance with the recommendations of Institutional Animal Care and Use Committee (IACUC) of Gansu Research Center for Swine Production Engineering and Technology. The protocol was approved by the College of Animal Science and Technology, Gansu Agricultural University.

Animal infection experiment and sample collection

The animal experiment progress was consistent as the description of Huang et al. (2019), the detail was as following: thirty healthy 7-day-old piglets (Landrace × Yorkshire, Xitai breeding co., LTD. Gansu Province) were chosen with detected serum antibodies negatively for E. coli (enterotoxin and K88), Salmonella (typhimurium, typhisuis and choleraesuis) and C. perfringens type C β toxin antibodies by enzyme-linked immunosorbent assay (ELISA).

The C. perfringens type C strain (CVCC 2032, China Veterinary Culture Collection Center) was shaking cultured 16 h at 37 °C in the bouillon culture-medium (HopeBio, Qingdao, China) before used for infection. The colony-forming units (CFUs) of C. perfringens type C was determined by plate colony counting method, and finally an expected concentration of 1 × 10⁹ CFU/mL C. perfringens type C medium was used to inoculate piglets.

Twenty-five piglets were randomly selected to suffer oral inoculate administration of 1 × 10⁹ CFU/mL C. perfringens type C medium, the remaining 5 piglets were orally inoculated sterile medium as control group (SC), the experiment lasted for 5 days for the research report (Songer & Uzal, 2005) and the result of our pre-experiment. During the 5-day experimental period, all piglets were housed separately by special environmental control equipment, bodyweight, mental state, shape and color of feces and degree of diarrhea for each piglet were detailly observed and recorded detailly in every day. The degree of diarrhea was evaluated according to the diarrhea scoring standard of piglets (Huang et al., 2019): 0 point for strip or granular feces; one point for soft forming; two points for thick and unformed feces; three points for liquid and watery excrement.

At the end of the test, the total diarrhea scores of each piglet were calculated by adding each defecation score of each piglet during the test period. Finally, 3 piglets with the highest and lowest total diarrhea scores were considered as the susceptible group (SS) and t the tolerance group (SR) (Fig. S1). In general, SS represents in sensitive to C. perfringens infection with serious diarrhea, and SR represents which resistant to C. perfringens infection with mild diarrhea.

Finally, nine piglet spleen tissues of SS, SR and SC groups were rapidly collected without RNA enzyme and washed with PBS buffer frozen by liquid nitrogen and transferred to −80 °C for storage.

Protein extraction

Spleen samples were added and mixed by lysis buffer (8 M urea, 1% Protease Inhibitor Cocktail) for standing still for 30 min, then followed by sonication on ice three times using a high-intensity ultrasonic processor (Scientz). After fully lysing on ice 30 min, the remaining precipitated debris was removed by centrifugation at 12,000 g at 4 °C for 10 min. The supernatant was collected and protein concentration was determined by BCA kit (Beyotime biotechnology, Shanghai, China).

Trypsin digestion

Protein solution was reduced with 5 mM dithiothreitol for 30 min at 56 °C and then alkylated with 11 mM iodoacetamide for 15 min. Then, the protein sample was diluted by adding 100 mM TEAB to urea concentration less than 2 M for trypsin digestion with first digestion of 1:50 trypsin-to-protein ratio overnight and the 4 h second digestion of 1:100 ratios. Approximately 100 µg protein for each sample was used for the following experiments.

TMT labeling

After trypsin digestion, the peptide was desalted by Strata X C18 SPE column (Phenomenex) and vacuum-dried. Briefly, peptide was reconstituted in 0.5M TEAB and processed according to the manufacturer’s protocol for TMT kit (Thermo Fisher Scientific, Waltham, Massachusetts). Then, one unit of TMT reagent (defined as the amount of reagent required to label 100 µg of protein) was thawed and reconstituted in 24 µL ACN. Finally, peptide mixtures were incubated for 2 h at 37 °C, following desalting and drying by vacuum centrifugation.

Peptide fractionation by high-performance liquid chromatography (HPLC)

The tryptic peptides were fractionated by high pH reverse-phase HPLC using C18 column (5 µm particles, 10 mm ID, 250 mm length). According the standard analytical procedure, peptides were firstly separated with a gradient of 8% to 32% acetonitrile (pH 9.0) over 60min into 60 fractions, then combined into 18 fractions and 6 fractions, respectively, and dried by vacuum centrifugation.

LC-MS/MS analysis

The tryptic peptides were dissolved in 0.1% formic acid (solvent A), directly loaded onto a home-made reversed-phase analytical column (15 cm length, 75 µm ID). The gradient was increased from 6% to 23% solvent B (0.1% formic acid in 98% acetonitrile) over 26min, from 23% to 35% in 8min and climbing to 80% in 3min then holding at 80% for the last 3min. The constant flow rate was 400 nL/min with the EASY-nLC 1000 UPLC system.

Intact peptides were detected in the Orbitrap at a resolution of 70,000. Peptides were selected for MS/MS using a normalized collision energy setting of 28; the Peptides and peptide fragments were detected in the Orbitrap at a resolution of 17,500. A data-dependent procedure that MS scan were followed by 20 MS/MS scans was applied for the top 20 precursor ions above a threshold ion count of 1.5E4 in the MS survey scan with 15.0s dynamic exclusion. An electrospray voltage of 2.0 kV was applied. Automatic gain control (AGC) was used to prevent overfilling of the iontrap; 5E4 ions were accumulated for generation of the MS/MS spectra. For MS scans, the m/z scan range was 350 to 1,600. Data are available via ProteomeXchange with identifier PXD030745.

Protein identification and analysis

Maxquant software (v 1.6.4) was used for MS/MS data peptide identification and quantitation against UniProt Proteomes-Sus scrofa (UP000008227) with the following parameters: enzyme—Trypsin, peptide mass tolerance—± 10 ppm, fragment mass tolerance—± 0.02 Da, max missed cleavage—2, fixed modification—carbamidomethyl (C), variable modification—oxidation (M) and deamidated (NQ), quantification method—TMT10 plex report ion MS2, database pattern—decoy. Maximum false discovery rates (FDRs) were set to 1% for peptide and protein identification.

Screening differentially expressed proteins

The MS data validation was by mass error distribution of all identified peptides and peptide length distribution. Firstly, we checked the mass error of all the identified peptides. The distribution of mass error is near zero and most of them are less than 0.02Da which means the mass accuracy of the MS data fit the requirement. Secondly, the lengths of most peptides were distributed between 8 and 20, which agree with the property of tryptic peptides, meaning sample preparation reach the standard. The proteins with a log 2 (fold change) ≥1.3 and a P value < 0.05 were considered as differentially expressed protein.

Quantification validation of MS data

Only peptides unique for a certain protein were considered for TMT relative quantification, which was normalized using the average ratio of all the unique peptides in each sample. A two-tailed Fisher’s exact test was employed to test the enrichment of the differentially expressed protein against all identified proteins. A corrected P value < 0.05 was considered significant. For further hierarchical clustering based on different protein functional classification, the cluster membership was visualized by a heat map using the “heatmap.2” function from the “gplots” R-package.

Functional enrichment analysis

The different expressed proteins were performed to Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways analysis using DAVID (version 6.7). The protein functional annotation was derived from the UniProt-GOA database (http://www.ebi.ac.uk/GOA/) (Ashburner et al., 2000). The functional annotation and descriptions of identified protein domains were annotated by InterProScan (a sequence analysis application) based on protein sequence the InterPro (http://www.ebi.ac.uk/interpro/) domain database. The identified proteins domain functional description was annotated by InterProScanbased (http://www.ebi.ac.uk/interpro/) on protein sequence alignment.

Protein-protein interaction network

All differentially expressed proteins interactions, were analyzed against the Sus scrofa database by STRING version 10. Only the confidence score ≥ 0.7 (high confidence) of interactions was fetched. Interaction network form STRING was visualized in R package “networkD3”. Cytoscape was used for the visualization of the networks.

Statistical analysis

One-way ANOVA and Duncan’s multiple comparisons in SPSS 26.0 software were used to test the significance of difference between diarrhea scores of piglets, the weight of piglets before euthanasia, and the weight of heart, liver, spleen, lung and kidney after euthanasia in the groups of SR, SS and SC, Differences with a P < 0.05, the log2-transformed ratio is larger than 1.3 were considered as significant difference. Results were shown as the means and standard errors (mean ± SE).

Phenotypic characterization of piglets infected with C. perfringens type C

After the infection of C. perfringens type C, the inoculated piglets presented the varying degrees of diarrhea symptom. As shown in Fig. 1, the bodyweight, heart, liver, spleen, and kidney of inoculated piglets in the SS and SR groups were significantly lower than those of control group (SC) (P < 0.01). Meanwhile, the weights of body and heart, spleen, kidney and liver in SS group were the lightest. In addition, lung weight in SS and SR groups were significantly higher than that in SC group (P < 0.01, P < 0.05), and SS group has the heaviest lung weight.

Quantitative identification of spleen proteins for piglets

To comprehensively profile the landscape of spleen protein expression before and after C. perfringens infection, we had identified and quantified the differential proteome dynamics in three replicates during the C. perfringens infection. According to the experimental workflow (Fig. S1), we achieved comprehensive proteome profiling with deep coverage using TMT-labeled LC-MS/MS combined with pre-fractionation by HPLC, allowing for identification and quantification simultaneously. In total, 6,145 proteins can be identified, among which 4,958 proteins were quantified in the global proteome (Supplemental Table S1). In addition, many uncharacterized proteins were identified according to bioinformatics comparison results and the comprehensive annotation (Supplemental Table S2).

QC validation of MS data

To validate the quality of MS and MS/MS profiling, the mass error of all the identified peptides were checked, results demonstrated that the mass accuracy of the MS data and sample preparation fit the experiment requirement standard. The repeatability among the three groups was assessed by principal component analysis (PCA), relative standard deviation (RSD) and Pearson correlation coefficient, results indicated that there were the higher aggregation degree and correlation among SS, SR and SC groups (Figs. S2A, S2B, S2C), which proved that the proteomic analysis was robust and all data would be used for the following analysis with high quality.

Differentially quantified proteins identified among SC, SR and SS

The differentially expressed proteins and specific number for differential proteins between SR, SS and SC were analyzed. The proteins exhibiting a log 2 (fold change) > 1.3 and a P value < 0.05 were regarded as DEPs, then a total of 115 DEPs were identified in SR vs SC, of which 48 were up-regulated proteins and 67 were down-regulated proteins, 176 DEPs were identified with 83 up-regulated proteins and 93 down-regulated proteins in SS vs SC, 83 DEPs were identified, with 40 were up-regulated proteins and 43 down-regulated proteins in SR vs SC (Fig. 2, Table S3). The above differentiation of protein expression indicated molecular changes and potential functional transformation underlined C. perfringens infected piglets.

Bioinformatic analysis of differentially expressed proteins

The proteins with diverse biological processes and distinct subcellular locations performed significantly different molecular functions. There was a wide range of functional distribution of the DEPs among SS, SR and SS groups for each module, the three top were mainly located in extracellular, cytoplasm and nucleus, and detailly, extracellular enriched the most differential proteins for SR vs SC, cytoplasm for SS vs SC, and nucleus for SS vs SR (Fig. 3).

The heatmap comparisons of SS vs SC, SR vs SC, SS vs SR groups were shown in Fig. 4 and Table S4, which divided into biological process, cellular component and molecular function. The DEPs with same expression pattern were clustered together.

Functional enrichment analysis of DEPs

To explore the function of DEPs among SS, SR and SC groups, GO and KEGG functional enrichment analysis were performed. GO catalogs analysis of the DEPs among SR vs SC, SS vs SC and SR vs SS groups revealed that the most significantly enriched biological processes as acute inflammatory response, defense response to bacterium, leukocyte aggregation, defense response, cellular oxidant detoxification, antimicrobial response, positive regulation of cellular metabolic process, the main cellular component functions as extracellular space, protein-lipid complex, and the main molecular functions as tyrosine kinase binding, kinase binding, iron ion transmembrane transporter activity, Toll-like receptor 4 binding, RAGE receptor binding, nucleosomal DNA binding (Fig. 5, Table S4).

Specially, in the SS vs SR groups, the biological process included many immune and metabolism associated functions, such as defense response to virus, inflammatory response, regulation of response to virus immune response, defense response to other organism, antimicrobial humoral response, regulation of cytokine-mediated signaling pathway, regulation of apoptotic process and so on. The cellular components were mainly cell and organelle. And the molecular function mainly enriched in binding and enzyme catalytic activity of methyltransferase, enzyme regulator, peptidase regulator and endopeptidase inhibitor, translation initiation factor receptor binding.

Functional domains related to immunoglobulin-like fold, MHC classes I/II-like antigen recognition protein, calcium S100/CaBP-9k-type, Ferritin/DPS protein domain, Fibrinogen were conspicuously enriched in these DEPs (Fig. 6, Table S5).

KEGG analysis was performed to investigate the enriched pathways participated by the DEPs. A total of 30 KEGG signaling pathways were enriched among them, detailly, 13 signaling pathways for SR vs SC, 20 signaling pathways for SS vs SC and 5 signaling pathways for SS vs SR (Fig. 7, Table S6). In which, PPAR signaling pathway, IL-17 signaling pathway were highly enriched in SR vs SC and SS vs SC after C. perfringens infection, antigen processing and presentation, intestinal immune network for IgA production, hematopoietic cell lineage was specifically significantly in SS vs SC, RIG-I-like receptor signaling pathway were specifically significantly enriched in SS vs SR, these identified significantly signaling pathways were involved in the process of piglet spleen immune responses against C. perfringens infection.

Analysis of protein interaction network

In order to clearly display the interactions between proteins, a network constitute of top 50 closest interactions proteins of functional protein-protein interactions was built using STRING v.10.0 online software against the Sus scrofa database (Ashburner et al., 2000). A total of 34, 47 and 10 known or predicted interactions (PPI enrichment P-value < 1.0e⁻¹⁶) were formed among DEPs in the PPI network (Fig. 8). The prediction of the protein interaction network of DEPs showed that P09571, A0A287AMK0, A0A287AQG3, C3S7K6, Q6S4N2, F1RQW9, and Q8SPS7 all had the pivotal role in each PPI network. The highest number of interactions was observed for Fibronectin 1 (FN1), Serum albumin (Manyes et al., 2014).