3 Results 3.1 Determination of the Optimal Inoculation Dose for PDCoV To obtain the optimal inoculation dose that could produce DEPs to the greatest extent, IPEC-J2 cells were inoculated with PDCoV at low, medium, and high MOIs (i.e., 0.01, 0.1, and 1 TCID50/cell, respectively) for 36 h before the formal experiment. By observing the CPE, we found that the high MOI resulted in a rapid synchronous infection of all the cells, causing majority of the cells detached and disintegrated within 12 hpi. This is not conducive to the interaction between the virus and the cells. By contrast, the cells inoculated with the low MOI only displayed minimal CPEs at the same time points, which might induce limited changes in the expression levels of diverse proteins. Therefore, the medium MOI of 0.1 was selected as the optimal dose for PDCoV inoculation in all subsequent experiments. 3.2 Selection of the Best Sampling Time for the Proteomic Analysis Following PDCoV Infection In order to determine the optimal time point for the proteomic analysis following PDCoV infection, IPEC-J2 cells were inoculated with PDCoV at an MOI of 0.1 and microscopically observed for CPE at 4, 8, 12, 24, and 36 hpi. Meanwhile, the cell supernatants at the same time points were collected and used to measure the proliferation dynamics of PDCoV in IPEC-J2 cells. Compared to mock-infected cells, PDCoV-inoculated IPEC-J2 cells began to exhibit slight CPE at 8 hpi, and the CPE gradually became increasingly evident as the infection progressed. Obvious CPEs were observed at 12 hpi and became more evident at 24 and 36 hpi, which were characterized by cell rounding, enlarging, and granular degeneration of the cytoplasm that occurred either singly or in different-sized clusters, usually forming cell masses, followed by cell shrinkage and increased detachment. These CPEs were in agreement with those reported by Jung et al.,8 and resembled those observed in PDCoV-infected swine testicular (ST) and LLC-PK1 cells.34 However, from 36 hpi onward, the majority of the cells became detached and floated in the medium (Figure 1A). The proliferation of PDCoV in IPEC-J2 cells was verified by IFA using a mAb raised against the PDCoV nucleocapsid protein, and the results demonstrated that almost all cells became infected at 24 hpi (Figure 1B). The one-step growth curve further revealed that the virus titer reached a plateau of ∼107.5 TCID50/mL at 24 hpi, followed by a gradual and continuous decline (Figure 1C). In general, the time point at which viral proliferation stays high but no obvious cellular membrane or cytoskeleton rearrangement occurs is the optimal sampling time for a proteomic analysis.22 On the basis of the above experimental results, we therefore chose 24 hpi as the optimal time point for the proteomic analysis of PDCoV-infected IPEC-J2 cells. Figure 1 Proliferation of PDCoV in IPEC-J2 cells. (A) Morphological changes in IPEC-J2 cells infected with the PDCoV CHN-HN-1601 strain at an MOI of 0.1 TCID50/cell or mocked infected for 4, 8, 12, 24, and 36 h, respectively. Scale bars, 10 μm. (B) Confirmation of PDCoV proliferation in IPEC-J2 cells by immunofluorescence assays using the mAb 1A3 specific for PDCoV (α-N) and an Alexa Fluor 488-labeled goat antimouse IgG, and mock-infected cells at 24 h was used as a negative control. Cell nuclei were counterstained with DAPI (blue). Scale bars, 10 μm. (C) One-step growth curve of PDCoV in IPEC-J2 cells at the indicated time points following viral infection. The titer of virus was presented as TCID50/mL, and the data were recorded as means ± SD from three independent experiments. 3.3 Identification of DEPs of IPEC-J2 Cells in Response to PDCoV Infection To identify the DEPs following viral infection, the total cellular proteins extracted from PDCoV- and mock-infected IPEC-J2 cells were processed for quantitative proteomics research using iTRAQ-coupled LC-MS/MS technique. In all, 5502 cellular proteins were identified in both PDCoV- and mock-infected IPEC-J2 cells at 24 hpi (Supplementary File S1). On the basis of the widely used criteria for judging DEPs (fold changes >1.2 or <0.83 and with p < 0.05),28,29 23 proteins were significantly upregulated and 55 proteins were markedly downregulated in PDCoV-infected IPEC-J2 cells in comparison with the mock-infected cells (Table 1). Meanwhile, six viral proteins, including the nucleocapsid protein, spike protein, membrane protein, 3C-like proteinase Nsp5, accessory proteins NS6 and NS7 (Supplementary File S2), were also identified in PDCoV-infected IPEC-J2 cells by searching against the porcine deltacoronavirus Uniprot database. In order to ensure the reliability of the obtained proteomic data, three biological replicates of PDCoV- or mock-infected cell samples were collected and three technical replicates were performed during the proteomic analysis. The difference was plotted against the percentage of the identified proteins, which suggested that the proteomic data had high credibility (Supplementary Figure S1). Notably, due to the fact that the current genome database of pigs is inadequately annotated in comparison to the human genome database, we found six uncharacterized or unassigned proteins among the 78 DEPs (Table 1). Therefore, a functional analysis of these proteins warrants further investigation. Table 1 Differentially Expressed Proteins Identified by iTRAQ Analysis of IPEC-J2 Cells in Response to PDCoV Infection protein name Uniprot accession no. log2 ratios (infection/control) peptides sequence coverage (%) p-values functions Upregulated proteins in PDCoV-infected cells ISG15 ubiqutin-like modifier I3LU39 1.24 18 30.1 3.62 × 10–06 RIG-1/MDA5 mediated induction of IFN-alpha/beta pathways Interferon-stimulated protein 60 F1SCY2 1.09 5 14.6 3.51 × 10–05 Defense response to virus 2′-5′ oligoadenylates synthetase 1, OAS1 F1RJN6 0.81 2 3.1 0.0001 Antiviral protein Radical S-adenosyl methionine domain-containing protein 2 F1S9L2 0.70 6 13.8 1.60 × 10–05 Defense response to virus Interferon induced protein with tetratricopeptide repeats 1, IFIT1 K7GN56 0.70 5 9.6 0.0002 Regulation of defense response to virus Family with sequence similarity 8 member A1 F1RUH9 0.69 1 2.2 0.0103 unknown uncharacterized protein I3LH89 0.658 1 4.7 0.0074 unknown Interferon-induced GTP-binding protein Mx1 K7GKN2 0.58 2 2.0 0.0013 GTPase activity; cellular response to type I interferon Nucleoporin 37 (NUP37) F1SRJ4 0.57 1 5.2 0.0016 Transporting macromolecules Nuclear transcription factor Y submit beta (NFYB) F1SG36 0.55 2 6.3 0.0013 DNA-binding transcription factor activity DNA excision repair protein ERCC-6-like isoform a (ERCC6L) I3LFY4 0.45 1 0.8 0.0496 DNA translocase activity Collagen type IV alpha 1 chain F1RLM1 0.45 1 0.6 0.0153 Component of glomerular basement membranes (GBM) Ribouclease T2 F1SBX8 0.35 2 4.4 0.0004 Ribinuclease T2 activity MRPL32 I3LTC4 0.34 1 11.7 0.0472 translation Phostensin Q767M0 0.32 3 6.8 0.0047 Phosphatase binding Bridging integrator 1 F1RXZ6 0.32 7 22.7 0.0012 Regulation of endocytosis Terpene cyclase/mutase family member I3L7C2 0.29 10 16.7 0.0012 Beta-amyrin synthase activity Secretagogin (SCGN) Q06A97 0.28 2 9.4 0.0214 Calcium binding Interleukin 13 receptor subunit alpha 2 K7GSC6 0.28 2 14.6 0.0370 Obsolete signal transducer activity and cytokine receptor activity FOS-like 1 (FOSL1) F1RU26 0.28 1 6.2 0.0214 DNA-binding transcription factor activity Prolactin regulatory element binding F1SED4 0.27 9 10.1 0.0229 GTPase activator activity CDK-activating kinase assembly factor MAT1 (MNAT1) F6Q8T7 0.27 1 6.4 0.0076 DNA-dependent ATPase activity Intraflagellar transport 81 F1RNN4 0.26 1 3.7 0.0033 Tubulin binding Downregulated proteins in PDCoV-infected cells Very low density lipoprotein receptor (VLDLR) E7CXS1 –0.27 21 22.8 0.0001 Apolipoprotein binding uncharacterized I3LFU8 –0.27 3 5.6 0.0218 unknown Myeloid leukemia factor 2 F1SLT1 –0.28 6 15.8 0.0488 DNA-binding Helicase-like transcription factor (HLTF) I3LM88 –0.28 1 1.9 0.0254 DNA-dependent ATPase activity Ras association domain family member 6 F1RUL8 –0.29 2 5.5 0.0011 Regulation of apoptotic process Nucleoside diphosphate kinase 7 (NME7) F1RPV8 –0.29 3 8.0 0.0279 Synthesis of nucleoside triphosphates Bromodomain adjacent to zinc finger domain 2A (BAZ2A) F1SLA2 –0.29 1 1.2 0.0189 Component of the nucleolar remodeling complex BMP2 inducible kinase I3LT15 –0.29 1 2.8 0.0319 Protein kinase activity RNA helicase (DDX55) F1RFL5 –0.29 1 1.5 0.0274 Catalytic activity TMEM55B F1S8H7 –0.29 1 3.9 0.0445 Catalytic activity Chromatin assembly factor 1 subunit A (CHAF1A) F1S7L7 –0.29 4 5.3 0.0252 Chromo shadow domain binding RNA polymerase II subunit A C-terminal domain phosphatase (CTDP1) F1RWS7 –0.29 2 5.5 0.0123 Promoting the activity of RNA polymerase II uncharacterized I3LIB8 –0.30 2 14.7 0.0373 unknown Intraflagellar transport 172 I3LPC6 –0.30 1 0.69 0.0383 Negative regulation of epithelial cell proliferation Myocyte enhancer factor 2D (MEF2D) F1RP31 –0.30 1 1.9 0.0418 Protein dimerization activity Poly(ADP-ribose) glycohydrolase F1SDW4 –0.31 1 1.3 0.0355 Poly(ADP-ribose) glycohydrolase activity Mitogen-activated protein kinase 4 F1STG5 –0.31 4 4.9 0.0233 Protein serine/threonine kinase activity Anaphase promoting complex subunit 7 (ANAPC7) I3L7Q8 –0.31 4 7.9 0.0218 Ubiquitin protein ligase activity Transmembrane protein 45A F1SKZ6 –0.31 1 3.6 0.0191 Modulates cancer cell chemosensitivity Katanin p60 ATPase-containing subunit A1 I3LVP8 –0.32 2 32.1 0.0181 ATPase activity Ubiquitin associated protein 2 (UBAP2) F1SEA5 –0.32 1 1.7 0.0237 Cadherin binding Ribosomal RNA adenine dimethylase domain containing 1 F1RHJ0 –0.32 1 2.3 0.0391 rRNA (adenine-N6,N6)-dimethyltransferase activity Lemur tyrosine kinase 2 F1RFL2 –0.32 1 1.0 0.0104 Protein serine/threonine kinase activity Poly(ADP-ribose) polymerase family member 10 F1RSK9 –0.32 1 1.8 0.0182 K63-linked polyubiquitin modification-dependent protein binding CDK9 C9E1C9 –0.32 4 11.0 0.0183 Cyclin-dependent protein serine/threonine kinase activity Cystatin C Q0Z8R0 –0.33 2 17.1 0.0282 Cysteine-type endopeptidase inhibitor activity Cyclin and CBS domain divalent metal cation transport mediator 2 F1S849 –0.33 1 1.1 0.0173 Adenylnucleotide binding Protein-serine/threonine kinase F1S069 –0.34 4 8.2 0.0247 Protein serine/threonine kinase activity Aamy domain-containing protein F1S5K2 –0.34 1 2.5 0.0390 Catalytic activity; amino acid transport GATOR complex protein WDR24 I3LF05 –0.34 1 1.4 0.0368 Regulation of autophagy; positive regulation of TOR signaling CD109 antigen isoform 1 preproprotein K7GKY0 –0.35 18 12.9 0.0105 Endopeptidase inhibitor activity Nonspecific serine/threonine protein kinase (AKT2) G9BWQ2 –0.35 7 12.0 0.0310 Transferase activity DNA polymerase delta interacting protein 3 (POLDIP3) F1SJQ4 –0.35 3 3.8 0.0287 RNA binding Exostosin-like glycosyltransferase 2 F1S568 –0.35 1 3.3 0.0215 Transferase activity 4F5 domain-containing protein I3LS25 –0.36 9 42.4 0.0015 Positive regulator of amyloid protein aggregation and proteotoxicity Integrator complex subunit 4 (INTS4) F1STY6 –0.37 3 2.4 0.0429 Involved in the small nuclear RNAs (snRNA) U1 and U2 transcription Smoothelin-like 2 F1RGN8 –0.37 1 2.7 0.0119 Actin cytoskeleton organization Elongation of very long chain fatty acids protein I3L7S8 –0.39 1 21.6 0.0444 Catalytic activity uncharacterized I3LBD1 –0.40 1 7.4 0.0085 unknown Secretory carrier-associated membrane protein F1SJ46 –0.41 7 13.1 0.0144 Protein transport Cadherin 6 F1SP42 –0.42 3 3.7 0.0122 Calcium ion binding Superoxide dismutase I3LUD1 –0.43 2 4.9 0.0139 Superoxide dismutase activity Probable ribosome biogenesis protein C16orf42 homologue (SMARCA2) F1RFZ6 –0.44 1 4.4 0.0470 Involved in ribosome biogenesis Protein zyg-11 homologue B F1S6I1 –0.45 1 3.4 0.0030 Positive regulation of proteasomal ubiquitin-dependent protein catabolic process BCL2 interacting protein 3 I3LDJ9 –0.45 2 6.6 0.0009 Protein homodimerization activity Alpha-(1,6)-fucosyltransferase (FUT8) F1SA54 –0.47 1 3.3 0.0492 Alpha-(1→6)-fucosyltransferase activity Signal peptide peptidase-like 2B F1S8G9 –0.52 2 1.9 0.0190 Protein homodimerization activity INTS3 and NABP interacting protein (INTS4) F1SNA6 –0.54 5 12.5 0.0006 DNA repair Trypsin domain containing 1 F1SUE6 –0.55 1 1.8 0.0215 Serine-type endopeptidase activity UDP-glucose glycoprotein glucosyltransferase 2 F1RP50 –0.62 2 2.3 0.0320 Glycoprotein glucosyltransferase activity Phosphatidylglycerophosphate synthase 1 I3LN95 –0.62 1 2.7 0.0440 Calcium ion binding Metallothionein-2A P79379 –0.63 9 32.8 0.0017 Metal ion binding uncharacterized I3LNY1 –0.66 1 6.5 0.0246 unknown Formin-like 1 I3LH80 –0.83 1 1.8 0.0191 Rho GTPase binding uncharacterized K7GL96 –1.13 1 10.9 0.0412 unknown 3.4 Validation of the DEPs by qPCR and Western Blot Analyses To validate the obtained LC-MS/MS data, qPCR was performed to evaluate the transcription levels of two randomly selected DEPs, the downregulated ANAPC7 and the upregulated IFIT1. To this end, IPEC-J2 cells were mock-infected or infected with PDCoV at an MOI of 0.1. At 24 hpi, total cellular RNA was extracted from the cells and subjected to qPCR assays. As shown in Figure 2A, the level of mRNA encoding ANAPC7 and IFIT1 proteins was significantly downregulated and upregulated in PDCoV-infected cells, respectively, as compared to the mock-infected cells (p < 0.05). The qPCR results were in agreement with the MS data which were acquired by the iTRAQ approach (Figure 2B). For further confirmation of the proteomic data, the expression level of ANAPC7 and IFIT1 proteins in IPEC-J2 cells, which were infected exactly as the aforementioned conditions, was also tested by Western blot analysis. To track the progression of PDCoV infection, the mAb 1A3 that specifically recognizes PDCoV was utilized. As shown in Figure 2C, compared with the mock-infected IPEC-J2 cells, PDCoV significantly decreased the expression of ANAPC7 protein and its relative ratio to β-actin in the cells, whereas the expression of IFIT1 protein and its relative ratio to β-actin in the cells were significantly increased as a consequence of PDCoV infection. The original images of the entire PVDF membranes containing the target Western blots were included in Supplementary Figure S2. The Western blot results were also consistent with the MS data (Figure 2D). Taken together, these experimental results reveal that our quantitative proteomics data are quite reliable. Figure 2 Validation of the LC-MS/MS results by Western blot analysis. (A) Quantitative real-time PCR (qPCR) analysis of the relative mRNA expression level of ANAPC7 and IFIT1 in IPEC-J2 cells upon PDCoV infection. IPEC-J2 cells were mock infected or infected with the PDCoV CHN-HN-1601 strain at an MOI of 0.1 TCID50/cell and collected at 24 hpi. Total RNA was extracted and reverse-transcribed into cDNA for the subsequent analysis via qPCR. Fold-change values were calculated based on the 2–ΔΔCt method, using β-actin as the housekeeping gene. Error bars indicate the standard error of three independent experiments (Student’s t test; *p < 0.05). (B) The relative ratio of ANAPC7 and IFIT1 mRNAs normalized to β-actin between PDCoV- and mock-infected cells was calculated based on the qPCR data. The iTRAQ ratio (PDCoV/Mock) obtained by MS analysis was simultaneously shown as a comparison. (C) Western blot (WB) analysis of the expression of ANAPC7 and IFIT1 proteins in IPEC-J2 cells upon PDCoV infection. IPEC-J2 cells were mock infected or infected with the PDCoV CHN-HN-1601 strain at an MOI of 0.1 TCID50/cell. At 24 hpi, the cells were harvested and processed for WB analysis using rabbit anti-ANAPC7, mouse anti-IFIT1 polyclonal antibodies and the mAb 1A3 specific for PDCoV. β-Actin was included as an internal loading control. The images shown are representatives of three independent experiments. (D) The optical intensity ratio between the corresponding bands (PDCoV-infected band/Mock band) was measured by densitometric scanning and normalized to the intensity of the β-actin bands in each experiment. The iTRAQ ratio (PDCoV/Mock) obtained by MS analysis was simultaneously shown as a comparison. 3.5 GO Functional Annotation of the DEPs To characterize the DEPs, GO analysis was conducted to annotate the proteins based on three major categories: biological process (BP), cellular component (CC), and molecular function (MF). Within the BP category, the proteins were predicted to be involved in 13 biological processes, including immune system process, reproductive process, biological adhesion, multiorganism process, detoxification, multicellular organismal process, developmental process, localization, and so on (Figure 3A; Supplementary File S3), among which those associated with multiorganism process, detoxification, and localization were significantly enriched (Figure 3B); within the CC category, the proteins were predicted to be primarily distributed within 9 different cellular components, such as synapse, extracellular region, membrane, organelle, and cell part (Figure 3A), with the significantly enriched being located in the membrane, organelle and cell part (Figure 3B); and within the MF category, the proteins were predicted to be linked with 8 molecular functions, for instance, structural molecule activity, transporter activity, and antioxidant activity (Figure 3A), but no GO term was identified as significantly enriched within this category (Figure 3B). Figure 3 GO functional annotation of the 78 differentially expressed proteins identified in IPEC-J2 cells upon PDCoV infection. (A) GO annotations for the upregulated and downregulated proteins. The proteins were annotated into three major categories: biological process (BP), cellular component (CC), and molecular function (MF). The abscissa text indicates the name and classification of GO terms. The pink and blue columns represent the upregulated and downregulated proteins, respectively, with the number of altered proteins being marked on top of each column. (B) GO enrichment analysis for the upregulated and downregulated proteins. The name and classification of each GO term are indicated in the abscissa. Each column denotes a GO term, and the height of the column represents the enrichment rate. The color implies the significance of the enrichment (p-value), and the darker the color, the more significant the enrichment of the GO term (Fisher’s exact test; *p < 0.05). 3.6 COG Function Classification of the DEPs To further characterize the DEPs, COG function classification was also applied to categorize the proteins. As shown in Figure 4, the DEPs could be further classified into 18 categories (Supplementary File S4). Among them, 9 proteins were related to general function prediction only; 8 proteins were involved in transcription; 6 proteins were relevant to signal transduction mechanisms; 5 proteins were associated with protein turnover, posttranslational modification, and chaperones; 4 proteins were linked to vesicular transport, intracellular trafficking, and secretion; 4 proteins were correlated with RNA processing and modification; 3 proteins were correlated with chromatin structure and dynamics; 3 proteins were associated with carbohydrate transport and metabolism. Seven proteins were respectively related to one of the following biological functions: nucleotide transport and metabolism; lipid transport and metabolism; translation, ribosomal structure, and biogenesis; cell wall/membrane/envelope biogenesis; inorganic ion transport and metabolism; extracellular structures; and cytoskeleton. Notably, another 7 proteins related to unknown function were also identified (Figure 4). Therefore, further investigation concentrating on the function of these cellular proteins is certainly worth trying in the future. Figure 4 COG function classification of the 78 differentially expressed proteins identified in IPEC-J2 cells upon PDCoV infection. The capital letters in abscissa denote the COG categories as marked on the right of the histogram and the ordinate indicates the number of proteins in each category. 3.7 KEGG Pathway Analysis of the DEPs To explore the underlying signaling pathways existing among the identified DEPs, KEGG pathway analyses were done to draw pathway maps.35 As shown in Figure 5A, the 78 identified DEPs were involved in 26 pathways, among which the top 5 involving more than three proteins were related to viral infectious diseases, signal transduction, immune system, digestive system and cancers. All the DEPs could be further classified into 6 KEGG pathway categories, including organismal systems, metabolism, human diseases, genetic information processing, environmental information processing, and cellular processes (Supplementary File S5). For the upregulated proteins, the top 20 relevant pathways were illustrated in Figure 5B and Supplementary File S6. The signaling pathways of interest included the RIG-I-like receptor signaling pathway, PI3K-AKT signaling pathway, endocytosis, pathways in cancer, etc. For the downregulated proteins, the top 20 relevant pathways were displayed in Figure 5C and Supplementary File S7. The signaling pathways of interest included the mTOR, MAPK, FoxO signaling pathways and so on. Interestingly, one upregulated protein (Secretagogin) and one downregulated protein (AKT2) were simultaneously involved in Fc gamma R-mediated phagocytosis. To further explore the possible involvement of the identified DEPs in the underlying signaling pathways, KEGG pathway enrichment analysis was performed. Our data demonstrated that the DEPs were primarily involved in the HIF-1 signaling pathway, HTLV-I infection, human papillomavirus infection, AGE-RAGE signaling pathway in diabetic complications, central carbon metabolism in cancer, influenza A, measles, Fc gamma R-mediated phagocytosis, small cell lung cancer, glycosaminoglycan biosynthesis, progesterone-mediated oocyte maturation, and relaxin signaling pathway (Figure 5D). These signaling pathways were mainly distributed in four distinct functional categories: environmental information processing, human diseases, metabolism, and organismal systems (Figure 5D). Figure 5 KEGG pathway analysis of the 78 differentially expressed proteins identified in IPEC-J2 cells upon PDCoV infection. (A) KEGG pathway classification of the 78 differentially expressed proteins. The ordinate text indicates the name of biological functions which were classified into 6 KEGG pathway categories, including organismal systems (OS), metabolism (M), human diseases (HD), genetic information processing (GIP), environmental information processing (EIP), and cellular processes (CP). The abscissa displays the number of proteins in each category. (B) The top 20 significant pathways of the significantly upregulated proteins. (C) The top 20 significant pathways of the significantly downregulated proteins. (D) KEGG pathway enrichment analysis of the differentially expressed proteins. The abscissa text displays the name and classification of the KEGG pathways. Each column represents a pathway, and the height of the column implies the enrichment rate. The color manifests the significance of the enrichment (p-value), and the darker the color, the more significant the enrichment of the pathway (Fisher’s exact test; *p < 0.05). 3.8 Protein–Protein Interaction Networks of the DEPs To explore the potential protein network connections between the identified DEPs, the web-tool STRING was applied to depict protein–protein interaction networks. As shown in Figure 6, the DEPs were mapped to three major functional interaction networks, among which two were tightly connected by a hub protein, ENSSSCG00000000860 (namely NUP37), while the third exists independently. For the two tightly connected networks, they were comprised of two groups of strongly interacted proteins, including SCGN-ISG15-OAS1-IFIT1-IFIT3-ANAPC7-NME7, which are associated with innate immunity, and POLDIP3-NUP37-ERCC6L-DDX55-SMARCA2-NFRB-BAZ2A-HLTF-CHAF1A-CDK9-MNAT1-CTDP1-INTS4, which are associated with cell cycle and cellular components. Of note, at least five proteins act as hub proteins in these two networks tightly connected, including IFIT1, IFIT3, NUP37, SMARCA2, and CDK9. Interestingly, MNAT1 interacted highly with CTDP1 and CDK9, and INTS4 was also well connected to CDK9 (Figure 6). For the third network, there were five proteins with strong interaction in response to PDCoV infection, including MEF2D, VLDLR, LOC780439, PDK1, and AKT2, which are related to cell death and survival. Taken together, these findings further indicate that various functional types of host proteins, various biological functions, and complicated protein networks were affected during PDCoV infection of IPEC-J2 cells, which should provide valuable clues for a better understanding of PDCoV pathogenesis. Figure 6 Protein–protein interaction networks of the 78 differentially expressed proteins identified in IPEC-J2 cells upon PDCoV infection. The networks were built using the STRING database with a minimum interaction score of 0.4 at medium confidence. Each node denotes a protein in the graph; each line is indicative of the interaction between two proteins, and the thicker the line, the closer the mutual relationship. The width of the edges represents the predicted strength of functional associations.