key: cord-0899056-qc7umdxz
authors: Zeng, Songlin; Zhang, Huan; Ding, Zhen; Luo, Rui; An, Kang; Liu, Lianzeng; Bi, Jing; Chen, Huanchun; Xiao, Shaobo; Fang, Liurong
title: Proteome analysis of porcine epidemic diarrhea virus (PEDV)‐infected Vero cells
date: 2015-03-09
journal: Proteomics
DOI: 10.1002/pmic.201400458
sha: 17747929d55f7d936082e7b1459b9b4a54deea1e
doc_id: 899056
cord_uid: qc7umdxz

Porcine epidemic diarrhea virus (PEDV) causes an acute, highly contagious, and devastating viral enteric disease with a high mortality rate in suckling pigs. A large‐scale outbreak of PED occurred in China in 2010, with PEDV emerging in the United States in 2013 and spreading rapidly, posing significant economic and public health concerns. In this study, LC–MS/MS coupled to iTRAQ labeling was used to quantitatively identify differentially expressed cellular proteins in PEDV‐infected Vero cells. We identified 49 differentially expressed cellular proteins, of which 8 were upregulated and 41 downregulated. These differentially expressed proteins were involved in apoptosis, signal transduction, and stress responses. Based on these differentially expressed proteins, we propose that PEDV might utilize apoptosis and extracellular signal regulated kinases pathways for maximum viral replication. Our study is the first attempt to analyze the protein profile of PEDV‐infected cells by quantitative proteomics, and we believe our findings provide valuable information with respect to better understanding the host response to PEDV infection.

where PEDV was considered an exotic before 2013. The PEDV variant has emerged and spread rapidly in the United States since May 2013 [9, 10] . By the end of April 2014, PEDV outbreaks had been reported in 27 states of the United States, with 2692 confirmed cases (www.aasv.org/news/story.php? id = 6989), resulting in significant economic and public health concerns.

Despite years of intensive research, the mechanisms of PEDV pathogenesis and immunomodulation remain largely unknown [11] . Emerging technologies, such as proteomic approaches, have become important tools for facilitating a comprehensive characterization of virus-host interactions involved in infection and pathogenesis [12, 13] . Several quantitative proteomics approaches have been used to provide specific insights into the gene expression profiles of coronaviruses including severe acute respiratory syndrome coronavirus (SARS-CoV) [14] , infectious bronchitis coronavirus [15] , and transmissible gastroenteritis coronavirus [16] . Generally, in host cells coronavirus infection gives rise to alterations in transcriptional patterns, the cell cycle, the cytoskeleton, apoptosis pathways, the immune system, stress responses, and likely also causes inflammation [17] . More data related to the host response to coronavirus infection should shed some light on potential targets for antiviral agents.

To date, proteomics has not yet been applied to analyze whole cell changes caused by PEDV infection. Since Hofmann and Wyler [18] reported for the first time that PEDV could be successfully propagated in Vero cell cultures, Vero cells have been widely used for PEDV isolation, propagation, vaccine production, and basic research. In this study, we used iTRAQ coupled with LC−MS/MS to quantitate and identify cellular proteins that were differentially expressed in Vero cells following infection with PEDV.

African green monkey kidney (Vero) cells were purchased from the American Type Culture Collection (ATCC). Cells were cultured in DMEM (Invitrogen, Carlsbad, CA, USA) containing 10% FBS and 1% penicillin-streptomycin at 37ЊC/5% CO 2 . The Chinese PEDV strain AJ1102 (GenBank Accession No. JX188454) used in this study was isolated in 2011 from the intestine of a suckling piglet with acute diarrhea [5] . Virus stock was harvested in Vero cells and virus titer was 10 6.4 PFU/mL determined by plaque assays.

Monolayers of Vero cells were grown to 80% confluence in 10 cm 2 cell culture dishes. Culture medium was removed and monolayers gently washed with serum-free DMEM prior to inoculation. Cells incubated in serum-free DMEM containing 10 g/mL trypsin (Promega) were inoculated with PEDV strain AJ1102 at a multiplicity of infection of 0.1. Uninfected cells served as the mock-infected group. Infected and mock groups were collected as biological replicates from three independent experiments at 24 h postinfection (hpi). Viral propagation was confirmed by the observation of cytopathic effect (CPE) and indirect immunofluorescence assay (IFA) using a monoclonal antibody against PEDV nucleocapsid (N) protein. This monoclonal antibody was produced from hybridoma cells derived from Sp2/0 myeloma cells and spleen cells of BALB/c mice immunized with recombinant N protein of PEDV strain AJ1102.

Cultures of PEDV-infected and control cells, including three independent biological replicates, were collected using a cell scraper, centrifuged (300 × g, 10 min), and washed twice with ice-cold PBS containing 1 mM pervanadate and 1 mM sodium fluoride. Cell pellets were resuspended in 500 L of lysis buffer (7 M urea, 2 M thiourea, 4% CHAPS, 40 mM Tris-HCl, pH 8.5) containing 1 mM PMSF, 2 mM EDTA, and 10 mM DTT. Further protein solubilization was achieved by sonicating, with cellular debris removed by centrifugation (25 000 × g, 20 min, 4ЊC). The protein concentration in supernatants was quantified using the Bradford protein assay. We digested 100 g of each sample at 37ЊC for 12 h with sequencing grade modified trypsin (Promega, Madison, WI, USA). Mock-infected samples were labeled with iTRAQ 113 (IT113), iTRAQ 116 (IT116), or iTRAQ 119 (IT119), and PEDV-infected samples were labeled with iTRAQ 114 (IT114), iTRAQ 118 (IT118), or iTRAQ 121 (IT121). The labeled digests were then pooled and vacuumdried.

Peptide mixtures were loaded onto an UltremexSCX column (4.6 × 250 mm) and separated by SCX chromatography using a 20AB HPLC system (Shimadzu, Kyoto, Japan). The labeled peptides were eluted with 4 mL of buffer A (25 mM NaH 2 PO 4 in 25% (ACN), pH 2.7) for 10 min, followed by a linear gradient of 5-35% and 35-80% buffer B (25 mM NaH 2 PO 4 , 1 M KCl in 25% ACN, pH 2.7) for 11 and 1 min at 1 mL/min, respectively. The process of elution was monitored by the absorbance at 214 nm. A total of 20 fractions that were desalted with a StrataX desalting column were collected and lyophilized in a vacuum concentrator.

The equal amounts of digested proteins were dissolved in buffer A (2% ACN, 0.1% formic acid) at the final concentration of 0.25 g/L and analyzed in an LC-MS/MS system. After that, 9 L supernatant was loaded on a Symmetry C18 trapping column (5 m particles, 180 m inner diameter × 20 mm; Waters) with the nanoACQUITY Sample Manager (Waters) and the peptides were eluted from the trapping column over a BEH 130 C18 column (1.7 m particles, 100 m inner diameter × 100 mm). The samples were loaded at 2 L/min for 15 min, and then gradient run at 300 nL/min were from 5 to 35% buffer B (98% ACN, 0.1% formic acid) in 40 min followed by a wash for 5 min at 80% B and a 2-min equilibration step. MS/MS was performed with a Triple TOF 5600 mass spectrometer (AB Sciex, Concord, ON). The ion source was a NanoSpray III source (AB Sciex), and the radiator was a needle made of quartz (New Objectives, Woburn, MA, USA). During data acquisition, the spray voltage was set to 2.5 kV, the temperature of the interface heater was 150ЊC, with the curtain gas 30 pounds per square inch. Survey scans were acquired in 250 ms and 30 product ion scans were collected under the condition of exceeding the threshold of 120 counts per second (counts/s). Total cycle time was fixed to 3.3 s. Dynamic exclusion was set for half of peak width (18 s), and the fragmentation energy was set to 35 ± 5 eV.

The acquired MS raw data files were converted into MGF files by using 5600 ms converter and the MGF files were further searched. MS/MS data were searched against the monkey sequence database from the NCBInr database (release April 2013, containing 29 736 sequences) and the proteins were identified using MASCOT search engine (Matrix Science, London, UK; version 2.3.02). The search was conducted using trypsin as a specific enzyme. A maximum of one missed cleavage was accepted and Gln→pyro-Glu (N-term Q), Oxidation (M), iTRAQ8plex (Y) were chosen as variable modifications. The data acquired from the Triple TOF 5600 mass spectrometer were searched with a fragment mass tolerance of 0.1 Da and a peptide mass tolerance of 0.05 Da. The charge states of peptides were set to +2 and +3. An automatic decoy database search was carried out in MASCOT through choosing the decoy checkbox in which a random sequence of database was generated and tested for raw spectra and the real database. Peptide spectra matches were confirmed through percolator based on q-values at a false discovery rate of 0.6% [19] . Peptide identifications were grouped into proteins according to the law of parsimony and filtered to 0.6% false discovery rate [20] . Each of the confident protein identification involves at least two peptides and only unique peptides were used to quantify proteins. The quantitative protein ratios were weighted and normalized by the median ratio in MASCOT. For accurate comparison between samples, fold change > 1.5 or <0.667 and a p-value < 0.05 were considered as significantly different expressions according to the t-test. Bioinformatics analysis was conducted using Ingenuity Pathway Analysis software (IPA, www.ingenuity.com), GO, and UniProt.

Vero cells were seeded on glass coverslips (NEST Biotechnology) that had been placed in 24-well plates until cultures were 70-80% confluent. Cells were infected with PEDV (multiplicity of infection = 0.1) and fixed with cold 4% paraformaldehyde at 12, 18, 24, 30, and 36 hpi, respectively. The PEDV-infected cells were detected using a mouse monoclonal antibody directed against the PEDV N protein and FITC-conjugated goat anti-mouse IgG antibody (Sigma). Cell nuclei were counterstained with 0.01% DAPI (Invitrogen). The fluorescent images were examined by using a confocal laser scanning microscope (LSM 510 Meta, Carl Zeiss).

We chose to analyze hydroxysteroid dehydrogenase like 2 (HSDL2), tyrosine 3-monooxygenase/tryptophan 5monooxygenase activation protein (14-3-3) gamma, and HSPB1 (HSP27) by Western blotting as significant changes in their protein expression levels were observed. Cell lysates for PEDV-and mock-infected cultures were harvested at 24 hpi and protein concentrations determined. Equivalent quantities of cell lysates were from the three independent biological replicates denatured in 5× sample loading buffer by heating at 100ЊC for 10 min and separated by 12% SDS-PAGE. Separated proteins were electroblotted onto 0.45 m PVDF membranes (Millipore, Billerica, MA, USA) and blocked with 10% w/v skim milk in Tris-buffered saline containing Tween 20 for 2 h at room temperature. Membranes were incubated with a rabbit polyclonal antibody against HSDL2, 14-3-3 gamma, and HSP27 (Proteintech, China) at 4ЊC overnight. After washing three times with Tris-buffered saline containing Tween 20, membranes were incubated with the appropriate HRPconjugated secondary antibodies (Beyotime, China) at 37ЊC for 1 h. Detection was performed using Clarity ECL reagents (Bio-Rad, Hercules, CA, USA).

The kinetics of PEDV propagation in Vero cells was determined by monitoring CPE and viral protein expression, as well as viral titers at 12, 18, 24, 30, and 36 hpi. As shown in Fig. 1A , minimal CPE was visible at 18 hpi and obvious CPE could be observed at 30 hpi; at 36 hpi, almost all cells fell off. The results of IFA for detecting the expression of PEDV N protein showed that the majority of cells were infected at 24 hpi (Fig. 1B) . The one-step growth curve showed that viral titer reached to 6.1 log 10 at 24 hpi, peaked at 6.8 log 10 at 30 hpi, and then gradually declined (Fig. 1C ). To make sure higher proportion of the infected cells and to avoid excessive CPE, we thus chose 24 hpi as the time point under our infection conditions for further proteomic analysis.

Protein extracts were prepared from PEDV-and mockinfected Vero cells, and then treated as described in Section 2. We detected and quantified 3418 proteins, of which 8 were significantly upregulated and 41 downregulated based on our criteria for the identification of differentially expressed proteins. Among these 49 proteins, 45 proteins have at least two unique peptides, while 4 proteins (mitogen-activated protein kinase 3-like (MAPK3), adenylyl cyclase associated protein 1 isoform 2, alpha-enolase isoform 3, transportin-2-like) have only one unique peptide and one nonunique peptide. The annotated tandem mass spectra and the detailed information regarding protein assignments based on single-peptide assignments of the four proteins are provided as Supporting Information Data 1. The molecular functional classes and subcellular locations of the 49 differentially expressed proteins were analyzed using UniProt and the GO database (Supporting Information Table 1 ). The eight upregulated proteins in infected cells were localized to the membrane (one molecule), mitochondria (three molecules), nucleus (two molecules), membrane and mitochondria (one molecule), and cytoplasm and nucleus (one molecule). The 41 downregulated proteins were located in the cytoplasm and nucleus (12 molecules), cytoplasm (6 molecules), ribosome (5 molecules), unknown cellular location (4 molecules), cytoskeleton (4 molecules), nucleus (3 molecules), membrane (2 molecules), cytoplasm and membrane (2 molecules), cytoplasm, extracellular and nucleus (2 molecules), and mitochondrion and nucleus (1 molecule).

Based on gene identifications of the proteins in Table 1 , protein GI numbers and ratios (infection/control) were imported into the IPA tool. Their functional classification and interacting pathways were analyzed and reconstructed based on the underlying biological evidence from the literature. The analysis of function classification by IPA showed that the 8 significantly upregulated and 41 downregulated proteins could be divided among four distinct functional sets: diseases and disorders; molecular and cellular functions; physiological system development; and functions including toxicity functions ( Fig. 2 and Supporting Information Table 2 ). These differentially regulated proteins are involved in biological processes relevant to gene expression, cell cycle, cell signaling, cell death and survival, cellular function, and maintenance. With goal of exploring the potential protein network connections for the differentially regulated proteins in detail, IPA tool was used. The differentially regulated proteins were mapped to three specific functional networks. Each network had at least five "focus" proteins (significantly up-or downregulated proteins). The three networks of interest correspond to cancer, cell death and survival, cell-to-cell signaling, and interaction (Fig. 3A) ; cancer, cardiovascular disease, and hereditary disorder (Fig. 3B) ; RNA post-transcriptional modification, cellular developmental, cellular growth, and proliferation (Fig. 3C ).

We analyzed expression levels of HSDL2 (upregulated), 14-3-3 gamma (downregulated), and HSP27 (downregulated) in PEDV-infected Vero cells (Fig. 4) by Western blotting. The proteins were selected on the basis of interest and on their different ratios. The ratios of these three proteins between infected and uninfected cells were consistent with those determined using the iTRAQ labeled LC-MS/MS system.

Proteomics has become an important methodology for determining cellular protein interactions and host cellular pathophysiological processes following virus infection [12, 21] . To date, no analysis has been reported for the differential proteomes of host cells infected with PEDV. Based on our findings, the expressions of 49 different proteins were significantly altered in PEDV-infected cells. These proteins were involved in a variety of processes including apoptosis, signal transduction, and stress responses of host cells. Our data are overview of the altered proteins during a host response to PEDV infection and should provide clues regarding PEDV pathogenesis. The apoptosis of host cells plays a major role in regulating the pathogenesis of many infectious diseases [22] . Apoptosis triggered by virus infection directly leads to viral pathogenesis; however, blocking apoptosis can avoid premature death of the infected cells, permitting a high titer of virus replication or a persistent infection [17] . In our analysis, two apoptosis-related genes, (14-3-3 and caspase-8) were identified following PEDV infection. The 14-3-3 proteins constitute a family of conserved adaptor and scaffold proteins present in all eukaryotic organisms. Previous reports showed that 14-3-3 proteins modulate the cell cycle and apoptosis at multiple steps, leading to an inhibition of the onset of apoptosis [23, 24] . Our results demonstrate that decreased expression of four 14-3-3 proteins (eta, theta, beta/alpha, gamma) was concomitant with PEDV infection, and thus predicted to promote apoptosis of infected cells. Caspase-8 is the initiator caspase that is activated by the correlation between external apoptotic elements and cell surface molecules, playing fundamental roles in initiating the apoptotic cascade response [25] . The downregulation of caspase-8 expression in PEDV-infected cells suggests a mechanism that attenuates virus-induced apoptosis at the early infection phase, thus favoring a virus to replicate to a higher titer. In the case of SARS-CoV, stimulation of both apoptotic and antiapoptotic elements has been seen, allowing for rapid multiplication of progeny virus before cell lysis [26] . The interplay between apoptosis-related genes in PEDV infection might contribute to it achieving maximum viral replication before cell death.

Viruses are known to develop the ability to manipulate a variety of host cellular signal transduction pathways to facilitate virus survival. Among these pathways is the extracellular signal regulated kinases (ERKs) pathway, comprising Raf, MEK1/2, and ERK1/2. These play central roles in the modulation of cell survival, differentiation, and proliferation [27] . Many viruses activate the ERK pathway for higher levels of cellular and viral gene production. By activating ERK1/2, enterovirus 71 prevents activation of the proapoptotic molecules caspase-9 and Bcl-2 associated death promoter [28] . Porcine reproductive and respiratory syndrome virus infection activates ERK1/2 at the early infection phase in PAM cells, with ERK activation essential for porcine reproductive and respiratory syndrome virus replication in vitro [29] . Some studies have shown a negative correlation between ERK activity and viral replication. The Ebola virus glycoprotein GP specifically inhibits ERK1/2 to cause increased cell toxicity and viral titers [30] . Yoshizuka et al. reported that overexpression of the human immunodeficiency virus (HIV) type 1 (HIV-1) Vpr protein downregulated gene expression and phosphorylation of ERK1/2 in the ERK/MAPK pathway. It was speculated that the decrease in ERK1/2 activation is important for HIV replication and pathogenesis [31] . SARS-CoV N induces apoptosis of COS-1 monkey kidney cells in the absence of growth factors through downregulation of the ERK pathway [32] . ERK1/2 regulation of viral replication functions occurs at specific points during the replication cycle [33] . The manipulation of the ERK pathway is a ubiquitous characteristic during viral infection and is not limited to specific viruses or viral events [34] . In our study, ERK1 was significantly downregulated during PEDV infection. The specific mechanism by which PEDV manipulates the ERK pathway for maximal viral replication is currently under investigation in our laboratory.

Activation of the heat-shock response might be a specific function that a virus exerts to guarantee appropriate synthesis of viral proteins [35] . In our study, HSP27 was significantly downregulated in Vero cells following PEDV challenge. HSP27, a ubiquitous stress-inducible cellular protein, is a multifunctional protein that plays a role in the inhibition of caspase-independent apoptosis, actin cytoskeletal remodeling, and might also participate in protein degradation [36, 37] . It has been implicated that HSP27 is upregulated during enterovirus 71, feline herpesvirus, and infectious bursal disease virus infections [38] [39] [40] . Based on our results, HSP27 was downregulated during PEDV infection, similar to previous reports regarding adenoviruses [41] , classical swine fever virus [42] , and porcine circovirus type 2 [21] . HSP27 is an intracellular antiviral factor against hepatitis B virus through the induction of type I interferon and downstream antiviral effectors [43] and inhibits caspase-independent apoptosis. We suspected that decreased expression of HSP27 might have a specific role in host cells during viral infection, however that specific mechanism requires further investigation.

In summary, the proteomic changes in PEDV-infected Vero cells were characterized using iTRAQ combined with LC-MS/MS. To the best of our knowledge, this is the first time proteomics has been used to explore the pathogenhost protein interaction network in PEDV-infected Vero cells. Since the PEDV variant emerged in China and the United States, PEDV has received increasing attention, and more researchers have undertaken the study of PEDV. Our analyses of the proteins discussed above were descriptive, and further functional investigations are required to elucidate the pathogenic mechanisms and cellular responses to PEDV infection. Such molecular insights into differentially expressed cellular proteins would be helpful in studying the infectivity and pathogenesis of PEDV.

Cloning and sequence analysis of the N gene of porcine epidemic diarrhea virus LJB/03

An apparently new syndrome of porcine epidemic diarrhoea

Virus-like particles associated with porcine epidemic diarrhoea

New variants of porcine epidemic diarrhea virus, China

Complete genome sequence of porcine epidemic diarrhea virus strain AJ1102 isolated from a suckling piglet with acute diarrhea in China

Genetic properties of endemic Chinese porcine epidemic diarrhea virus strains isolated since

Complete genome sequence of a porcine epidemic diarrhea virus variant

Outbreak of porcine epidemic diarrhea in suckling piglets

Isolation and characterization of porcine epidemic diarrhea viruses associated with the 2013 disease outbreak among swine in the United States

Emergence of Porcine epidemic diarrhea virus in the United States: clinical signs, lesions, and viral genomic sequences

Epimedium koreanum Nakai water extract exhibits antiviral activity against porcine epidermic diarrhea virus in vitro and in vivo

Viral proteomics

Using SILAC and quantitative proteomics to investigate the interactions between viral and host proteomes

Proteomic analysis of SARS associated coronavirus using twodimensional liquid chromatography mass spectrometry and one-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by mass spectroemtric analysis

Comparative proteome analysis of tracheal tissues in response to infectious bronchitis coronavirus, Newcastle disease virus, and avian influenza virus H9 subtype virus infection

Identification of cellular proteome using two-dimensional difference gel electrophoresis in ST cells infected with transmissible gastroenteritis coronavirus

Biochemical aspects of coronavirus replication and virus-host interaction

Propagation of the virus of porcine epidemic diarrhea in cell culture

Enhanced peptide identification by electron transfer dissociation using an improved Mascot Percolator

Interpretation of shotgun proteomic data: the protein inference problem

Differential proteome analysis of host cells infected with porcine circovirus type 2

Classical swine fever virus infection protects aortic endothelial cells from pIpC-mediated apoptosis

14-3-3 proteins: key regulators of cell division, signalling and apoptosis

14-3-3 proteins block apoptosis and differentially regulate MAPK cascades

Caspases: the executioners of apoptosis

Comparative host gene transcription by microarray analysis early after infection of the Huh7 cell line by SARS coronavirus and human coronavirus 229E

The MAPK signalling pathways and colorectal cancer

Phosphorylation of PI3K/Akt and MAPK/ERK in an early entry step of enterovirus 71

Porcine reproductive and respiratory syndrome virus replication is suppressed by inhibition of the extracellular signal-regulated kinase (ERK) signaling pathway

The ERK mitogen-activated protein kinase pathway contributes to Ebola virus glycoprotein-induced cytotoxicity

Human immunodeficiency virus type 1 Vpr-dependent cell cycle arrest through a mitogen-activated protein kinase signal transduction pathway

The SARS coronavirus nucleocapsid protein induces actin reorganization and apoptosis in COS-1 cells in the absence of growth factors

Suppression of coronavirus replication by inhibition of the MEK signaling pathway

Suppression of astrovirus replication by an ERK1/2 inhibitor

Activation of heat-shock response by an adenovirus is essential for virus replication

Heat shock protein 27 (HSP27): biomarker of disease and therapeutic target. Fibrogenesis Tissue Repair

Biological and clinical implications of heat shock protein 27000 (Hsp27): a review

Transcriptomic and proteomic analyses of rhabdomyosarcoma cells reveal differential cellular gene expression in response to enterovirus 71 infection

Proteomics analysis of host cells infected with infectious bursal disease virus

Mass spectrometry reveals specific and global molecular transformations during viral infection

The expression of heat shock protein hsp27 and a complexed 22-kilodalton protein is inversely correlated with oncogenicity of adenovirus-transformed cells

Proteomic alteration of PK-15 cells after infection by classical swine fever virus

HSPB1 is an intracellular antiviral factor against hepatitis B virus