The authors have declared that no competing interests exist.
Conceived and designed the experiments: ARB JAW AS KMC. Performed the experiments: ARB. Analyzed the data: ARB JPC OK JAW AS KMC. Contributed reagents/materials/analysis tools: JPC OK. Wrote the paper: ARB JPC OK JAW AS KMC.
All viruses are dependent upon host cells for replication. Infection can induce profound changes within cells, including apoptosis, morphological changes, and activation of signaling pathways. Many of these alterations have been analyzed by gene arrays to measure the cellular “transcriptome.” We used SILAC (stable isotope labeling by amino acids in cell culture), combined with high-throughput 2-D HPLC/mass spectrometry, to determine relative quantitative differences in host proteins at 6 and 24 hours after infecting HEK293 cells with reovirus serotype 1 Lang (T1L). 3,076 host proteins were detected at 6hpi, of which 132 and 68 proteins were significantly up or down regulated, respectively. 2,992 cellular proteins, of which 104 and 49 were up or down regulated, respectively, were identified at 24hpi. IPA and DAVID analyses indicated proteins involved in cell death, cell growth factors, oxygen transport, cell structure organization and inflammatory defense response to virus were up-regulated, whereas proteins involved in apoptosis, isomerase activity, and metabolism were down-regulated. These proteins and pathways may be suitable targets for intervention to either attenuate virus infection or enhance oncolytic potential.
An understanding of how host cells respond to an invading pathogen may provide important clues about how to either attenuate pathogenesis mediated by the agent, or may facilitate attempts to subvert the pathogen into a ‘beneficial’ organism. At any given time, a cell's genome remains constant. However, during a cell's life cycle, proteins that are expressed within the cell (proteome) will vary depending on numerous external influences that alter genomic biochemical interactions. A cell's proteome is dependent on the location of the cell, different stages of its life cycle or environmental conditions. When a cell becomes exposed to an invading microorganism, such as a virus that requires the host cell's machinery and metabolism to replicate, the cell's proteome reflects the specific alterations of the pathways induced by infection.
Previous analyses of how cells respond to various stimuli have used microarray technologies which measure the cellular “transcriptome”
Recent advances in mass spectrometry and bioinformatics have now provided several general ways that allow in-depth quantitative analysis of large numbers of proteins, and these are being used in several systems to examine the proteome. These methods include 2D difference gel electrophoresis (2D DIGE), isotope-coded affinity tagging (ICAT) and the similar iTRAQ (Isobaric tag for relative and absolute quantitation), as well as stable isotope labeling by amino acids in cell culture (SILAC)
The mammalian reoviruses (MRV) are the prototype viruses in the family Reoviridae. This family currently contains 12 genera
In this study we used SILAC to examine host protein responses in human embryonic kidney (HEK)-293 cells after 6 and 24 hours of infection by MRV strain T1L. The two time points were chosen to represent an early time point when viral particles are being assembled and a later time point when the mature virus particles are starting to be released. These time points in the viral life cycle are therefore potentially important in host cell and virus interactions that would be represented by differences in host protein abundances in the cell. We identified thousands of proteins at both time points, and a small percentage of these proteins were significantly differentially regulated at each time point. From these data, biological functions and network analysis were performed using DAVID bioinformatics resource
Reovirus serotype 1 Lang (T1L) and serotype 3 Dearing, Cashdollar strain (T3D) are laboratory stocks. The viruses were grown and titrated in mouse L929 cell monolayers in Joklik's Suspension Modified Minimal Essential Medium (J-MEM) (GIBCO products, Grand Island, NY, USA) supplemented with 5% fetal bovine serum (FBS), and 200 mM l-glutamine, as previously described
All cell lines that were used, HEK293, L929, A549, CaCo2 and HeLa cells, were acquired from ATCC, catalogue numbers CRL-1573, CCL-1, CCL-185, HTB-37 and CCL-2, respectively. For SILAC preparation, HEK293 (Human embryonic kidney) cells were grown in D-MEM (Dulbecco's Modified Eagle Medium) supplied in a SILACTM Phosphoprotein Identification and Quantification Kit (Invitrogen Canada Inc.; Burlington, Ontario). D-MEM was supplemented to contain 10% dialyzed FBS, and either 100 mg/L ‘Light’ (normal) lysine and arginine or ‘Heavy’ (13C6-lysine and 13C6-/15N4-arginine). One set of HEK293 cells were allowed to double 7 times in L media and another set of HEK293 cells were allowed to double 7 times in H media. Three biologic replicates were performed.
Approximately 107 L-labeled cells in T75 flasks were infected with T1L at a multiplicity of infection (MOI) of 5 PFU per HEK293 cell. Equivalent numbers of H cells were mock-infected as control. The flasks were incubated at 4°C for one hour, with gentle rocking every 10–15 minutes, to allow virus to adsorb and to synchronize infections. An overlay of 10 mls of pre-warmed appropriate L or H media was then added to each flask, and the infected cell cultures were incubated at 37°C for 6 and 24 hours.
The titers of samples were determined by plaque assays as previously described
Infected and mock-infected cells in T25 flasks, and 12 well plates, were examined microscopically for cytopathic effect (CPE) at 0, 3, 6, 9, 12, 15, 18, 24, 30, 36, 48 and 72 hours post infection with a Nikon TE-2000, and cells were photographed with a Canon- A700 digital camera. Images were imported into PowerPoint and slight adjustments made in brightness and contrast to the exact same degree for all of the pictures, which did not alter image context with respect to each other.
HEK293 and L929 cells were infected with either T1L or T3Dc, mock infected, or treated with 100 μg/ml puromycin to act as a positive control. At 0, 3, 6, 9, 12, 15, 18, 24, 30, 36, 48 and 72 hours post infection, cells were harvested, and aliquots were combined with Trypan Blue solution (Sigma, cat#T8154-100ML) at 1 1 ratio. A hemocytometer was used to count a total of 200 cells, as well as all dead cells, in each sample.
HEK293 cells were mounted on autoclaved 12-spot slides, letting the cells adhere to the slides overnight at 37°C. Once adhered, the spots were washed two times with 1X PBS. The cells were counted and virus was added to each spot at an MOI of 5 in serum free media. The virus was allowed to adsorb on the cells for one hour on ice, to ensure synchronization of the infection, then complete media was used for overlay. The spots were incubated for 0, 4, 8, 12, 18 and 24 hours at 37°C. At each time point, spots were washed (3×) with PBS and fixed with paraformaldehyde (∼15 mins).
Once fixed, all spots were washed and cells were permeabilized with 0.2% TritonX-100 in PBS for 5 mins. Spots were blocked using PBS with 1% BSA and 5% FBS, followed with primary antibody (in-house rabbit anti-reovirus), then with Alexa Fluor® 488 Goat anti-Rabbit (Invitrogen, cat#A11008) secondary antibody, DAPI (Invitrogen, cat#D1306), and Alexa Fluor® 546 Phalloidin (Invitrogen, cat#A22283). Anti-fade reagent (Invitrogen, cat#P36935) was added to each spot before slides were covered with coverslips. Slides were examined on a Zeiss Axio Observer Z1 inverted microscope using 20× objective and fluorescence illumination using Exfo Xcite. Images were acquired using AxioVision 4.8.2 software.
At 6 and 24 hours post infection (hpi), both L infected and H mock-infected cells were collected and counted. Aliquots of cultures were also saved for virus titration to confirm infection status. Equal numbers of L and H cells were mixed, and mixed cells were washed three times with ice-cold Phosphate Buffered Saline (PBS). The washed cells were re-suspended in NP-40 buffer (140 mM NaCl, 1.5 mM MgCl2, 10 mM Tris [pH 7.4], 0.5% NP-40) at a concentration of 2.0×108 cells/ml, with the addition of Pepstatin A to a concentration of 1.1 µM.
The mixed, washed cells were then incubated on ice for 30 minutes with gentle mixing after 15 minutes. The nuclei were pelleted by centrifuging at 300xg for 10 minutes. The cytosol (supernatant) was transferred to a fresh microfuge tube; and the two fractions (nuclear pellet and cytosol) were frozen at −80°C until further processing took place.
Protein concentrations in cytosolic lysates were determined using a BCATM Protein Assay Kit (Pierce; Rockford, IL). Aliquots corresponding to 300 µg of protein were diluted 6-fold with 100 mM ammonium bicarbonate. Samples were reduced with 10 mM (final concentration) of dithiothreitol (DTT) for 45 minutes at 60°C, alkylated with 50 mM of iodoacetic acid for 30 min at room temperature in the dark. Excess alkylating agent was removed by addition of 16 mM DTT. Samples were digested with 6 μg of sequencing grade trypsin (Promega, cat#V5111) overnight at 37°C. A different protein standard was used in the first experiment compared to the second and third experiments, which resulted in less protein being processed, loaded and identified in the second and third experiments.
An Agilent 1100 Series HPLC system with UV detection at 214 nm was used for off-line first dimension fractionation. 100 μg of each digested protein sample was loaded onto a 1×100 mm XTerra C18 column (Waters, Milford, MA) and separated using a linear 0–35% water-acetonitrile gradient in 60 minutes at 150 μL/min flow rate. Both eluents contained 20 mM ammonium formate pH 10
The 30 concatenated fractions collected from the first dimension high pH reversed-phase (RP) fractionations were analyzed separately by low-pH RP LC with on-line ESI/TOF Quadupole MS/MS detection (QStar Elite) as was previously described
Within each experiment, proteins identified with L:H ratios were normalized using z-score analysis, described previously
DAVID analysis was performed using lists of proteins up and down regulated previously generated from Protein Pilot analysis
Data sets derived from the Protein Pilot analysis, containing gene identifiers and corresponding expression values, were uploaded into the Ingenuity® Systems application. Each gene identifier was mapped to its corresponding gene object in the Ingenuity Pathways Knowledge Base. A z-score cutoff of ±3.29, ±2.58 and ±1.96 (representing 99.9%, 99% and 95% confidence scores, respectively) was set to identify proteins whose expression was significantly differentially regulated. These proteins, called focus proteins, were overlaid onto a global molecular network developed from information contained in the Ingenuity Pathways Knowledge Base. Networks of these focus genes were then algorithmically generated based on their connectivity.
The Functional Analysis of a network identified the biological functions and/or diseases that were most significant to the genes in the network. The network genes associated with biological functions and/or diseases in the Ingenuity Pathways Knowledge Base were considered for the analysis. Fischer's exact test was used to calculate a p-value determining the probability that each biological function and/or disease assigned to that network was due to chance alone.
Western blot analysis of infected HEK 293 cells was performed as described previously
To determine which virus subtype and cell line to use for proteomic analyses, we performed an initial growth assay of both MRV strains T1L and T3D in five different cell lines; mouse L929 which are usually used for reovirus work, and in human A549, HEK293, CaCo2 and Hela cells (
Each of five different cell lines (L929, A549, HEK293, CaCo2 and Hela) were infected at MOI = 1 PFU/cell with T1L (a) or T3D (b). Cell lysates were harvested at 0, 24, 48 and 72hpi and titrated. Experiments were performed in triplicate; error bars represent standard error. Virus titers were greatest in the L929 and HEK293 cells for both virus strains. HEK293 (c) and L929 (d) cells were then re-analyzed as in (a) and (b) after infection at MOI = 5 and at additional time points. Aliquots of the infections in (c) and (d) were also assessed for cell viability by trypan blue exclusion (e and f, respectively), with 100 μg/ml puromycin used as a positive cell killing control. Experiments were performed in duplicate; error bars represent standard error.
Cell viability was also measured by Trypan Blue exclusion in both HEK293 (
Because HEK293 cells are human and therefore potentially more relevant, we decided to use these cell lines in subsequent experiments. Many reovirus studies have been performed with both T1L and T3D, and although most oncolytic studies have focused on strain T3D, a recent study indicates that T1L may also have oncolytic relevance
Cytopathic effects in HEK293 cells were examined microscopically (
(a) photomicrographs of 293 cells mock-infected (top row), or infected with T1L (middle row) or T3D (bottom row) for various times (indicated at top). Scale bar = 40 μm. (b) HEK293 cells were infected with MRV strain T1L at MOI 5. Cells were stained for f-actin (red), reovirus (green), and DAPI (blue). At 8, 12, 18 and 24 hpi, 5.31%, 64.0%, 83.2% and 100% of cells were infected, respectively (indicated in upper rightmost corner of each image). Scale bar = 40 μm.
To ensure that an MOI of 5 T1L would infect the majority of HEK 293 cells, and that virus proteins were being expressed to affect host processes by the later chosen time point, immunofluorescence microscopy was performed at 8, 12, 18 and 24 hpi. Almost ⅔ of the cells demonstrated infection by 12 hpi and >99% of the cells showed virus infection by 24 hpi (
We quantitatively examined protein alterations in HEK293 cells at early (6 hpi) and late (24hpi) times after infection with MRV T1L. We performed three separate biological replicates of the infection for both time points. 2237, 1920 and 1062 proteins were identified in three experiments at the 6 h time point, with a total of 3076 unique identified proteins (
Venn diagram summaries of the three separate T1L reovirus SILAC experiments at (a) 6hpi, when a total of 3076 unique proteins were identified and at (b) 24hpi where a total of 2992 unique proteins were identified. Overlapping numbers represent those proteins identified in more than one biological replicate (c) Population distribution represented by the log2 of L:H ratios of identified proteins plotted against the protein counts. This population distribution is used to determine the z-scores indicating those proteins considered significantly up or down regulated. Most proteins are seen at a 1∶1 ratio after infection, indicated by the peak of the population distribution indicated at 0. For clarity, only experiment 1 at 24hpi is shown, however population distributions were generated at both time points, for all biological replicates.
To facilitate inter-experiment comparisons, each protein's L:H ratio was converted into log2 space. These log ratios were plotted in a population distribution graph and used to convert every L:H ratio into a z-score that measures how far each particular protein's L:H ratio lies in relation to the population mean and standard deviation. A representation of this for experiment 1 at 24hpi is shown in
Accession | HGNC | Name | 6h L:H ratio |
24h L:H ratio |
|
||||
34416 | LTF | Lactotransferrin |
|
|
6013427 | ALBU | serum albumin precursor |
|
|
17318569 | K2C1 | keratin 1 |
|
|
1683637 | BAD | Bcl-2 binding component 6 |
|
0.76 |
25058739 | ALB | ALB protein |
|
|
9963783 | Q9HBZ9 | RNA helicase |
|
|
4557325 | APOE | apolipoprotein E precursor |
|
|
6563288 | UBQL2 | ubiquitin-like product Chap1/Dsk2 |
|
1.00 |
16553095 | ASPC1 | unnamed protein product |
|
|
56204388 | MEA1 | male-enhanced antigen |
|
0.88 |
20336350 | HCD2 | endoplasmic reticulum-associated amyloid beta peptide-binding protein |
|
|
4885217 | XPF | excision repair cross-complementing rodent repair deficiency, complementation group 4 |
|
|
55959029 | UBE2T | ubiquitin conjugating enzyme |
|
1.19 |
22671717 | HBA2 | hemoglobin alpha-2 |
|
|
55660909 | SC23B | Sec23 homolog B (S. cerevisiae) |
|
1.21 |
29792115 | SMAD3 | SMAD, mothers against DPP homolog 3 (Drosophila) |
|
|
48734878 | Q6IPH7 | RPL14 protein |
|
|
7020506 | UB2R2 | unnamed protein product |
|
1.08 |
4505571 | SQSTM | sequestosome 1 |
|
1.39 |
57208697 | REPS1 | RALBP1 associated Eps domain containing 1 |
|
|
14042287 | NRBP | unnamed protein product |
|
1.02 |
4589628 | PALLD | KIAA0992 protein |
|
|
55962212 | CTBP2 | C-terminal binding protein 2 |
|
0.72 |
55749758 | DIP2B | DIP2 disco-interacting protein 2 homolog B |
|
1.36 |
5803123 | PSMF1 | proteasome inhibitor subunit 1 isoform 1 | 1.64 |
|
21961219 | DNJB4 | DnaJ (Hsp40) homolog, subfamily B, member 4 |
|
0.88 |
12654999 | DPOA2 | Polymerase (DNA directed), alpha 2 (70kD subunit) |
|
0.94 |
13543689 | SLD5 | GINS complex subunit 4 (Sld5 homolog) |
|
|
793843 | RL29 | ribosomal protein L29 |
|
|
3970842 | AL1A2 | RALDH2 |
|
1.36 |
56202415 | STX12 | syntaxin 12 |
|
1.18 |
28336 | ACTB | mutant beta-actin (beta'-actin) |
|
|
56205198 | SNAPN | SNAP-associated protein |
|
1.01 |
55962100 | B1AN89 | eukaryotic translation initiation factor 4 gamma, 3 |
|
0.94 |
13477351 | CP080 | Gene trap locus 3 (mouse) |
|
0.95 |
54696534 | ATPO | ATP synthase, H+ transporting, mitochondrial F1 complex, O subunit (oligomycin sensitivity conferring protein) |
|
|
7705411 | YTHDF2 | YTH domain family, member 2 |
|
0.99 |
9623363 | DPOE3 | DNA polymerase epsilon p17 subunit |
|
|
55962775 | S10AD | S100 calcium binding protein A13 |
|
1.34 |
7672784 | HSP7E | heat shock protein HSP60 |
|
|
7576229 | VIME | vimentin |
|
0.99 |
21749793 | Q8NB80 | unnamed protein product |
|
0.82 |
7705431 | CCD72 | coiled-coil domain containing 72 |
|
|
48146451 | DBNL | HIP-55 |
|
1.18 |
21750170 | Q8NB11 | unnamed protein product |
|
1.29 |
55663125 | SET | SET nuclear oncogene |
|
0.95 |
9966827 | PCNP | PEST-containing nuclear protein |
|
|
6633995 | SK2L2 | KIAA0052 protein | 1.16 |
|
6274552 | STAT1 | signal transducer and activator of transcription 1 isoform alpha | 1.11 |
|
56203909 | DDI2 | DNA-damage inducible protein 2 (DDI2) | 1.09 |
|
862457 | ECHA | enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase alpha-subunit of trifunctional protein | 1.04 |
|
34532659 | Q6ZTT1 | unnamed protein product | 1.03 |
|
5738608 | NFU1 | HIRA-interacting protein HIRIP5 | 1.00 |
|
3287825 | DC1L2 | Cytoplasmic dynein 1 light intermediate chain 2 (Dynein light intermediate chain 2, cytosolic) (LIC53/55) (LIC-2) | 0.93 |
|
31652283 | ACO13 | thioesterase superfamily member 2 | 0.87 |
|
31880783 | PELO | pelota homolog | 0.77 |
|
18490917 | SCG2 | Secretogranin II (chromogranin C) |
|
|
4731861 | OAS3 | 2′–5′oligoadenylate synthetase 3 |
|
|
14550514 | ISG15 | ISG15 ubiquitin-like modifier |
|
|
34192824 | IFIT2 | Interferon-induced protein with tetratricopeptide repeats 2 |
|
|
2653424 | SHIP2 | inositol polyphosphate 5-phosphatase |
|
|
49574526 | IFIT1 | interferon-induced protein with tetratricopeptide repeats 1 |
|
|
47123412 | Q6NSF2 | RPLP0 protein |
|
|
51476787 | Q68D64 | hypothetical protein |
|
|
51476787 | Q68D64 | hypothetical protein |
|
Protein is included if at least half of the biologic z-score values are ≥1.960σ (indicated by bolding) and there are no major disagreements between biological replicates.
L/H ratio refers to the geometric mean of all log2 L/H values for each given gi number, expressed as relative protein quantity in infected cultures.
Accession | HGNC | Name | 6h L:H ratio |
24h L:H ratio |
|
||||
35700 | PRPS2 | unnamed protein product |
|
|
7239177 | SHPK | CARKL |
|
1.08 |
6063691 | VDAC1 | porin isoform 1 |
|
1.14 |
16307253 | SNP29 | Synaptosomal-associated protein, 29kDa |
|
|
9295343 | MRRP1 | HNYA |
|
1.31 |
13112017 | Q9BTT9 | GYS1 protein |
|
0.88 |
45501009 | Q6NWZ1 | CKAP4 protein |
|
0.91 |
13938553 | KAD6 | TAF9 RNA polymerase II, TATA box binding protein (TBP)-associated factor, 32kDa |
|
0.91 |
30354279 | AKTS1 | AKT1S1 protein |
|
1.02 |
4885579 | RCD1 | RCD1 required for cell differentiation1 homolog |
|
|
55959177 | VPS45 | vacuolar protein sorting 45A (yeast) |
|
0.91 |
577315 | AMPM1 | KIAA0094 |
|
1.00 |
48145521 | RNF14 | RNF14 |
|
|
48146107 | VAMP8 | VAMP5 |
|
1.68 |
7741063 | NUDT4 | diphosphoinositol polyphosphate phosphohydrolase type 2 beta |
|
0.93 |
6537210 | PPP6 | serine/threonine protein phosphatase catalytic subunit |
|
0.88 |
55957551 | Q5TB52 | 3′-phosphoadenosine 5′-phosphosulfate synthase 2 |
|
0.88 |
57997480 | BAG6 | hypothetical protein |
|
0.88 |
31542862 | GRWD1 | glutamate-rich WD repeat containing 1 |
|
|
50949588 | LS14A | hypothetical protein |
|
0.84 |
14041989 | UBP47 | unnamed protein product |
|
|
51258951 | XPO6 | XPO6 protein |
|
|
5542014 | DKC1 | dyskerin |
|
1.01 |
27769298 | TRI25 | Tripartite motif-containing 25 |
|
1.42 |
51476912 | Q68D08 | hypothetical protein | 0.81 |
|
47605963 | ROCK2 | Rho-associated protein kinase 2 (Rho-associated, coiled-coil-containing protein kinase 2) (p164 ROCK-2) (Rho kinase 2) | 0.81 |
|
36122 | RPB2 | RNA polymerase II 140 kDa subunit | 0.84 |
|
27503838 | TPPC5 | Trafficking protein particle complex 5 | 0.85 |
|
15559560 | BGLR | Glucuronidase, beta | 0.87 |
|
56203612 | ARFG3 | OTTHUMP00000028526 | 0.88 |
|
50415798 | Q6DC98 | LMNB1 protein | 0.89 |
|
14124914 | GLYC | Serine hydroxymethyltransferase 1 (soluble) | 0.90 |
|
4503141 | CTSC | cathepsin C isoform a preproprotein | 0.95 |
|
56204970 | A8YXX4 | glutamate-ammonia ligase (glutamine synthase) | 0.96 |
|
47496673 | Q6ICN0 | GRB2 | 1.10 |
|
55958754 | PHP14 | RP11-216L13.10 | 1.19 |
|
55960176 | LAMC1 | laminin, gamma 1 (formerly LAMB2) | 1.22 |
|
5679449 | OTTHUMP00000016179 |
|
||
34535789 | Q6ZR81 | unnamed protein product |
|
Protein is included if at least half of the biologic z-score values are ≥1.960σ (indicated by bolding) and there are no major disagreements between biological replicates.
L/H ratio refers to the geometric mean of all log2 L/H values for each given gi number, expressed as relative protein quantity in infected cultures.
Several of the SILAC-identified up-regulated and non-regulated proteins were examined by Western blotting (
(a) Western blot analyses of selected proteins. Mock-infected and T1L-infected cells were harvested at 24hpi, lysed with 0.5% NP-40, and 20–80 μg of each cytosolic fraction resolved in each lane of 10×6.5×0.1 cm 10% SDS-mini-PAGE. Proteins were transferred to PVDF membranes, blocked, probed with various indicated primary antibodies, developed with appropriate secondary antibodies, and visualized with an Alpha Innotech FluorChemQ MultiImage III instrument. (b) Densitometry analysis comparison to SILAC L:H average ratios for the ten different host proteins in (a). Most of the Western blot results correlated to the regulation of the proteins observed in SILAC (whether or not they are up or down regulated). All of the proteins tested by Western blot were also identified at 24hpi in SILAC and were therefore used for confirmation. Some of these proteins were also identified at 6hpi in SILAC and these were used for WB comparison. (GAPDH – glyceraldehyde-3-phosphate dehydrogenase, LTF – lactotransferrin, ISG15 – interferon-stimulated protein 15kDa, OAS3 – 2′–5′oligoadenylate synthetase 3, SCG2 – secretogranin II, hnRNPA1 – heterogeneous nuclear ribonucleoprotein A1, IFIT2 – Interferon-induced protein with tetratricopeptide repeats 2, BAD – Bcl-2 binding component 6, PTPN12 – protein tyrosine phosphatase, non-receptor type 12, STAT1 – signal transducer and activator of transcription 1).
Lists of proteins observed to be either up- or down-regulated after averaging z-scores from all three biological replicates were uploaded into the DAVID analysis web-based tool
Proteins were also analyzed by using Ingenuity Pathways Analysis (Ingenuity® Systems,
Graphs represent host cell functions with highest score (x-axis) based on the number of differentially regulated proteins observed in that network. The higher the score, the greater the number of proteins differentially regulated in that particular function network.
The top network functions identified in the previous figure are shown in more detail with interconnecting protein relationships indicated by solid (direct interaction) or dashed (indirect interaction) lines. Red/pink molecules represent proteins up-regulated; green molecules represent down-regulation. (a) Network 1, 6hpi; top functions include cell death (b) Network 2, 6hpi; top functions include cellular growth and proliferation. (c) Network 1, 24hpi; top functions include molecular transport. (d) Network 4, 24hpi; top functions include infectious disease and inflammatory response.
Metabolic and cellular canonical signaling pathways were also used for analysis of uploaded datasets. Activation of IRF by cytosolic pattern recognition receptors is one such signaling pathway that was significantly altered by 24hpi. This pathway had an enrichment p-value of 3.7×10−5 based on the average z-score analysis at 24hpi.
Protein regulation expression patterns overlaid from first biological replicate. Red indicates up-regulation, grey represents proteins not changed in abundance after infection and white represents molecules not identified in SILAC experiment but are part of the known canonical pathway.
SILAC is an effective way to observe differences in the relative protein concentrations of thousands of proteins that may differ in expression levels between two separate experimental cell cultures. Our study quantified changes in the physiological levels of several thousand cellular proteins during T1L infection. We analyzed all data from all three experiments, both separately and in combination, by IPA and DAVID analysis to identify a large list of cell metabolic and biological pathways that were affected by MRV T1L infection. This approach was also used to confirm inter-experiment reproducibility. Comparing the top network (functions in molecular transport, RNA trafficking and lipid metabolism) generated after 24hpi from one experiment to another showed a significant up regulation of the whole network, with numerous proteins up regulated in more than one experiment (Supplemental Figure S2). By studying the difference in networks generated from multiple time points, we examined how the host cell's response to infection changed during the virus life cycle. At the earlier time point, cell death and proliferation as well as antigen presentation are seen as major pathways affected which correlated with DAVID analysis that indicated apoptosis was influenced. At 24hpi, molecular transport, DNA replication and repair, post-translational modification and inflammatory response are more prominently affected, correlating with positive regulation of interferon-alpha production and defense response to virus identified with DAVID. In general, the host response to viral infection produced more up regulated proteins than down regulated ones. Detection of more up regulated proteins than down regulated ones correlates with a previous genomic assay performed on T1L in HEK293 cells that observed this same trend in differential regulation of mRNA
Differentially regulated apoptosis and DNA repair pathway mRNA is observed after infection with T1L and T3D by 24hpi
Known canonical pathways were also examined and activation of IRF by cytosolic pattern recognition receptors pathway was shown to be significantly altered (p value 3.7×10−5) after 24 h indicating this pathway is important in reovirus infection. Up-regulated proteins in this pathway are RIG-1, ISG15, MAVS, STAT1 and ISG-54 (also known as IFIT2). ISG15 has not previously been identified as affected by MRV infections, but has been identified in numerous other viral infections
In addition to RIG-1 and MAVS, numerous other proteins involved in apoptosis were identified as differentially regulated. A major hub at 6hpi in network 1 (
Another hub at 6hpi in network 2 (
The top differentially regulated network identified at 24hpi had functions in molecular transport, with TNF alpha as a major hub (
In conclusion, we have used SILAC to identify a number of additional novel candidate proteins, more focused study of which should provide greater understanding of MRV-induced pathogenesis as well as a better understanding of how to use this virus as a more effective research tool.
DAVID analysis of up (a, b) or down regulated (c, d) proteins using average z-scores from all three biological replicates. Metabolic and biological functions are shown that are enriched from the differentially regulated protein lists at 95, 99, and 99.9% confidence cutoffs (indicated by blue, red and green bars, respectively). At 6hpi (a, c) top functions included transforming growth factor β-receptor, activation of pro-apoptotic gene products, cis-trans isomerase activity, late endosome functions and serine protease inhibitor processes. At 24hpi (b, d) top biological processes included positive regulation of interferon-alpha production, defense response to virus by host, interferon-mediated immunity and procollagen-lysine 5-dioxygenase activity.
(TIF)
Inter-experiment reproducibility of 95% confidence IPA network analyses. Each of the three biologic replicates of the top 24hpi network; molecular transport, RNA trafficking and lipid metabolism, are depicted (a, b, c). Numerous proteins are up regulated in more than one replicate.
(TIF)
Table showing all host proteins observed as up-regulated (at 95% confidence) after infection with T1L reovirus at both 6 and 24 hpi. Proteins at the top of the lists are those identified more than one time; proteins in the bottom half are those identified only once. Table shows the individual protein z-scores found for each of the three biological replicates, as well as the average L:H ratios for all three experiments. A protein is included in this table if at least half of the biologic z-score values are ≥1.960σ (indicated by bolding; at least 95% confidence score) and there are no major disagreements between biological replicates.
(XLS)
Table showing all host proteins observed as down-regulated (at 95% confidence) after infection with T1L reovirus for both 6 and 24 hpi. Proteins at the top of the lists are those identified more than one time; proteins in the bottom half are those identified only once. Table shows the individual protein z-scores found for each of the three biological replicates, as well as the average L:H ratios for all three experiments. A protein is included in this table if at least half of the biologic z-score values are ≥1.960σ (indicated by bolding; at least 95% confidence score) and there are no major disagreements between biological replicates.
(XLS)