Fig 1.
Proteomic analysis of BGLF4-induced protein phosphorylation.
(A) Schematic illustration of the SILAC-based quantitative proteomic approach. Akata (EBV+)-tet-BGLF4 and Akata (EBV+)-tet-Vector cells were cultured in “light” (12C6 14N2-K and 12C6 14N4-R) and “heavy” (13C6 15N2-K and 13C6 15N4-R) medium respectively. The cells were then treated with doxycycline for 48 hrs and equal amounts of cells were mixed, fractionated and the nuclear fractions were lysed, digested, desalted and then subjected to lyophilization. The peptide mixture was further fractionated by bRPLC and concatenated to 12 fractions. One peptide aliquot was used for phosphopeptide enrichment by TiO2 beads and the other was used for protein level analysis. The samples were analyzed by LC-MS/MS and the resulting high resolution mass spectra revealed BGLF4-induced changes in phosphorylation level and protein level. See also S1 Fig. (B and C) Quantitation of phosphopeptide and protein levels after BGLF4 induction. Log2 (BGLF4:Vector) plots for quantified phosphopeptides (B) and protein level (C) for Akata (EBV+)-tet-BGLF4 vs Akata (EBV+)-tet-Vector cells. See also S1 and S2 Tables. (D) Summary of phosphoproteomic and proteomic data obtained in the MS analysis (E) Overlap of up-regulated phosphoproteins (> = 2-fold increase) and total proteins quantified.
Fig 2.
BGLF4-induced kinase and phosphatase phosphorylation.
(A) Examples of kinase and phosphatase modulation by BGLF4. *: The maximum BGLF4/Vector ratio was set as 100-fold. (B) Western blot analysis of cell extracts from Dox inducible Akata (EBV+) cells carrying empty vector or BGLF4 using anti-phospho-CDK1, anti-CDK1, anti-HA (BGLF4) and anti-β-actin antibodies as indicated. The cells were untreated (-) or treated with doxycycline (Dox) for 24 hrs or 48 hrs. (C) Immunoblot analysis of cell lysates from Dox inducible Akata cells carrying empty vector, wild type BGLF4 (BGLF4WT), SUMO binding-deficient mutant (BGLF4mSIM-N), or kinase-dead mutant (BGLF4KD) using the antibodies indicated. Cells were treated with doxycycline for 48 hrs to induce BGLF4 expression. (D) Immunoprecipitation assay performed on extracts from transfected cells showing co-precipitation of V5-PP1α with Flag-BGLF4 (lane 1) and with Flag-TIP60 (lane 2). In the control cells V5-PP1α was co-transfected with Flag-vector (lane 3). (E) Summary of a potential BGLF4-triggered signaling cascade.
Fig 3.
Motif analysis of phosphopeptides identified in BGLF4 expressing cells.
(A-B) Dominant phosphorylation motifs showing up-regulation (A) or no change (B) in BGLF4 expressing Akata (EBV+) cells. (C) Dominant phosphorylation motif derived from hyperphosphorylated peptides extracted from the shared proteins identified by MS analysis and in vitro protein microarray screening. Residues above the midline were overrepresented and those below were underrepresented. (D-K) Phospho-serine (D-H)/threonine (I-K) motifs from BGLF4 up-regulated peptides (n = 2,927) were extracted and ranked using the Motif-X algorithm.
Fig 4.
BGLF4-induced DNA damage response signaling.
BGLF4 up-regulated phosphoproteins that are linked to the DNA damage response pathway were identified using the DAVID bioinformatics resource and manual literature curation. See also S3 Table.
Fig 5.
BGLF4-induced mitotic phosphorylation network.
Linkage of BGLF4 up-regulated phosphoproteins to mitosis was obtained using the DAVID bioinformatics resource and manual literature curation. See also S4 Table.
Fig 6.
BGLF4 binds to and phosphorylates key proteins of the spindle assembly checkpoint.
(A) BGLF4 induced phosphorylation of proteins involved in SAC (spindle assembly checkpoint) activation and APC/C (anaphase promoting complex /cyclosome) inhibition. P: phosphorylation; Ac: acetylation. (B) Western blot analysis showing co-precipitation of BGLF4 with MPS1, CDC20 and p31Comet but not BUB1. BGLF4 was co-transfected with MPS1, BUB1, CDC20 or p31Comet in 293T cells as indicated. Cell lysates were immunoprecipitated with anti-HA antibody, anti-Myc or IgG control followed by immunoblotting with anti-BGLF4, anti-HA, anti-Flag or anti-Myc antibodies as indicated. (C) BGLF4 phosphorylates TIP60, MPS1, PP1α and CDC20. (Upper) Purified wild-type (WT) and kinase-dead (KD) BGLF4 were mixed with GST-TIP60 (aa 1–290), GST-MPS1 (a.a. 410–517), GST-PP1α, GST-CDC20 and GST. Autoradiography showing only WT BGLF4 phosphorylates TIP60, MPS1, PP1α and CDC20. GST-TIP60 (a.a. 1–290) and GST were included as positive and negative controls respectively. Arrows indicate the positions of the phosphorylated substrates. (Lower) Immunoblot showing purified WT and KD BGLF4 proteins (Elute) used in the assay. Arrows indicate the positions of purified untagged BGLF4 (Elute) and Halo-BGLF4 proteins in the total cell lysate (Lysate). See also S5 Fig.
Fig 7.
EBV lytic reactivation triggers the phosphorylation of PP1α and a mitosis-related protein.
(A) EBV reactivation triggers the phosphorylation of PP1α. Western blot analysis of cell extracts from Akata-BX1 (EBV+) and Akata-4E3 (EBV-) cells using anti-phospho-PP1α T320, anti-PP1α, anti-BGLF4 and anti-β-actin antibodies as indicated. The cells were untreated (0 hr) or treated with IgG (1:200) for 24, 48 and 72 hrs as indicated. Arrow heads indicate the positions of phospho-PP1α T320. (B) EBV reactivation and BGLF4 induction trigger the phosphorylation of a mitosis-related protein. Western blot analysis of cell extracts from Akata-BX1 (EBV+) and Akata-4E3 (EBV-) cells using anti-phospho-Ser/Thr-Pro MPM2 antibody, anti-BGLF4 and anti-β-actin antibodies as indicated. The cells were untreated (0 hr) or treated with IgG or Doxycycline as indicated. Arrow head indicates the position of a phosphorylated mitotic protein.
Fig 8.
Regulation of known APC/C substrates by BGLF4.
(A) Fold change in protein level of APC/C substrates as measured by MS. See also S2 Table. (B) Western blot validation of BGLF4 mediated up-regulation of individual APC/C substrates in Dox inducible Akata (EBV+) cells expressing wild-type BGLF4, but not SUMO binding-deficient and kinase-dead mutants (BGLF4mSIM-N and BGLF4KD). Immunoblot analysis of the same cell lysates from Fig 2C using antibodies as indicated. (C) EBV replication induces the accumulation of TOP2A and Aurora B. Western blot analysis of cell extracts from Akata-BX1 (EBV+) treated as indicated using anti-Aurora B, anti-TOP2A and anti-β-actin antibodies as indicated. (D) and (E) SAC inhibition suppresses production of extracellular EBV virus. Supernatant virion DNA (D) and cell associated viral DNA (E) from Akata (EBV+) cells treated as indicated was determined by PCR. The experiments were carried out in three biological replicates with similar results and the representative results are presented. Relative PCR product intensity was quantified by ImageJ software.
Fig 9.
BGLF4-induced phosphorylation of nuclear pore associated proteins.
Localization of BGLF4 up-regulated phosphoproteins associated with the nuclear pore as identified using the DAVID bioinformatics resource and manual literature curation. See also S6 Table.