Fig 1.
Structural and functional domains of ORF50 protein.
Numbers in the diagram correspond to amino acid positions. NLS, nuclear localization signal; Basic, basic-rich region; LR, leucine heptapeptide repeat; S/T, serine/threonine-rich region. Distinct functional domains or regulatory motifs are shown, including the DNA-binding domain, activation domain, ubiquitin E3 ligase domain, homo-multimerization domain, SUMO-binding domain, DNA-binding inhibitory sequence (DBIS), and protein abundance regulatory signal (PARS). The PARS contains two components located in aa 490–535 (PARS-I) and aa 590–650 (PARS-II), respectively.
Fig 2.
Correlation between the subcellular localization and protein abundance of PARS-I mutants.
(A) Left, schematic diagram of PARS-I mutants of ORF50. Specific deletion or amino acid substitutions in the PARS-I motif are indicated in red color in the diagram. The protein abundance and subcellular localization of each PARS-I mutant protein are summarized. The degree of protein abundance is indicated by “+”. N: nucleus; C: cytoplasm; N/C: both nucleus and cytoplasm. Right, confocal microscopy images of PARS-I mutant proteins in 293T cells. (B) Abundance of PARS-I mutants in 293T cells. Cells were transfected with the indicated plasmids encoding FLAG-tagged ORF50 mutants for 24 hr. Cell lysates were then immunoblotted using anti-FLAG antibody. ORF50-A: hyperphosphorylated ORF50; ORF50-B: hypophosphorylated ORF50.
Fig 3.
Nuclear translocation of ORF50(KK/EE) and its abundance.
(A) Schematic diagram of GFP-ORF50 and GFP-ORF50(KK/EE) constructs with or without an appendage of the SV40 NLS. The subcellular localization and protein abundance of each GFP fusion construct determined by confocal microscopy (B) and by immunoblot analysis (C) are summarized. (D) Confocal images of both F-ORF50(KK/EE) and GFP-tagged proteins expressed in 293T cells. Cells were cotransfected with plasmids expressing F-ORF50(KK/EE) and GFP or GFP-ORF50 for 24 hr. Intracellular localization of F-ORF50(KK/EE) (red) and GFP or GFP-ORF50 (green) in cells was analyzed by confocal microscopy. Arrows indicate the cells expressing both F-ORF50(KK/EE) and GFP-ORF50. (E) Changes in the expression of F-ORF50(KK/EE) by coexpression with GFP-ORF50. Various amounts of the GFP-ORF50 expression plasmid (0, 25, 50, 100 and 400 ng) were cotransfected with 400 ng of the F-ORF50(KK/EE) expression plasmid into 293T cells. The expression levels of F-ORF50(KK/EE) and GFP-ORF50 in cells were determined by immunoblotting using anti-FLAG and anti-GFP antibody, respectively. The expression of wild-type F-ORF50 in 293T cells was also included in the experiment (lane 8). (F) Summary of phenotypic changes of F-ORF50(KK/EE) in the presence of GFP-ORF50.
Fig 4.
The PARS-II motif functions in the nucleus in the control of ORF50 abundance.
(A) Left, diagram of PARS-I and/or PARS-II mutants of ORF50. A series of ORF50 C-terminal deletions with or without the KK-to-EE mutation in the PARS-I motif were included. All ORF50 mutants contain a FLAG tag at their N terminus. The subcellular localization and protein abundance of each of ORF50 mutants in 293T cells are summarized in the diagram. The degree of protein abundance is indicated by “+”. N: nucleus; C: cytoplasm; N/C: both nucleus and cytoplasm. Right, representative confocal images showing the subcellular localization of these ORF50 mutants. (B) Immunoblot analysis of ORF50 deletion mutants in 293T cells.
Fig 5.
Phenotypic changes of F-ORF50(KK/EE) by different GFP-ORF50 deletion effectors.
(A) Diagram of GFP-ORF50 deletion constructs and a summary of their characteristics. (B) Confocal images of 293T cells coexpressing F-ORF50(KK/EE) and GFP-ORF50 mutants. The subcellular localization of both F-ORF50(KK/EE) (red) and GFP-tagged proteins (green) in transfected cells was analyzed by confocal microscopy. (C) Effect of GFP-ORF50 deletion mutants on the expression of F-ORF50(KK/EE). 293T cells were cotranfected with equal amounts (400 ng) of plasmids encoding F-ORF50(KK/EE) and the indicated GFP-tagged proteins. Cell lysates were immunoblotted with either anti-FLAG antibody or anti-GFP antibody.
Fig 6.
The protein abundance of ORF50 is controlled through the ubiquitin-proteasomal degradation pathway.
(A) Effect of MG132 on ORF50 abundance in cells. 293T cells were transfected with a plasmid encoding F-ORF50 or F-ORF50(1–564). At 16 hr after transfection, cells were untreated or treated with 5 μM of MG132 for another 12 hr and 24 hr. Cell lysates were analyzed by immunoblotting using anti-FLAG antibody. (B and C) Ubiquitination of ORF50 in cells. 293T cells were transfected with plasmids encoding HA-ubiquitin (HA-Ub) or/and F-ORF50. At 16 hr posttransfection, cells were treated with MG132 for another 24 hr. Denatured lysates were immunoprecipitated (IP) with either anti-HA antibody (B) or anti-FLAG antibody (C). Cell lysates (input) and the immunoprecipitated proteins were analyzed by immunoblotting (IB) using anti-ORF50 or anti-HA antibody. Asterisks indicate the cross-reaction of ORF50 with the used antibodies (S4 Fig).
Fig 7.
Both N-terminal and C-terminal regions in ORF50 are required for ubiquitination.
(A) Deletion constructs of ORF50 and a summary of their intracellular localization, protein abundance and ubiquitination status. (B) Immunoblot analysis of F-ORF50, F-ORF50(1–590) and F-ORF50(357–691) in 293T cells. (C and D) Mutants F-ORF50(1–590) and F-ORF50(357–691) were evaluated for ubiquitination in 293T cells. Cells were transfected with the indicated plasmids. At 16 hr after transfection, cells were treated with MG132 for another 24 hr. Cell lysates were immunoprecipitated and immunoblotted as described in Fig 6B. Asterisks indicate the cross-reaction of ORF50 with the used antibodies (S4 Fig). (E) Proposed model for ORF50 ubiquitination. The PARS-I is responsible for the nuclear translocation of ORF50, and the PARS-II motif is required for the binding of specific ubiquitin enzymes. The ubiquitin acceptor sites are likely to be located in the N-terminal 356-aa region.
Fig 8.
Human MDM2 promotes ORF50 degradation.
(A) Effect of MDM2 expression on ORF50 abundance. Increasing amounts of an MDM2 expression plasmid, pCMV6-XL-MDM2, were cotransfected with a plasmid expressing F-ORF50, F-590 or F-390 in 293T, 293 or HKB5/B5 cells. At 24 hr after cotransfection, the expression of these ORF50 proteins was examined by immunoblotting. (B) Evaluation of the half-life of ORF50 in MDM2-transfected cells. 293T cells were cotransfected with an F-ORF50 expression plasmid and pCMV6-XL-MDM2 or control vector. At 19 hr after transfection, cells were untreated or treated with cycloheximide for another 6, 12 and 24 hr. The expression of F-ORF50 and MDM2 were analyzed by immunoblotting with anti-FLAG and anti-MDM2 antibody, respectively. The relative levels of F-ORF50 from immunoblots (left panel) were quantified by densitometry and normalized to tubulin, which are depicted in the right panel.
Fig 9.
MDM2 interacts with ORF50 in cells.
(A) 293T cells were transfected with the expression plasmid for GFP or GFP-ORF50. Cell lysates were harvested and immunoprecipitated with anti-MDM2 or anti-FLAG antibody. Cell lysates (input) and the resulting immunoprecipitates were probed with anti-MDM2 or anti-GFP antibody. (B) Same as in (A), except that the immunoprecipitation was carried out using anti-GFP antibody. (C) Coimmunoprecipitation of MDM2 and endogenous ORF50 in HH-B2 cells. HH-B2 cells were untreated or treated with 3 mM sodium butyrate (SB). At 6 hr after SB treatment, cells were then treated with MG132 (5 μM) for 20 hr. After the cell lysates were immunoprecipitated with anti-MDM2 or anti-FLAG antibody, the immunoprecipitated proteins were analyzed for the existence of MDM2 and ORF50 by immunoblotting. (D) Colocalization of MDM2 and ORF50 in HH-B2 cells. Confocal immunofluorescent analysis was performed using HH-B2 cells treated with MG132, SB or both SB and MG132. Unique ORF50/MDM2 colocalized complexes were observed in nucleolar areas when cells were treated with both SB and MG132 (d, arrows). The bottom sides (e and f) show high magnification of the selected cells from (d).
Fig 10.
Mapping of the interaction domains between MDM2 and ORF50, and ubiquitination of ORF50 by MDM2 in vitro.
(A) Defining the ORF50-interacting domain of MDM2. The purified His-tagged ORF50 (His-ORF50) was incubated with GST, GST-MDM2 or GST-MDM2 deletion mutants (1–220, 100–290 and 221–491) expressed in E. coli. Following pull-down with glutathione beads, the pull-down lysates were immunoblotted with anti-ORF50 or anti-GST antibody. (B) Defining the MDM2-interacting domain of ORF50. Glutathione beads conjugated to GST-MDM2 were incubated with His-ORF50 or His-ORF50 deletions as shown in the diagram. The GST pull-down precipitates were immunoblotted using anti-His antibody. (C) Detection of the interaction between GST-MDM2(1–220) and His-ORF50(490–691) in pull-down assay. (D) Effect of MDM2 on ORF50 ubiquitination in vitro. The in vitro ubiquitantion assay was performed using purified components as indicated. In the reactions, purified His-ORF50 protein was used at a final concentration of 120 nM and GST-MDM2 (or GST-MDM2(1–220)) at 100 nM or 200 nM. Reaction mixtures were analyzed by immunoblotting with anti-ORF50 or anti-Ub antibody.
Fig 11.
Lysine residues at positions 152 and 154 in ORF50 are critical for MDM2-mediated degradation.
(A) Schematic diagram of ORF50 and ORF50 lysine mutants. The positions of 25 lysines in ORF50 are shown in the diagram (circles). Black circles represent the substitutions of lysine (K) with arginine (R) in F-ORF50. (B) Immunoblot analysis of lysine substitution mutants of ORF50. Cell lysates of 293T cells that were transfected with the indicated expression plasmids were probed with anti-FLAG antibody. (C) Effects of single- or double-lysine mutations from K124 to K243 (the middle 8 lysine clusters) on ORF50 abundance. (D) Susceptibility of ORF50 mutants to MDM2-mediated degradation. The expression plasmids encoding ORF50 mutants were cotransfected with an MDM2 expressing plasmid or control vector into 293T cells. The expression of these ORF50 mutants in the presence or absence of exogenous MDM2 was determined by immunoblotting with anti-FLAG antibody. (E) Relative protein levels of ORF50 mutants in the presence or absence of overexpressed MDM2. The bar graph summarizes densitometry data from three independent experiments. Error bars: standard deviation.
Fig 12.
MDM2 negatively regulates ORF50 expression in KSHV-infected cells.
The expression kinetics of MDM2 and viral lytic proteins (including ORF50, K8, ORF45, K8.1) were analyzed in HH-B2 (A) and BC3 cells (B) after treatment with sodium butyrate (SB). (C and D) The lentiviral vector-mediated MDM2 knockdown was performed in HH-B2 and BC3 cells. At 24 hr after lentiviral infection, HH-B2 or BC3 cells were either untreated or treated with SB for another 18 hr. The expression of MDM2, ORF50, K8 and ORF45 was determined by immunoblotting with the indicated antibodies.