Fig 1.
Identification of novel US2 specific substrates by plasma membrane profiling.
US2 downregulates cell surface expression of α and β integrins in addition to other proteins. (A) Scatter plots of proteins identified in PMP and quantified by 2 or more unique peptides. Fold change (single gene expressing THP-1 cells / control THP-1 cells) is shown as log2 ratio on the x axis and the summed peptide intensity on the y-axis as log10. Proteins unaltered by viral gene expression locate to the center of the plots (0-fold change), whereas proteins left or right of center represent respectively proteins down- or up-regulated by viral gene expression. Significance B was used to estimate p values. (B) Verification of integrin downregulation in US2-expressing THP-1. Cytofluorometric analysis of the indicated proteins in THP-1 cells stably expressing US2 with or without TRC8 depletion, and control THP-1 cells. (C) Representation of the integrin family. Integrins downregulated by US2 are highlighted in red. See also S1 and S2 Figs and S1 Table.
Fig 2.
US2 induces the proteasomal degradation of integrin α4 and α5 and prevents maturation of the β1 integrin.
(A and B) US2 induces proteasomal degradation of integrins α4 and α5. THP-1 cells stably expressing US2, US11 or depleted of TRC8 (shTRC8) were analyzed by immunoblots for the indicated proteins. Cells were treated with the proteasome inhibitor lactacystin for 18 hours before harvest as indicated. (C-D) Integrin 4 is proteasomally degraded in US2-expressing cells. THP-1 cells expressing integrin α4-HA were labeled with [35S]methionine-cysteine and pulse-chased for the indicated times in the presence or absence of MG132, followed by integrin α4 immunoprecipitation using the HA-tag (E) β1 integrin accumulates in its precursor form in the presence of US2. THP-1 cells were pulse-chase labelled as above and endogenous β1 integrin was immune precipitated using anti-integrin β1 antibody. (F and G) US2 associates with integrin α4 and recruits the TRC8 E3 ligase via its cytosolic tail. Cells stably expressing full-length US2 (US2) or US2 with Δ186–199 residues (US2ΔC') in combination with TRC8-HA or integrin α4-HA were solubilised in 1% digitonin, immunoprecipitated with anti-HA antibody and visualized by immunoblot. (H) US2 induces integrin α4 ubiquitination. THP-1 cells expressing integrin α4-HA in combination with US2 or empty vector were pulse-chase labeled for the indicated times. Integrin α4 was immune precipitated using the HA-epitope tag, eluted from beads, fully denatured to dissociated non-covalently linked proteins and re-precipitated using α-HA or α-ubiquitin antibodies.
Fig 3.
US2-mediated downregulation of integrins inhibits downstream integrin-mediated signalling, cell adhesion and migration.
(A) US2 inhibits signaling downstream of integrin α4β1. THP-1 cells were seeded on tissue culture plates coated +/- fibronectin as indicated for 1 hour at 37°C. Total cell lysates of adherent and non-adherent cells were analyzed by immunoblotting with phospho-specific antibody to tyr118 paxillin and total paxillin. (B) US2 inhibits THP-1 cell adhesion. THP-1 cells were seeded in triplicate on tissue culture plates coated with fibronectin as indicated for 1 hour at 37°C and washed 3 times with PBS. Adherent cells were quantified by CyQUANT-NF (Invitrogen). The percent adherent cells was normalized against control cells expressing empty vector (vector/US2 n = 3; US11/shbeta1 n = 2). (C) US2 inhibits cell migration. THP-1 cell migration +/- chemotaxis agent MCP-1 (10 nM) was examined in cells stably expressing US2, US11, control or ITGB1 shRNA (shbeta1) using fibronectin-coated Transwell plates. Number of cells migrated to the lower compartment is indicated (average of triplicates, n = 4). Data are represented as mean ± SEM. p-values were calculated using paired Student’s t-test.
Fig 4.
PMP comparing wild-type and US2-deficient HCMV shows a requirement for US2 in the downregulation of cell surface targets.
(A) Scatter plots of proteins identified in PMP and quantified by 2 or more unique peptides. Fold change (HFF infected with wild-type HCMV versus HFF infected with ΔUS2 HCMV) is shown as log2 ratio on the x axis and the summed peptide intensity on the y-axis as log10. Proteins unaltered by HCMV US2 gene deletion locate to the center of the plots (0-fold change), whereas proteins left or right of center represent respectively proteins down- or up-regulated by HCMV in a US2-dependent manner. Significance B was used to estimate p values. (B) US2 is required for integrin and thrombomodulin downregulation during HCMV infection. HFF cells infected with GFP tagged HCMV ΔUL16-18 or ΔUL16-18ΔUS1-11 (moi 0.5) were analyzed by cytofluorometric analysis of the indicated proteins at 72 hours post-infection. (C) HFFs infected with HCMV wild-type or ΔUS2 (moi 0.5) were analyzed by flow cytometry at 72 hours post-infection. Cell surface staining for MHC-I enabled gating for HCMV infected (MHC-Ilo) cells (left panel) and subsequent analysis of the integrin and thrombomodulin expression (right panels). See also S3 Table.
Fig 5.
US2/TRC8-mediated degradation of α integrins reduces cell adhesion of lytically HCMV infected THP-1 cells.
(A,B) TRC8 depletion abolishes US2-induced α integrin degradation in HCMV infected THP-1 cells. THP-1 cells stably expressing a control shRNA (shCtrl) or shRNA targeting TRC8 (shTRC8) were PMA activated and infected with HCMV TB40 UL32-GFP (moi 25). Expression of indicated integrins was assessed at 96 hours post-infection by flow cytometry (A) or immunoblotting (B). (C) TRC8/US2-mediated degradation of α integrins reduces cell adhesion of HCMV-infected THP-1 cells. THP-1 cells were infected as above and at 96 hours post-infection counted, and seeded on tissue culture plates coated with fibronectin, recombinant VCAM-1-Fc, collagen or uncoated. After a 1 hour (fibronectin, VCAM-1, uncoated; n = 4) or 2 hours (collagen n = 2) incubation, loosely attached cells were washed off and the number of adherent cells was determined by CyQUANT-NF assay. Data show adhesion relative to fibronectin binding of control cells as mean ± SEM. p-values were calculated using paired Student’s t-test.
Fig 6.
UL141 and US2 are both essential for surface down-regulation of CD112.
(A,B) Scatter plots of proteins identified in PMP and quantified by 2 or more unique peptides, for HFF infected with (A) HCMV wild-type vs. ΔUL141 and (B) HCMV wild-type vs. a ΔUL141ΔUS2 double deletion mutant. Fold change (wild-type/specific gene deletion HCMV infected human fibroblasts) is shown as log2 ratio on the x axis and the summed peptide intensity on the y-axis as log10. Proteins unaltered by HCMV UL141 or combined US2/UL141 gene deletion locate to the center of the plots (0-fold change), whereas proteins left or right of center represent respectively proteins down- or up-regulated by HCMV in a UL141 (A) or combined US2/UL141-dependent (B) manner. Significance B was used to estimate p values. (C) Schematic showing verified substrates of US2 and UL141. Substrates identified by PMP and confirmed by us or previous studies are indicated. (D) HCMV downregulation of cell surface CD112 requires US2, UL141 and TRC8. Cell surface expression of CD112 was determined in HFF infected with mock or HCMV wild-type, ΔUS2 or ΔUL141 (moi 5) at 72 hours post infection. HFFs were stably transduced with scrambled shRNA (HFF + shCtrl) (top panel) or shTRC8 (HFF + shTRC8) prior to HCMV infection (lower panel). Dotted lines indicate the mean CD112 expression in HFFs with control shRNA infected by indicated virus. See also S3 Table.
Fig 7.
UL141 retains CD112 in the ER and enhances US2 mediated degradation.
ER retention of CD112 is mediated by UL141 and only detected upon depletion of US2 or TRC8. (A) HFF infected with mock, HCMV wild-type, ΔUS2 or ΔUL141 (moi 5) were analyzed for the indicated proteins at 72 hours post infection, in the presence (shTRC8+) or absence (shTRC8-) of TRC8 depletion. (B) UL141 retains CD112 in the ER prior to TRC8-dependent degradation by US2. THP-1 cells stably expressing US2, US11, UL141-Myc or shTRC8 were solubilized in 1% digitonin, immunoprecipitated with anti-Myc antibody and the indicated proteins visualized by immunoblot. See also S4 Fig.
Fig 8.
Model for US2 action in HCMV-infected cells.
US2 targets a wide variety of novel cell surface substrates, including integrins, the NK cell ligand CD112, the anti-coagulation factor thrombomodulin and the IL-12 receptor β 1. The majority of these targets are directly targeted for proteasomal degradation in a TRC8 E3 ligase dependent pathway. β1 integrin is not itself degraded, but accumulates in an ER-resident immature form following US2-induced degradation of its α integrin interaction partner. CD112 is retained in the ER by UL141 from where US2 promotes the TRC8-dependent degradation. The functional consequences of US2 substrate degradation include diminished cell adhesion, migration, NK cell activation and antigen presentation.