Downregulation of natural killer cell-activating ligand CD155 by human cytomegalovirus UL141

Peter Tomasec, Eddie C Y Wang, Andrew J Davison, Borivoj Vojtesek, Melanie Armstrong, Cora Griffin, Brian P McSharry, Rebecca J Morris, Sian Llewellyn-Lacey, Carole Rickards, Akio Nomoto, Christian Sinzger, Gavin W G Wilkinson, Peter Tomasec, Eddie C Y Wang, Andrew J Davison, Borivoj Vojtesek, Melanie Armstrong, Cora Griffin, Brian P McSharry, Rebecca J Morris, Sian Llewellyn-Lacey, Carole Rickards, Akio Nomoto, Christian Sinzger, Gavin W G Wilkinson

Abstract

Natural killer (NK) cells are crucial in the control of cytomegalovirus infections in mice and humans. Here we show that the viral UL141 gene product has an immunomodulatory function that is associated with low-passage strains of human cytomegalovirus. UL141 mediated efficient protection of cells against killing by a wide range of human NK cell populations, including interferon-alpha-stimulated bulk cultures, polyclonal NK cell lines and most NK cell clones tested. Evasion of NK cell killing was mediated by UL141 blocking surface expression of CD155, which was previously identified as a ligand for NK cell-activating receptors CD226 (DNAM-1) and CD96 (TACTILE). The breadth of the UL141-mediated effect indicates that CD155 has a key role in regulating NK cell function.

Figures

Figure 1
Figure 1
Resistance to NK cell attack maps to the HCMV UL/b′ sequence. (a,b) Human fetal foreskin fibroblasts (HFFFs) were mock infected (Mock) or were infected for 72 h with HCMV strains AD169, Toledo, Merlin or 6397 or a strain AD169ΔUL40 deletion mutant (multiplicity of infection, 10). NKL cells were used as effectors in allogeneic chromium release assays. (c) HFFFs were infected with strains AD169, Towne, Toledo or the recombinant virus Towne/Tol11 1.1 (T/T11 1.1) and NKLs were used as effectors in allogeneic NK cell assays. (d) Autologous assay with D9SF targets (primary skin fibroblast from donor 9) infected as described in c with IFN-α-activated bulk cultures derived from D9 PBMCs (D9NKb) as effectors. Results are mean ± s.d. of triplicate or quadruplicate cultures and are representative of at least three and up to eight independent experiments. E:T, effector:target ratio.
Figure 2
Figure 2
UL141 induces protection against NK cell attack. Cytotoxicity assay with NKLs as the effectors against stable HEK 293 cell lines expressing UL141-GFP, UL14-GFP or UL16-GFP fusion proteins or GFP alone. To ensure homogeneous expression, we sorted cell lines on the basis of GFP fluorescence before the assay. Far right, immunoblot with a GFP-specific rabbit polyclonal antibody, indicating fusion proteins were being expressed by the transfected cell lines. Left margin, molecular sizes (in kDa). Data are representative of three experiments.
Figure 3
Figure 3
Characterization of gpUL141 expression. Cells were infected with recombinant adenoviruses or various HCMV strains. GpUL141 expression was detected with the M550 series of mAbs. (a) Expression of gpUL141, after infection of HFFFs with RAdUL141. The protein gpUL141 is sensitive to both Endo H (+E) and peptide-N-glycosidase F (+P). M, mock digest. (b) Time course of gpUL141 expression through the lytic replication cycle of HCMV strains Toledo and AD169. (c) Endo H (+E) and peptide-N-glycosidase F (+P) sensitivity of gpUL141 expressed from strain Toledo or Merlin. M, mock digest. (d) Immunoblot of the expression of gpUL141 in HFFFs infected with various HCMV strains (above lanes). (e) NK cell cytolysis assay with NKL as effectors, comparing the susceptibility of HFFFs after infection with strains TB40E, Toledo and AD169. (f) Immunoblot of strain TB40E that contained two viruses: UL141-replete (Lisa) and a virus without UL141 expression (Bart). Immunoblot with the M550 mAbs shows expression of gpUL141 in HFFFs infected with TB40E-Lisa or TB40E-Bart. MOI, multiplicity of infection. (g) NK cell cytolysis assay comparing the sensitivity to killing with NKL shown by cells infected with TB40E-Bart (dashed line) or TB40E-Lisa (dotted line). NK cell cytotoxicity results are mean ± sd. Data are one representative experiment of three. (a–d,f) Left margin, molecular sizes (in kDa).
Figure 4
Figure 4
UL141-mediated downregulation of cell surface CD155 detected by flow cytometry. Flow cytometry with mAb D171 of cell surface CD155 on mock-infected cells or cells after infection with HCMV strains AD169 (UL141−) or Toledo (a), after infection with TB40E-Bart (UL141−) or TB40E-Lisa (b), after infection with adenovirus vectors encoding UL141 or UL16 (c), and in continuous cell lines expressing UL141-GFP or UL16-GFP fusion proteins (d). Dotted lines, staining with an isotype-matched control antibody. mIgG, mouse IgG.
Figure 5
Figure 5
UL141-mediated downregulation of cell surface CD155 detected by fluorescence microscopy. CD155 expression monitored by staining with the monoclonal antibodies D171 or 5D1 in the presence or absence of UL141. The gpUL141 was expressed with an adenovirus vector (RAdUL141-GFP), infection with strain Toledo-GFP in HFFFs or in UL141-GFP–transfected HEK 293 cells. Similarly, gpUL16 was expressed with an adenovirus vector (RAdUL16-GFP) or in UL16-GFP-transfected cells. HCMV strain AD169 is UL141 negative. HCMV and adenovirus vectors contain an intact GFP gene to identify infected cells. In HEK 293 cells, gpUL16-GFP and gpUL141-GFP fusion proteins traffick to the endoplasmic reticulum and concentrate in intracytoplasmic foci.
Figure 6
Figure 6
UL141 inhibits maturation of CD155. Immunoblots of extracts of cells infected with viruses with the 5D1 mAb to CD155, the gpUL141-specific M550 mAbs or the HC-10 mAb to HLA class I. The sensitivity of the CD155 protein to Endo H and peptide-N-glycosidase F (PNGaseF) digestion is analyzed in parallel. HLA class I (Class I) downregulation is a surrogate marker of efficient HCMV infection. Left margin, molecular sizes (in kDa).

Source: PubMed

3
Abonner