Steady-state plasma membrane expression of human cytomegalovirus gB is determined by the phosphorylation state of Ser900 - PubMed (original) (raw)

Steady-state plasma membrane expression of human cytomegalovirus gB is determined by the phosphorylation state of Ser900

K N Fish et al. J Virol. 1998 Aug.

Abstract

Human cytomegalovirus (HCMV) infection of an astrocytoma cell line (U373) or human fibroblast (HF) cells results in a differential cell distribution of the major envelope glycoprotein gB (UL55). This 906-amino-acid type I glycoprotein contains an extracellular domain with a signal sequence, a transmembrane domain, and a 135-amino-acid cytoplasmic tail with a consensus casein kinase II (CKII) site located at Ser900. Since phosphorylation of proteins in the secretory pathway is an important determinant of intracellular trafficking, the state of gB phosphorylation in U373 and HF cells was examined. Analysis of cells expressing wild-type gB and gB with site-specific mutations indicated that the glycoprotein was equally phosphorylated at a single site, Ser900, in both U373 and HF cells. To assess the effect of charge on gB surface expression in U373 cells, Ser900 was replaced with an aspartate (Asp) or alanine (Ala) residue to mimic the phosphorylated and nonphosphorylated states, respectively. Expression of the Asp but not the Ala gB mutation resulted in an increase in the steady-state expression of gB at the plasma membrane (PM) in U373 cells. In addition, treatment of U373 cells with the phosphatase inhibitor tautomycin resulted in the accumulation of gB at the PM. Interestingly, the addition of a charge at Ser900 trapped gB in a low-level cycling pathway at the PM, preventing trafficking of the protein to the trans-Golgi network or other intracellular compartments. Therefore, these results suggest that a tautomycin-sensitive phosphatase regulates cell-specific PM retrieval of gB to intracellular compartments.

PubMed Disclaimer

Figures

FIG. 1

Confocal images of gB staining in HCMV-infected cells. HF (A), U373 cells (B), and MDM (C) were infected with HCMV as described in Materials and Methods. HF were fixed at 3 days p.i., U373 cells were fixed at 7 days p.i., and MDM were fixed at 14 days p.i. The cells were permeabilized and stained with a monoclonal mouse anti-gB antibody. Magnifications, ×294.

FIG. 2

In vitro CKII phosphorylation of a GST-gB C-terminal tail chimeric protein. A GST-gBwt or gBAla C-terminal tail chimera was attached to microbeads and treated with either CKII or CKI and [32P]orthophosphate followed by magnetic bead purification and SDS-PAGE. (A) gBwt but not gBAla was phosphorylated by CKII. However, neither of the chimeric proteins was phosphorylated by CKI. (B) Both GST-gBwt and GST-gBAla, were stably expressed as determined by analysis by denaturing SDS-PAGE.

FIG. 3

gB is phosphorylated in HF and U373 cells in vivo. HCMV-infected (A), RVV gBwt-infected (B), or RVV gBAla-infected (B) HF and U373 cells were labeled with inorganic 32P and then subjected to immunoprecipitation with gB-specific rabbit antisera and a preimmune (pre.) control serum. In vivo-phosphorylated gB was detected in HCMV- and RVV gBwt-infected but not in RVV gBAla-infected HF and U373 cells.

FIG. 4

Subcellular localization of gB in HF and U373 cells. Confocal images of gB staining of cells infected with RVV gBwt, RVV gBAla, or RVV gBAsp were obtained. (A to C) Infected HF; (D to F) infected U373 cells. The cells were stained before permeabilization or at 4°C (representing surface gB) with mouse anti-gB (green) and after permeabilization (representing total gB) with mouse anti-gB (red). (A and D) RVV gBwt infections; (B and E) RVV gBAla mutant infections; (C and F) RVV gBAsp mutant infections. Surface gB staining is a combination of prepermeabilization at 4°C and postpermeabilization (yellow) (A, B, C, and F). (G and H) Confocal microscopy images demonstrating the presence of gB in VV-infected cells. U373 cells were infected with either RVV gBwt (G) or RVV gBAla (H) and subsequently treated with the phosphatase inhibitor tautomycin. The cells were stained with mouse antibody to gB prepermeabilization or at 4°C (green) (surface gB) and with mouse antibody to gB postpermeabilization (red) (total gB). The accumulation of gB trafficking to the surface was observed only in the RVV gBwt infection (yellow) (G). To demonstrate that transport of gB to the cell surface is not affected by the state of phosphorylation, we coinfected U373 cells with RVV gBwt (I), RVV gBAsp (J), or RVV gBAla (K) and RVV dynK44A. Magnifications, ×303 for panels A to F and I to K and ×473 for panels G and H.

FIG. 5

The steady-state cell surface expression of gBAsp is greater than that of gBwt and gBAla on the PM of U373 cells. U373 cells were infected with RVV gBwt, RVV gBAla, or RVV gBAsp and then subjected to surface biotinylation to analyze differences in gB PM expression. Specifically, 35S-labeled U373 cells infected with RVV gBwt, RVV gBAla, or RVV gBAsp were pulsed with NHS-SS-biotin and the immunoprecipitated surface gB was analyzed by SDS-PAGE. This figure demonstrates that substantially more gBAsp than gBAla is detected at the surface of U373 cells. These results support the hypothesis that a charge at position 900 in the gB cytoplasmic tail increases surface expression in U373 cells. c-term, C terminus.

FIG. 6

Endocytosis and intracellular targeting of gB in U373 cells. gB antibody uptake experiments were performed in RVV gBwt-infected (A) or RVV gBAsp-infected (B) U373 cells. At 6 h p.i., mouse anti-gB N-terminus antibody was applied to the cells for 30 min. The cells were then rinsed and incubated for a 30-min chase period followed by fixation. Nonpermeabilized cells were stained with a cyanine-5–anti-mouse secondary conjugate (blue; stable surface gB), followed by rinsing, permeabilization, and staining with a TRITC–anti-mouse secondary conjugate (red; internalized gB). The cells were rinsed again and exposed to rabbit anti-gB C-terminus antibody followed by an FITC–anti-rabbit secondary conjugate (green; total gB). Therefore, internalized gB is both green and red (yellow vacuoles), PM gB is blue, red, and green (white PM staining), and gB that was absent from the PM during the 30-min mouse anti-gB N-terminus antibody exposure is green. Magnifications, 473×.

References

1. Bogner E, Anheier B, Offner F, Smuda C, Reschke M, Eickmann M, Radsak K. Nuclear translocation of mutagenized forms of human cytomegalovirus glycoprotein B (gpUL55) J Gen Virol. 1997;78:1647–1651. - PubMed
1. Bogner E, Reschke M, Reis B, Reis E, Britt W, Radsak K. Recognition of compartmentalized intracellular analogs of glycoprotein H of human cytomegalovirus. Arch Virol. 1992;126:67–80. - PubMed
1. Bos K, Wraight C, Stanely K K. TGN38 is maintained in the trans-Golgi network by a tyrosine-containing motif in the cytoplasmic domain. EMBO J. 1993;12:2219–2228. - PMC - PubMed
1. Bresnahan P A, Leduc R, Thomas L, Thorner J, Gibson H I, Brake A J, Barr P J, Thomas G. Human fur gene encodes a yeast KEX2-like endoprotease that cleaves pro-beta-NGF in vivo. J Cell Biol. 1990;111:2851–2859. - PMC - PubMed
1. Britt W J. Neutralizing antibodies detect a disulfide-linked glycoprotein complex within the envelope of human cytomegalovirus. Virology. 1984;135:369–378. - PubMed

Steady-state plasma membrane expression of human cytomegalovirus gB is determined by the phosphorylation state of Ser900 - PubMed (original) (raw)