Core protein phosphorylation modulates pregenomic RNA encapsidation to different extents in human and duck hepatitis B viruses

Abstract

To clarify the role of core protein phosphorylation in pregenomic-RNA encapsidation of human and duck hepatitis B viruses (HBV and DHBV, respectively), we have examined the phosphorylation states of different forms of intracellular HBV core protein and the phenotypic effects of mutations in the phosphorylation sites of HBV and DHBV core proteins. We show that HBV core protein is phosphorylated to similar extents in the form of protein dimers and after further assembly in pregenomic RNA-containing capsids. Individual and multiple substitutions of alanine and aspartic acid for serine in the phosphorylation sites of HBV core protein resulted in site-specific and synergistic effects on RNA encapsidation, ranging from 2-fold enhancement to more than 10-fold inhibition. Core protein variants with mutations in all phosphorylation sites exhibited dominant-negative effects on RNA encapsidation by wild-type protein. The results suggest that the presence of phosphoserine at position 162 of HBV core protein is required for pregenomic-RNA encapsidation, whereas phosphoserine at position 170 optimizes the process and serine might be preferable in position 155. Examination of the pregenomic-RNA-encapsidating capacities of DHBV core protein variants, in which four phosphorylation sites were jointly mutated to alanine or aspartic acid, suggests that phosphorylation of DHBV core protein at these sites may optimize pregenomic-RNA encapsidation but that its impact is much less profound than in the case of HBV. The possible mechanisms by which RNA encapsidation may be modulated by core protein phosphorylation are discussed in the context of the observed differences between the two viruses.

FIG. 1

Phosphorylation of HBV core protein. (A) HepG2 cells were transfected with the indicated plasmids and labeled with [³³P]phosphoric acid for 6 h (³³P) or [³⁵S]methionine-cysteine for 3 h (³⁵S). Labeled cells were lysed and subjected to immunoprecipitation with an anti-HBV core antibody followed by SDS-PAGE and autoradiography. (B) HepG2 cells transfected with plasmid HBV RT⁻ were labeled with [³³P]phosphoric acid or [³⁵S]methionine-cysteine for 50 h in the presence of a 1,000-fold molar excess of cold phosphate or methionine-cysteine, respectively. Cells were then lysed, and the lysates were layered onto 10 to 60% (wt/wt) sucrose step gradients. Eight fractions collected after ultracentrifugation of the gradients were analyzed by immunoprecipitation followed by SDS-PAGE and autoradiography. Fraction numbers are shown on the top. Fractions corresponding to unassembled and assembled forms of core protein are indicated. Left lanes in panel A and right lanes in panel B, marker proteins; molecular masses in kilodaltons are indicated on the right.

FIG. 2

Pregenomic-RNA encapsidation by HBV core protein variants. HepG2 cells were cotransfected with plasmids expressing the indicated core protein variants and a plasmid providing pregenomic RNA and polymerase (HBV RT⁻ C⁻). WT, wild-type core protein. Negative controls were cells transfected with HBV RT⁻ C⁻ alone (lane HBV RT⁻ C⁻) and cells cotransfected with the plasmid encoding wild-type core protein and plasmid HBV C⁻ Pstop, providing pregenomic RNA but no polymerase (lane Pstop). Pregenomic RNA encapsidation within core protein particles produced in transfected cells was analyzed by electrophoresis of the particles through a nondenaturing agarose gel followed by transfer to nylon and nitrocellulose membranes by capillary action. The amount of core protein particles in the sample was estimated by ECL-Western blotting performed on the nitrocellulose membrane (A). The HBV RNA content of the particles was estimated by hybridization with a ³²P-labeled HBV DNA probe performed on a nylon membrane after destruction of the particles in situ with alkali (B).

FIG. 3

Pregenomic-RNA encapsidation by coassembling HBV core protein variants. HepG2 cells were cotransfected with plasmids expressing the indicated core protein variants in the indicated ratios and plasmid HBV RT⁻ C⁻. (A) The amount of total pregenomic RNA produced in transfected cells was estimated by Northern blotting performed on cytoplasmic RNA. (B and C) Pregenomic RNA encapsidation within core protein particles was analyzed as described in the legend to Fig. 2, using duplicate flasks of cells from the same transfections as in panel A. (B) Immunostaining of core protein particles. (C) Detection of encapsidated pregenomic RNA.

FIG. 4

Pregenomic-RNA encapsidation by DHBV core protein variants. LMH cells were cotransfected with plasmids expressing the indicated core protein variants and a plasmid providing pregenomic RNA and polymerase (DHBV RT⁻ C⁻). WT, wild-type core protein. Negative controls were cells transfected with DHBV RT⁻ C⁻ alone (lane DHBV RT⁻ C⁻) and cells cotransfected with the plasmid encoding wild-type core protein and plasmid DHBV C⁻ Pstop, providing pregenomic RNA but no polymerase (lane Pstop). Pregenomic-RNA encapsidation within core protein particles produced in transfected cells was analyzed as described in the legend to Fig. 2, except that a DHBV DNA probe was used for hybridization (B) and an anti-DHBV core antibody was used for immunostaining (A).

FIG. 5

The C-terminal domains of HBV and DHBV core proteins. Phosphorylation sites are boldfaced (10, 20); arginine and lysine residues are boxed. Amino acid numbering refers to HBVayw and DHBV 16 (3, 12). Arrows indicate which portions of each core protein are dispensable and which are indispensable for RNA encapsidation (2, 13, 17).