Persistence of bovine viral diarrhea virus is determined by a cellular cofactor of a viral autoprotease

Abstract

Polyprotein processing control is a crucial step in the life cycle of positive-strand RNA viruses. Recently, a vital autoprotease generating an essential viral replication factor was identified in such a virus, namely, the pestivirus bovine viral diarrhea virus. Surprisingly, the activity of this protease, which resides in nonstructural protein 2 (NS2), diminishes early after infection, resulting in the limitation of viral RNA replication. Here, we describe that a cellular chaperone termed Jiv (J-domain protein interacting with viral protein) acts as a cofactor of the NS2 protease. Consumption of the intracellular Jiv pool is responsible for temporal regulation of protease activity: overexpression of Jiv interfered with regulation and correlated with increased accumulation of viral RNA; downregulation of the cellular Jiv level accelerated the decline of protease activity and reduced intracellular viral RNA levels and virion production. Accordingly, the amount of a cellular protein controls pestiviral replication by limiting the generation of active viral protease molecules and replication complexes. Importantly, this unique mechanism of replication control is essential for maintenance of the noncytopathogenic phenotype of the virus and thereby for its ability to establish persistent infections. These results add an entirely novel aspect to the understanding of the molecular basis of viral persistence.

FIG. 1.

Influence of the molar ratio of GST-Jiv90 and Flag-NS2-3 on NS2-3 cleavage efficiency. (A) BHK-21 cells were cotransfected with a constant amount of pflag-NS2-4A and increasing amounts of pGST-Jiv90 in the molar ratios indicated above lanes 2 to 8; molar ratios of plasmids were calculated according to their molecular masses. After metabolic labeling of the transfected cells, cell lysates were immunoprecipitated with anti-Flag MAb and further analyzed by SDS-PAGE and autoradiography. After transfection of increasing amounts of pGST-Jiv90, an about proportional increase in GST-Jiv90 was observed (data not shown). (B) Signals of Flag-NS2-3 and Flag-NS2 in the gel shown in panel A were quantified by phosphorimaging, and NS2-3 cleavage efficiency was calculated (see Materials and Methods). (C) The signals of Flag-NS2 and GST-Jiv90 in the gel shown in panel A were quantified by phosphorimaging, and protein amounts were calculated, taking into account the relative number of cysteine and methionine residues in both proteins (see also B); the highest value was set to 1.

FIG. 2.

NS2-3 cleavage induction by GST-Jiv90 and its truncated derivatives. Flag-NS2-4A was transiently expressed without GST-Jiv90 (lane 1), together with GST-Jiv90 (lane 2), or together with truncations of Jiv90 as GST fusion proteins (lanes 3 to 7) in BHK-21 cells; T7-vaccinia virus-infected cells served as a control (lane 8). The size of the N- and/or C-terminal truncation of Jiv90 is indicated above lanes 3 to 7. After metabolic labeling of the transfected cells, cell lysates were immunoprecipitated with anti-Flag MAb and further analyzed by SDS-PAGE and autoradiography. wt, wild type.

FIG. 3.

Effect of amino acid exchanges in Jiv90 on NS2-3 cleavage efficiency. (A) Alanine scan. Flag-NS2-4A was transiently expressed without GST-Jiv90, together with the GST-Jiv90 wild type (wt) or together with GST-Jiv90 mutants in BHK-21 cells. The positions of the substitutions to alanine within Jiv90 are indicated at the bottom of the graph. After metabolic labeling of the transfected cells, cell lysates were immunoprecipitated with anti-Flag MAb and further analyzed by SDS-PAGE. The signals of Flag-NS2-3 and Flag-NS2 were quantified by phosphorimaging. NS2-3 cleavage efficiency was calculated (see Materials and Methods); NS2-3 cleavage efficiency using GST-Jiv90 wt was set to 100%. (B) Effect of single amino acid exchanges at position 39 of Jiv90 on NS2-3 cleavage efficiency. Tryptophan at position 39 of Jiv90 was replaced by the amino acids indicated above lanes 3 to 9. After metabolic labeling of cells transfected with pflag-NS2-4A and GST-Jiv90 wt or mutants, cell lysates were immunoprecipitated with anti-Flag MAb and further analyzed by SDS-PAGE and autoradiography. A low-level NS2-3 cleavage induction by GST-Jiv90 mutant W39H was reproducibly detected in independent experiments. (C) Quantification of Flag-NS2-3 and Flag-NS2 shown in panel B by phosphorimaging. NS2-3 cleavage efficiency was calculated as described in Materials and Methods; the highest value was set to 100%. (D) Summary of the data obtained in the Jiv90 mutagenesis study. Numbers indicate the position within the amino acid sequence of Jiv90. White background, N- or C-terminal amino acid blocks of which one at a time is dispensable for NS2-3 cleavage induction; grey background, amino acids which can be replaced by alanine without a complete loss of NS2-3 cleavage induction; black background, aromatic residue at position 39 of Jiv90 which is essential for NS2-3 cleavage induction; underlining, minimal part of Jiv90 which is essential but not sufficient for a stable interaction with pestiviral NS2 (Fig. 4).

FIG. 4.

Binding of truncated GST-Jiv90 derivatives to NS2. Flag-NS2 was transiently expressed without GST-Jiv90 (lane 1) or together with truncated Jiv90 derivatives fused to the C terminus of GST (lanes 2 to 13) in BHK-21 cells. The size of the N- or C-terminal truncation of Jiv90 is indicated above lanes 2 to 13. After metabolic labeling of the transfected cells, cell lysates were immunoprecipitated with anti-Flag MAb (α-flag) (upper part) or anti-GST MAb (α-GST) (lower part) and further analyzed by SDS-PAGE and autoradiography.

FIG. 5.

Influence of single amino acid mutations in the NS2 protease on NS2-3 cleavage efficiency in the presence of GST-Jiv90. The Flag-NS2-4A wild type (wt) (lane 2) or mutants (lanes 3 to 6) were expressed together with GST-Jiv90 in BHK-21 cells. Mutations within NS2 are indicated above lanes 3 to 6. After metabolic labeling of the transfected cells, cell lysates were immunoprecipitated with anti-Flag MAb and further analyzed by SDS-PAGE and autoradiography.

FIG. 6.

Effect of the Jiv level on the kinetics of NS2-3 processing in noncp BVDV-infected cells. Cells were metabolically labeled for 1-h periods postinfection (p.i.), as indicated above the lanes. The cell lysates were immunoprecipitated with an anti-NS3 MAb, which also recognizes NS2-3, and were further analyzed by SDS-PAGE and autoradiography. (A) Effect of Jiv overexpression. MDBK cells (left part) or MDBK_tet-onJiv cells induced with 10 μM Dox for 16 h prior to infection (right part) were infected with noncp BVDV at a multiplicity of infection of 10. (B) Effect of Jiv knockdown. PT_tet-on cells (left part) or PT-Jiv-kd cells with a reduced Jiv level (right part) were infected with noncp BVDV at a multiplicity of infection of 10.

FIG. 7.

Effect of the Jiv level on viral replication. (A and B) Effect of the Jiv level on intracellular viral RNA amounts in noncp BVDV-infected cells. Cells were infected with noncp BVDV at a multiplicity of infection of 5. Total cellular RNA was prepared at indicated times postinfection, and viral RNA was quantified by quantitative real-time RT-PCR (see Materials and Methods). The graph shows mean values from three independent experiments. The amount of BVDV RNA measured at 12 h postinfection in the control cells MDBK (A) or PT_tet-on (B) was set to 1. (A) Overexpression of Jiv. Quantification of intracellular viral RNA isolated at different time points postinfection from MDBK cells or MDBK_tet-onJiv cells which were either not induced or induced by 10 μM Dox for 16 h prior to infection. (B) Jiv knockdown. Quantification of intracellular viral RNA isolated at the indicated time points postinfection from cell lines PT_tet-on, PT-Jiv-kd, and PT-Jiv-kd-rescue is shown. (C) Growth kinetics for noncp BVDV in PT-Jiv-kd cells compared to the parental PT_tet-on cells and to PT-Jiv-kd-rescue cells. The graph shows the mean values from three independent experiments. Cells were infected at a multiplicity of infec-tion of 5, and the culture supernatants were harvested at the indicated time points postinfection. Virus titers are given as log 50% tissue culture infective dose per milliliter.