Disulfide linkages mediating nucleocapsid protein dimerization are not required for porcine arterivirus infectivity

Abstract

The nucleocapsid (N) proteins of the North American (type II) and European (type I) genotypes of porcine reproductive and respiratory syndrome virus (PRRSV) share only approximately 60% genetic identity, and the functionality of N in both genotypes, especially its role in virion assembly, is still poorly understood. In this study, we demonstrated that the ORF7 3' untranslated region or ORF7 of type I is functional in the type II PRRSV background. Based on these results, we postulated that the cysteine at position 90 (Cys90) of the type II N protein, which corresponds to an alanine in the type I protein, is nonessential for virus infectivity. The replacement of Cys90 with alanine confirmed this hypothesis. We then hypothesized that all of the cysteines in the N protein could be replaced by alanines. Mutational analysis revealed that, in contradiction to previously reported findings, the replacement of all of the cysteines, either singly or in combination, did not impair the growth of either type II or type I PRRSV. Treatment with the alkylating agent N-ethylmaleimide inhibited cysteine-mediated N dimerization in living cells but not in released virions. Additionally, bimolecular fluorescence complementation assays revealed noncovalent interactions in living cells among the N and C termini and between the N-terminal and C-terminal regions of the N proteins of both genotypes of PRRSV. These results demonstrate that the disulfide linkages mediating the N dimerization are not required for PRRSV viability and help to promote our understanding of the mechanism underlying arterivirus particle assembly.

Fig 1

Type I N protein can function in the type II PRRSV background. (A) Schematic diagram of the chimeras. The infectious cDNA clones of the type II PRRSV strain APRRS and the type I strain SHE are indicated as white and gray boxes, respectively. The sequences after the boxes show the junction of ORF6 and ORF7. Chimeric constructs were generated as detailed in Materials and Methods. For pA-SHE73′, the ORF7 3′UTR segment of SHE was placed immediately downstream of ORF6 of APRRS, the overlap between ORF6 and ORF7 was separated, and the start codon AUG for the type II ORF7 in the duplication of the overlap was mutated to ACG. For pA-OSHE73′, the overlap between ORF6 and ORF7 was maintained as in SHE, and pA-OSHE7 was derived from pA-OSHE73′ by replacing the 3′UTR with its counterpart from APRRS. (B) Indirect immunofluorescence analysis was conducted to verify heterologous N protein expression. Two days after transfection, MARC-145 cells were fixed and stained with a MAb that can specifically recognize both genotypes of PRRSV N protein, further incubated with an Alexa Fluor 568-labeled goat anti-mouse antibody, and then visualized by inverted fluorescence microscopy. (C) Plaque morphology. The P1 viruses were used to inoculate MARC-145 cells for 1 h, overlaid with 1% agarose, and incubated for about 4 days until plaques developed. Plaques were stained with 1% crystal violet. (D) Growth kinetics. MARC-145 cells were infected with P1 chimeric viruses at an MOI of 0.1, and the supernatants were harvested at the indicated time points. Virus titer was determined by the TCID₅₀/ml, and the results are presented as mean values from three independent experiments. (E) The subgenomic RNA transcription profiles were analyzed by Northern blotting. Two days after the transfection of MARC-145 cells, RNAs from each sample were loaded and separated by 1% agarose gel electrophoresis and then transferred to the membrane. sgmRNAs were hybridized with a type I-specific PRRSV probe located in ORF7. 18S rRNA was used as an internal control. The nonspecific band is indicated by an asterisk.

Fig 2

Sequence alignment of PRRSV N proteins and the construction of cysteine mutants. (A) Alignment of N sequences from type II PRRSV strains (VR2332, GenBank accession no. U87392; APRRS, GQ330474) with the corresponding sequences from type I strains (LV, M96262; SHE, GQ461593). Completely conserved residues are indicated in black boxes, and partially conserved residues are in white boxes. The cysteines at positions 23, 75, and 90 of the type II N protein are indicated by black triangles. The scissors indicate the split site of the N protein for the BiFC assay. The secondary structure elements above the sequence show the structure of the C terminus of the VR2332 N protein (11). This figure was generated by ESPript (17), with slight modifications. (B) The cysteine residues in the N protein of APRRS and SHE were replaced with alanine or serine, either alone or in combination. The number above the boxes indicates the location of the cysteine or alanine. PCR-based site-directed mutagenesis was used to replace the codon for cysteine (TGC) with that of alanine (GCG) or serine (AGC).

Fig 3

All of the type II N protein cysteines can be knocked out without effect. (A) IFA showing the expression of N protein. The clusters of stained cells represent the spread of virus infection, whereas the single cell indicates the absence of released live virus. (B) Plaque morphology of cysteine mutants. (C) Growth kinetics were examined, and the results are presented as mean values from three independent experiments.

Fig 4

Cysteine knockout mutants of type I PRRSV show virological properties similar to those of the parental virus. (A) Expression of N protein was verified by IFA. (B) Plaque morphology. (C) Growth curves for cysteine mutants and the wild-type virus. The results are presented as mean values from three independent experiments.

Fig 5

Dimerization of cellular N proteins. Virus-infected or expression vector pCAGGS-transfected cells were lysed in the presence (+) or absence (−) of 1 mM NEM, separated by SDS-PAGE under reducing (+DTT) or nonreducing (−DTT) conditions, and subjected to Western blot analysis. (A) N proteins expressed from wild-type and mutant type II PRRSV APRRS. (B) N proteins expressed from wild-type and mutant type I PRRSV SHE. (C) Wild-type and cysteine mutant APRRS N proteins expressed from the mammalian vector pCAGGS. (D) Wild-type and cysteine mutant SHE N proteins expressed from pCAGGS. (E) Cellular APRRS N proteins expressed from virus or pCAGGS were treated with 1 mM NEM before cell lysis. Cells then were lysed without NEM and subjected to SDS-PAGE under reducing or nonreducing conditions, followed by Western blotting. (F) Cellular APRRS N proteins expressed from virus or pCAGGS were lysed at pH 5.5 without NEM treatment. Cell lysates then were subjected to SDS-PAGE under reducing or nonreducing conditions and analyzed by Western blotting. Monomers (N) or dimers (2N) of the N protein are indicated by arrows.

Fig 6

Dimerization of N proteins in extracellular virions. Virions from culture supernatants were purified in the absence or presence of NEM as described in Materials and Methods. Purified particles were resolved by SDS-PAGE under reducing or nonreducing conditions and subjected to Western blot analysis. (A) Purified wild-type and mutant APRRS particles without NEM treatment. (B) Purified wild-type and mutant SHE particles without NEM treatment. (C) Purified APRRS and SHE particles with NEM treatment. Monomers (N) or dimers (2N) of the N protein are indicated by arrows.

Fig 7

BiFC analysis of full-length N proteins. (A) Schematic diagram of BiFC constructs. The Venus protein was split between amino acid residues 173 and 174, resulting in fragments VN (N-terminal 173 residues) and VC (C-terminal 66 residues). The so-called Target protein was fused to the N terminus of VN or VC with a linker sequence to generate the BiFC pair. (B) Visualization of interactions between full-length N proteins in living cells. Fragments VN and VC or α-tubulin gene-fused VN and VC were coexpressed in MARC-145 cells, and the fluorescent signals were examined as negative controls to assess the specificity of the BiFC method (panels a to c). Full-length N proteins from both genotypes of PRRSV with or without cysteine mutations were fused to the N terminus of VN or VC, and cells were cotransfected and examined for fluorescence (panels d to i).

Fig 8

BiFC analysis of truncated N proteins. Full-length N proteins with or without cysteine mutations were split between amino acid residues 61 and 62 for APRRS or between 62 and 63 for SHE. The resulting halves were ligated to VN or VC, MARC-145 cells were transfected, and fluorescent signals were visualized. The arrows indicate the nucleoli with no fluorescence.