Mutagenesis analysis of the nsp4 main proteinase reveals determinants of arterivirus replicase polyprotein autoprocessing

Abstract

Nonstructural protein 4 (nsp4; 204 amino acids) is the chymotrypsin-like serine main proteinase of the arterivirus Equine arteritis virus (order Nidovirales), which controls the maturation of the replicase complex. nsp4 includes a unique C-terminal domain (CTD) connected to the catalytic two-beta-barrel structure by the poorly conserved residues 155 and 156. This dipeptide might be part of a hinge region (HR) that facilitates interdomain movements and thereby regulates (in time and space) autoprocessing of replicase polyproteins pp1a and pp1ab at eight sites that are conserved in arteriviruses. To test this hypothesis, we characterized nsp4 proteinase mutants carrying either point mutations in the putative HR domain or a large deletion in the CTD. When tested in a reverse genetics system, three groups of mutants were recognized (wild-type-like, debilitated, and dead), which was in line with the expected impact of mutations on HR flexibility. When tested in a transient expression system, the effects of the mutations on the production and turnover of replicase proteins varied widely. They were cleavage product specific and revealed a pronounced modulating effect of moieties derived from the nsp1-3 region of pp1a. Mutations that were lethal affected the efficiency of polyprotein autoprocessing most strongly. These mutants may be impaired in the accumulation of nsp5-7 and/or suffer from delayed or otherwise perturbed processing at the nsp5/6 and nsp6/7 junctions. On average, the production of nsp7-8 seems to be the most resistant to debilitating nsp4 mutations. Our results further prove that the CTD is essential for a vital nsp4 property other than catalysis.

FIG. 1.

Alternative processing pathways for the EAV replicase pp1a. The figure was modified from references and . The three EAV proteinases (PCP1β, CP, and SP) and their cleavage sites are depicted; the EAV nsp nomenclature is used. Prominent hydrophobic domains (H) are indicated. PCP1β, papain-like Cys proteinase; CP, Cys proteinase; SP, nsp4 Ser proteinase, Z, zinc finger; Hel, helicase; N, nidovirus-specific endoribonuclease (NendoU). The association of cleaved nsp2 with nsp3-8 (and probably also nsp3-12) was shown to be a cofactor in the cleavage of the nsp4/5 site by the nsp4 proteinase (major pathway). Alternatively, in the absence of nsp2, the nsp5/6 and nsp6/7 sites are processed and the nsp4/5 junction remains uncleaved (minor pathway). The status of the small nsp6 subunit (fully cleaved or partially associated with nsp5 and/or nsp7) remains to be elucidated.

FIG. 2.

(A) Ribbon diagram of the X-ray crystal structure of the EAV nsp4 proteinase. The ribbon diagram was made using DeepView/Swiss-PdbViewer v3.7 (21). Members of the catalytic triad (His-39, Asp-65, and Ser-120) are indicated. Four residues of the hinge region (amino acids 154 to 161) are colored black (see panel B). Residues Thr-157 and Ser-161 may act as a hinge to facilitate the rotation of the C terminus relative to the rest of the proteinase (4). Residues Ala-155 and Asp-156 were targeted for mutagenesis to change the structural flexibility of the nsp4 hinge region. (B) Multiple sequence alignment of the hinge region of main proteinase of arteriviruses. The alignment was generated for a representative set of arteriviruses using the MUSCLE program (14) integrated in the Viralis platform (A. E. Gorbalenya, A. A. Kravchenko, A. M. Leontovich, V. K. Nikolaev, D. V. Samborskiy, and V. A. Sorokin, unpublished). The positions of the N-terminal and C-terminal residues of the alignment are indicated using proteinase numbering. The HR is indicated with a black line. NC_002532_EAV-Buc and AY349167_EAV-CW96, equine arteritis virus strain Bucyrus (11) and CW96 (3), respectively. NC_003092_SHFV, simian hemorrhagic fever virus (56). NC_002534_LDV and LDU15146_LDV-P, lactate dehydrogenase-elevating virus neurovirulent type C (18) and strain Plagemann (36), respectively. AY588319_PRRSV-LV4, AY366525_PRRSV-Eur, AF325691_PRRSV-NVS, AY032626_PRRSV-CH1, AF494042_PRRSV-P12, and AF184212_PRRSV-SP signify porcine reproductive and respiratory syndrome virus strain LV4.2.1, strain EuroPRRSV (38), isolate NVSL 97-7985 IA 1-4-2, strain CH-1a, isolate P129, and strain SP (44), respectively.

FIG. 3.

(A) Plaque assay of wild-type control virus EAV1a and nsp4 mutants EAV1a-G1 and EAV1a-G2. BHK-21 cells were transfected with in vitro-transcribed infectious (wt or mutant) RNA. The representative plaque morphology obtained with progeny virus harvested at 48 h p.t. is shown. (B) Virus growth curves. BHK-21 cells were transfected with infectious RNA generated from wt (pEAV1a) and mutant (pEAV1a-G1 or -G2) full-length cDNA clones. Samples of the transfected cell culture supernatants were taken at 14, 23, 32, 38, and 45 h p.t. and subjected to virus titration as described in Materials and Methods.

FIG. 4.

Expression and processing of the full-length EAV pp1a using the recombinant vaccinia virus/T7 RNA polymerase expression system (see Materials and Methods). Immunoprecipitation analysis of the pL1a-ΔC mutant and the pL1a HR mutants after a 5-h labeling interval by use of an anti-nsp7-8 serum (panel A) and a mixture of anti-nsp3 and anti-nsp4 sera (panel B). Because nsp2 is present in this system, pp1a processing occurs mainly via the major pathway (in which the nsp5/6 and nsp6/7 sites are not cleaved), although small amounts of the products of the minor pathway (no cleavage of the nsp4/5 site, but cleavage of the nsp5/6 and nsp6/7 sites instead) can also be detected. Mock-transfected cells, wt pL1a, and the pL1a-N control mutant were included as controls. nsp2 is coprecipitated due to complex formation with nsp3 and/or nsp3-containing processing intermediates. The positions of the molecular mass markers used during SDS-PAGE and the various processing products are indicated.

FIG. 5.

Pulse-chase experiment to analyze the in vitro processing of wt full-length EAV pp1a and mutants G3, P, and ΔC. Proteins were ³⁵S labeled for 15 min, chased up to 300 min, and then immunoprecipitated with an anti-nsp7-8 serum. nsp4-5 and nsp4-6 are presumably coimmunoprecipitated by an unknown interaction with the precursor or other products brought down by the anti-nsp7-8 serum. For more information, see also the legend for Fig. 4.

FIG. 6.

Expression and processing of the EAV nsp4-8 polyprotein using the recombinant vaccinia virus/T7 RNA polymerase expression system (see Materials and Methods). The nsp4-8 polyprotein is processed via the minor pathway exclusively. In the absence of nsp2, the nsp4 proteinase is unable to process the nsp4/5 junction, but the other cleavage sites can be processed (52). nsp4-5 and nsp4-6 are presumably coimmunoprecipitated due to an uncharacterized interaction with nsp7- and/or nsp8-containing polypeptides. Wild-type pL1a3′, pL1a3′-N, and mock-transfected cells were included as controls. The positions of the molecular mass markers used during SDS-PAGE and the various processing products are indicated.

FIG. 7.

Pulse-chase experiment to analyze the in vitro processing of the wt EAV nsp4-8 polyprotein and mutants G3, P, and ΔC. Proteins were ³⁵S labeled for 15 min, chased up to 300 min, and then immunoprecipitated with an anti-nsp7-8 serum. For more information, see also the legend for Fig. 6.