Introduction

Porcine reproductive and respiratory syndrome virus (PRRSV) is an enveloped single-stranded positive-sense RNA virus, which is a member of the order Nidovirales, family Arteriviridae accompanying lactate dehydrogenase-elevating virus (LDV), simian hemorrhagic fever virus (SHFV), and equine arteritis virus (EAV) [1, 2]. Generally, the syndromes of arteritis virus infection exhibit an persistent acute infection to stillbirth or lethal hemorrhagic fever [3]. Despite the differences in virion size, morphology, and pathogenesis, the evolutionary relationship among nidoviruses is obvious from the organization and composition of the large replicase gene [2, 4, 5]. PRRSV is classified into two genotypes, the European genotype 1 and the North American genotype 2 [5, 6], which share about only 60 % nucleotide sequence identity [7–9]. PRRSV contains a large genome of approximately 15.4 kb, consisting of ten open reading frames (ORFs). ORF1a and ORF1b encode polyproteins, which are processed to 14 nonstructural proteins (nsps) by viral protease [10]. ORF2a, ORF2b, ORF3, ORF4, ORF5a, and ORF5 to ORF7 encode eight structural proteins, namely GP2a, 2b (E), GP3, GP4, ORF5a, GP5, matrix protein (M), and nucleocapsid (N) [2, 11–13]. As it is reported before, nonstructural proteins have been shown to play critical roles in virion replication or proteolytic cleavage [11, 14, 15].

Despite recent advances in Arterivirus replicase dissection, various aspects of nsps are still unknown, particularly in the nsp5-to-nsp8 region [6]. The largest subunit in this region contains nsp7 (269 aa in genotype IP RRSV, 259 aa in genotype II PRRSV), which is flanked by nsp6 and nsp8 that possibly modulate its function, Recently, nsp7 was found to be internally cleaved by the nsp4 protease, yielding small amounts of nsp7α and nsp7β [16]. As the virus infects the host, nsp-specific antibodies are produced in high titers, particularly against nsp1α, nsp1β, nsp2, and nsp7 [10, 17, 18]. Therefore, these antibodies could be used to differentiate between post-infectious and post-vaccination antibodies in animals vaccinated with inactivated vaccine [19]. However, these nonstructural proteins including nsp7, have yet to be fully researched and their major role in virus replication and production still remains unclear. Manolaridis et al. [20] have determined the solution structure of nsp7α of EAV by nuclear magnetic resonance (NMR) spectroscopy, which revealed an interesting fold for this protein. Also, they substantiated the importance of nsp7 for viral RNA synthesis by structure-based reverse genetics studies. Furthermore, some pivotal amid acid sites constituting the secondary structure of EAV were screened and confirmed by site-directed mutagenesis.

The reverse genetics technique provides a very valuable tool for exploring the function of viral proteins by targeted gene manipulation. In our study, we predicted the nsp7 protein spatial structure (Fig. 1a) of PRRSV with the help of I-TASSER server online tool (http://zhanglab.ccmb.med.umich.ed), which displayed the most adjoining one in published protein spatial structures (Fig. 1b) matching that of PRRSV. Consequently, three infectious clones were constructed (Fig. 3a, b): one with deletion of nsp7α, another with deletion of nsp7β, and the third with deletion of nsp7. Furthermore, based on the partial nsp7 amino acid sequence alignments (Fig. 2) between PRRSV and other Arteriviruses, groups of amino acids corresponding with that in the previous report have been found [21]. Further analysis with the software DNAStar Lasergene (Supplemental data) showed that deletions in nsp7 of PRRSV potentially affected viral folding and packaging. Based on the observation, a series of mutants depending on an infectious clone of NA genotype strain XH-GD were constructed (Fig. 3a, c). Finally, the mutant viruses were rescued, and their growth properties were evaluated. The results demonstrated that nsp7α and nsp7β were critical and necessary proteins for viral growth in vitro, as a cDNA infectious clone neither with nsp7α deletion nor with nsp7β deletion could rescue live virus.

Fig. 1
figure 1

The nsp7 spatial structure comparison between EAV and PRRSV. a A flash cartoon representation of the NMR structure of EAV nsp7α [21], α-helices are blue colored, β-strands magenta, and β-turns pink. The N and C termini are labeled, as are the residues (Leu17, Asn31, Phe39, Lys67, Phe71, and Asp81) referred to in the text discussing the mutagenesis results [21]. b A flash cartoon representation of the PRRSV nsp7α spatial structure predicted by I-TASSER server online tool (http://zhanglab.ccmb.med.umich.ed) and analyzed by Swiss-Pdb Viewer 3.7. α-Helices are colored light blue, β-strands and-turns are colored deep pink. The mutated residues (Leu18, Asn32, Phe40, Lys68, Phe72, and Asp84) in text are indicated by the green color (Color figure online)

Full size image
Fig. 2
figure 2

Alignment of partial amino acid sequences of Arterivirus nsp7α proteins. The alignment is based on sequence data for the North American porcine reproductive and respiratory syndrome virus (NA-PRRSV, accession no. NC001961), European porcine reproductive and respiratory syndrome virus (EU-PRRSV, accession no. M962622), Lactate dehydrogenase-elevating virus strain Plagemann (LDV-P, accession no. NC001639), simian hemorrhagic fever virus (SHFV, accession no. NC003092), and equine Arterivirus (EAV, accession no. NC002532). The gray strips indicate conserved residues, the braces display amino acid regions constituted secondary structure of nsp7α, published in a reference paper [14]. Asterisks indicate the PRRSVnsp7 residues that were probed by reverse genetics in this study (Leu18, Asn32, Phe40, Lys68, Phe72, and Asp84), which correspond with residues in EAV nsp7 (Leu17, Asn31, Phe39, Lys67, Phe71, and Asp81)

Full size image
Fig. 3
figure 3

Schematic diagram of PRRSV nsp7 protein deletion and mutation mutants. a Schematic drawing of the complete genome of PRRSV. b Construction strategy of the full-length cDNA clone of PRRSV XH-GD containing AflII and NheI sites in fragment B1 and three internal deletion mutants. Dashed squares indicate the part of deletion in nsp7. c Graphic representation of the selected mutation amino acids in nsp7, arrows indicate the mutated residues

Full size image

Materials and methods

Cells, viral strains, and plasmids

MARC-145 and BHK-21 were maintained in Eagle’s minimal essential medium (EMEM) supplemented with 10 % fetal bovine serum at 37 °C with 5 % CO2. PRRSV XH-GD (GenBank accession no. EU624117), belong to HP-PRRSV. The infectious PRRSV cDNA clone pOKXH-GD (unpublished data) was constructed by the XH-GD strain and plasmid pOKq [20]. It served as the backbone for all nsp7 deletions and mutations used in the study. Nsp7 was located in fragment B1 [20] based on the process of constructing the cDNA clone. Therefore, all deleted and mutanted operations were solely performed in fragment B1 by an overlapping extension PCR technique, and every deletion region and mutation site are shown in Fig. 3a–c. Primers used in the construction of each deletion and mutation mutant are displayed in Table 1. In two separate PCRs, two separated fragments A and B of the target gene region were amplified in the first cycle. In the second-cycle of PCR, by using two internal overlapping primers, fragments A and B were fused together by denaturing and annealing them in a subsequent primer-extension reaction. The overlap region allowed one strand from each fragment to act as a primer on the other, and extension of these overlaps resulted in the deletion or mutation product. The PCR products were double digested with restriction enzymes AflII and NheI, then ligated into the infectious clone plasmid pOKXH-GD.

Table 1 Primers used in this study, two pairs of primers augmenting deletion nsp7 fragment are B1-F/Δnsp7-R and Δnsp7-F/B1-R, other deletion and mutation fragments are also augmented by similar methods
Full size table

DNA transfection and recovery of deletion and mutation viruses

Deletion and mutant plasmids were purified using a QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany). BHK-21 cells were grown to 70–80 % confluence in six-well plates and the 2 μg plasmid was transfected by using 5.88 μl Lipofectamine™ 2,000 reagent (Invitrogen) diluted in 250 μl OPTI-MEM (GIBCO) according to the manufacturer’s instructions. Rescued viruses were designated as the primary passage (P0), and used to infect MARC-145 cells for twelve subsequent passages (P1–P12) using 103-fold dilutions at each passage. In order to avoid the occasionality of the rescued process, the recovered operation on mutants was repeated at least three times independently.

Indirect immunofluorescence assay (IFA)

IFA was performed for the detection of viral antigens in infected cells. MARC-145 cells were infected with individual mutant viruses. At 48 h post-inoculation (h.p.i), infected cells were washed five times with PBS, followed by fixation in 4 % paraformaldehyde. The cells were incubated at 37 °C for 2 h with the monoclonal antibody against PRRSV N protein (mAbN), which is a gift kindly provided by Dr. Guangzhi Tong from Shanghai Veterinary Research Institute, Chinese Academy of Agricultural Sciences, China. After extensively washing with PBS, the cells were incubated for 1 h with fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (1:100 dilution). The cells were washed five times with PBS. Finally, the fluorescence was observed under an Olympus inverted fluorescence microscope.

Multi-step growth curve

Subconfluent MARC-145 cells in six-well plates were infected with rescued viruses (P3) at a multiplicity of infection (MOI) of 0.01. After 1-h incubation at 37 °C, cells were washed three times with PBS and incubated at 37 °C in 2 ml EMEM containing 2 % FBS in a CO2 incubator. At certain time points (12, 24, 36, 48, 72, 96, and 108 h) post-infection, supernatants were collected and frozen at −80 °C till use. The viral titers (TCID50/ml) were measured in MARC-145 cells and calculated using the Reed–Muench method [22].

Viral plaque assay

To examine the plaque size of the mutant viruses, tenfold serial dilutions of virus suspensions (P3) were inoculated into MARC-145 cells in six-well plates, respectively. After 1 h adsorption, cell monolayers were washed and then covered with a mixture of EMEM containing 2 % FBS and 1 % low-melting agarose (Cam-brex, Rockland, ME, USA). When the agarose overlay solidified, the plates were inverted and incubated at 37 °C for 4–5 days in a humidified 5 % CO2 incubator. The resulting plaques were stained with crystal violet (5 % w/v in 20 % ethanol).

Quantitative RT-PCR and RT-PCR

In order to analyze the effect on genomic mRNA synthesis by these mutations, SYBR Green real-time PCR was performed by using specific primers based on the sequence of the nsp7 gene, which was: nsp7Y-F:5′-CACTGACGACGTCGTGAGATCC-3′, nsp7Y-R:5′-ACAATCGCACTCGCGAGAG -3′. β-actin served as an internal reference gene and the specific primers were as follows: actin-F:5′-TCGATCATGAAGTGCGACGTG-3′, actin-R:5′-GTGATCTCCTTCTGCATCCTGTC -3′. Viral particles were added at a multiplicity of infection of 0.01 and 36 h after infection of 0.01 MOI, the cells were lysed using TRIzol Reagent (Invitrogen). Total RNA was extracted according to the manufacturer’s instruction. Reverse transcription was carried out by using MLV-RT (Promega) in a 20 μl reaction mixture containing 2 μg RNA each according to the manufacturer’s recommendation. Amplification was carried out in a 25 μl reaction volume containing 1 μl cDNA, 12.5 μl Power SYBR Green PCR Master Mix (Takara Bio Inc., Japan), 0.5 μl Rox, 0.4 μM of each primer. The reaction procedure occurred at 95 °C for 10 min, followed by 40 cycles at 95 °C for 15 s, and 62 °C for 1 min. The real-time PCR was performed in an ABI PRISM 7500 sequence detection system and analyzed with ABI PRISM 7500 software (Applied Biosystems). Cycle times of internal reference that varied by >1.0 unit in triplicate were discarded. Cycle times of the target gene were averaged and normalized to the average cycle time of β-actin accordingly.

Sequencing of RT-PCR products was carried out to analyze the introduced mutations and deletions in nsp7. The nsp7 fragment was amplified by Taq DNA polymerase (TaKaRa Dalin, China) using forward primer nsp7-F and reverse primer nsp7-R. The PCR products were purified using a TIANgel Mini Purification Kit (TIANGEN, China) and sequenced.

Statistical analysis

Data were expressed as means ± standard deviations. The significance of the variability among the trials was determined by one-way analysis of variance using GraphPad Prism (version 5.0) software.

Results

Deletions of nsp7α/nsp7β and nsp7 were fatal for recovery of virus

After the MARC-145 cell monolayers were inoculated with the supernatants obtained from BHK-21 cells transfected with the constructed plasmids, the cell culture supernatants were harvested and passaged into MARC-145 cells. As visualized in IFA, a specific fluorescence against N protein of PRRSV would be observed, provided deletion mutants were rescued successfully. When compared with wild virus XH-GD as a positive control, nsp7, nsp7α/nsp7β deletion mutants did not appear to display representative fluorescence with the exception of vOKXH-GD (Fig. 4).

Fig. 4
figure 4

The viability of nsp7 deletion mutants by intracellular N expression. MARC-145 cells were transfected with plasmids of pOKXH-GD, pOKXH-Δnsp7α, pOKXH-Δnsp7α, and pOKXH-Δnsp7 as indicated. The wild virus was used as positive control. Expression of N protein was visualized by immunofluorescence staining at 72 h.p.t

Full size image

Mutations in individual amino acid sites in nsp7 have diverse effects on the infectivity of the rescued viruses in MARC-145 cells

Based on the results from the previous reports [21] and results of the online protein spatial structure prediction software on nsp7 of PRRSV, eight amino acid sites were selected, as they were predicted to be the key positions of the secondary structure of nsp7 and might play an important role in protein folding. Each amino acid site was substituted to generate the targeted amino acid residues as shown in Fig. 3c. As a result, eight amino acid sites mutant plasmids were found: pOKXH-nsp7(L18D), pOKXH-nsp7(N32A), pOKXH-nsp7(F40A), pOKXH-nsp7(K68A), pOKXH-nsp7(F72D), pOKXH-nsp7(D84A), pOKXH-nsp7(D84E), and pOKXH-nsp7(D84N). These mutant plasmids were transfected into BHK-21 cells. At 72 h after transfection, supernatants of the transfected cells were collected and used to infect fresh MARC-145 cells. As visualized in IFA, only vOKXH-nsp7(F40A), vOKXH-nsp7(K68A), and positive control vOKXH-GD and wild virus XH-GD expressed the PRRSV N protein and spread into the neighboring cells at 72 h.p.i (Fig. 5). These results indicated that mutant viruses vOKXH-nsp7(L18D), vOKXH-nsp7(N32A), vOKXH-nsp7(F72D), vOKXH-nsp7(D84A), vOKXH-nsp7(D84E), and vOKXH-nsp7(D84N) containing single mutations at L18, N32, F72, and D84 in nsp7 were fatal for cellular infectivity. On the contrary, mutant viruses vOKXH-nsp7(F40A) and vOKXH-nsp7(K68A) retained their cellular infection, suggesting that all eight mutant amino acid sites had diverse effects on virus viability.

Fig. 5
figure 5

The viability of nsp7 deletion mutants by intracellular N expression. MARC-145 cells were transfected with plasmids of pOKXH-GD, pOKXH-nsp7(L18D), pOKXH-nsp7(N32A), pOKXH-nsp7(F40A), pOKXH-nsp7(K68A), pOKXH-nsp7(F72D), pOKXH-nsp7(D84A), pOKXH-nsp7(D84N), and pOKXH-nsp7(D84E) and the wild virus was used as positive control. Expression of N protein was visualized by immunofluorescence staining at 72 h after infection

Full size image

Mutation at amino acid position 40 in nsp7 did not affect virus viability in MARC-145 cells but did decrease virus titer and reduced plaque

To further characterize the mutants vOKXH-nsp7(F40A) and vOKXH-nsp7(K68A) carrying a mutation in nsp7, multi-step growth kinetic was determined in infected MARC-145 cells. As shown in Fig. 6a, the growth kinetic of the mutant vOKXH-nsp7(K68A) was similar to that of wild virus XH-GD and positive control vOKXH-GD while the mutant vOKXH-nsp7(F40A) has a remarkable difference compared with others. The time needed to reach the highest virus titer of mutant vOKXH-nsp7(F40A) was 36 h longer than that of positive control vOKXH-GD and wild virus XH-GD. In addition, the overall yield of the mutant vOKXH-nsp7(F40A) was nearly two log less than that of positive control vOKXH-GD virus while that of the mutant vOKXH-nsp7(K68A) was not significantly different from positive control vOKXH-GD virus. The viral plaque size was also determined on MARC-145 cells to monitor any phenotype changes of mutants. The plaques produced by positive control vOKXH-GD virus, wild virus XH-GD, and mutant vOKXH-nsp7(K68A) viruses were similar while the plaque generated by mutant vOKXH-nsp7(F40A) was slightly smaller than that generated by wild virus and positive control vOKXH-GD (Fig. 6b).

Fig. 6
figure 6

Virological properties of nsp7 protein mutation mutants. a Multi-step growth kinetics of wt and mutant viruses in MARC-145 cells. Cells in six-well plates were infected with PRRSV at an MOI of 0.5. The culture supernatants were collected at the indicated time points and titrated. The geometric mean titers with standard deviations (error bars) from three independent experiments are shown. b Viral plaque size. Transfectant supernatants were serially diluted tenfold and inoculated into young MARC-145 cells in six-well plates. The MARC-145 cell monolayers were cultured in EMEM containing 1 % agarose overlay, fixed at 4 d.p.i, and stained with 1 % crystal violet

Full size image

Confirmation of the introduced mutations and stability of the mutants

To determine whether the introduced mutations were indeed maintained in nsp7 of the mutant vOKXH-nsp7(F40A) and vOKXH-nsp7(K68A), supernatants were collected from MARC-145 cells infected with each passage of the mutants. The complete sequence of different generations containing the nsp7 region was amplified in RT-PCR followed by nucleotide sequencing. The results confirmed the engineered deletions in all of the mutants. To assess the stability of mutants vOKXH-nsp7(F40A) and vOKXH-nsp7(K68A), mutant viruses collected from the infected MARC-145 cells and supernatants were serially passaged till the 12th passage. The 1st, 3rd, 6th, 9th, and 12th passage viruses of mutants were selected for measuring the viral titers (TCID50/ml) in MARC-145 cells and calculated using the Reed–Muench method. Data showed that the overall yields in the first three passage viruses (1st, 3rd, and 6th) of mutants were slightly less than that of wt virus but the overall yields of the remaining two (9th, 12th) were equitable to the positive control vOKXH-GD and wild virus (Fig. 7).

Fig. 7
figure 7

The comparison of stability among nsp7 protein mutants. Cells in six-well plates were infected with PRRSV at an equal number of viral particles. Culture supernatants were collected at 24 h.p.i and titrated. The geometric mean titers (TCID50/ml) with standard deviations (error bars) from three independent experiments are shown

Full size image

Impact on mRNA level of virus produced by mutation in amino acid position 40 in nsp7

In order to measure the effect on the gene level generated by mutation in amino acid positions 40 and 68 in nsp7, total RNA of the cell culture in 24-well plates was extracted at 36 h after infection at a multiplicity of infection (MOI) of 0.5, and used for quantitative real-time PCR detection with the primers directed to nsp7, and β-actin as an internal reference gene. The results indicated that in the groups infected with vOKXH-nsp7(F40A), the level of nsp7-specific mRNA was reduced by about 50-fold in comparison with that in the vOKXH-GD infected group. In the group transfected with OKXH-nsp7(K68A), the level of nsp7-specific mRNA was 45-fold higher than that in the vOKXH-nsp7(F40A) transfected group, but was also more or less reduced compared to the vOKXH-GD transfected group (Fig. 8a).

Fig. 8
figure 8

The effect on the intensity level of viral mRNA and translation by mutants. a The levels of nsp7 mRNA were normalized to the level of β-actin mRNA in the same sample. Error bars represent deviation of the median from three experiments. The MARC-145 cells were infected with 0.5 MOI. About 36 h post-infection, total RNA was extracted and subjected to real-time RT-PCR analysis. vOKXH-GD transfected cells were used as control. The mRNA of β-actin served as an interval reference. b The IFA was used to monitor protein expression level. Pictures were taken at 36 h post-transfection with the same amount of virus using an AxioCam MRc5 camera on an inverted fluorescence microscope (ZEISS Axiovert 200)

Full size image

Impact on viral protein translation produced by mutation in amino acid position 40 in nsp7

In order to determine if the mutation in amino acid positions 40 and 68 in nsp7 could effect N protein translation of PRRSV GD-XH strain, MARC-145 cells were infected with equal amounts of mutant vOKXH-nsp7(F40A) and vOKXH-nsp7(K68A). The density and intensity of the GFP fluorescence in the infected cells were measured by an Olympus inverted fluorescence microscope at 48 h after infection. The results showed that cells infected with mutant vOKXH-nsp7(F40A) had relatively less fluorescence compared to the positive control group infected with vOKXH-GD, while the fluorescence of mutant vOKXH-nsp7(K68A) was similar to the positive control group (Fig. 8b).

Discussion

Among all nonstructural proteins of Arterivirus (PRRSV, EAV), nsp7 is one of the most conservative proteins and nsp7α could be a critical protein domain for EAV RNA synthesis [21]. In the study, we not only demonstrated that nsp7 and nsp7α/nsp7β are essential for viral viability and recovery, but also found that several amino acid sites contribute to the secondary structure of nsp7 and have divergent effects on virus survival in vitro. Single mutations at amino acid positions 18, 32, 72, and 84 in nsp7 may be fatal for recovery of the infectious virus, but mutations at positions 40 and 68 of nsp7 will not affect viral viability. To make sure that no other alterations were introduced, the full-length of the plasmid were sequenced, the result revealed that no other mutations were found in anywhere else in the genome except the designated mutations.

Previous report demonstrated that nsp-specific antibodies are produced in high titers in response to virus replication, especially against nsp1a, nsp1b, nsp2, and nsp7 [17]. However, numerous aspects of nsp7 have yet to be clarified and probed. Manolaridis et al. [20] demonstrated the function of nsp7 in EAV through reversed genetic manipulation technique. Furthermore, some significant amino acids in the spatial structure of nsp7 had been located and proved to be the key sites for EAV RNA synthesis. These residues were Leu17, Arg24, Arg33, Phe39, Lys67, Phe71, and Asp81 (Fig. 1a). Therefore, based on an online protein structure prediction tool and alignments among Arteriviruses, we found the corresponding residues in nsp7 protein of PRRSV. The sites selected for mutagenesis were thought to be important for protein folding, as deduced from the structure (Leu18, Phe72, and Asp84) or residues with solvent-accessible side chains(Arg32, Phe40,and Lys68). Moreover, all amino acids applied for substitution were selected based on criteria. Ala (A) was the simplest and smallest amino acid with R group “H” of all polar amino acids Ala (A), Val (V), Leu (L), Ile (I), Pro (P), and Phe (F). On the contrary, Asp (D), Glu (E) are negatively charged and Lys (L) is positively charged while Gln (Q) is a neutral, all of which are nonpolar amino acids. Hence, the substitutions occurred among amino acids that have the ability to undergo radical transformation in their protein spatial structures. Especially in substitutions of Ala and other amino acids, it is predicted that changes occurred in the protein folding and formation.

According to a previous report by Manolaridis et al. [20], only these mutations Arg33-Ala, Lys67-Ala, Asp81-Asn, and Asp81-Glu in all amino acids, which constituted the secondary structure of nsp7 in EAV, had been rescued successfully for viable virus, but none of the mutations in Leu17-Asp, Phe39-Ala, Phe71-Asp, and Asp81-Ala was recovered. Through nsp7 sequence alignments between EAV and PRRSV, the corresponding amino acids were Leu18, Arg32, Phe40, Lys68, Phe72, and Asp84 in nsp7 of PRRSV. Our results revealed that only two mutants with single mutations at Phe40-Ala and Lys68-Ala, which were predicted as composing spatial structure, had been rescued out, while mutations that occurred in Leu18-Asp, Arg32-Ala, Phe72-Asp, Asp84-Ala, Asp84-Asn, and Asp84-Glu failed. Therefore, the negative consequences from mutations at Leu18-Asp, Phe72-Asp, and Asp84-Ala in nsp7 of PRRSV were similar to those of mutations at Leu17-Asp, Phe71-Asp, and Asp81-Ala in nsp7 of EAV. Meanwhile, the positive result from mutation at Lys68-Ala of nsp7 in PRRSV was consistent with that of Lys67-Ala of nsp7 in EAV. The inconformity in several mutations between the two viruses is most likely due to the specificity of the two viruses. Moreover, our results from analyzing the recovered viruses were in accordance with the predictions based on the spatial structure of nsp7 in PRRSV by DNAStar Lasergene software. Those predictions showed that there were noticeable differences in the secondary structure of nsp7 when mutations occurred in Leu18, Arg32, Phe72, and Asp84, but not in Phe40 and Lys68 (see supplement material). Leu18 and Phe72 were predicted to be buried within the hydrophobic cores, since the mutation L18D will lead to the increased possibility of turn region and decreased α-region while both decreased in the mutation (F72D),which might disrupt the formation of the core (see supplement material). Meanwhile, positions like Arg32, Phe40, and Lys68 are exposed on the protein surface and therefore could be of functional important, as it is in accordance with the secondary structural prediction, no noticeable differences in the secondary structure were found in the mutation of Phe40 and Lys68, however, Arg32 mutation will definite affect the secondary structural (see supplementary document), that might be the reason why the mutations of F40A, K68A is not fatal for the recovery of the virus but it is for the mutation of N32A. As is predicted, Asp81 formed a salt bridge that keeps the α-helical bundle stacked against the β-sheet in EAV, it is the same to the Asp84 in PRRSV, thus, the mutations will obviously affected the structure of the protein, as is in accordance with the secondary structural prediction [21].

To further analyze the differentiation between rescued viruses and the parental virus, we performed a multi-step growth curve and a plaque assay. Results revealed that the highest viral titer of mutant vOKXH-nsp7(F40A) was approximately two log less than that of parental virus vOKXH-GD, and the time of reaching its highest viral titer was almost 36 h more compared to the parental virus. Furthermore, in order to assess if mRNA and protein translation levels had any alterations caused by mutant vOKXH-nsp7(F40A), quantitative RT-PCR and IFA were adapted. The data presented in Fig. 8a illustrats that in the groups transfected with vOKXH-nsp7(F40A), the level of nsp7-specific mRNA was reduced by about 50-fold in comparison with that in the vOKXH-GD infected group. However, in the group transfected with vOKXH-nsp7(K68A), the level of nsp7-specific mRNA was approximately five times less than that in the vOKXH-GD transfected group. These results showed mutations at amino acid positions 40 and 68 had inhibited the replication and RNA synthesis of the virus, but the mutation in position 40 had a more obvious effect than that in position 68. More importantly, data from the IFA displayed that the MARC145 cell infected with mutant vOKXH-nsp7(F40A) had less fluorescence than the cells infected with vOKXH-GD and vOKXH-nsp7(K68A).Consequently, this demonstrated that a mutation in amino acid position 40 could impact on the infectivity rate of PRRSV and expression of the viral protein to some degree. These results were also in accordance with the previous data from growth kinetic curves and plaque assays.

In the study, the structural prediction defines PRRSV nsp7α, the most conserved region of nsp7, is likely preserved in the C-terminally extended full-length nsp7. All deletions (nsp7, nsp7α/nsp7β) in nsp7 and the majority of substitutions of nsp7 residues affected the replicate of PRRSV indicating that they are structurally or functionally important. What’s more, the mutant vOKXH-nsp7(F40A) indeed caused some variation compared with the parental virus, postponing the time of reaching the highest viral titer, decreasing the highest viral titer, and reducing the level of intensity of mRNA and protein translation of the virus. Even though the nsp7α structure prediction and reverse genetics analysis provided us limited clues on the hypothetical function of this protein, some valuable results possibly provide the first direct evidence that nsp7 could be a critical protein domain for PRRSV RNA synthesis and viral protein translation. In summary, nsp7α is part of many replicase-processing intermediates. As in other plus-stranded RNA virus systems, e.g., picornaviruses and alphaviruses [23–27], such intermediates may perform specific functions themselves and proteolytic cleavages may serve as a switch to activate them.