The current study aimed to determine the humoral immune response to the PRRSV nsps and to develop new tools for identification of PRRSV-infected animals. Previous studies of the humoral immune response to PRRSV focused mainly on detection of antibodies to viral structural proteins, especially N (9
). Several studies showed that certain nsps, such as nsp1 and nsp2, are highly immunogenic (4
). Antibody responses to linear epitopes in nsp2 have been reported to appear within 1 to 4 weeks of infection in type I and type II PRRSV strains (4
). Johnson et al. (13
) observed robust and rapid cross-reactive antibody responses induced by nsp1 and nsp2 to vaccine and field isolates and substantially higher levels of immunoreactivity related to conformational epitopes. In this study, our data demonstrated that nsp7 is also highly immunogenic. Analysis of the kinetics of antibody response showed that response to nsp7 is comparable to antibody responses to nsp1 and nsp2, as well as antigens used in the commercial Idexx ELISA. As indicated by Johnson et al. (13
), nsps are available from the earliest time of infection for presentation to the immune system in the context of major histocompatibility complex class I antigen presentation pathways. As cytolytic infection also releases viral proteins into interstitial spaces, it is hypothesized that a pronounced antibody response, equivalent to the immune response to structural proteins, would be generated to nsps. One intriguing feature of the antibody response to nsp antigens was the sustained antibody titers over a 202-day period of infection, while the antibody response to Idexx antigen, N protein, showed a gradual decay in titers after 126 days postinoculation. The mechanism for sustained levels of nsp antigen may reflect the long-term retention and presentation of nsp to the immune system.
To select an antigen for diagnostic-test development, we compared the correlation between the PRRSV nsp ELISA and the Idexx ELISA. Our results showed that nsp2- and nsp7-based ELISAs had higher correlation with the Idexx ELISA. We further compared the amino acid sequences of nsp2 and nsp7. Our previous studies showed that the PRRSV nsp2 region is highly variable within and between genotypes, with 70.6% to 91.6% amino acid identity within type I PRRSV and 74.9% to 95.6% amino acid identity within type II PRRSV, but only 33.8% identity between type I and type II genotypes (7
). The central region of nsp2 contains hypervariable domains with insertions and deletions (10
), and most identified B-cell epitopes are located in these regions (4
). In contrast, nsp7 is relatively conserved within each genotype and is divergent between genotypes. Amino acid sequence comparisons showed that nsp7 shares 96.7% to 97.4% amino acid identity within type I PRRSV and 84.9% to 100% amino acid identity within type II PRRSV, but only about 45% identity between type I and type II genotypes (7
). These results suggest that the nsp7-based ELISA could be able to detect genotype-specific anti-nsp7 antibody responses.
Further validation of the nsp7-based ELISA showed good sensitivity and specificity of the assay as determined by ROC analysis. The two-graph ROC plots of both type I and type II nsp7 ELISAs display the histograms of the uninfected and PRRSV-infected populations and demonstrate minimal overlap of the two populations (Fig. ). The overlap between the two populations was attributed to eight samples from the type I PRRSV-infected population and nine samples from the type II PRRSV-infected population that had values below the established cutoff. Closer examination of these 17 samples revealed that all demonstrated strong background on the negative control well of the ELISA plate, which suggests that the serum may contain other nonspecific components that interacted with the secondary antibody. In addition, eight of these samples were hemolyzed, which indicates that the serum collection and processing steps were not completed under optimal conditions. There were four samples from the negative population that demonstrated positive results on the type I PRRSV ELISA and three samples from the negative population showing positive results on the type II PRRSV ELISA. The Idexx ELISA S/P values of these seven samples ranged from 0.2 to 0.3. This observation may support the practice by some veterinarians of using follow-up testing for any samples having an S/P value greater than 0.20. We suspected that these samples might be from a herd that had a history of PRRSV infection, since the nsp7 ELISA was able to detect an antibody response up to 202 days postinoculation.
Serology is a standard diagnostic and surveillance method for determining if pigs have been exposed to PRRSV. Currently, the Idexx PRRS ELISA is the most widely used serological assay for determining the serostatus of swine herds. However, positive Idexx ELISA results in otherwise seronegative herds cause concern for producers, which necessitates a variety of follow-up assays to verify that the result is either positive or negative. This indicates that there is still a need for a reliable assay to identify the serological status of single reactors compared to herd reactors. While there is no standard protocol to verify false-positive serological results for PRRSV, most diagnostic laboratories use the IFA and/or virus neutralization assays. However, the results from both of these assays are affected by antigenic variation, and they may not detect a serological response against antigenically diverse PRRSV isolates, such as the European-like PRRSV strains, known as North American type I isolates. The appearance of the type I PRRSV isolates in the United States also complicates the diagnosis of PRRSV, as there is presently no standard serological assay that clearly differentiates between type I and type II strains of PRRSV (7
). The movement of the swine industry toward strategies to eliminate or eradicate PRRSV will require an adequate serological diagnostic assay that can detect acutely and persistently infected pigs, detect various strains of PRRSV, and have the capacity to differentiate between type I and type II PRRSV isolates. The results generated in this study suggest that PRRSV nsp7 could be a potential new antigen for use in ELISA-based diagnostic assays. In particular, using a target other than the N protein, any false positives specifically associated with the N antigen would be avoided.
In summary, our results showed that nsp1, nsp2, and nsp7 induced high levels of antibody response during the course of PRRSV infection. Among these three proteins, nsp7 is the most suitable for diagnostic development, with the following characteristics: (i) nsp7 is expressed as a soluble recombinant protein in bacterial culture, which is convenient for ELISA antigen preparation, especially when applied to diagnostic tests dealing with massive numbers of diagnostic samples; (ii) the PRRSV nsp7 protein coding region is more homologous among different strains within the genotype than those of the other two immunogenic proteins, nsp1 and nsp2; (iii) it is able to detect antibody responses later than 126 days postinoculation. The nsp7-based ELISAs showed good sensitivity and specificity for identification and differentiation of type I and type II PRRSV. Furthermore, the nsp7 dual ELISA resolved 98% of samples with suspected false-positive Idexx ELISA results. Therefore, nsp7-based ELISA has the potential to serve as an alternative or follow-up test for the Idexx ELISA.