Neutralizing antibody formation in swine infected with seven strains of porcine reproductive and respiratory syndrome virus as measured by indirect ELISA with peptides containing the GP5 neutralization epitope.

An indirect ELISA with peptides containing the GP5 neutralization epitope was used to measure the time courses of formation of neutralizing antibodies in sera of groups of 10 pigs infected with seven different strains of porcine reproductive and respiratory syndrome virus (PRRSV) generated in an earlier study (Johnson et al., Vet.Immunol. Immunopath. 102:233-247. 2004). The neutralizing antibody responses varied greatly between individual pigs infected with each PRRSV strain. Some pigs generated high titers of neutralizing antibodies between 7 and 28 days post infection (p.i), whereas other pigs had not generated a significant response by 42 days p.i. The heterogeneity in antibody formation in individual pigs also pertained to anti-N-protein antibody formation as measured by both HerdCheck and peptide ELISAs and there was no correlation between anti-N-protein and neutralizing antibody formation. Also, viremia was disconnected from neutralizing antibody formation. Viremia peaked 7-15 days p.i. and then precipitously declined to undetectable levels in most infected pigs by 21-28 day p.i. whether or not they had generated neutralizing antibodies. Furthermore, contrary to an earlier conclusion from this study the formation of neither neutralizing nor anti-N-protein antibodies was related to viral load as measured by infectious virus levels in serum.

Neutralizing antibody responses of pigs infected with natural GP5 N-glycan mutants of porcine reproductive and respiratory syndrome virus.

In order to assess the effect of the N-glycans associated with the GP5 neutralization epitope of porcine reproductive and respiratory syndrome virus (PRRSV) on the neutralizing antibody (Ab) response of swine, groups of young pigs were infected with PRRSV strains differing in N-glycosylation pattern. The humoral immune response to strain VR-2332, harboring four potential N-glycan sites, was compared to that of two natural field isolates carrying mutations either abolishing the N-glycosylation site at position 44 (N44) or the two N-glycosylation sites in the hypervariable region upstream of the neutralization epitope (HV-1). The pigs were bled at intervals and their sera were assayed for neutralizing Abs by indirect and competition ELISAs using peptides containing the GP5 neutralization epitope, and selectively for infectivity neutralization of a number of PRRSV strains. In addition, viremia was monitored by quantitative RT-PCR, and anti-N-protein Ab formation was measured by HerdChek ELISA. The neutralizing Ab responses as measured by peptide ELISA varied greatly between individual pigs infected with each PRRSV strain. Some pigs generated high titers of peptide binding Abs between 7 and 28 days post infection (p.i.), whereas other pigs had not generated a response by 90 days p.i. However, the HV-1-infected pigs generated Abs to the neutralization epitope more rapidly and to a 5-10 times higher level than VR-2332-infected pigs, and the Abs neutralized the homologous HV-1 virus 10-20 times more efficiently than PRRSV strains VR-2332, N44, MN184, or SDSU73. In contrast, most N44-infected pigs generated neutralizing Abs only after 42 days p.i. and only to low levels. The results suggest that the deletions of the N-glycans or other amino acid substitutions in the GP5 ectodomains of the mutants affect the immunogenicity of the neutralization epitope and the specificity of the Abs raised to it but not the sensitivity of the virions to Ab neutralization.

Peptide ELISA for measuring antibodies to N-protein of porcine reproductive and respiratory syndrome virus.

Indirect and competition ELISAs with synthetic peptides were used to characterize the epitopes of the N-protein of porcine reproductive and respiratory syndrome virus (PRRSV) that are recognized by a battery of monoclonal antibodies (mAbs) and by antibodies from infected pigs. Four linear epitopes recognized by mAbs have been identified in the most hydrophilic segment of the N-protein (AA25-57). Similarly, at least four linear epitopes in this segment are immunogenic in PRRSV-infected pigs, but only one corresponds to an epitope recognized by one of the mAbs (AA36-45). Antibody formation to these epitopes varied greatly between individual pigs. Most infected pigs generated antibodies that bound to both peptides and HerdChek plates, which are commonly used in the sero-diagnosis of PRRSV infections, but the time course of formation of peptide binding antibodies and antibodies that react with HerdChek plates differed greatly between pigs. This suggests that, although the peptide and HerdChek ELISAs may detect antibodies to some of the same epitopes, they also seem to detect antibodies to epitopes that are uniquely expressed by one and not the other. Some mAbs fail to bind to HerdChek ELISA plates and this is also the case for certain pig antibodies. Peptide ELISA results identified four herds in which most or all pigs possessed N-protein peptide binding antibodies, even though they were HerdChek ELISA sero-negative and exhibited no other signs of PRRSV infection. Thus PRRSV infections may be more widespread than presently realized involving strains that cause asymptomatic infections. The peptide ELISA is useful as an adjunct to the HerdChek ELISA or it could replace it since only two serum samples among 450 tested were HerdChek ELISA positive but peptide ELISA negative. The peptide ELISA is also considerably cheaper than the HerdChek ELISA, more flexible and can provide information on the epitope specificity of the reacting antibodies.

Polyclonal hypergammaglobulinemia and formation of hydrophobic immune complexes in porcine reproductive and respiratory syndrome virus-infected and uninfected pigs.

Infection of young conventional, domestic pigs with porcine reproductive and respiratory syndrome virus (PRRSV) strains VR2332 and JA142 resulted in a rapid, progressive increase in serum IgG reaching maximum levels of 20-30 mg/mL at about 3 weeks post infection (p.i.), which were maintained until at least 63 days p.i., whereas the level of serum IgG remained at 4-6 mg/mL in sham infected pigs. In most of the VR2332 and JA142-infected pigs hypergammaglobulimenia was associated with the formation of hydrophobic, 150-300-kDa IgG-containing immune complexes that bound in the presence of 0.1% Tween 20 to ELISA plates that were not coated with any antigen. The ELISA plate-binding activity remained low in most infected pigs, but reached high levels in some JA142-infected pigs. Binding of the immune complexes was also observed, but at a lower level, to uncoated ELISA plates in the peptide ELISA for anti-PRRSV Abs. The immune complexes bound to uncoated ELISA plates with a much lower affinity than Abs to plates coated with peptides containing the appropriate epitopes. The immune complexes also bound to HerdChek ELISA plates, but because of low binding affinity for these plates, the bound complexes were removed by the repeated washes with Tween 20 solution. Overall the PRRSV-induced hypergammaglobulinemia and generation of ELISA plate-binding immune complexes resembled those observed in mice infected with the closely related lactate dehydrogenase-elevating virus (LDV) and thus, like the latter, seem a result of a polyclonal activation of B cells. We also found that sera of a group of older sows possessed high levels of IgG as well as of ELISA plate-binding immune complexes, in spite of being PRRSV infection negative by all criteria presently available. On the other hand, sera from wild hogs contained no ELISA plate-binding IgG in spite of possessing high total serum IgG levels.

Epitope specificity of monoclonal antibodies to the N-protein of porcine reproductive and respiratory syndrome virus determined by ELISA with synthetic peptides.

In an attempt to develop an alternate to ELISAs using recombinant N-proteins as antigen for the sero-diagnosis of porcine reproductive and respiratory syndrome virus (PRRSV) infections of pigs I have measured the binding of nine anti-N-protein mAbs, which had been previously generated by various investigators, to overlapping peptides encompassing amino acids 19-70 of the N-proteins of the North American prototype (VR2332) and the European prototype (Lelystad virus, LV) of PRRSV. I also measured the binding of the mAbs to HerdChek ELISA plates coated with recombinant N-protein. All mAbs bound in an indirect ELISA to some of the peptides whether the mAbs had previously been reported to recognize continuous or discontinuous epitopes, but with different specificity and titer. Three mAbs bound with high titer to different linear epitopes located in amino acid segments 23-33, 31-50 and 43-56 and also with similar high titers to HerdChek plates. mAb SDOW17 bound with high titer to HerdChek plates but poorly to any of the peptides. In contrast, four mAbs bound with broad specificity to peptides containing an epitope(s) in amino acid segment 30-48, but poorly, or not at all, to HerdChek ELISA plates. Thus, this epitope is missing on the antigens of the HerdChek ELISA or is destroyed during immobilization of the antigens on the plate. A mAb to the N-protein of the closely related mouse arterivirus lactate dehydrogenase-elevating virus bound to the same epitope. Abs that bound with broad specificity to an epitope(s) in the 30-50 amino acid segment were also detected by the peptide ELISA in sera of 25 field sera that were sero-positive in the HerdChek ELISA, but also in sera of pigs from two out of three herds tested that were sero-negative by this test.

The primary GP5 neutralization epitope of North American isolates of porcine reproductive and respiratory syndrome virus.

I have used indirect ELISA with overlapping synthetic peptides representing the GP5 ectodomain to study the generation and specificity of peptide-binding Abs in pigs that were infected in utero with porcine reproductive and respiratory syndrome virus (PRRSV) strain VR2332 and in North American field sera submitted for PRRSV infection diagnosis. Peptide-binding Abs appeared in sera of the VR2332-infected pigs within about 30 days post-farrowing (dpf), reaching maximum titers 100-200 dpf and then decreasing slowly to about half of maximum titer by about 400 dpf. The formation of peptide-binding Abs and of virus neutralizing Abs correlated and their initial appearance coincided with disappearance of virus from the circulation. The Abs were specific for VR2332-specific peptides. In contrast, anti-N-protein Abs as measured by HerdCheck ELISA appeared within 7 dpf, reached maximum levels at about 100 dpf and had decreased below detectable levels by about 200 dpf. Twenty-seven field serum samples with virus neutralizing activity all possessed high levels of peptide binding Abs, but the Abs bound about equally to VR2332 and strain Lelystad virus (LV)-specific peptides. The indirect ELISA results using various large peptides and competition ELISA using small peptides (8 or 9 amino acids long) confirmed that the epitope recognized by the Abs is located in the GP5 ectodomain sequence 37SHLQLIYNL of VR2332. Use of mutated peptides in the competition ELISA showed that 42I to T and 38HL to TY substitutions blocked Ab recognition, whereas deletion of 41L had no effect. In addition, 26 serum samples submitted by two farms for diagnostic tests were found to possess low levels of Abs that bound to GP5 ectodomain peptides, even though the sera were sero-negative in the HerdChek ELISA and lacked neutralizing activity. Competition ELISA showed that the Abs recognized one or more epitopes located downstream of the PRRSV neutralization epitope. An epitope(s) located in the same area was recognized by Abs generated in mice by immunization with a GP5 ectodomain peptide conjugated to BSA. These Abs also lacked neutralizing activity.

GP5 ectodomain epitope of porcine reproductive and respiratory syndrome virus, strain Lelystad virus.

Sera from pigs infected with the European porcine reproductive and respiratory syndrome virus (PRRSV), strain Lelystad virus (LV), were screened by indirect ELISA for antibodies that bound to a series of overlapping synthetic peptides covering amino acids (AA) 19-60 of the primary envelope glycoprotein (GP)5. Antibodies were detected that bound to an epitope(s) located in an ectodomain segment composed of AA 38-54. The antibodies strongly cross-reacted with peptides specific for the North American PRRSV VR2332. Antibodies with the same specificity were present in sera of pigs infected in the US with a European-like PRRSV. Competitive ELISA using overlapping 8-10-AA-long peptides confirmed that the epitope recognized by the antibodies from LV-infected pigs is located in the same segment as the neutralization epitopes previously identified for PRRSV VR2332 and the closely related arterivirus, lactate dehydrogenase-elevating virus (LDV). No antibodies were detected that bound to synthetic peptides representing further upstream GP5 segments that have been reported to carry neutralization or non-neutralization epitopes of some PRRSV isolates.

Porcine reproductive and respiratory syndrome virus: origin hypothesis.

Porcine reproductive and respiratory syndrome is a serious swine disease that appeared suddenly in the midwestern United States and central Europe approximately 14 years ago; the disease has now spread worldwide. In North America and Europe, the syndrome is caused by two genotypes of porcine reproductive and respiratory syndrome virus (PRRSV), an arterivirus whose genomes diverge by approximately 40%. My hypothesis, which explains the origin and evolution of the two distinct PRRSV genotypes, is that a mutant of a closely related arterivirus of mice (lactate dehydrogenase-elevating virus) infected wild boars in central Europe. These wild boars functioned as intermediate hosts and spread the virus to North Carolina in imported, infected European wild boars in 1912; the virus then evolved independently on the two continents in the prevalent wild hog populations for approximately 70 years until independently entering the domestic pig population.

Glucocorticoid regulation of lactate dehydrogenase-elevating virus replication in macrophages.

Lactate dehydrogenase-elevating virus (LDV) is a macrophage-tropic arterivirus which generally causes a persistent viremic infection in mice. LDV replication in vivo seems to be primarily regulated by the extent and dynamics of a virus-permissive macrophage population. Previous studies have shown that glucocorticoid treatment of chronically LDV-infected mice transiently increases viremia 10-100-fold, apparently by increasing the productive infection of macrophages. We have further investigated this phenomenon by comparing the effect of dexamethasone on the in vivo and in vitro replication of two LDV quasispecies that differ in sensitivity to immune control by the host. The single neutralizing epitope of LDV-P is flanked by two N-glycans that impair its immunogenicity and render LDV-P resistant to antibody neutralization. In contrast, replication of the neuropathogenic mutant LDV-C is suppressed by antibody neutralization because its epitope lacks the two protective N-glycans. Dexamethasone treatment of mice 16 h prior to LDV-P infection, or of chronically LDV-P infected mice, stimulated viremia 10-100-fold, which correlated with an increase of LDV permissive macrophages in the peritoneum and increased LDV infected cells in the spleen, respectively. The increase in viremia occurred in the absence of changes in total anti-LDV and neutralizing antibodies. The results indicate that increased viremia was due to increased availability of LDV permissive macrophages, and that during a chronic LDV-P infection virus replication is strictly limited by the rate of regeneration of permissive macrophages. In contrast, dexamethasone treatment had no significant effect on the level of viremia in chronically LDV-C infected mice, consistent with the view that LDV-C replication is primarily restricted by antibody neutralization and not by a lack of permissive macrophages. beta-Glucan, the receptor of which is induced on macrophages by dexamethasone treatment, had no effect on the LDV permissiveness of macrophages.

Hydrophobic IgG-containing immune complexes in the plasma of autoimmune MRL/lpr mice, lactate dehydrogenase-elevating virus-infected mice, and pigs: association with transforming growth factor-beta and pH-dependent amplification.

Persistent infection of mice with lactate dehydrogenase-elevating virus (LDV) is associated with polyclonal B cell activation, autoimmunity, and circulating hydrophobic IgG-containing immune complexes (ICs), which bind to the surfaces of uncoated ELISA plates in the presence of 0.05% Tween 20. We demonstrate here that hydrophobic IgG-containing ICs also appear naturally in the plasma of autoimmune MRL/lpr mice. These and the similar hydrophobic ICs of LDV-infected mice as well as pigs coincide on ELISA plate surfaces with TGF-beta, apparently in the form of an IgG-TGF-beta complex. Circulating hydrophobic IgG-containing ICs are also susceptible to considerable amplification in vitro by exposure to alkaline conditions. By this latter method, the fraction of in vivo hydrophobic IgG, relative to the maximum in vitro chemically inducible IgG, was found to be about 20% in the plasma of LDV-infected mice, 5% in normal mouse plasma, and less than about 2% in pig plasma. These results indicate the potential for both chemically induced and protein-binding contributions to the generation of hydrophobic IgG-containing molecules, and have implications for immunopathological mechanisms in autoimmunity and persistent virus infections.

N-glycans on the short ectodomain of the primary envelope glycoprotein play a major role in the polyclonal activation of B cells by lactate dehydrogenase-elevating virus.

The common biologically cloned isolates of lactate dehydrogenase-elevating virus (LDV-P and LDV-vx) invariably cause a polyclonal activation of B cells in immunocompetent mice. It is recognized by an at least 10-fold increase in plasma IgG2a levels and the de novo formation of immune complexes that most likely consist of autoantibodies and their antigens. The present study indicates that three closely spaced N-glycans on the short ectodomain of the primary envelope glycoprotein, VP-3P, of LDV-P/vx, play a major role in inducing the polyclonal proliferation of B cells. IFN-gamma then seems to mediate the differentiation of the activated B cells to IgG2a-producing plasma cells. These conclusions are based on the finding that the IgG2a hypergammaglobulinaemia and immune complex formation were much lower in mice that were infected with LDV variants (LDV-C and LDV-v) whose VP-3P ectodomains lack two of the three N-glycans than in LDV-P/vx infected mice. In contrast, the VP-3P ectodomains of three neutralization escape variants of LDV-C/v whose VP-3P ectodomains possess three N-glycosylation sites caused a polyclonal activation of B cells comparable to that of LDV-P/vx.