The investigation into the specificity of the antibody response upon PRRSV infection, with a focus on identifying both NEs and non-NEs, holds significant importance not only for the improving of knowledge on PRRSV immunology but also for it practical applications. Firstly, the identification of NE relevant for protection would facilitate the development of rapid diagnostic tests with high sensitivity and specificity, enabling the prompt and automated assessment of NA responses. Additionally, the identification of relevant epitopes would encourage the development of next-generation vaccines with improved formulations that incorporate these epitopes to enhance the efficacy of existing PRRSV vaccines. Consequently, numerous studies have been carried out with the objective of pinpointing biologically relevant epitopes [14,15,16, 18, 21, 22, 32].
The outcomes of these investigations suggest that NEs, which could be potentially crucial for robust protection against reinfections, are located in GP2, GP3, GP4, GP5, and M proteins [14]. Furthermore, in vivo studies utilizing chimeric viruses containing mosaic envelope protein sequences from various isolates identified the collaborative role of these proteins in fostering broadly NAs and in improving cross-protection [33].
Based on the existing knowledge, our study has focused on the analysis of the envelope proteins GP2, GP3, GP4, GP5 and M, due to their significance in the neutralizing antibody response, with the objective of identifying NEs, and more specifically linear NEs, that could be conserved and relevant for the induction of broadly reactive NA. To accomplish this aim we employed a panel of PRRSV isolates characterized on the basis of their susceptibility or resistance to neutralization and a collection of hyperimmune monospecific sera encompassing both, broadly reactive sera and sera with poor cross-reactivity against heterologous viruses in SN assays. Specifically, the use of sera with known cross-neutralization capacity is pivotal for distinguishing conserved NEs responsible for neutralizing heterologous isolates from those involved solely in neutralizing the immunization (i.e. homologous) virus.
The immunodominant peptides of various PRRSV envelope proteins have been successfully identified using the Pepscan technique [14]. Different systems have been proposed in the literature for the analysis of Pepscan signals. In this study, we have used the system established for PRRSV by Vanhee et al. [14] in which the signal of a particular peptide is related to that obtained for the set of peptides of the protein studied. The main advantage of this system is that it ensures that the signals identified are very specific. However, and with the objective of further confirming the specificity of the reactivity identified, the signals were evaluated using two other systems widely used for the analysis of the reactivity of different sera against specific peptides in Pepscan systems. The first one consists of the establishment of a cut-off point that corresponds to the average value of the optical density obtained with the negative sera plus two times the standard deviation [34]. The second one consists of qualifying as specific the signals whose value is at least three times higher than the average value of the negative controls [35]. The results obtained in our study using these systems are very similar to those obtained using the method proposed by Vanhee et al. [14] (data not shown). The consistency of the results across different methods suggests the reliability of Pepscan for consistent and reliable outcomes and supports the immunogenic potential of the peptides identified.
The results of our study reveal that the linear peptides of the ectodomains of the PRRSV-1 main envelope proteins exhibit a relatively low immunogenicity, as evidenced by the relatively low number of reactive peptides and, more importantly, by the low number of sera that typically recognize the reactive peptides. Thus, most synthetic peptides were recognized by only one or two sera, regardless the protein or the protein region considered and none of them was recognized by all sera. This observed lack of broad reactivity has been previously described in the literature [14, 36] and challenges the antigenic relevance of these areas. The reasons for this low recognition rate have not been identified but they can be diverse. On one hand, it is possible that these regions are poorly exposed to the immune system as many of the studied proteins are minor virion proteins and their relative expression in the infected cell is low, compared to other immunodominant proteins, such as the N protein or nsp-7, which induce a high antibody response in infected individuals [37,38,39]. These differences in immune exposure might motivate the development of a weaker immune response to these antigenic determinants. However, although it might be the case for proteins poorly recognized by convalescent pigs, such as the minor envelope glycoproteins GP2, GP3 and GP4 [40], it is unlikely to be the case for the GP5 or M proteins, which are generally well recognized by infected pigs [41]. An alternative explanation is that the antigenic variability of PRRSV makes the recognition of heterologous strains quite difficult, to the point that hyperimmune monospecific sera might not be able to recognize some antigenic determinant present in heterologous strains. In this line of thinking, the lack of cross-reactivity in serological studies carried out using polyclonal and monoclonal antibodies has been repeatedly demonstrated in the literature and constitutes a hallmark of PRRSV [23, 42,43,44,45]. Finally, it should be kept in mind that the capacity for specific epitope recognition is genetically determined by the individual’s B-cell repertoire [46]. The results of our study indicate that the reactive peptides are not generally recognized, not even by all homologous sera. Thus, it is likely that the poor recognition of the reactive linear peptides identified is the result of the combination of the PRRSV variability, that might prevent the recognition of antigenic determinants of heterologous viruses, and the individual variability in B cell repertoires, that may lead to non-recognition by even most individuals within a population.
Despite the overall low immunogenicity of the peptides studied, certain regions with greater immunogenicity have been identified in some of the protein studied. Among them, an AR identified in the ectodomain of GP3 stands out for the high number of sera of different specificities that recognize it in all three viruses analyzed. These results are in agreement with the results of previous studies in which the same region has been widely recognized by convalescent sera in both PRRSV-1 and PRRSV-2 [14, 34, 36]. Remarkably, a NE has been described within this AR for PRRSV-1 [14]. As the amino acidic sequence of this region is fairly well conserved between PRRSV isolates and recognition by sera of different specificity has been systematically described, it could be speculated that this epitope is relevant for cross-neutralization. Nevertheless, the data obtained in this study do not suggest a crucial involvement of this epitope in the effective neutralization of heterologous strains, as this region is recognized by broadly reactive sera and by sera with low cross-reactivity in SN assays. Finally, it is noteworthy to mention that another AR, previously described within PRRSV-1 and PRRSV-2 GP3 [14, 34, 36], has been identified in our study. However, the main epitope in this region has been classified as non-neutralizing by others [12, 14] and despite it recognition by a significant number of sera of different specificities, its significance in protection seems to be inconsequential.
The results of our study confirm that the ectodomain of GP5 also harbors a widely accepted and fairly conserved AR, that is recognized by both homologous and heterologous sera with varying degrees of cross-reactivity. This particular region has been deemed crucial for protection, given the identification of one of the first NEs in this region [19]. Later, the role of this epitope as NE has been questioned for PRRSV-1 isolates [14] and even for PRRSV-2 [47]. However, recent studies utilizing machine learning techniques has revealed that specific modifications in the GP5 NE sequence could significantly alter the antigenic distance, measured by SN assays, between PRRSV-1 isolates, suggesting a potential role for this epitope as a NE, despite previous uncertainties [27].The results of our study indicate that, shall the region contain a NE, it is not preeminently involved in broad-spectrum neutralization, as the proportion of broadly-reactive and poorly-reactive sera that react with these peptides is similar.
In the case of GP4 ectodomain, a highly immunogenic NE has been described in PRRSV-1 isolates [32]. Consistently, in our study, all homologous sera recognized the corresponding peptides in EU-14 and EU-21 isolates, irrespectively of their cross-reactivity. On the contrary, no recognition by heterologous sera was recorded. This finding was not unexpected as the most outstanding characteristic of this epitope is its high variability, which may contribute to the nearly unique sequence observed in each virus isolate studied [23], potentially explaining the limited cross-reactivity exhibited by the viruses in most cases. This variability might be an escape mechanism of PRRSV as it allows the virus to undergo evolution during in vivo infection as animals develop specific NA [48]. This adaptive process leads to the creation of mutants resistant to neutralization [32] that have the potential to continue circulating in immune populations.
An unexpected observation was the lack of recognition of the GP4 NE of the EU-24 isolate, not even by any of the homologous sera used in the study. Although this epitope seems to be present in all PRRSV-1 virus isolates studied so far, it has not been documented in PRRSV-2 isolates [34]. It is plausible that certain PRRSV-1 isolates, particularly those more diverse and distant from the LV, the PRRSV-1 prototype strain, such as the Italian strain used in this study, may lack this epitope. This absence could lead to the adaptative immune response of infected animals targeting alternative epitopes. In fact, a peptide located upstream of the typical NE of GP4, and previously undescribed for PRRSV-1, has been identified in EU-24 by a variety of sera, both homologous and heterologous. The biological relevance of this peptide and its existence in alternative viral isolates deserves further investigation.
Finally, even though a NE and other immunodominant regions have been previously described in GP2 [14, 34, 36], linear peptides from this protein were poorly recognized by the hyperimmune sera used in this study. These results are consistent with those obtained by others [14] and confirm the very poor immunogenicity of this protein, despite its biological relevance [11].
On the other hand, it should be mentioned that some sera were able to recognize several peptides from proteins of different virus isolates. Specifically, one of the sera against the EU-9 isolate showed a very broad reactivity, recognizing a high number of peptides. This phenomenon, which has been previously described [24], could be due to non-specific reactions, so additional studies would have to be carried out to verify the specificity of the reaction.
The Pepscan results were used to make a selection of peptides, targeting the most immunogenic ARs in each envelope protein and virus. In addition, peptides previously identified as NE were also selected, regardless of their recognition rate. All these peptides were used to further study the corresponding AR and determine their role in virus neutralization. To do so, two different approaches were followed. In the first place, a competition assay was carried out between the selected linear peptides and the virus for binding to cellular receptors. If the peptides bind to cellular receptors, mimicking virus binding, they will compete with the virus and a reduction in virus infectivity will be observed [31]. However, the results obtained in our study indicate that the addition of either individual peptides or combinations of peptides to the cell culture before infection with two of the virus isolates used was not sufficient to block infection. These results are in agreement with those obtained by Robinson et al. [31] using a PRRSV-2 isolate and different peptide concentrations. However, and surprisingly, some peptides or combinations of peptides from the third isolate used in our study (i.e. EU-24) were able to partially or even completely block infectivity. Specifically, the addition of a GP3 peptide led to a very marked inhibition of viral infectivity and incubation with one of the GP4 peptides or with a mix of GP2 to GP5 peptides completely inhibited virus replication, indicating that these peptides might play a significant role in virus-cell interaction. Although the reasons for these discrepancies have not been elucidated, our results seem to indicate that different PRRSV isolates might differ in their interactions with cell receptors. In this line of thinking, some differences have been observed between PRRSV-1 and PRRSV-2 isolates in their interaction with CD163 [49] and also in the range of susceptible cell subpopulations depending on PRRSV pathogenicity [50].
Finally, the selected peptides were incubated with the hyperimmune sera with the objective of blocking the specific antibodies against those peptides, following previously established protocols [14, 31]. If the peptides were the target of the NAs, they would compete with the virus in binding to them, thus inhibiting their neutralizing activity and increasing the infectivity of the virus in SN assays [31]. However, the results obtained indicate that the peptides used do not have any effect on the neutralizing capacity of the selected hyperimmune sera. These results are in line with those obtained by Robinson et al. [31] using a PRRSV-2 isolate, but contrast with the observations made by Vanhee et al. [14]. These authors demonstrated that antibodies directed against some of the peptides in the ARs identified in GP2, GP3 and GP4 have neutralizing activity. Although the results of both studies seem to be contradictory, they can be explained by the different methodological approaches followed. Thus, Vanhee et al. [14] purified the antibodies specific to the assayed peptides by affinity column and revealed that these purified antibodies had neutralizing activity in SN assays. This fine experimental design allows demonstrating that those epitopes can elicit NAs but does not exclude the presence or the role of NAs of different specificities in PRRSV immune sera. In our study, and in the study carried out by Robinson et al. [31], polyclonal hyperimmune sera were incubated with selected peptides. In this experimental approach, it is possible that the concentration of NAs specific for the assayed peptides was lower than in the study carried out by Vanhee et al. [14], preventing their detection. Besides, in polyclonal sera the presence of other antibodies, of different specificities, might have produced steric hindrance or another type of impediment in the binding to the NE studied abrogating their biological effect. However, another possible, and more likely, explanation is that, although NAs specific for the peptides tested have been blocked, other PRRSV-specific NAs present in the sera, directed against other NEs, either linear or, more likely, conformational, might have acted blocking the virus infectivity and concealing the presence of NAs against the peptides tested in this study. In this line of thinking, it should be kept in mind that the minor envelope proteins are folded and interact during virus assembly [11] and thus, the conformational nature of the peptides in the native protein might be different and non-contiguous or conformation-dependent epitopes might span different regions or even proteins. Finally, the cell type used in the SN assays might have influenced the results. Thus, in our study, and in the study carried out by Robinson et al. [31], SN assays were carried out in MARC-145, a cell line highly permissive to PRRSV infection [51] that contains the monkey CD163, instead of swine, and does not contain Siglec-1 but other Siglecs [52] while Vanhee et al. [14] used porcine alveolar macrophages. The use of MARC-145 cells for PRRSV functional analysis may be controversial. Firstly, PRRSV often requires adaptation, both at non-structural and structural proteins, especially at GP2-GP3, to replicate in MARC-145 cells [53, 54]. This adaptation may alter the virus-host cell interaction and may not accurately represent what occurs in vivo infections. Additionally, the interaction between PRRSV-1 and MARC-145 cells is not well understood; although initial binding is facilitated by heparan sulfate, the mechanisms of internalization remain unclear. One possibility is that PRRSV-1 may enter MARC-145 cells through spontaneous internalization, a process described in other viruses like rotaviruses and HIV [55, 56], which have limited host cell specificity but can still grow in stable cell lines. This spontaneous internalization might bypass critical receptor-mediated steps that are essential in natural PRRSV infections of macrophages. Finally, using MARC-145 grown viruses for functional analysis may introduce biases. Antibodies developed against viruses grown in MARC-145 or other non-macrophage cells may not effectively bind viral structures as they appear in the natural host, further compromising the relevance of such assays. For these reasons, while MARC-145 cells offer a convenient experimental platform, they present certain limitations for PRRSV functional analysis. Future studies could be carrying in porcine macrophages to confirm the validity of the results of this study.
Add Comment