In silico prediction and ex vivo evaluation of potential T-cell epitopes in glycoproteins 4 and 5 and nucleocapsid protein of genotype-I (European) of porcine reproductive and respiratory syndrome virus

Abstract

T-cell epitopes of porcine reproductive and respiratory syndrome virus (PRRSV) glycoproteins 4 (GP4), 5 (GP5) and nucleocapsid (N) were predicted using bioinformatics and later tested by IFN-γ ELISPOT in pigs immunized with either a modified live vaccine (MLV) or DNA (open reading frames 4, 5 or 7). For MLV-vaccinated pigs, immunodominant epitopes were found in N but T-epitopes were also found in GP4 and GP5. For DNA-immunized pigs, some peptides were differently recognized. Using a large set of PRRSV sequences it was shown that N contains a conserved epitope and that for GP5, the genotype-I counterparts of previously reported epitopes of genotype-II strains were also immunogenic.

Introduction

Porcine reproductive and respiratory syndrome virus (PRRSV) was firstly isolated in 1991 by Dutch researchers [1] and thereafter classified as a member of the genus Arteriviridae[2]. At present, two genotypes of PRRSV are known: genotype-I (European) and genotype-II (American). Antigenic and genetic diversity appears to be high within each genotype, particularly within European strains [3], [4], [5], [6], [7], [8].

PRRSV genome is composed of nine open reading frames (ORFs). ORFs 1a and 1b account for approximately 75% of the viral genome and encode the non-structural proteins (nsp). ORFs 2a, 2b and 3–7 encode the PRRSV structural proteins of which resulting proteins 2a, 3, 4 and 5 are glycosylated (namely GP2a, GP3, GP4 and GP5, respectively). ORF6 encodes the viral matrix protein (M) and ORF7 encodes the nucleocapsid (N) [9], [10], [11], [12]. Neutralizing antibodies are thought to be mainly induced by GP5 and GP4 [13]; in contrast, a very limited knowledge is available regarding T-cell epitopes of PRRSV. Recently, two T-cell epitopes have been identified in GP5 of American-type strains based on their ability to induce IFN-γ responses in cultures of peripheral blood mononuclear cells (PBMC) obtained from PRRSV-immunized and later challenged pigs [14].

Determination of T-cell epitopes is a difficult task. The systematic approach based on the synthesis and testing of large sets of overlapping peptides is indicated when location of T-epitopes can, somehow, be estimated; however, when this knowledge lacks, that is a very expensive and cumbersome way. In contrast, bioinformatic prediction is extremely cheap and may help to restrict the number of peptides to be screened. Combination of T-cell epitope prediction and classical immunization experiments can be a useful strategy that permits to speed research in this area [15]. In the present study, bioinformatics were used to predict potential T-cell epitopes in GP4, GP5 and N proteins of PRRSV and were later tested on immunized animals.