We constructed foot-and-mouth disease virus (FMDV) mutants bearing independent deletions of the two stem-loop structures predicted in the 3 noncoding region of viral RNA, SL1 and SL2, respectively. Deletion of SL2 was lethal for viral infectivity in cultured cells, while deletion of SL1 resulted in viruses with slower growth kinetics and downregulated replication associated with impaired negative-strand RNA synthesis. With the aim of exploring the potential of an RNA-based vaccine against foot-and-mouth disease using attenuated viral genomes, full-length chimeric O1K/C-S8 RNAs were first inoculated into pigs. Our results show that FMDV viral transcripts could generate infectious virus and induce disease in swine. In contrast, RNAs carrying the ⌬SL1 mutation on an FMDV O1K genome were innocuous for pigs but elicited a specific immune response including both humoral and cellular responses. A single inoculation with 500 g of RNA was able to induce a neutralizing antibody response. This response could be further boosted by a second RNA injection. The presence of the ⌬SL1 mutation was confirmed in viruses isolated from serum samples of RNA-inoculated pigs or after transfection and five passages in cell culture. These findings suggest that deletion of SL1 might contribute to FMDV attenuation in swine and support the potential of RNA technology for the design of new FMDV vaccines.
Foot-and-mouth disease virus (FMDV) is a member of thePicornaviridae family and the causative agent of an acute vesicular disease considered a major animal health problem worldwide, affecting pigs, ruminants, and other cloven-hoofed livestock (32, 53). The virus consists of a nonenveloped particle enclosing a single-stranded positive-sense RNA molecule of about 8.5 kb in length, with the viral protein VPg covalently linked to the 5Ј end and a poly(A) tract at the 3Ј end. The viral genome contains a single open reading frame flanked by two highly structured noncoding regions (NCRs) at their 5Ј and 3Ј termini, respectively (7). The 5Ј NCR, approximately 1,300 nucleotides in length, includes sequences required for the initiation of replication and translation, comprising the S fragment, a 360-nucleotide-long region predicted to form a large hairpin structure (23, 62), a poly(C) tract, multiple pseudoknots, the cis replication element (cre) (38), and the internal ribosome entry site (IRES) (37). The 3Ј NCR is a sequence of about 100 nucleotides predicted to be structured in two stem-loops, SL1 and SL2 (55). We have previously shown the strict requirement of the 3Ј NCR for FMDV infectivity and replication (52) and the stimulatory effect of this region on IRES-dependent translation, even in the absence of a poly(A) tail (36). Moreover, the 3Ј NCR interacts with the IRES and S regions located at the 5Ј end through direct RNA-RNA interactions (55). Cellular proteins binding to the 3Ј NCR have also been described (36,48). Some of these factors are also able to interact with the S region and undergo proteolytic cleavage upon FMDV infection (48). The pu...