Genetic characterization of food-and-mouth disease virus WFL strain

1 College of Animal Science and Veterinary Medicine, Jilin University, 5333 Xi’an Road, Changchun, China. 2 Department of Animal Food, The 11th Institute, Academy of Military Medical Sciences, 1068 Qinglong Road, Changchun, China. 3 Department of Pharmacology, School of Pharmacy, Second military medical university, No.800, Xiangyin Road, Shanghai, China. 4 Heping Campus, Jilin University, 5333 Xi’an Road, Changchun, 130062, China.


INTRODUCTION
Foot and mouth disease (FMD) is a highly contagious and economically devastating disease of cloven-hoofed livestock, characterized by the appearance of vesicles on the feet and mouth (Marvin and Barry, 2004;Salguero et al., 2005;Sobrino et al., 2001).
The main goal of the present study was to obtain the entire genome sequence of food-and-mouth disease virus WFL strain, including the 3′-and 5′-terminal noncoding regions of the genome.

Viral isolates, RT-PCR, and sequencing
Foot-and-mouth disease virus WFL strain was isolated from swine host in 1999 in Yunnan province, China, and adapted to BHK-21 cells. Total RNA was extracted using the RNeasy Mini kits (QIAGEN) according to the manufacturer's instructions. Subsequently, RNA was reverse-transcribed into cDNA using the primer IRT and SuperScript® III Reverse Transcriptase (Invitrogen). The first-strand cDNA was then subjected to PCR amplification using primer pairs, A-B, C-F, and G-H (Table 1), to amplify 3 separate overlapping PCR products containing the complete genome of FMDV using LA Taq polymerase (Takara Biotechnology (Dalian) CO., LTD). The PCR products were purified and sequenced (Sangon Biological Engineering Technology and Service, Shanghai, China). The primers were designed based on the complete reference sequence obtained from GenBank.

Sequence analysis
The RNA structure was depicted according to the RNA-fold prediction program (Gruber et al., 2008). The reference sequences included in the analysis were obtained from GenBank (Table 2). Phylogenetic and molecular evolutionary analyses were conducted using MEGA version 4.1 (Tamura et al., 2007).

Full-length genomic sequence of WFL
Here, we obtained the full-length genome of the WFL strain by RT-PCR. Using a total of 7 primers (Table 1), the complete genome sequence of the WFL strain was amplified as 3 separate overlapping PCR products. The result showed that the complete genome sequence of the WFL strain was 8155 nucleotides (nt) in length, including poly(C). The full-length sequence was submitted to GenBank (GenBank ID: EF175732).
The genomic organization of WFL was shown in Table  3. The complete sequence was divided into sixteen  fragments except poly (A). 5' UTR (non-translated region) and 3'UTR were located in 1-1059 and 8029-8155, respectively (Table 3).

Characteristics of UTR
5'UTR played an important role in replication and selective translation of the viral RNA. The FMDV 5'UTR contains a short fragment called S-fragment, a poly (C) tract of variable length, followed by a large fragment (LF) of over 700 bases in length (LF-5' UTR) that can form a number of highly conserved secondary structures that include randomly repeated pseudoknots (PKs), a cisacting replication element (cre) and an internal ribosome entry site (IRES) (Mohapatra et al., 2009;He et al., 2011). RNA helicase A (RHA) and 3C pro specifically bind the FMDV S fragment. RHA interacts with the S fragment of the FMDV 5′ NTR (Lawrence and Rieder, 2009). The result showed (Figure 1) that the IRES element of the WFL strain was about 458 nt in length and had five domains, which participated in the viral protein translation in a cap-independent manner. There were two PKs followed by an inverted repeats CCCGTTT/AAACGGG and cre (Figure 2).
The cre region was essential for RNA genome replication (Marvin and Barry, 2004). And a conserved 'AAACA' motif in the cre/bus region has been recently shown to be involved in VPg uridylylation (López et al., 2001;He et al., 2011). In this study, cre was 55nt with 45.5% of G/C, and had a stem-loop (Figure 2).
IRES including domain 2 to domain 5, and four direct repeat motifs, GGTGACA, were located in IRES region. It was reported that the conserved motifs and the structural domains in the IRES interact with an array of cellular factors involved in host translation initiation (Ramos et al., 1999;Pacheco et al., 2010) and some motifs were also crucial in maintenance of the tertiary structure of the IRES through RNA-RNA interaction (Fernández et al., 2006). Domain 4 followed by domain 5 in the IRES displayed highest degree of conservation (Mohapatra et al., 2009). The GNRA tetraloop was a thermostable tetraloop which can exist within a RNA structure solely on its own, or take place in an interaction with a receptor. The 'GNRA' tetraloop in domain 3, which plays critical role in determining the tertiary structural conformation of the IRES element (Mohapatra et al., 2009), was found to be 'GTGA' in the WFL strain. The cleavage site for RNase P within the 'GNRA' stem-loop was 'T 377 CC' motif. In this study the conserved 'motif A', which interacts with 'GNRA' motif to maintain structural organization of the central domain of IRES (Nayak et al., 2006), was found to be 'G 448 CACG' (Figure 1). The eIF4C binding domain was GACTAA, and the eIF4B interation domain was ACCGGAGG. The 3' UTR, composed of two stem-loops and a poly(A) tract, was required for viral infectivity and stimulates IRES activity (Serrano et al., 2006). The 3' end established two distinct strand-specific, long-range RNA-RNA interactions, one with the S region and another with the IRES element (Serrano et al., 2006). The S region was recognized by each of the separate stem-loops. S-3'UTR interaction was dependent on a structural conformation induced by the presence of the poly(A) tract (Serrano et al., 2006). Here, it was found that 3' UTR of the WFL strain was 127nt, including 93nt stem-loops and poly (A). The 93nt stem-loops region can fold into two stem-loops, SL1 and SL2 (Figure 3).   (Feng et al., 2004;Ma et al., 2006). And the strain O/ES/2001 was a recombinant of serotype O and Asia 1 (Wu et al., 2009). It was obvious that WFL strain had a close relationship to LZ strain (GenBank ID: DQ248888), indicating that the WFL strain belonged to serotype O. The close link between WFL and these three isolates was consistent with our previous finding using the VP1 sequence (data not shown). Complete amino acids of the WFL strain and the LZ strain were compared. The results showed that there were 16 different deduced amino acid residues between WFL and LZ (Table 4). Detailed comparison of WFL with other strains is still doing.