Background

Foot-and-mouth disease (FMD) is a severe and extremely contagious viral disease that causes vesicular disease in cloven-hoofed domestic and wild animals. Outbreaks can lead to serious economic losses due to decreased livestock productivity and restrictions on movement and trade of animals and their products [1]. The causative agent of the disease is foot-and-mouth disease virus (FMDV), an Aphthovirus from the family Picornaviridae, with seven antigenically distinct serotypes including the EuroAsiatic serotypes A, O, C, and Asia 1, and the Southern African territories (SAT) serotypes 1, 2, and 3. Cross-protection is not conferred between serotypes following infection or vaccination and may not be conferred between different subtypes or variants of the same serotype [2]. The key factor to prevent the disease and control its spread is early detection and identification of the serotype(s) and subtype(s) of outbreak strains so that directed vaccination may be initiated. The antigenic determinants of FMDV reside within the structural proteins of the viral capsid (VP1-VP4), encoded by the genes 1A (VP4), 1B (VP2), 1C (VP3), and 1D (VP1). Historically, FMDV molecular epidemiology of outbreaks has focused on analyses of partial or full-length sequences of the 1D gene that encodes the structural protein VP1 [3]. Currently, over 60% (9571/14952 in December 2023) of FMDV records in the GenBank database have a total nucleic acid length of 400–700 bp and cover the partial or full 1D gene. Unfortunately, relying solely on the VP1 coding sequence does not provide all of the information required to fully characterize important phenotypic traits of FMDV strains, since several antigenic determinants are located on other viral capsid proteins, such as VP2 and VP3 [4,5]. Moreover, during shorter epidemic time scales, sequencing of 1D cannot provide enough resolution to discriminate the viruses collected from adjacent premises within the same outbreak clusters due to the viral populations not having diverged substantially [6]. In this regard, numerous reports have been published using the whole P1 region, encoding all structural proteins (i.e., VP1 to VP4) for strain genotyping and phylogenetic analysis [7,8]. The characterization of the whole P1 region (1A to 1D) is important for tracking the emergence or spread of FMD and for selecting vaccines in case of an outbreak. However, restricting analyses to P1 sequences may not detect recombination events that can occur between multiple serotypes or topotypes in the event of co-circulating FMDV within infected animals [5,9,10]. Next-generation sequencing (NGS) techniques offer much promise to perform viral whole genome sequencing (WGS) without prior knowledge about the target sequence and provide a huge amount of data from a limited quantity of starting material in a rapid, cost-effective manner. Valdazo-González and colleagues (2012)[11] have demonstrated that WGS of FMDV is a powerful tool for the reconstruction of transmission trees.

WGS of FMDV has been previously conducted on different platforms including Oxford Nanopore Technology (ONT) and Illumina [6,10,12–14]. These methods use reverse transcriptase to convert RNA to cDNA and/or go through one or two FMDV-specific or non-specific PCR amplifications. Some are labor-intensive with cumbersome procedures; others may have relatively longer turnaround times. Moreover, viral RNA manipulation can introduce biases into the data [15], which may not represent the actual diversity within samples [6]. Here, we introduce for the first time an FMDV direct RNA sequencing method using Nanopore sequencing on Flongle flow cells with MinION Mk1C (or GridION). Compared to the original ONT SQK-RNA002 protocol, we omitted the optional reverse transcription step to make the process even shorter by saving ~35 minutes as the ONT device only sequences the RNA molecule, not the cDNA. Our method is low-cost, fast, and reliable and can be deployed to low-resource settings when using an Mk1C device, such as a remote or field laboratory with alternative RNA extraction methods. This POC method requires less than 50 min of manipulation time from RNA to NGS run (Graphical Abstract), followed by a rapid identification of the whole FMDV genome within minutes.