4 | DISCUSSION
During the study period (2009-2018), clinical cases in cattle were observed year-round over a wide geographical distribution with relatively more cases to be reported in the winter months (December-February). So far, four FMDV serotypes have been reported in Sudan: O, A, SAT 1 and SAT 2 (Abu Elzein, 1983), although no cases due to the SAT 1 serotype were detected in this study or have been recorded in Sudan since 1976 (Abu Elzein and Crowther, 1979). The low incidence of SAT1 in Sudan is supported by serological data for SAT1-specific antibodies in comparison to serotypes O and A (Raouf et al., 2009; Habiela et al., 2010a). Serotype O was the most frequently detected; findings which are also supported by recent serological data (Raouf et al., 2016) showing higher prevalence of serotype O specific antibodies (60.2%) compared to serotype A (30.0%) and serotype SAT 2 (12.3%). Nucleotide distances of more than 15% (for serotypes O and A) or 20% (for serotypes SAT 1 and SAT 2) are used to classify isolates into different topotypes (Samuel and Knowles, 2001; Knowles and Samuel, 2003), while nucleotide differences of between 5% and 15% indicate distinct virus lineages (Bronsvoort et al., 2004). The Sudanese sequences reported here differed from the prototype strains of the relevant topotypes (Knowles et al., 2016) by 7.7-12.1% (O/SUD/2/86), 13.8-16% (O/ETH/3/2004), 15.2-17.8% (O/ETH/2/2006), 12.5-15.5% (O/ETH/1/2007), 13.4-16% (A/SUD/3/77), 7.4-9.9% (SAT2/SAU/6/2000) and 11.9-14.4% (SAT2/CAR/8/2005). These differences indicated the continued divergence of virus lineages within each topotype.
Serological findings from 2013 (Raouf et al., 2016) demonstrate that serotype O is present along the Nile basin up to Khartoum and in Western, Eastern and Northern Sudan. A previous study (Habiela et al., 2010b) highlighted that within-country circulation is an important mechanism by which serotype O virus is maintained in Sudan. New FMDV sequence data generated here continues to demonstrate that serotype O is maintained in the country over successive years without any requirement for the introduction of viruses via transboundary pathways. This point is supported by the presence of at least two viral lineages within the phylogenetic tree that only contain sequences from Sudan (2004-2011 and 2008-2010). The three lineages containing Sudanese and Egyptian sequences are indicative of the northerly spread of FMDV O/EA-3 from Sudan, which based on the dates occurred as separate independent events during 2011-2012, 2013-2016 and 2016-2017. One of these clades also contains a sequence from Eritrea, highlighting other potential transboundary transmission pathways to other countries in the neighbourhood.
Data for serotype A sequence also support the concept that FMD virus lineages are maintained solely within Sudan. The phylogenetic tree contains two lineages that comprise sequences only from Sudan collected from different years (2013-2014 and 2018). In common with the serotype O data, there was also evidence for the northerly spread of A/AFRICA/G-IV into Egypt (during 2011-2012). Epidemiological links to FMD cases in Ethiopia are highlighted in a second clade that contained Egyptian sequences (from 2015-2017) and more recent sequences collected from Sudan (from 2018), where the Ethiopian sequence (A/ETH/19/2015) pre-dated those for the Sudanese samples. More broadly, the serotype A (and serotype O) phylogenetic trees highlight the relationship between recent sequences for FMD viruses collected in Ethiopia (Gizaw et al., 2020) and contemporary data from Sudan. For both serotypes, there are examples where Sudanese sequences have a more basal ancestral location in the tree; for example, the relationships between O/Sudan/2012 and O/Ethiopia/2018-19 and A/Sudan/2013-14 and A/Ethiopia/2018-19, respectively. These results indicate that these FMD viruses have spread from Sudan into Ethiopia; however, this interpretation of the data should be treated cautiously in view of the likely high number of unsampled FMD cases in both countries where the sampling density is very low and susceptible livestock populations are high.
Serotype SAT 2 was the last FMDV serotype to be detected in Sudan in 1976 (Abu Elzein and Crowther, 1979) and 18 Sudanese isolates of serotype SAT 2 were detected during four of the years during this study. All of these sequences from 2010-2017 were characterised as belonging to the SAT 2/VII/Alx-12 lineage that has spread widely in the region to cause FMD outbreaks in countries such as Egypt, Palestine and Israel. In common with serotypes O and A, sequences for viruses from this lineage collected from Sudan are earlier compared to those for viruses sampled in Ethiopia. The SAT 2/VII/Alx-12 lineage is distinct to earlier viral clades detected in Sudan such as those that have spread previously from East Africa into West Africa (Habiela et al., 2010b; Ularamu et al., 2017: highlighted by presence of SAT2/SUD/1/2007 in the phylogenetic tree).
No previous reports have described clinical FMD in Sudan in species other than cattle (Abu Elzein, 1983; Raouf et al., 2010; Habiela et al., 2010a). Recent serological studies suggested a relatively low seroprevalence of non-structural protein antibodies among small ruminants at around 14%, except in Khartoum (Habiela et al., 2009) and Blue Nile State (Raouf, 2015; Raouf et al., 2017). This may indicate a varied role of sheep in the epidemiology of FMD. FMD infection in wild ruminants in Sudan has not been investigated.
In summary, FMD infection in Sudan remains regionally significant and this study highlights the epidemiological connections between FMDV sequences collected in Sudan and neighbouring countries such as Ethiopia and Egypt. Due to incomplete and convenience-biased sampling, it should be recognised that there are limitations in our understanding of the transboundary connectivity in the region and the sequences reported here provide only a crude snapshot survey of the underlying FMD transmission events. Recognition of border areas as particularly risky hotspots for the introduction of FMD viruses is of high importance in developing a risk-based control strategy especially when resources are limited. In this context, continued studies are warranted to improve our understanding of FMD epidemiology in Sudan and risk-pathways in the region.