Phylogeographic analysis of foot‐and‐mouth disease virus serotype O dispersal and associated drivers in East Africa

Abstract

The continued endemicity of foot and mouth disease virus (FMDV) in East Africa has significant implications for livestock production and poverty reduction, yet its complex epidemiology in endemic settings remains poorly understood. Identifying FMDV dispersal routes and drivers of transmission is key to improved control strategies. Environmental heterogeneity and anthropogenic drivers (e.g., demand for animal products) can impact viral spread by influencing host movements. Here, we utilized FMDV serotype O VP1 genetic sequences and corresponding spatiotemporal data in order to (i) infer the recent dispersal history, and (II) investigate the impact of external factors (cattle density, human population density, proximity to livestock markets, and drought) on dispersal velocity, location, and direction of FMDV serotype O in East Africa. We identified statistical evidence of long-distance transmission events, and we found that FMDV serotype O tends to remain circulating in areas of high cattle density, high human population density, and in close proximity to livestock markets. The latter two findings highlight the influence of anthropogenic factors on FMDV serotype O spread in this region. These findings contribute to the understanding of FMDV epidemiology in East Africa and can help guide improved control measures.

Ecological and Anthropogenic Spatial Gradients Shape Patterns of Dispersal of Foot-and-Mouth Disease Virus in Uganda

“…In addition, pastoral cattle herds, herds in areas of low rainfall, and herds in areas of high cattle density had relatively higher numbers of seropositive animals. The phylogeographic modeling of sequences obtained from the same cross-sectional study, combined with previous sequence data from Uganda, Kenya, Tanzania, and Ethiopia, re-affirmed the transboundary nature of the spatiotemporal spread of FMDV and identified a tendency for FMDV to remain circulating near livestock markets, in areas of high cattle density, and in areas of high human population density [ 10 ]. However, using available methods, no directional trends in FMDV spread were elucidated.…”

“…In a previous study of FMDV phylogenetics using seraphim , we identified several characteristics of areas where FMDV is likely to remain circulating (near livestock markets, areas of high cattle density, and high human population density), but we were unable to elucidate directional patterns in FMDV spread [ 10 ]. Using the RGF model, we identified directional trends related to rainfall, cattle density, and proximity to livestock markets.…”

“…Using the RGF model, areas less than 14.8 km from livestock markets were 1.88 times more likely to be classified as a branch-start location (95% CI 1.05–3.39), indicating that these areas may act as a source of FMDV dissemination. Utilizing seraphim , we found that FMDV serotype O lineages in East Africa have a tendency to remain circulating near livestock markets [ 10 ]. Applying phylogeographic methods to FMDV serotype O in Ecuador, Carvalho et al found that circulating strains had originated in areas of increased animal commerce [ 49 ].…”

A scoping review of foot‐and‐mouth disease risk, based on spatial and spatio‐temporal analysis of outbreaks in endemic settings

“…FMD outbreaks were linked to areas in which livestock markets operate or that are close to human consumption centres as evidenced by a link to the location of abattoirs and high human density locations. A growing body of evidence of viral circulation across cattle markets, highly populated areas and slaughterhouses backs up this epidemiological linkage (Buckle et al., 2021 ; Colenutt et al., 2022 ; Di Nardo et al., 2021 ; Gunasekara et al., 2021 ; Munsey et al., 2021 ). Cross‐border trade was also represented and was often connected to a higher outbreak risk.…”

“…Outbreak risk attributed to presumed contact with wildlife was confirmed in some models, although further investigations are required. Di Nardo et al, 2021;Gunasekara et al, 2021;Munsey et al, 2021).…”