Epidemiology of African swine fever at the wildlife-livestock interface of the Gorongosa National Park, Mozambique

Abstract

African swine fever (ASF) is a fatal and devastating viral disease that hampers pig production in most sub-Saharan African countries. The infection causes high mortality and morbidity rates in susceptible pigs. The disease is caused by a DNA Arbovirus classified as Asfivirus, the exclusive member of the family Asfarviridae. The natural reservoirs of ASF in Africa are argasid ticks of the genus Ornithodoros, mainly if not entirely those of the Ornithodoros moubata complex, and wild suids, primarily warthogs (Phacochoerus africanus) and bush pigs (Potamochoerus spp) in which infection is unapparent. Three different scenarios for virus circulation were found to exist in southern Africa: a sylvatic cycle among wildlife involving argasid ticks and wild suids, a domestic cycle between soft ticks and domestic pigs and a domestic pig cycle without soft tick involvement. The overall aim of this work was to elucidate the epidemiology of the disease at the wildlife/domestic interface of Gorongosa National Park (GNP), central Mozambique, by determining the prevalence of soft ticks (Ornithodoros spp) found in domestic and sylvatic habitats, contacts between wild and domestic pigs and soft ticks, the presence of ASF virus (ASFV) in different hosts and to demonstrate the genetic differences of ASFV isolates using PCR, nucleotide sequencing and phylogenetic analysis. Addictionaly, the performance of currently available serological diagnostic tests in their ability to detect ASFV antibodies in an endemic area for ASF in Mozambique was determined. PCR-based virus detection and virus isolation methods were used to detect the presence of ASFV in 1662 Ornithodoros porcinus porcinus and 203 of O. p. domesticus specimens recovered in 29 warthog burrows in the GNP and 2 pig pens from GNP Buffer Zone respectively. A total of 72.4 %, 95% CI [52.8; 92.0] of ticks from the warthog burrows and all ticks from pig pens tested positive for ASFV DNA using nested PCR. However, using primary cell culture, haemadsorbing ASFV could only be isolated from 47.4 %, 95% CI [3.4; 60.4] of these burrows and half of the domestic habitats. Twenty nine heamadsorbing viruses isolated from soft ticks were genotyped using a combination of partial p72, p30 and p54 gene sequencing and analysis of tetrameric amino-acid tandem repeats within the B602L gene. The isolates in this study clustered in three different genotypes: II (together with isolates from outbreaks that had occurred in Tete province-Mozambique, Mauritius, Malawi, Madagascar and the recent outbreaks in the Caucasus and Russia), V (consisting of isolates circulating in Mozambique and Malawi) and one newly identified genotype XXIII. The control of ASF in endemic settings relies on reliable diagnostic tools and the application of surveillance and control measures to prevent the spread of disease. Therefore, the sensitivity and specificity of available diagnostic tests under field conditions should be estimated during the design of an efficient surveillance program. Serum samples of 273 local breed domestic pigs from small scale pig farmers of Gorongosa District were tested simultaneously with an immunoblotting test (considered the gold standard), the OIE recommended Indirect ELISA (IELISA), a p30 recombinant protein detection system (rp30 ELISA) and a commercially available ELISA based on the p72 protein detection (Blocking p72 ELISA). The I-ELISA had sensitivity and specificity values of 88.2%, 95% CI [77.4; 99.1] and 97.5%, 95% CI [95.5; 99.5] respectively. The sensitivity and specificity for the rp30 ELISA was 26.5%, 95% CI [11.6; 41.3] and 96.7%, 95% CI [94.4; 98.9] and for the Blocking p72 ELISA these values were estimated at 50%, 95% CI [33.2; 66.8] and 98.3%, 95% CI [96.7; 100] respectively. An almost perfect agreement was observed between the I-ELISA and immuno-blotting assays (kappa value of 0.84, 95% CI [0.72; 0.93]), while the comparison of these tests with the Blocking p72 ELISA showed good agreement with kappa values of 0.51, 95% CI [0.40; 0.67] and 0.58, 95% CI [0.46; 0.73] respectively. Moderate agreement was observed when the rp30 ELISA was compared to the IELISA, immuno-blotting and Blocking p72 ELISA (kappa values of 0.28, 95% CI [0.17; 0.42] 0.29, 95% CI [0.18; 0.44] and 0.38, 95% CI [0.26; 0.55]) respectively. The highest sensitivities were obtained with the parallel combination of the I-ELISA with the Blocking p72 ELISA (sensitivity of 94.1%, 95% CI [91.3; 96.9]) and the I-ELISA combined with the rp30 ELISA (sensitivity of 91.3%, 95% CI [87.9; 94.6]). The combination of the Blocking p72 ELISA and the rp30 ELISA gave a modest sensitivity estimate of 60.8%, 95% CI [55.0; 66.6]. To assess the sero-prevalence of ASF in pig populations within the study area, 634 domestic pigs from 314 small scale farms scattered in the different villages of Gorongosa District and 12 warthogs from GNP were bled and tested by means of I-ELISA for the presence of antibodies against ASFV. The seroprevalence to ASFV in domestic pigs was approximately 9%, 95% CI [7.0; 11.7], in contrast with 75%, 95% CI [42.8; 94.5] in warthogs inside the GNP. Approximately 33%, 95% CI [29.2; 36.9] of pigs tested sero-positive to salivary antigens of Ornithodoros spp. ticks, while in warthogs the sero-prevalence was just above 77%, 95% CI [40.0; 97.2]. This work confirms the existence of an ASFV sylvatic cycle at the wildlife/livestock interfaces in the central parts of Mozambique and provides evidence that this cycle acts as a permanent source of ASFV to domestic pigs in Mozambique. The presence of a new genotype in the ticks represents a potential risk for the occurrence of outbreaks caused by previously unknown genotypes. This could potentially contribute to failure of vaccines once they are developed based on current knowledge and warrants more research on the disease in Mozambique. The results of this study will also aid in the design and implementation of appropriate control strategies by the veterinary authorities of Mozambique.