African horse sickness outbreak investigation and disease surveillance using molecular techniques

Abstract

African horse sickness (AHS) is a life-threatening disease of equids caused by African horse sickness virus (AHSV), a member of the genus Orbivirus in the family Reoviridae. The virus is transmitted by midges (Culicoides spp.) and the disease is most prevalent during the time of year, and in areas where vector Culicoides spp. are most abundant, namely in late summer in the summer rainfall areas of endemic regions. The disease is of importance to health and international trade in horses worldwide. Effective surveillance is critical in order to establish transparent criteria for animal trade from a country or region where AHS occurs.

The 2011 outbreak of African horse sickness in the African horse sickness controlled area in South Africa: An outbreak of AHS caused by AHSV type one (AHSV1) occurred in the surveillance zone of the AHS controlled area of the Western Cape during the summer of 2011. The epicentre of the outbreak was the town of Mamre in the magisterial district of Malmesbury, and the outbreak was confined to a defined containment zone within this area through movement control of all equids and a blanket vaccination campaign. A total of 73 confirmed cases of AHS were reported during this outbreak, which included four subclinical cases confirmed by virus isolation (VI). The estimated morbidity rate for the outbreak was 16% with an estimated mortality rate of 14% and a case fatality rate of 88% based on the figures above. Outbreak disease surveillance relied on agent identification using AHSV group specific reverse transcriptase quantitative polymerase chain reaction (GS RT-qPCR) based assays, which was novel for an AHS outbreak in South Africa. The source of this outbreak was not confirmed at the time, but was believed to be associated with an illegal 2 movement of an infected animal into the Mamre area. A detailed description of the outbreak is given in Chapter 2, and the outbreak provided an opportunity to assess decision making in future AHS outbreaks in the AHS controlled area of South Africa and in countries where AHS is an exotic or emerging disease. This outbreak further highlighted deficiencies and complications of available AHSV diagnostic testing and surveillance methods, and the need for further refinement of these assays and strategies.

Development of three triplex real-time reverse transcription PCR assays for the qualitative molecular typing of the nine types of African horse sickness virus: The typing of the specific AHSV involved in the Mamre outbreak was initially done by partial, direct sequencing of the S10 gene (encoding the non-structural protein NS3) and the L2 gene (encoding the type-specific outer capsid protein VP2) which confirmed the virus to be AHSV1. This process is time consuming and it became evident that a faster alternative was needed. This led to the development of type specific RT-qPCR (TS RT-qPCR) assays to supplement the GS RT-qPCR assay that had already been developed, characterized and validated. Blood samples collected during routine diagnostic investigations from South African horses with clinical signs suggestive of AHS were subjected to analysis with the GS RT-qPCR assay and VI with subsequent serotyping by plaque inhibition (PI) assays using AHSV type-specific antisera. Blood samples that tested positive by AHSV GS RT-qPCR were then selected for analysis using AHSV TS RT-qPCR assays. The TS RT-qPCR assays were evaluated using both historic stocks of the South African reference strains of each of the 9 AHSV types, as well as recently derived stocks of these same viruses. Of the 503 horse blood samples tested, 156 were positive by both AHSV GS RT-qPCR and VI assays, whereas 135 samples that were VI negative were positive by AHSV GS RT-qPCR assay. The virus isolates made from the various blood samples included all 9 AHSV types, and there was 100% agreement between the results of conventional serotyping of individual virus isolates by PI assay and AHSV TS RT-qPCR typing results. Results of this study confirmed that the AHSV TS RT-qPCR assays for the identification of individual AHSV types are applicable and practicable and therefore are potentially highly useful and appropriate for virus typing in AHS outbreak situations in endemic or sporadic incursion areas, which can be crucial in determining appropriate and timely vaccination and control strategies.

Evaluation of the use of foals for active surveillance in an AHS containment zone during the season following an AHS outbreak: In order to further evaluate the AHS status of horses in the Mamre area after the outbreak of 2011, a targeted surveillance strategy was developed. Serial serum and whole blood samples were collected on a monthly basis from January to June, 2012 from foals (identified by microchip) that were born in the Mamre 3 district after the end of the outbreak. Sera were evaluated using traditional serological methods and the results were compared to the results obtained using the newly developed molecular assays for virus detection and identification. This study confirmed that AHSV was eradicated in the Mamre area after the outbreak and, therefore, that the control measures implemented in the area by the State Veterinary Authorities were effective.

Characterization of the dynamics of African horse sickness virus in horses by assessing the RNAaemia and serological responses following immunisation with a commercial polyvalent live attenuated vaccine: As was shown in the 2011 Mamre outbreak, detection of AHSV during outbreaks has become more rapid and efficient with the recent development of quantitative GS RT-qPCR assays to detect AHSV nucleic acid. Use of this assay together with the TS RT-qPCR assays described in Chapter 3, will not only expedite diagnosis of AHS but also facilitate further evaluation of the dynamics of AHSV infection in the equine host. A potential limitation to the application of these assays is that they detect viral nucleic acid originating from any AHSV live attenuated vaccine (AHSVLAV), which is the vaccine type routinely administered to horses in South Africa. A study was, therefore, designed to characterize the dynamics and duration of the RNAaemia as compared to the serological responses of horses following vaccination with a commercial AHSV-LAV, using GS and TS RT-qPCR assays and serum neutralisation tests. This study provided baseline data on the GS and TS nucleic acid dynamics in weanling foals vaccinated for the first time, yearlings vaccinated for a second time and adult mares following a booster to multiple previous vaccinations. These data are fundamental to interpreting results of AHSV GS RT-qPCR testing of vaccinated horses within an area where virological surveillance is being applied.

African horse sickness caused by genome reassortment and reversion to virulence of live, attenuated vaccine viruses, South Africa, 2004 - 2014: In 2014 a further outbreak of AHS caused by AHSV1 occurred in the Porterville area of the AHS protection zone (PZ), spreading into the Wellington area in the AHS surveillance zone (SZ). Further involvement of the Robertson area (AHS PZ) subsequently also occurred. The case fatality rate was much lower than that of the Mamre outbreak. The clinical signs in infected horses were also generally milder in the 2014 outbreak, as compared to the 2011 outbreak. Whole genome sequencing of samples from the Porterville outbreak confirmed that causative virus was a recombination (reassortant) of AHSV types 1 and 4, with genes derived from the relevant vaccine strains contained in OBP comb1 of the commercial polyvalent AHSV-LAV used in South Africa. This led to further analysis of 39 AHSV strains from field cases of AHS that originated from outbreaks within the controlled area, which confirmed reversion to virulence 4 of AHSV type 1 vaccine in two outbreaks (2004 and 2011) and multiple reassortment events in two outbreaks (2004 and 2014) with genes derived from all three AHSV vaccine strains (types 1, 3 and 4). This study provided a molecular and epidemiological comparison of the five unique AHSV type 1 outbreaks in the AHS controlled area. It was shown that all the outbreaks in the AHS controlled area attributed to AHSV type 1 since the inception of the area in 1997, have been due either to reversion to virulence of the AHSV type 1 vaccine strain, or recombination of AHSV type 1 vaccine strain with one or both of the other vaccine strains in OBP comb1 of the commercial AHSV-LAV.