Genetic diversity of Anaplasma marginale in cattle and in putative novel Anaplasma species from wildlife in Mpumalanga, South Africa

Abstract

Bovine anaplasmosis is amongst the three most important tick-borne diseases (TBDs) of ruminants in southern Africa and results in major economic losses in food and animal production not only in southern Africa but also on a global scale. This disease is mainly caused by the obligate intracellular rickettsia, Anaplasma marginale, which is currently widespread in South Africa. Anaplasma centrale can also cause disease in cattle although this is rare. Other Anaplasma species have been identified in cattle in South Africa, but it is not known if they cause disease. The economic impact of bovine anaplasmosis in South Africa has been estimated at approximately R115 million ($US9.6 million) per year due to mortalities and the cost of controlling the disease, as well reduced production. In other parts of the world, costs arising from anaplasmosis have been estimated from $US300 to $US800 million. Clinical signs caused by infection with A. marginale are characterized by fever, progressive anaemia, weight loss, abortion in pregnant cows and lowered milk production, as well as icterus that may result in mortality. Animals under one year of age are usually asymptomatic to infection with A. marginale. However, older animals are more likely to react severely and fatally upon challenge. Several other species of Anaplasma which infect cattle have been reported in South Africa: these include A. centrale, A. bovis, A. platys and Anaplasma sp. (Omatjenne). This study centres around assessing the diversity of Anaplasma species harboured by African wildlife and the possible impact thereof in humans and livestock situated at the wildlife-livestock interface.

The rapid advancement of high-throughput sequencing technologies in the 21st century has resulted in the discovery of a plethora of genetic material ascribed to the genus Anaplasma worldwide, with the proposal of over 20 new species with unique 16S rRNA sequences from various hosts since the last formal organization of the genus. The relationship of these newly detected agents to known pathogens and their ability to serve as a source of cross-reaction in detection assays, have not been well assessed. Third-generation sequencing and bioinformatics tools were used to profile Anaplasma populations in wildlife species roaming in the Kruger National Park (KNP) and surrounding game reserves, Mpumalanga Province, South Africa, situated adjacent to the resource-poor rural area, the Mnisi community in Mpumalanga Province, thus resulting in the wildlife-livestock-human interface in the area. In a comprehensive screening of 343 wildlife samples using an Anaplasma genus-specific real-time PCR assay, Anaplasma species were detected in 70.0% (21/30) of African buffalo (Syncerus caffer), 86.7% (26/30) of impala (Aepyceros melampus), 36.7% (11/30) of greater kudu (Tragelaphus strepsiceros), 3.2% (1/31) of African wild dog (Lycaon pictus), 40.6% (13/32) of Burchell’s zebra (Equus quagga burchelli), 43.3% (13/30) of warthog (Phacochoerus africanus), 22.6% (7/31) of spotted hyena (Crocuta crocuta), 40.0% (12/30) of leopard (Panthera pardus), 17.6% (6/34) of lion (Panthera leo), 16.7% (5/30) of African elephant (Loxodonta africana) and 8.6% (3/35) of white rhinoceros (Ceratotherium simum) samples. Microbiome sequencing data from the Anaplasma-positive samples revealed four genotypes that phylogenetically group with known and previously published Anaplasma 16S rRNA sequences, as well as 13 novel Anaplasma 16S rRNA sequences. Our findings reveal a greater genetic diversity of Anaplasma sequences and potentially novel species circulating in wildlife hosts in South Africa than are currently classified within the genus Anaplasma which might be transmitted to livestock or companion animals. Furthermore, these putative species are phylogenetically similar to known Anaplasma spp. and may possibly serve as a source of cross-reaction in the current detection assays. Our findings further highlight the need for additional genetic data and genome sequencing of these putative species for correct Anaplasma species classification and further assessment of their occurrence in livestock and companion animals.

Data collected previously in the study area of the Mnisi Community in the Mpumalanga Province indicated the presence of A. marginale, with occasional bovine anaplasmosis cases reported at villages close to the wildlife-livestock interface. In an attempt to understand the clinical cases of bovine anaplasmosis in the study area, the infection dynamics and A. marginale strain diversity during a 12-month period were examined in ten calves in a peri-urban area and at a wildlife-livestock interface. The composition of Anaplasma species circulating in these calves was also assessed. Anaplasma marginale was detected in all five calves in the peri-urban area from the first month, but in only two calves at the wildlife-livestock interface and only after six months. Msp1α genotype analysis revealed 42 A. marginale genotypes in calves in the peri-urban area and ten genotypes in calves at the wildlife-livestock interface, with superinfections evident in calves from both areas. The 16S microbiome sequencing data revealed the presence of four Anaplasma species circulating in the ten calves. Of the total number of Anaplasma 16S rRNA sequences detected, 87% were identified in calves in the peri-urban area and 13% in calves at the wildlife-livestock interface. The 16S rRNA sequencing data consisted mostly of A. platys-like 16S rRNA sequences (83.3%), followed by A. marginale (16.6%) and A. boleense (<0.1%). Our findings therefore suggest that the occasional bovine anaplasmosis cases observed at the wildlife-livestock interface in the Mnisi communal area might be attributed to a localised lack of endemic stability since calves at the wildlife-livestock interface are not continually infected with A. marginale in their first year when natural immunity is higher. Our findings further highlight complex A. marginale infection in infected cattle driven by co-infection and superinfection by distinct A. marginale strains in both areas within the 12-month study period, indicating continuous challenge with multiple strains that should lead to robust immunity in infected animals. Other Anaplasma species detected in the calves might be due to proximity with wildlife hosts and might confer cross-protection against A. marginale thus contributing to endemic stability, but this requires further investigation.

Considering the strain variation detected in A. marginale in the Mnisi community and other South African provinces (based on previous studies) and the difficulty of isolating and culturing Anaplasma species, alternative methods of genome sequencing of Anaplasma species from cattle and wildlife in South Africa are required. An attempt was therefore made to obtain A. marginale genome sequence data directly from infected carrier cattle. Cattle at the Innovation Africa @ University of Pretoria Experimental Farm were screened for the presence of A. marginale and other haemoparasites and the msp1α genotypes in A. marginale-positive animals were determined. Blood was drawn from cattle infected with a single A. marginale strain and no other haemoparasites, red blood cells were separated and washed seven times with phosphate-buffered saline. High molecular weight DNA was extracted directly from the washed red blood cells. Three rounds of microbial enrichment were conducted to deplete the host DNA in the sample, followed by whole genome amplification. The resulting DNA sample was sent for whole genome sequencing on a Pacific Biosciences (PacBio) sequencing platform. A total of 298 058 raw PacBio reads were retrieved from the PacBio single-molecule real-time (SMRT) analysis 2.3.0 software, which were mainly bovine host reads. Anaplasma reads mapped to the A. marginale St Maries and A. marginale Florida reference genomes resulted in two different incomplete A. marginale assemblies, each informed by the reference sequence. Further sequencing data is thus needed for full closure of the genome sequence. Advances in molecular techniques for microbial DNA enrichment and sequencing, and assortment of contigs into species-specific bins and assembly of binned data could be incorporated in this study to complete the A. marginale genome. Such a technique could then be used to obtain the whole genome sequences of the different Anaplasma spp. circulating in livestock, wildlife and companion animals without the need to culture. Therefore, there is a need for molecular techniques for microbial DNA enrichment and metagenomics to generate more genome sequences of A. marginale and the different Anaplasma spp. circulating in livestock, wildlife and companion animals. This will allow for correct classification in the Anaplasma taxonomy and to study the natural rate of variation between the different Anaplasma species and their specific genotypes and to fully understand their evolution and diversity. This will further assist with the identification of species-specific targets for the development of more specific serological nucleic-acid-based detection methods suitable for examining the epidemiology of all Anaplasma spp. from various hosts.