Haemoplasma prevalence, diversity, and transmission dynamics in terrestrial small mammals from South Africa

Abstract

Within South Africa, 9.1 million people live in resource-poor environments where they have direct and indirect (through shared ectoparasites) contact with both invasive and indigenous rodent species. Rodents are carriers of multiple pathogens, inclusive of the emergent and potentially zoonotic group of bacteria known as haemotropic mycoplasmas (or haemoplasmas). Prior to this study, data on rodent-associated haemoplasma occurrence and diversity in South Africa was limited to a single study on subterranean African mole-rats of the family Bathyergidae. Information regarding haemoplasma occurrence, diversity and modes of transmission in South Africa are presently lacking for invasive and indigenous rodent species occupying informal human settlements in densely populated urban environments or communities living at the wildlife/livestock/human interface. In this study, we assessed the occurrence, diversity, and potential routes of transmission of haemoplasma genotypes cycling in 486 rodents, comprising of three invasive Rattus species (Rattus rattus, R. tanezumi and R. norvegicus) and six indigenous rodent species (Aethomys ineptus, Mastomys natalensis, Mastomys coucha, Gerbilliscus leucogaster, Lemniscomys rosalia and Saccostomus campsestris). Where possible, data on co-infections with Anaplasma, another genus of bloodborne bacteria, seasonal variation and variation across a human land-use gradient were assessed. Invasive Rattus samples were collected across the urban Gauteng metropole, the peri-urban Roodeplaat Dam Provincial Nature Reserve (RDPNR) in the Gauteng Province and in the Mnisi community area in the Mpumalanga Province (representative of agropastoral informal settlements as well as the bordering Manyeleti Nature Reserve). Indigenous species were either sampled in the peri-urban RDPNR (A. ineptus only) or in the Mnisi community area (all indigenous species assessed). Targeted 16SrRNA PCR assays were used to determine haemoplasma occurrence. PCR assays targeting the 16S rRNA and Rnase P gene regions allowed for phylogenetic placement of detected haemoplasma genotypes. Where available, DNA extracted from buccal swabs, ectoparasites and foetal samples were screened for haemoplasma presence. Haemoplasma occurrence rates were found to vary between the species assessed, ranging from 0% in S. campestris in the Mnisi area to 58.33% in R. rattus samples collected in the Gauteng metropole. Four haemoplasma genotypes, grouping in three well-supported phylogenetic clusters, were detected in the Rattus species and were closely related to haemoplasmas previously detected in Rattus species sampled in Japan, Hungary and Brazil. This suggests pathogen co-invasion during historic introduction of Rattus hosts into South Africa. A higher diversity of Rattus-associated genotypes was detected in the Gauteng metropole compared to RDPNR and the Mnisi area, however, sample size of invasive rats was lower in both areas. In total, ten distinct genotypes, grouping in four well-supported phylogenetic clusters, were detected in the indigenous rodent species assessed in the peri-urban RDPNR (n = 5) and the Mnisi community area (n = 7). Two genotypes, isolated from M. coucha samples in the Mnisi area had close relationships (p-distance of 1.5% and 1.8%) to a genotype detected in R. norvegicus samples collected in the Gauteng metropole, suggestive of host-switching events at some point. In terms of transmission routes, ticks (43.5%), fleas (18.2%) and lice (14.5%) collected from rodents sampled in the Mnisi area tested positive for haemoplasmas. While buccal swabs showed low occurrence rates (0.3%), the detection of haemoplasma in a R. tanezumi sample, collected in the Gauteng metropole, represents the first record in a mouth swab of a free-living rodent species. The foetal samples of 30% of haemoplasma-positive dams tested positive for haemoplasma presence, representing the first report of vertical transmission of rodent-associated haemoplasmas. These results indicate the use of multiple transmission routes by rodent-associated haemoplasmas. No differences in haemoplasma occurrence between host sex or age was found. Haemoplasma occurrence did not vary seasonally or between transformed and natural sampling sites. Low climatic variations between sampling seasons and proximity between natural and transformed sites highlight the need for fine-scale seasonal and spatial sampling when assessing haemoplasma occurrence. No haemoplasma/Anaplasma co-infections were found in Aethomys rodents from RPNR. The presence of multiple peaks at specific sites of the generated nucleotide sequence chromatograms (ranging between 0.29% and 7.1% of sites per nucleotide sequence), indicated that 13.9% of the positive samples collected from the Mnisi community area showed co-infection with more than one haemoplasma genotype. Together, these results provide a baseline for data on potentially zoonotic haemoplasmas cycling in indigenous and invasive rodent communities within South Africa and expand on the knowledge of rodent-associated haemoplasmas worldwide.