Blood was collected from immobilised wild animals in Laikipia County and Maasai Mara National Reserve between February 2014 and October 2015. Laikipia County is about 9500 km² and is located in the central region of Kenya to the northwest of Mt. Kenya between 0.88N 36.18E and 0.2667S 37.38E. The Maasai Mara National Reserve is approximately 1510 km² between 1.22S 34.75E and 1.75S 35.42E within Narok County in southwestern Kenya along the border with Tanzania. The main human populations in both areas are pastoralists whose livelihoods are dependent on livestock. Other forms of land use include agriculture, commercial ranching, wildlife conservation and ecotourism. The two areas comprise some of the most important areas for biodiversity in Kenya, and they have large populations of free-ranging wildlife, which share habitats and other resources with humans and domestic animals providing a likely interface for disease transmission. The sampling sites in each study area were selected based on high interaction of livestock and wildlife and accessibility to enable darting of the animals. The coordinates of each sampling site were recorded using a Global Positioning System (GPS) (Garmin GPS 12 XL, Garmin Olathe, KS, USA) and entered into a Geographical Information System (GIS) database.

Convenience sampling of the animals was employed because of the difficulties of constructing a sampling frame in wildlife to allow for random sampling. This method allowed for readily available animals of the target species to be sampled. The target species were those most common in the study areas with the high tendency to interact with livestock. These included buffalo (Syncerus caffer), zebra (Equus burchellii), Grant’s gazelle (Nanger granti), common waterbuck (Kobus ellipsiprymnus ssp. ellipsiprymnus), impala (Aepyceros melampus), Topi (Damaliscus lunatus ssp. jimela), Coke’s hartebeest (Alcelaphus buselaphus) and wildebeest (Connochaetes taurinus). To facilitate sample collection, the animals were immobilised following the protocols recommended by McKenzie (1993) by experienced personnel to ensure a humane exercise as much as possible. At least 30 mL of blood was collected from each animal by jugular venipuncture into EDTA-coated tubes and split into four aliquots. Each aliquot was labelled with information identifying the sample number, date, location and animal species and stored frozen in liquid nitrogen (-196 °C) until required for processing.

Genomic DNA was extracted from preserved EDTA blood samples using the manufacturer’s instructions for the DNeasy^® Blood and Tissue Kits (QIAGEN GmbH, Hilden, Germany). However, these instructions were modified slightly in order to optimise the amount of DNA extracted by increasing the amount of blood from 50 µL to 100 µL as recommended by the manufacturer to 200 µL and reducing the amount of AE buffer from 200 µL to 150 µL. Extracted DNA quality was evaluated using the agarose gel electrophoresis protocol in which an aliquot of the extracted DNA was run on 1.2% agarose gel.

Host DNA was screened for evidence of Rickettsia using PCR assay to amplify the intergenic spacer rpmE-tRNA^fMet and the citrate synthase-encoding gene gltA using previously described primer sets shown in Table 1. The amplifications were carried out in a total volume of 25 µL reaction mix containing 1 µL deoxyribonucleotide triphosphates solution (dNTPs), 2.5 µL standard Taq buffer (Biolabs^®, New England, UK), 1 µL each of reverse and forward primers, 17.25 µL of DNase/RNase-Free^® PCR grade water (QIAGEN, Hilden, Germany), 0.25 µL of Taq DNA polymerase (Biolabs^®, New England, UK) and 2 µL of template DNA. The amplifications were performed in Applied Biosystems Veriti^® 96-well thermocycler (Applied Biosystems, California, USA). The PCR conditions were as follows: 3 min initial denaturation at 95 °C, 35 cycles of 30 s denaturation at 95 °C, 30 s primer annealing at temperatures specific for each of the primers (Table 1), one minute extension at 72 °C and a final 10 minute extension at 72 °C. The mixture was then maintained at 4 °C. A negative control using DNase/RNase-Free^® PCR grade water (QIAGEN, Hilden, Germany) was included for quality control. The positive control used was DNA extracted from Rickettsia africae isolated from a tick in Kenya. The PCR products were visualised using agarose gel electrophoresis to check for the presence of band and size of amplicon.

The gltA-positive PCR amplicons of, or close to, the expected product sizes were purified using QIAquick^® purification kit (QIAGEN, Hilden, Germany) following the manufacturer’s instructions. Sequencing was done by direct cycle sequencing using the ABI PRISM BigDye Terminator V3.1 cycle sequencing kit and the sequences analysed in an ABI310 DNA analyser (Applied Biosystems, California, USA). Traces were assembled and primer regions trimmed using Geneious v 8.1.6 software. Consensus nucleotide sequences were used to query the GenBank database, and the highest similarity was identified by Basic Local Alignment Search Tool (BLASTN) available from the National Center for Biotechnology Information (Bethesda, MD). This was used to assign identity to the recovered species. The study sequences along with those with closest match in GenBank were aligned using MUSCLE (Edgar 2004). The evolutionary history was inferred by using the maximum likelihood method based on the Tamura 3-parameter model (Tamura 1992). The phylogeny tree with the highest log likelihood (-1233.9127) was developed. The percentage of trees in which the associated taxa clustered together was shown next to the branches. Initial tree(s) for the heuristic search were obtained automatically by applying neighbour-joining and BioNJ algorithms to a matrix of pair-wise distances estimated using the maximum composite likelihood (MCL) approach and then selecting the topology with superior log likelihood value. The tree was drawn to scale with branch lengths measured in the number of substitutions per site. The analysis involved 15 nucleotide sequences. Codon positions included were 1st+2nd+3rd+Noncoding. There were a total of 774 positions in the final dataset. Evolutionary analyses were conducted in MEGA6 (Tamura et al. 2013).

Animals were sampled in 17 locations in Laikipia County and 19 locations in Maasai Mara National Reserve and the neighbouring areas during six separate expeditions between February 2014 and October 2015 (Table 2). The sampled areas in Laikipia County included nine sites in Ol Pejeta conservancy, five sites in ADC Mutara Ranch and two sites in Mpala Ranch all of which incorporate cattle ranching and wildlife conservation as well as one site in Kiamariga sub-location, which is a community land where free-ranging wildlife interacts with livestock (Figure 1). In Maasai Mara, sampled areas included seven sites inside the National Reserve and 12 sites in neighbouring community group ranches where wildlife interacts freely with livestock (Figure 1).

In total, 152 animals comprising of eight different species were sampled in both areas. These comprised of 79 in Laikipia and 73 in Maasai Mara. All the animals responded well to the immobilisation drugs and induction times ranged between 8 and 12 min. No complications were encountered during immobilisation and handling except for a few animals, which had slightly elevated body temperatures that were attributed to physical exertion during darting as well as psychological stress and fear. These animals were cooled by applying copious amounts of water on the whole body.

Two of the 79 (2.5%) animals in Laikipia were found infected with SFG rickettsioses. Infections were found in a zebra (Equus burchellii) and a buffalo (Syncerus caffer) representing a prevalence of 2.6% and 3.2% in these species, respectively. Four of the 73 (5.5%) animals in Maasai Mara were infected with SFG rickettsioses. Infections were found in 1 of the 35 wildebeests (Connochaetes taurinus) and 3 of the 30 Topi (Damaliscus lunatus ssp. jimela) representing a prevalence of 2.9% and 30%, respectively. Infections were detected by amplification of the intergenic spacer rpmE-tRNA^fMet and gltA gene. These findings are summarised in Table 3. A representative gel image of PCR amplification of gltA gene is shown in Figure 2.

Rickettsia sibirica was identified in one sample obtained from a Topi (Damaliscus lunatus ssp. jimela) in Maasai Mara National Reserve. The isolate yielded a partial gltA sequence of 779 bp that had 99% identity to R. sibirica isolated from China Accession number KM288711 in the GenBank. The study sequence was submitted to the GenBank and provided accession number KX244606. The maximum likelihood phylogeny tree drawn using the study sequence and those similar to it in GenBank is shown in Figure 3. From the tree, the study sequence has a close relationship with other sequences from the same species in GenBank, such as R. sibirica from Senegal, while having a more distant relationship with the same species from other continents.