Molecular evidence of Rickettsia felis infection in dogs from northern territory, Australia

The prevalence of spotted fever group rickettsial infection in dogs from a remote indigenous community in the Northern Territory (NT) was determined using molecular tools. Blood samples collected from 130 dogs in the community of Maningrida were subjected to a spotted fever group (SFG)-specific PCR targeting the ompB gene followed by a Rickettsia felis-specific PCR targeting the gltA gene of R. felis. Rickettsia felis ompB and gltA genes were amplified from the blood of 3 dogs. This study is the first report of R. felis infection in indigenous community dogs in NT.


Introduction
Rickettsioses are important emerging vector-borne diseases in humans [1], and some have been reported to infect dogs [2,3]. Rickettsioses that are purported to be endemic in Australia include murine typhus (Rickettsia typhi), Queensland tick typhus (Rickettsia australis), Flinders Island spotted fever (Rickettsia honei), scrub typhus (Orientia tsutsugamushi) and Q fever (Coxiella burnetti). Cat flea typhus or flea-borne spotted fever, caused by Ricketsia felis, which was first described in humans in the USA [4], is an emerging zoonosis that has been reported from throughout the world and was recently reported in a cluster of family members in Victoria, Australia [5].
The cat flea, Ctenocephalides felis, is a reservoir and biological vector of R. felis [6]. In Australia, R. felis DNA was first identified in fleas in 2006 [7]. Very recently, a molecular study in Australia detected R. felis in 9% of 100 tested Southeast Queensland (SE QLD) pound dogs, suggesting that dogs may act as mammalian reservoir hosts for R. felis [8]. Given the close bond that exists between humans and canines, it is possible that dogs may be a potential source of R. felis infection for humans. Dogs are also important in indigenous community life, and it is therefore prudent to better understand the public health risks that may be associated with the human-animal bond in these communities. To this end, we sought to investigate the prevalence of spotted-fever group organisms in dogs from the indigenous community of Maningrida in the Northern Territory (NT), using PCR assays. Data on gender were unavailable for 5 dogs. One mL of whole blood was obtained from each dog by venipuncture, and the blood was immediately transferred to a QIAcard FTA ® Four Spots (Qiagen, Hilden, Germany) to facilitate preservation of DNA during transport.

Materials and methods
DNA extraction from blood applied to FTA cards was carried out using the MasterPure™ DNA purification kit (Epicentre Biotechnologies, Madison, USA), following the protocol described by Abd Rani et al. [9].
All extracted DNA was subjected to a previously described PCR assay which amplifies a portion of the ompB gene in SFG rickettsiae [10]. DNA extracted from a Rickettsia conorii culture obtained from the Australian Rickettsial Reference Laboratory was used as a positive control, whilst nuclease-free water (NFW) was used as a negative control. Blood samples found to be positive for SFG rickettsial DNA were further subjected to a speciesspecific nested PCR for R. felis, which amplified the gltA gene (Table 1). Rickettsia felis specific primers were designed by aligning gltA sequences retrieved from  (Table 1). However, the potential cross-reaction of this newly designed PCR was not tested with other Rickettsia spp. This PCR that was developed relies on confirming diagnosis by DNA sequencing to ensure there is no cross reactivity with closely related organisms.
Primary PCR amplification of the gltA gene of R. felis was performed in a 25 μl reaction mixture containing 2 μl of DNA, 5 μl 5× PCR buffer, 200 μmol/L dNTP, 2.0 mmol/L MgCl 2 , 0.5 units of GoTaq polymerase (Promega, Madison, WI, USA), 10 pmol of each forward and reverse primer made to the final volume with NFW. Two μl of distilled water and R. felis DNA isolated from an infected cat flea were used as negative and positive control, respectively. PCR cycling conditions comprised an initial activation step at 95°C for 2 min, followed by 40 cycles of 95°C for 45 s, 63°C for 30 s and 72°C for 45 s with a final extension step of 72°C for 7 min. The secondary round of PCR amplification was performed as for the primary round except that the annealing temperature was 59°C.
All amplified PCR products of both SFG rickettsial and R. felis-specific PCR were visualised via transillumination on 1.6% agarose gel stained with SYBR ® Safe (Invitrogen, Eugene, USA) DNA gel stain. All PCR products of the correct size were purified using QIAquick™ PCR Purification Kit (Qiagen, Hilden, Germany) according to manufacturer's protocol and subjected to DNA sequencing using both forward and reverse primers. DNA sequences were assessed for homology with those in GenBank using the NCBI Blast program http://blast.ncbi.nlm.nih.gov/Blast.cgi. This project was approved by the University of Queensland Animal Ethics Committee.

Results
Of the total 130 tested samples, three (

Discussion
This is the first study to report the presence of R. felis infection in indigenous community dogs in the NT. In Australia, R. felis was first identified in fleas collected from dogs and cats in Western Australia in 2006 [7]. Recently, 19.8% of fleas collected from cats in Brisbane, Sydney and Melbourne [11] and 9% of pound dogs in SE QLD in Australia were reported to harbour R. felis DNA as identified by PCR [8]. Cat flea typhus was also reported in a cluster of human patients in Victoria [5]. The findings of these studies supported the presence of a peridomestic cycle and wide geographical distribution for this pathogen in Australia.
In the previous molecular study, the infection of R. felis in pound dogs in SE QLD was confirmed by PCR assay targeting only the ompB gene and confirmed by phylogenetic analysis. In this study, the gltA gene of R. felis in dogs' blood was amplified in addition to the ompB gene by a novel R. felis-specific nested PCR. This study strengthens the evidence for dogs harbouring active rickettsaemias of this emerging zoonotic pathogen, and supports the role of dogs as important reservoir hosts. This finding was also supported by a study reporting concurrent R. felis infections in a PCR-positive dog and its infected owners in Spain [12].
In this study, the flea infestation in indigenous community dogs was not evaluated in terms of flea species and burden. However, the cat flea, C. felis, is the most common flea infesting dogs in Australia [7,13]. These indigenous community dogs lead a semi-domesticated life. They are mostly poorly cared for and often infested by ectoparasites. Hence, it is likely that the dogs in this study were infested with fleas.
The prevalence of R. felis infection in indigenous community dogs (2.3%) is lower than those reported in pound dogs in SE QLD (9%) [8]. The transportation of blood using FTA cards and different DNA extraction technique may have partially affected the prevalence detected in indigenous community dogs. This is supported by studies showing amplification of DNA extracted from whole blood is more sensitive and reliable than filter paper-based technologies [14,15]. It should also be noted that neither study was particularly comprehensive with respect to sampling across different sites in each region, hence caution is required in making detailed comparisons with respect to prevalence. In this study, all dogs also appeared physically healthy, a common feature that is usually a characteristic of reservoir hosts, however further study is needed to determine the pathogenesis and pathogenic potential of R. felis infections in dogs. Table 1 Sequences and primer sets for the amplification of a partial region of the ompB gene of SFG rickettsiae and gltA gene of R. felis

Conclusion
This study further supports the potential role of dogs as reservoir hosts of R. felis and suggests that this pathogen has a wide geographical distribution in Australia.