Unravelling the Diversity of Microorganisms in Ticks from Australian Wildlife

Ticks and tick-borne pathogens pose a significant threat to the health and welfare of humans and animals. Our knowledge about pathogens carried by ticks of Australian wildlife is limited. This study aimed to characterise ticks and tick-borne microorganisms from a range of wildlife species across six sites in Victoria, Australia. Following morphological and molecular characterisation (targeting 16S rRNA and cytochrome c oxidase I), tick DNA extracts (n = 140) were subjected to microfluidic real-time PCR-based screening for the detection of microorganisms and Rickettsia-specific real-time qPCRs. Five species of ixodid ticks were identified, including Aponomma auruginans, Ixodes (I.) antechini, I. kohlsi, I. tasmani and I. trichosuri. Phylogenetic analyses of 16S rRNA sequences of I. tasmani revealed two subclades, indicating a potential cryptic species. The microfluidic real-time PCR detected seven different microorganisms as a single (in 13/45 ticks) or multiple infections (27/45). The most common microorganisms detected were Apicomplexa (84.4%, 38/45) followed by Rickettsia sp. (55.6%, 25/45), Theileria sp. (22.2% 10/45), Bartonella sp. (17.8%, 8/45), Coxiella-like sp. (6.7%, 3/45), Hepatozoon sp. (2.2%, 1/45), and Ehrlichia sp. (2.2%, 1/45). Phylogenetic analyses of four Rickettsia loci showed that the Rickettsia isolates detected herein potentially belonged to a novel species of Rickettsia. This study demonstrated that ticks of Australian wildlife carry a diverse array of microorganisms. Given the direct and indirect human–wildlife–livestock interactions, there is a need to adopt a One Health approach for continuous surveillance of tick-associated pathogens/microorganisms to minimise the associated threats to animal and human health.


Introduction
Australia is home to more than 3000 species of wild vertebrates, and new species are being added to the list frequently [1]. Many of these vertebrate hosts are susceptible to tick infestation and its harmful effects such as skin damage, toxicosis, irritation, paralysis, allergies, and anaemia [2,3]. Ticks are obligate haematophagous ectoparasites commonly associated with the transmission of infectious agents in wild and domestic animals [2]. Additionally, 17% of human infections are vector-borne, and there is an increasing incidence of tick-borne zoonoses [4,5] that has been primarily ascribed to climate change, landscape modifications, and changes in mammalian host populations due to habitat fragmentation and degradation [6][7][8][9][10]. For example, an increase in the prevalence of tick-associated Lyme disease in North America has been attributed to the increased population of the key reservoir host (i.e., white-footed mice (Peromyscus leucopus)) due to forest fragmentation and associated reduced mammalian diversity [11]. 2 of 16 While risks to human and animal health due to ticks and tick-borne pathogens (TBPs) are of longstanding concern in Australia, our knowledge of specific pathogens posing such risks is limited [12]. For example, although Borrelia burgdorferi s.l. has not been reliably identified in Australia, anecdotal evidence supports the association of tick bites with a Lyme-disease-like illness known as the Debilitating Symptom Complexes Attributed to Ticks (DSCATT) [12,13]. Infections with Rickettsia (R.) are also common in Australia, and clinical cases of Queensland tick typhus (caused by Rickettsia australis) [14,15], Flinders Island spotted fever (R. honei) [16], and Australian spotted fever (R. honei subsp. marmionii) were reported earlier [17].
Traditionally, PCR-based diagnostic tools have been used for the identification of TBPs. However, these conventional tools are not ideal for large-scale surveillance programs due to limitations such as the capacity to simultaneously target only a few pathogens (usually known), requiring large volumes of target nucleic acid, and consuming a longer time for testing multiple pathogens. Additionally, conventional diagnostic tools may not target important commensals or endosymbionts within ticks that play a critical role in transmitting TBPs [33,34]. Such limitations can be overcome through the use of a novel microfluidic-based technique [35] that uses a few millilitres of the DNA template to test 48 or 96 targets on a microfluidic system (BioMark™ dynamic array system, Fluidigm) [35][36][37][38]. These chips dispense 48 (or 96) samples and 48 (or 96) PCR mixes into individual wells on a microfluidic chip, thereby performing 2304 or 9216 real-time PCRs [35].
This study aimed to characterise ticks and tick-borne microorganisms from a range of wildlife species across six sites in Victoria, Australia. We used the microfluidic PCRbased technique for screening microorganisms followed by amplifying multiple loci using conventional PCRs to further characterise pathogens/microorganisms.

Morphological Identification of Ticks
Ticks were morphologically identified at the genus and species level (where possible) under a dissecting microscope (Olympus, Tokyo, Japan) following dichotomous keys [39,40]. Some of the tick specimens could not be reliably identified by using only their morphological characteristics but were subsequently identified along with other specimens via molecular characterisation as described below.

Molecular Identification of Ticks
A subset of representative tick specimens (n = 45) was selected for the extraction of genomic DNA and subsequent molecular identification in such a way that multiple specimens of each tick genus/species from each host species per location were included in the subset. Additionally, DNA was also extracted from those specimens (n = 20) for which morphological characterisation to the species level was not possible. Briefly, each tick was longitudinally cut into two halves; one half was preserved in ethanol and the other half was washed thrice (45 min each time) in distilled water. Subsequently, finely chopped

Morphological Identification of Ticks
Ticks were morphologically identified at the genus and species level (where possible) under a dissecting microscope (Olympus, Tokyo, Japan) following dichotomous keys [39,40]. Some of the tick specimens could not be reliably identified by using only their morphological characteristics but were subsequently identified along with other specimens via molecular characterisation as described below.

Molecular Identification of Ticks
A subset of representative tick specimens (n = 45) was selected for the extraction of genomic DNA and subsequent molecular identification in such a way that multiple specimens of each tick genus/species from each host species per location were included in the subset. Additionally, DNA was also extracted from those specimens (n = 20) for which Pathogens 2023, 12, 153 4 of 16 morphological characterisation to the species level was not possible. Briefly, each tick was longitudinally cut into two halves; one half was preserved in ethanol and the other half was washed thrice (45 min each time) in distilled water. Subsequently, finely chopped pieces of washed specimens were subjected to genomic DNA extraction using a DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) following the manufacturer's protocol except that the 56 • C incubation step with proteinase K digestion lasted for 24-48 h. The quality and concentration of the extracted DNA were assessed using a spectrophotometer (ND-1000 UV-vis spectrophotometer v.3.2.1; NanoDrop Technologies, Inc., Wilmington, DE, USA) and stored at −20 • C until further testing.

Microfluidic Detection of Microorganisms in Tick DNA Samples
The individual tick DNA samples (n = 45) were subjected to microfluidic real-time PCR for amplification of specific regions of DNA of the target microorganisms using a 48.48 dynamics array in a Bio-Mark™ real-time PCR system (Fluidigm, San Francisco, CA, USA), as described previously [35,37,38]. The details of the tested microorganisms and targeted genomic markers are given in Supplementary Table S1. Negative (no DNA), microorganism spike control (DNA of Escherichia coli, EDL933 strain), and tick DNA extraction controls were included in each microfluidic chip to ensure the efficient and valid amplification/detection of targets as described previously [35,37]. Microfluidic PCR results were confirmed (where only genus-level identification was achieved) using conventional PCR followed by Sanger sequencing as described previously [37].

qPCR Detection of Rickettsia spp.
The tick DNA samples were also screened for Rickettsia spp. using a real-time qPCR assay targeting citrate synthase gene (gltA) sequences as described by Tadepalli et al. [18]. A subset (n = 7) of ticks DNA-positive for Rickettsia gltA qPCR was further examined as described previously [18,43]. Briefly, the longer fragments of genes encoding (i) gltA, (ii) outer membrane protein A (ompA), and (iii) the 17 kDa outer membrane antigen were amplified [18].

DNA Sequencing and Phylogenetic Analyses
Conventional PCR amplicons generated for ticks and target microorganisms were purified using shrimp alkaline phosphate and exonuclease I (Thermo Fisher Scientific, Melbourne, Australia) and subjected to Sanger sequencing using PCR primers. Nucleotide sequences obtained for target loci of ticks (cox1 and 16S rRNA) and tick-borne microorganisms/pathogens (outer membrane protein B (ompB), ompA, gltA, and 17 kDa) were assessed for various quality parameters (such as the desired sequence length, base quality score, etc.) and assembled in Geneious Prime 2019.0.4 software using the de novo assembly function (Biomatters Ltd., Auckland, New Zealand; www.geneious.com). Unique sequences for each locus were identified via the "find duplicates" function in Geneious. Subsequently, these unique sequences were subjected to online searching within the National Center for Biotechnology Information (NCBI) (https://blast.ncbi.nlm.nih.gov/Blast.cgi) database to match their identities with known published reference sequences. All protein-coding nucleotide sequences were assessed for open reading frames, and then all unique nucleotide sequences were submitted to the NCBI GenBank. Pairwise comparisons were also conducted for each sequence dataset using BioEdit [44]. The reference sequences for each locus (of ticks or tick-borne microorganisms) were retrieved from GenBank and aligned in MEGA 11 [45] using MUSCLE v.3.8.31 [46]. Aligned sequences were trimmed to uniform lengths of 612 (cox1), 370 (16S rRNA), 352 (gltA), 434 (ompA), 391 (17 kDa), and 267 (ompB) bp.
Phylogenetic analyses were performed on aligned sequence datasets of individual loci for ticks (cox1 and 16S rRNA) and tick-borne microorganisms (ompA, ompB, gltA, and 17 kDa) using Bayesian Inference (BI), Maximum Likelihood (ML) and Neighbour Joining (NJ) methods. The NJ and ML analyses were performed in MEGA 11, whereas BI was performed using the MrBayes plugin [47] in Geneious. The jModelTest v.3.7 [48] was used to estimate the best-fit evolutionary models for individual sequence alignments based on the Akaike information criteria (AIC). The best-fit models for nucleotide sequences of ticks included the Tamura-Nei with Gamma distribution (cox1) and Tamura 3-parameter with Gamma distribution (16S rRNA) [49,50]. For nucleotide sequences of the microorganisms, the best-fit models were the Tamura 3-parameter (ompB and gltA), Tamura 3-parameter with Gamma distribution (ompA), and Kimura-2 parameter (17-kDa) datasets [49,51]. The nodal supports in the ML and NJ phylogenies were tested using the bootstrap method (10,000 replicates,) whereas the posterior probabilities of the BI analyses were calculated for 2,000,000 generations (ngen = 2,000,000) while saving every 200th tree (samplefreq = 200). Corresponding reference sequences of Ornithodoros moubata (GenBank accession: KJ133594), Rickettsia bellii (LAOIO1000001) and R. felis (CP000053, AF210692 and AF210694) were used as outgroups for the phylogenetic trees of ticks and microorganisms, respectively.
For the 16S rRNA sequences, I. kohlsi had the highest similarity (99.7%) to those of I. eudyptidis/kohlsi (KM488490 and KM488491) from Australia, while the I. antechini sequences showed the highest similarity (97.1%) to a sequence (AF113929) of the same species from an unknown origin. No 16S rRNA reference sequence was available for Ap. auruginans and I. trichosuri; however, their sequences determined herein showed the highest similarities (90.4% and 92%, respectively) to those of Bothriocroton concolor (JN8663727) and I. holocyclus (MH043264), respectively, from Australia. In concordance with the BLAST search results obtained for the I. tasmani cox1 sequences, the 16S rRNA sequences also showed a wide range of similarities (87.75-98.27%) with those reported previously (U95906 and MH043271). The pairwise comparisons of the 16S rRNA and cox1 sequences of I. tasmani isolates characterised in this study also showed a high level of intraspecific variation (16S rRNA: 0.3-14.3% and cox1: 2-11.3%) ( Supplementary Tables S3 and S4).
Separate phylogenetic trees for the 16S rRNA and cox1 sequences estimated using the three methods (BI, ML, and NJ) showed identical tree topologies; therefore, only ML trees are presented along with posterior probabilities (pp) and bootstrap support values (bs) for BI, and NJ and ML, respectively (Figure 2A,B). For both 16S rRNA and cox1, the clustering of sequences was similar and showed only minor differences potentially related to the unavailability of 16S rRNA and cox1 reference sequences for I. trichosuri and I. antechini, respectively. The cox1 sequences of I. trichosuri grouped with the previously published sequence (MN686568) of I. trichosuri from Australia with strong nodal support (pp = 1; bs for NJ = 100%; bs for ML = 99%) ( Figure 1B), whereas the corresponding 16S rRNA sequences clustered together as a separate clade with strong nodal support (pp = 1; bs for NJ = 99%; bs for ML = 99%) ( Figure 1A).
tering of sequences was similar and showed only minor differences potentially related to the unavailability of 16S rRNA and cox1 reference sequences for I. trichosuri and I. antechini, respectively. The cox1 sequences of I. trichosuri grouped with the previously published sequence (MN686568) of I. trichosuri from Australia with strong nodal support (pp = 1; bs for NJ = 100%; bs for ML = 99%) ( Figure 1B), whereas the corresponding 16S rRNA sequences clustered together as a separate clade with strong nodal support (pp = 1; bs for NJ = 99%; bs for ML = 99%) ( Figure 1A).

Microfluidic Detection of Tick-Borne Microorganisms
The microfluidic real-time PCR detected the DNA of at least one of the targeted microorganisms in 88.9% of tested ticks (40/45), and a total of seven microorganisms were detected ( Figure 3; Table 2; Supplementary Table S2 (Table 2; Supplementary Table S2). Ixodes tasmani, I. trichosuri, and I. antechini ticks were positive for five (out of seven) detected microorganisms, whereas one and three were found in I. kohlsi and Ap. auruginans ticks, respectively ( Table 2). Ticks collected from mountain brushtail possums, southern brown bandicoots, and agile antechinus had the highest number of microorganisms (n = 5) followed by those from bare-nosed wombats and koalas (n = 3) and little penguins (n = 1) ( Table 2; Supplementary  Table S2). The number of microorganisms detected in a tick species also varied in different host species. For example, I. tasmani ticks were positive for five, three, two, and one microorganisms in specimens collected from southern brown bandicoots, koalas, agile antechinus, and mountain brushtail possums, respectively. Similarly, I. trichosuri were positive for five and two microorganisms in specimens collected from mountain brushtail possums and southern brown bandicoots, respectively (Table 2).   Bartonella sp.

Co-Occurrence of Microorganisms
The DNA of single microorganisms was detected in 13 ticks (I. kohlsi, I. tasmani, and I. trichosuri), 27 ticks tested positive for DNA of multiple microorganisms, and only 5 ticks were found to be negative for DNA of all microorganisms tested herein (Figure 3; Supplementary Table S2). Among ticks positive for multiple microorganisms, most of them carried two (n = 14), three (n = 9), four (n = 2), and five (n = 2) microorganisms. The highest numbers of co-occurring microorganisms were detected in I. trichosuri and I. antechini ticks, which tested positive for five microorganisms (Figure 3; Supplementary Table S2).

Co-Occurrence of Microorganisms
The DNA of single microorganisms was detected in 13 ticks (I. kohlsi, I. tasmani, and I. trichosuri), 27 ticks tested positive for DNA of multiple microorganisms, and only 5 ticks were found to be negative for DNA of all microorganisms tested herein (Figure 3; Supplementary Table S2). Among ticks positive for multiple microorganisms, most of them carried two (n = 14), three (n = 9), four (n = 2), and five (n = 2) microorganisms. The highest numbers of co-occurring microorganisms were detected in I. trichosuri and I. antechini ticks, which tested positive for five microorganisms (Figure 3; Supplementary Table S2).

Genetic Relationship of Rickettsia Species
The BLAST search revealed variable results for different sets of the rickettsial nucleotide sequences determined herein (data not provided). The ompA and ompB sequences showed the highest similarity to those of Candidatus R. antechini (DQ372955; 100%) and Candidatus R. tasmanensis (GQ223393; 98.88%), respectively, that were previously published from Australia. Similarly, the 17 kDa sequence showed the highest similarity to that of R. honei (AF060704; 99.49%) from Australia, and the gltA sequence showed the highest similarity (99.72%) to that of R. conorii subsp. raoultii from China.
The phylogenetic analyses of unique sequences of four different genetic markers (ompA, ompB, gltA, and 17 kDa) of Rickettsia sp. determined herein also produced different topologies (Figure 4). Separate phylogenies based on gltA and ompA sequences showed that the rickettsial sequences determined in this study were grouped with respective sequences of Candidatus R. antechini (DQ372954, DQ372955) with various nodal supports (gltA: pp = 1, bs for NJ = 65%, and bs for ML = 75%; and ompA: pp = 0.97, bs for NJ = 89%, and bs for ML = 99%), respectively ( Figure 4A,B). For the phylogenetic tree based on the 17 kDa gene, the sequences determined herein grouped separately along with R. honei and R. japonica with low nodal support (pp = 0.97, bs for NJ = 43%, and bs for ML = 55%) ( Figure 4C). The genetic relationships of the ompB gene sequences revealed their grouping with those of Candidatus R. tasmanensis (GQ223393) from Australia with low to strong nodal support (pp = 1, bs for NJ = 77%, and bs for ML = 75%) ( Figure 4D).

Genetic Relationship of Rickettsia Species
The BLAST search revealed variable results for different sets of the rickettsial nucleotide sequences determined herein (data not provided). The ompA and ompB sequences showed the highest similarity to those of Candidatus R. antechini (DQ372955; 100%) and Candidatus R. tasmanensis (GQ223393; 98.88%), respectively, that were previously published from Australia. Similarly, the 17 kDa sequence showed the highest similarity to that of R. honei (AF060704; 99.49%) from Australia, and the gltA sequence showed the highest similarity (99.72%) to that of R. conorii subsp. raoultii from China.
The phylogenetic analyses of unique sequences of four different genetic markers (ompA, ompB, gltA, and 17 kDa) of Rickettsia sp. determined herein also produced different topologies (Figure 4). Separate phylogenies based on gltA and ompA sequences showed that the rickettsial sequences determined in this study were grouped with respective sequences of Candidatus R. antechini (DQ372954, DQ372955) with various nodal supports (gltA: pp = 1, bs for NJ = 65%, and bs for ML = 75%; and ompA: pp = 0.97, bs for NJ = 89%, and bs for ML = 99%), respectively ( Figure 4A,B). For the phylogenetic tree based on the 17 kDa gene, the sequences determined herein grouped separately along with R. honei and R. japonica with low nodal support (pp = 0.97, bs for NJ = 43%, and bs for ML = 55%) ( Figure  4C). The genetic relationships of the ompB gene sequences revealed their grouping with those of Candidatus R. tasmanensis (GQ223393) from Australia with low to strong nodal support (pp = 1, bs for NJ = 77%, and bs for ML = 75%) ( Figure 4D).

Discussion
In this study, we utilised, for the first time, a high-throughput microfluidic PCRbased technique to simultaneously detect bacterial, rickettsial, and protozoal microorganisms in ticks collected from six wild animal species in Australia. We found a diverse range of microorganisms, including several that were potentially pathogenic and of zoonotic importance (e.g., species of Rickettsia and Bartonella). The simultaneous detection of DNA of up to seven species of potentially pathogenic microorganisms highlighted the significance of such studies for the surveillance of ticks and TBPs from a One Health perspective. The molecular characterisation of I. tasmani revealed genetic variation among different isolates, which indicated the potential presence of cryptic species. Furthermore, the findings of our study added significantly to the growing literature on the diversity of ticks and tick-borne microorganisms of Australian wildlife.
Given that the scope of this study was limited to DNA-based detection of microorganisms, it did not exclude the possibility of the occurrence of RNA-based microorganisms (such as retroviruses), which have been reported previously in Australian ticks

Discussion
In this study, we utilised, for the first time, a high-throughput microfluidic PCR-based technique to simultaneously detect bacterial, rickettsial, and protozoal microorganisms in ticks collected from six wild animal species in Australia. We found a diverse range of microorganisms, including several that were potentially pathogenic and of zoonotic importance (e.g., species of Rickettsia and Bartonella). The simultaneous detection of DNA of up to seven species of potentially pathogenic microorganisms highlighted the significance of such studies for the surveillance of ticks and TBPs from a One Health perspective. The molecular characterisation of I. tasmani revealed genetic variation among different isolates, which indicated the potential presence of cryptic species. Furthermore, the findings of our study added significantly to the growing literature on the diversity of ticks and tick-borne microorganisms of Australian wildlife.
Given that the scope of this study was limited to DNA-based detection of microorganisms, it did not exclude the possibility of the occurrence of RNA-based microorganisms (such as retroviruses), which have been reported previously in Australian ticks [12,32]. It should also be noted that no corresponding serological testing of wildlife hosts was conducted for the detection of infections. Therefore, the evidence described in this paper did not imply that there was a transmission of these microorganisms between ticks and their hosts. However, these ticks may serve as reservoirs or vectors of infection for wildlife hosts, domestic animals, and human populations.
Microfluidic-based real-time PCR is a powerful screening tool that has the capacity to perform thousands of parallel real-time PCRs simultaneously, thereby automating rotework and reducing the costs of time, labour, and consumables [35,36]. In conventional molecular amplification assays, there are a restricted number of targets that can be chosen at a time, thus limiting the screening of TBPs to targeting only pathogens (primarily those that were previously reported as being transmitted by the ticks in an area) [52]. Moreover, the very small volume of DNA template (1.3 µL) required for testing either 48 or 96 targets allows for a far greater variety of pathogens to be tested in tick DNAs (which is usually very-low-yield DNA) [35]. It is also more useful to comprehensively identify co-occurring microorganisms in ticks in situations where conventional PCR and Sanger sequencing lack capacity [22]. While next-generation sequencing is also high-throughput and exhaustive, it ultimately requires time and labour, larger volumes of templates, more optimisation steps, and complex data analyses [53,54]. The ease of application and subsequent associated analyses of microfluidic technology makes it highly suitable for use in field-based rapid diagnostics and point-of-care testing in hospital settings [55].
The high level of intraspecific sequence variation (pairwise nucleotide differences: 16S rRNA = 0.3-14.1%; cox1: 2-11.3%) and phylogenies based on the 16S rRNA and cox1 sequences of ticks demonstrated that the I. tasmani sequences determined herein either belong to two distinct species or that there are subspecies. These findings concurred with those of previous studies that also reported that I. tasmani might be a cryptic species or that there could be subspecies [40,56]. Further sampling and molecular investigations that employ longer read sequencing such as complete mitochondrial and/or whole genome sequencing would be useful to resolve the unclear phylogeny for this tick species.
It has been previously demonstrated that tick infestations lead to irritation, ill-thrift, and anaemia in wild animals such as koalas [57] and bandicoots [58]. However, little is known about the pathogenicity of the diverse array of microorganisms reported within Australian ticks in this study or in previous studies [12,29]. Certain species of Anaplasma, Babesia, and Ehrlichia have been known to cause a clinical disease characterised by severe anaemia, lethargy, neurologic signs, and death in kangaroos [59,60] and wild canids [61]. These pathogens along with Theileria sp. are also a concern in agricultural losses because they are capable of causing haematological disease and associated production losses in Australian cattle [62][63][64][65]. However, these diseases are primarily limited to bovines inhabiting northern tropical regions of Australia and are usually transmitted by cattle ticks, including Haemaphysalis longicornis [62] and Rhipicephalus microplus [65]. There is a need to test the competency of ticks characterised herein for vector potential, particularly in Victoria where other tick vectors are non-endemic.
The detection of the DNA of Hepatozoon sp. in 2.2% of the ticks tested in the present study underpinned the potential role of this tick in the transmission, prevalence, and consequent spillage of hepatozoonosis to domestic and wild canids. Several Hepatozoon species (e.g., H. canis, H. banethi n. sp., and H. ewingi n. sp.) have been previously reported in ticks and wildlife in Australia [22]. Hepatozoonosis, which is caused by the apicomplexan blood parasite H. canis, was first reported in 1905 in dogs from India and is transmitted by the ingestion of infected tick vectors [66,67]. Until now, autochthonous infections of H. canis have been reported in more than 60 countries across Africa, Americas, Asia, Europe, and the Middle East. Recently, an H. canis infection with a potential tick origin was detected for the first time in a dog in Australia [68]. Our findings and previous reports of H. canis in Australia could be highly significant because following initial reports of autochthonous infections of this pathogen from Germany, Austria, Slovakia, and the Czech Republic, it was later demonstrated that H. canis was actually endemic in these countries [22,68]. Such findings of H. canis in previously non-endemic countries could be attributed to the invasion of infected ticks and canids in these countries as well as the non-existence of TBP surveillance systems [68,69]. Although the DNA of this pathogen has been previously demonstrated in several tick genera including Amblyomma, Dermacentor, Haemaphysalis and Ixodes, its primary vectors are members of Rhipicephalus sanguineus s.l. [68]. Given the availability of tick vectors for this pathogen in Australia, further investigations are needed to determine its level of spread and associated risks.
Coxiella burnetii, the cause of Q fever, is usually carried by various tick species found on macropods, domestic animals, bandicoots, and birds [10]. Infections with C. burnetii in these animals usually remain subclinical [10], but the clinical disease has been reported to cause significant production losses in sheep [64]. This pathogen is also of high zoonotic significance because it may be transmitted to people working closely with animals, such as farmers, abattoir workers, or veterinary staff [70]. Notably, the Coxiella-like sp. DNA detected in 6.7% of ticks tested in the present study potentially belongs to the non-pathogenic endosymbionts [71]. This significant finding highlighted the need for further exploration of the microbiota of Australian ticks because the evolutionary origin of highly pathogenic and infectious C. burnetii from a maternally inherited endosymbiont is still unclear [71]. Additionally, I. tasmani, Ap. Auruginans, and I. trichosuri (species found positive for Coxiella-like sp. DNA) are not known to transmit C. burnetii.
In the present study, 55.6% of the ticks tested positive for the DNA of Rickettsia species. These positive ticks included four species (Ap. auruginans, I. tasmani, I. antechini, and I. trichosuri) collected from five different host species. Previous studies reported various levels of occurrence of Rickettsia species in ticks from different hosts. For example, Izzard et al. [72] and Tadepalli et al. [18] found that 55% and 92% of I. tasmani and Amblyomma albolimbatum ticks were positive for rickettsial DNA, respectively. There are four known rickettsial species (including a novel subspecies) that cause clinical disease in mammals in Australia. These include R. australis (the cause of Queensland tick typhus), R. honei (Flinders Island spotted fever), and R. australis subsp. marmionii (Australian spotted fever) and R. gravesii, which are transmitted by several tick species [21]. The Blast analyses of the nucleotide sequences of multiple metabarcodes of the rickettsial isolates characterised herein did not provide conclusive evidence about their identity. Likewise, the phylogenies based on nucleotide sequences of four markers showed a close association of the rickettsial isolates determined herein with the spotted fever group's rickettsial species (Candidatus R. tasmanensis and Candidatus R. antechini). These results suggested that the rickettsial isolates identified in this study might belong to novel species or subspecies of unknown pathogenic and/or zoonotic potential. However, these findings should be further investigated through comparative molecular typing, isolation, and whole-genome characterisation.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/pathogens12020153/s1, Table S1: Details of target genetic markers used for testing microorganisms in tick DNA; Table S2: Host and location details of ticks tested and the real-time PCR cycle threshold (ct) values for microorganisms detected in this study; Table S3: Pairwise percentage nucleotide similarities of 16S ribosomal RNA sequences of Ixodes tasmani determined in this study; Table S4: Pairwise percentage nucleotide similarities of cox1 sequences of Ixodes tasmani determined in this study.