Tick-Borne bacterial and protozoan animal pathogens shape the native microbiome within Hyalomma anatolicum anatolicum and Rhipicephalus microplus tick vectors

Background Ticks vector a variety of bacterial, viral, and protozoan pathogens of public and animal health significance. Ticks also harbor a diverse community of microbes linked with their biological processes like hematophagy and hence vector competence. The interactions between bacterial and/or protozoan pathogens and their tick vector microbiome are yet to be investigated. In lieu of this, this study was designed to define the microbial composition of uninfected and infected Hyalomma (H.) anatolicum anatolicum and Rhipicephalus (R.) microplus tick species. Methodology/Principal findings A total of 320 H. anatolicum and R. microplus were screened for the presence of the protozoan (Theileria sp.), and bacterial (Anaplasma marginale) pathogens by PCR. Subsequently, the microbiome of uninfected and infected individual H. anatolicum and R. microplus were analyzed. The highly conserved V1-V3 region of the 16S rRNA gene was sequenced using the MiSeq Illumina platform. The microbiome of female H. anatolicum anatolicum ticks was dominated by the endosymbiont Candidatus Midichloria mitochondrii (CMM) and Francisella-like endosymbiont (FLE) which were not affected by pathogen infection. Ehrlichia species was detected in A. marginale-infected male H. anatolicum anatolicum (6.2%) as opposed to the Theileria sp.-infected female H. anatolicum anatolicum. Coxiella sp. was also detected in uninfected (2.96%) and A. marginale-infected (4.25%), but not in Theileria sp.-infected R. microplus ticks. Analysis of the eukaryote composition in the respectively ticks also revealed the presence of operational taxonomic units (OTUs) belonging to Plasmodium (P.) falciparum in Theileria sp.-infected H. a. anatolicum and R. microplus ticks, while Hepatozoon americanum detected from Theileria sp.-infected and uninfected H. a. anatolicum. Conclusion and Significance This study establishes the extent of the diversity of microbial community of two important tick species from Pakistan and also revealed the presence of Theileria and A. marginale and additional pathogenic bacteria that could be of public health significance. We hypothesized that infection with either a protozoan or bacterial pathogen will alter the microbial composition within these tick species. Interestingly, we reported the detection of the malarial parasite (P. falciparum) from ticks infected with the protozoan pathogen (Theileria sp.). Further validation experiments are required on endosymbionts and pathogens of ticks to investigate how they could be important in the epidemiology of human and animal pathogens.


Introduction
Ticks are known obligate, blood feeding ectoparasites of vertebrate animals that depend on the host's blood to carry out nutritive, reproductive and other physiologic functions. In addition to causing significant blood loss, they are also known to transmit infectious pathogens such as viruses, bacteria, protozoa, and fungi to their animal or human hosts during the feeding process. Recent studies have demonstrated ticks harbor microbial communities, members of which undergoes mutualistic and commensal relationships with their tick hosts, most of which have been commonly overlooked or considered as potential tick-borne pathogens (1)(2)(3)(4)(5). Therefore, it is posited that a constant wave of interaction will continually ensue between tick-borne pathogens (TBPs) and bacterial endosymbionts. TBPs and bacterial endosymbionts could directly compete for nutrient or niche within the tick hosts. Indirect inhibition or elimination could also occur through excretory molecules inhibiting growth of a competitor (6). With this knowledge comes the understanding that the microbial communities including TBPs must continuously coevolve to maintain equilibrium within the tick vectors.
The most reported tick-borne disease in Pakistan includes Theileriosis, Babesiosis and Anaplasmosis caused by Theileria (T.) annulata, Babesia (B.) bigemina, B. divergens, Anaplasma (A.) marginale and A. centrale, which have all been found to affect water buffaloes and cattle (7)(8)54). The maintenance of these diseases has been made possible due to the availability and endemicity of suitable tick vectors belonging to the Ixodid group (9,55) (9).
While recent studies have been centered on TBPs, little attention has been paid to the entire microbial community, with only one such study determining the microbiome and pathogen diversity in ticks from Pakistan (10). These non-pathogenic microorganisms were either overlooked or considered to be potential TBPs (11). However, these nonpathogenic bacterial communities could confer multiple beneficial or detrimental effects on the tick host, interfere with the basic reproductive and fitness functions, and also alter the dynamics of TBP transmission (11).
A number of studies on the epidemiology and distribution of tick-borne pathogens have been published in Pakistan (7,(9)(10)(17)(18), there is a dearth of information regarding the microbiome of these ticks and how they interact with these pathogens within the same tick host. Here, we report the microbial communities residing within the uninfected, Theileria species and Anaplasma marginale-infected ticks.

Pathogen prevalence
A total of 320 hard ticks pulled from cattle, sheep, and goat comprising of 198 H. anatolicum anatolicum and 122 R. microplus. All ticks were screened for the presence of Theileria sp. and A. marginale by PCR amplification using primers specific for the 18S rRNA and 16S rRNA gene respectively. Of the 320 samples, 23 (7%) and 90 (28%) were A. marginale and Theileria sp. positive respectively.

Microbiome composition
Analysis of the demultiplexed paired-end-reads generated 2902862 reads which ranged from 17063 to 123731 with an average of 65609 reads. Sequences from samples of R.

Hyalomma anatolicum anatolicum
The phylum Proteobacteria was the most abundant in both Theileria sp.-infected and uninfected ticks (~70%). The phylum Firmicutes had much more abundance within the Theileria-infected ticks representing about 20% of the total OTUs in the infected ticks. Other OTUs found to be present in relative amounts include phyla Actinobacteria, Tenericutes, and Bacteroidetes. OTUs with relative abundance below 1% were grouped into others. Similarly, the microbiome of A. marginale-infected male H. anatolicum anatolicum was dominated by the phylum Proteobacteria (~48%) followed by the phyla Actinobacteria (24.5%), Firmicutes (18.9%), Bacteroidetes (4.7%) and Cyanobacteria (1.4%). In contrast, the representative phyla in uninfected male H. anatolicum anatolicum were evenly distributed among the phyla Proteobacteria, Actinobacteria and Firmicutes, while phyla Bacteroidetes and Cyanobacteria had the least abundance (Fig. 1A).

Rhipicephalus microplus
The taxonomic abundance of microbial composition in uninfected Rhipicephalus ticks at the phylum level was represented by the phylum Proteobacteria (

Genetic relationship of selected Coxiella, Anaplasma, and Ehrlichia sequences
The two Coxiella sequences from this study uniquely grouped with a Coxiella burnetii sequence (GenBank: NR 104916.1)

Beta Diversity
Principal coordinate analysis (PCoA) of the distance matrixes showed clustered separation between Hyalomma a. anatolicum and R. microplus ticks (Fig 8A and 8B).
Distinct clustering was also observed on the PCoA plot of infected ticks using the unweighted and weighted distance matrices (Fig 9A and 9B). Additional information can be seen on the supplementary data (S1; Table 2 and 3)

Discussion
To the best of our knowledge, this study is the first to evaluate the microbiome composition of pathogen-infected and uninfected H. anatolicum anatolicum and R.  (13)(14)(15)(16), few of those only investigated the microbial patterns in infected and uninfected ticks (16).
In this study, the prevalence of Theileria species as seen in previous studies, was higher in H. anatolicum anatolicum ticks, while A. marginale had similar prevalence in both tick species (7,9,17,18) (Table 1). With the exception of the R. microplus ticks, microbiome composition at the phylum taxonomic level was predominantly dominated by the phylum Proteobacteria in all tick groups tested in this study (Fig 1A and 2A). The ability of two or more pathogen to co-infect male ticks with subsequent transmission to the host still needs to be investigated, the genera Anaplasma and Ehrlichia, both of which are Anaplasmataceae and obligate intracellular pathogens could both occur synergistically thus favoring co-existence within the same tick vector. Another source of the coinfection could be from the blood of the host animal the ticks were feeding on as at the time of collection.
We also revealed the Francisella genus in uninfected male H. anatolicum anatolicum ticks. This is possibly an endosymbiont that has also been previously reported in other tick species and has been shown to have parts of its genome that encodes for biosynthesis of B vitamins (24), hence helping the tick in compensating for the nutrient poor blood meal. Since male ticks rarely spend lesser time blood-feeding when compared to their female counterparts, in the presence of Francisella, it could be hypothesized that male ticks also utilize the B vitamins for reproductive development such as reproductive hormone production and nutrition of the sperm cells. Since male ticks have also been shown to be early arrivers for a blood meal in order to release pheromones that will subsequently attract females, this process could place a lot of physiological stressors (25, 26)  The endosymbionts CMM and FLE were both detected in uninfected and Theileriainfected female H. anatolicum anatolicum at relatively unchanged abundances ( Fig. 2A).
Although direct competition for nutrient and niche between endosymbionts and pathogen has been reported to take place in ticks (6), our finding did not support this as the abundances of these endosymbionts were not altered in the presence or absence of While alpha diversity analysis of uninfected H. anatolicum anatolicum and R. microplus ticks suggested no significant differences in the bacterial richness ( Fig. 6A and B), cluster analysis revealed that the two ticks have unrelated microbial composition based on the number of OTUs and the phylogenetic distances of individual bacterial species (Fig 8A and B). This finding was further supported by the differences in the bacterial composition at the genus level (Fig 2C and 3C). This difference in the microbial composition could be as a result of their feeding habit, Hyalomma ticks are known to be two/three-host ticks while Rhipicephalus ticks are generally a one-host tick.
Unsurprisingly, significant differences were seen in the alpha diversity analyses of pathogen infected ticks. Theileria-infected R. microplus ticks were the least diverse with a lesser amount of OTUs as well as a relatively low Faith_pd value (7A and 7B). This further supported our previous pathogen-associated dysbiosis that was seen in Theileria-infected R. microplus ticks (Fig 3D).
Other bacteria species found to be abundant in  (40)(41). In the United States, it is vectored by the Amblyomma maculatum nymphs or adults (39).
Pakistan is a malaria endemic country and we observed an increase in the relative abundance of P. falciparum in Theileria-infected H. anatolicum anatolicum when compared to the uninfected ticks (Fig. 5A). Similar observation was made in Theileriainfected R. microplus ticks (Fig 8A). The detection of P. falciparum the protozoan parasite, a causative agent of the human malaria, in ticks was an unaccepted finding in this study. The detection of P. falciparum in ticks from this study could have risen from the ticks accidentally feeding on an infected human host. While this is a possibility in a multi-host ticks as seen in Hyalomma species, this is highly an unlikely occurrence in R.
microplus which is a one-host tick.  marginale and additional pathogenic bacteria that could be of public health significance.
We hypothesized that infection with either a protozoan or bacteria pathogen will alter the microbial composition within these tick specie. Limitation faced during this study was the difficulty in getting tissue samples of ticks as they were field collected. Future tick developmental and tissue-specific studies warrant new insights in specific interactions between tick-borne pathogens and their associated microbiome.

Tick species identification
Tick identification was performed by an expert taxonomist (Dmitry A. Apanaskevich) at the United States National Tick Collection (USNTC) according to the criteria used in previously published reports (42)(43)(44). All stages were examined on an Olympus SZX16 stereoscopic microscope.

Genomic DNA Extraction
Prior to DNA extraction, individual ticks were surface sterilized in a series of steps.
Briefly, a 10% solution of sodium hypochlorite was used to clean the individual surfaces of ticks followed by rinsing in 70% ethanol. A final cleaning was done using sterile water. Genomic DNA was extracted from each individual tick homogenate using a DNeasy blood and tissue kit (Qiagen, Valencia, CA, USA) following the manufacturer's protocol. The concentrations of the extracted genomic DNA samples were quantified using a Nanodrop ND-100 instrument and DNA stored in -20 o C till further needed.

Detection of Theileria species
To detect Theileria species from the collected ticks, forward and reverse primers specific for the Theileria genus 18S rRNA gene were used in a 25

Detection of Anaplasma marginale
The detection of Anaplasma marginale was carried out as previously described (45).

Phylogenetic Analyses
The Anaplasma, Ehrlichia and Coxiella partial 16S rRNA sequences were obtained and genetic relationships was compared with similar sequences from NCBI. Briefly, downloaded reference sequences were downloaded, aligned using the ClustalW and phylogenetic tree was constructed (S1, Fig. 1) The evolutionary history was inferred using the Neighbor-Joining method (56). The bootstrap consensus tree inferred from 1000 replicates (57)

Data Analyses
Sequence analysis was done using the Quantitative Insights into Microbial Ecology (QIIME 2) (46), unless stated otherwise. Briefly, processing of raw FASTQ files were demultiplexed. The Atacama soil microbiome pipeline was incorporated for quality control of demultiplexed paired-end reads using the DADA2 plugin as previously described (47). Sequence alignment and subsequent construction of phylogenetic tree from representative sequences was done using the MAFFT v7 and FasTree v2.1 plugin (48). Operational taxonomic assignment was done using the qiime2 feature-classifier plugin v7.0 (49) which was previously trained against the SILVA 132 database preclustered at 99% (50).
Diversity analysis was done by rarifying individual sequences to a depth of 1000 to get adequate coverage of all samples analyzed and samples with insufficient reads subsequently screened out (Fig. 5). Diversity analysis was only estimated for female tick samples as the number of male ticks were small to give a confident result. Faith phylogenetic distance (Faith_pd) and the number of observed OTUs were used in assessing alpha diversity, while beta diversity was estimated using PERMANOVA analysis of unweighted and weighted UniFrac distance matrix. Raw data from this analysis were submitted deposited and assigned the GenBank BioProject number PRJNA600935.