High-Throughput Microfluidic Real-Time PCR for the Detection of Multiple Microorganisms in Ixodid Cattle Ticks in Northeast Algeria

Ixodid ticks are hematophagous arthropods considered to be prominent ectoparasite vectors that have a negative impact on cattle, either through direct injury or via the transmission of several pathogens. In this study, we investigated the molecular infection rates of numerous tick-borne pathogens in ticks sampled on cattle from the Kabylia region, northeastern Algeria, using a high-throughput microfluidic real-time PCR system. A total of 235 ticks belonging to seven species of the genera Rhipicephalus, Hyalomma, and Ixodes were sampled on cattle and then screened for the presence of 36 different species of bacteria and protozoans. The most prevalent tick-borne microorganisms were Rickettsia spp. at 79.1%, followed by Francisella-like endosymbionts (62.9%), Theileria spp. (17.8%), Anaplasma spp. (14.4%), Bartonella spp. (6.8%), Borrelia spp. (6.8%), and Babesia spp. (2.5%). Among the 80.4% of ticks bearing microorganisms, 20%, 36.6%, 21.7%, and 2.1% were positive for one, two, three, and four different microorganisms, respectively. Rickettsia aeschlimannii was detected in Hyalomma marginatum, Hyalomma detritum, and Rhipicephalus bursa ticks. Rickettsia massiliae was found in Rhipicephalus sanguineus, and Rickettsia monacensis and Rickettsia helvetica were detected in Ixodes ricinus. Anaplasma marginale was found in all identified tick genera, but Anaplasma centrale was detected exclusively in Rhipicephalus spp. ticks. The DNA of Borrelia spp. and Bartonella spp. was identified in several tick species. Theileria orientalis was found in R. bursa, R. sanguineus, H. detritum, H. marginatum, and I. ricinus and Babesia bigemina was found in Rhipicephalus annulatus and R. sanguineus. Our study highlights the importance of tick-borne pathogens in cattle in Algeria.


Introduction
Ixodid ticks are blood-sucking arthropods, and are considered to be prominent vectors of pathogens for humans as well as domestic and wild animals. They are known to transmit a wide variety of causative agents such as bacteria, protozoa, and viruses that may subsequently infect the mammal host. Globally, the incidence of tick-borne diseases is growing, mostly due to increased interactions between pathogens, vectors, and hosts [1,2]. Furthermore, climate change, including the prolongation of seasons, global warming, and changing precipitation patterns, extends the geographic range of a number of tick species and the pathogens they carry [3].
Ticks are regarded as the primary vectors of pathogens affecting livestock [4] and up to 80% of cattle worldwide are at risk of coming into contact with ticks and contracting diseases caused by transmitted tick-borne pathogens (TBPs) [5]. An individual animal can be infested with hundreds or even thousands of ticks, which clearly magnifies their effect on the host, either by direct injury or by the transmission of pathogens [4]. In addition, the co-transmission of several pathogens may lead to co-infection in animals, which aggravates their vital prognosis or, in some cases, gives rise to atypical forms, thereby complicating diagnosis [6]. Thus, it is very important to detect and identify TBPs, so that veterinarians can predict the risk of infection and subsequently implement appropriate control measures.
Hard ticks belonging to the genera Hyalomma, Rhipicephalus, Ixodes, and Haemaphysalis have been identified feeding on grazing cattle in Algeria [7]. In addition, various studies using classical molecular techniques have detected multiple pathogens in these ticks, e.g., the genera Rickettsia, Anaplasma, Coxiella, and Theileria, in cattle ticks from Algeria [8][9][10][11]. All of these studies used classical molecular methods that can only detect a few pathogens simultaneously, are time-consuming, and require large volumes of DNA for the detection of multiple pathogens. Moreover, evidence of non-pathogenic commensal microorganisms called endosymbionts is poorly documented in Algeria but provides useful information because they may influence the transmission of other tick-borne microorganisms or become pathogenic for humans and/or animals [6,12,13].
To do so, microfluidic-based high-throughput PCR systems have been described by various studies as the most sensitive approach to detect TBPs [6,14,15]. These systems allow the rapid and simultaneous detection of numerous microorganisms using a small volume of DNA, thereby making it possible to carry out large-scale epidemiological investigations on TBPs in ticks [14]. Here, we successfully employed this approach to investigate the distribution of TBPs in bovine ticks from the Kabylia region of Algeria for the first time.

Taxonomical Identification of Collected Tick Species
A total of 518 male and 537 female hard ticks were manually detached from bovines. The females varied in size due to different feeding durations. We did not find any immature tick stages in the collected samples. Three common tick genera were identified: Rhipicephalus (646/1055, 61.2%), Hyalomma (390/1055, 36.9%), and Ixodes (19/1055, 1.8%). Among these genera, seven different hard tick species were recognized. R. bursa and H. detritum were the two most common species, followed by H. marginatum and H. lusitanicum and finally I. ricinus, R. sanguineus, and R. annulatus. For morphologically deformed Hyalomma and Rhipicephalus ticks, specimens were determined to the genus level only. The difference in frequencies among identified species is described in Figure 1. In addition, co-infestation with different species of ticks was observed on the sampled bovines.

Infection Rates of Microorganisms and Their Co-Infection Rates in Ticks
Among all investigated ticks (

Infection Rates of Microorganisms and Their Co-Infection Rates in Ticks
Among all investigated ticks (Table 1) Anaplasma marginale was the most prevalent species of the genus Anaplasma (16/235, 6.80%), followed by A. centrale detected in 0.4% (1/235) of samples. The remaining 50% (17/34) of samples were confirmed by nested PCR, with the sequencing of nine positive samples revealing identity with unidentified Anaplasma spp. BLAST searches using the 16S rRNA gene sequence showed 99.6% identity with uncultured Anaplasma sp. clone AMCRO1 (Accession no. MN187218.1) in five samples, 98% with uncultured Anaplasma sp. clone AR2-1 (Accession no. MH250195.1) in two samples and 98% with uncultured Anaplasma sp. Oriente CuBov140 clone (Accession no. MK804764.1) in two samples. A. marginale was found in all three identified tick genera (Rhipicephalus, Hyalomma, and Ixodes), and A. centrale was detected only in Rhipicephalus ( Table 1).
The genus Borrelia was detected in 6.8% (16/235) of the investigated ticks. None of the eight species-specific primer/probe sets used for high-throughput microfluidic PCR gave a positive signal; therefore, these samples were confirmed by nested PCR followed by sequencing. All PCR-positive samples (16/235) were confirmed by nested PCR: 10/16 were positive on gel and 4/10 were sequenced, from which only two sequences were obtained. The BLAST analysis on the fla gene sequence showed 92.37% identity with an unidentified Borrelia species (Accession no. KR677091.1).
Pathogens belonging to the genus Bartonella were detected in 6.8% (16/235) of the sampled ticks using the high-throughput microfluidic PCR system. Species identification was attempted on all positive samples with conventional PCR target ftsZ gene, but no results were obtained by sequencing.
DNA of Babesia bigemina was found in 6/235 (2.55%). The positive specimens of B. bigemina detected in R. annulatus and R. sanguineus ticks were confirmed by nested PCR and sequencing with an identity of 97.32% (Accession no. MH257721.1).

Discussion
In this study, we identified three tick genera: Rhipicephalus, Hyalomma, and Ixodes. Among these, the thermophilic species R. bursa, H. detritum, and H. marginatum predominated, and the mesophilic species I. ricinus was less abundant (Figure 1). These results corroborate the previous reports from northern Algeria [7,11,16]. In future studies, a combination of morphological and molecular identification of ticks should be performed to identify ticks at the genus level. In North Africa, TBP detection is usually carried out using classic methods, such as PCR or real-time PCRs, which are based on the use of specific primers and/or probes [11,17,18]. This approach is limited by the characterization of a single pathogen. However, recent studies have described the importance of co-infections in the transmission of pathogens and the expression of disease severity. In this study, a new approach, based on high-throughput microfluidic technology, was used to detect 36 different microorganisms (pathogens and symbionts), and to monitor TBP circulation in hard ticks infesting cattle in northeastern Algeria.
It is important to notice that some of the ticks collected on bovines were engorged. Therefore, we cannot conclude that these ticks are vectors for the detected pathogens. The latter may have become infected with the microorganism while feeding on previously infected animals, and/or through co-feeding. Moreover, detection of DNA does not indicate that the pathogen is alive; it simply corresponds to potentially inert traces of the microorganism in the engorged tick. Here, we identified four different genera of bacteria (Rickettsia, Anaplasma, Borrelia, Bartonella) and two genera of protozoans (Theileria and Babesia) with an overall frequency of 80.4% of ticks infected with at least one of these TBPs. For example, Rickettsia spp. had the highest infection rate and four different species (i.e., R. aeschlimannii, R. massiliae, R. monacensis, and R. helvetica) were identified. Belonging to the pathogenic spotted fever group, R. aeschlimannii and R. massiliae cause infections in animals and humans worldwide [19]. The presence of R. aeschlimannii in H. marginatum ticks confirmed previous studies on the association of this vector with this bacteria species [19,20]. Surprisingly, we detected R. aeschlimannii DNA in H. detritum and R. bursa ticks also (Table 1). This species was first isolated from H. marginatum in Morocco and then in other African and Mediterranean countries [10,15,19,[21][22][23]. Ticks of the genus Hyalomma have been reported as vectors of R. aeschlimannii, including H. marginatum, H. marginatum rufipes, H. aegyptium, and H. truncatum [19,23]. In addition, this bacteria species has been found in R. appendiculatus in South Africa and Haemaphysalis ticks in Spain [24], and R. turanicus in Greece and China [25,26]. Nevertheless, Rickettsia spp. is considered as an endosymbiont of Hyalomma ticks, and several other tick genera. R. aeschlimannii has been identified in a large percentage of Hyalomma ticks, with unknown clinical relevance [27]. According to this finding and our results, we suggest that tick species other than Hyalomma can also be a suitable carrier for this bacterial species.
In addition, we detected R. massiliae in R. sanguineus. This tick has been described as a vector of this species in the Mediterranean region [8], but we also amplified the DNA of this species for the first time in Hyalomma ticks from Algeria. A similar study in Pakistan also using the microfluidic technique amplified R. massiliae in Hyalomma hussaini and H. anatolicum ticks [28]. The implication of Hyalomma species in the transmission cycle of this pathogen needs to be clarified. In our study, I. ricinus harbored R. monacensis and R. helvetica that were not detected in the other tick species examined (Table 1). This confirms the previous results observed in North Africa as well as Europe [19,20,[29][30][31][32].
Regarding the protozoan Theileria and Babesia species, we detected T. orientalis and B. bigemina. Although the primary vectors of T. orientalis are Haemaphysalis spp. ticks [33], in our study, R. bursa, H. detritum, H. marginatum, and I. ricinus ticks were found to be positive for this pathogen. The association of this pathogen with R. bursa and R. annulatus ticks have also been confirmed in Romania and Algeria, respectively [11,34]. Moreover, a strain closely related to T. buffeli has been detected in R. sanguineus, R. bursa, R. annulatus, H. marginatum, Dermacentor marginatum, and Haemaphysalis punctata ticks in Sardinia, Italy [35]. Taking these results together, the transmission of this TBP does not appear to be limited exclusively to Haemaphysalis spp., suggesting that other tick species may be involved in the transmission cycle worldwide. In the present study, we confirmed the presence of B. bigemina in Algerian R. annulatus, which is its principal vector [36]. Furthermore, we also found this protozoan species in R. sanguineus, lending support to previous reports from Sardinia and Iran [35,37]. The biological transmission of A. marginale involves at least 20 species of ticks mainly of the genera Dermacentor and Rhipicephalus [38]. Here, we report the presence of A. marginale in R. bursa, H. detritum, and I. ricinus. Similarly, A. marginale has been reported in R. bursa in Corsica and Portugal [39,40], and in I. ricinus in Hungary [41]. A. centrale, which is transmitted by Rhipicephalus simus [38], was confirmed individually on one Rhipicephalus sp. tick. Bartonella spp. was identified in multiple species of the tick genera Hyalomma and Rhipicephalus. There is little information on Bartonella transmission in cattle, but based on our recent study reporting B. bovis in ticks from Algeria, these ticks may play a critical role in the transmission of this pathogen in this area [42]. We also amplified DNA from Borrelia sp. in 16 specimens of different tick species. Usually, ticks from the genus Ixodes are the vectors of the zoonotic bacteria species B. burgdorferi s.l. from the Lyme disease group, whereas B. theileri, from the relapsing fever group that causes bovine borreliosis, is transmitted by Rhipicephalus ticks [43,44]. In previous research conducted in Algeria, DNA from these bacteria has been amplified in different tick genera in Algeria using quantitative PCR [31].
Coxiella, Francisella, and Rickettsia are the three major endosymbionts reported in ticks [6,45]. In addition to TBPs, we reported for the first time in Algeria a high rate of infection with Francisella-like endosymbionts in all tick species tested except R. annulatus. Furthermore, 60% of investigated ticks harbored unidentified Rickettsia spp., whereas all the specimens were negative for Coxiella-like endosymbionts. It has been reported that symbionts previously considered non-pathogenic may turn out to be pathogenic, as demonstrated for R. helvetica, R. slovaca, and R. monacensis [20,46]. However, the reasons that make one bacterial species become pathogenic while others remain non-pathogenic are still unclear [45]. In future studies, phylogenetic analysis targeting several genes for pathogenic and non-pathogenic microorganisms will allow us to better answer this question. Our study showed co-infection of cattle ticks with a large variety of pathogenic and non-pathogenic microorganisms. High co-infection rates of TBPs in livestock ticks has been reported in other countries [15,23,28,47]. Co-infections represent a significant risk of the cumulative effect of pathogen transfer and subsequent development of the associated diseases [48]. In addition, co-infections can cause severe complications in the treatment of tick-borne illnesses. Therefore, the study of the associations of multiple microorganisms within the same tick is of high importance, and can help better identify potential clinical co-infections to improve the epidemiological knowledge and control of TBPs [6].

Ethical Statement
Sample collection for this study was authorized by the National Veterinary School of Algiers, Algeria and by the Veterinary Services Department of the Tizi-Ouzou province, Algeria. All bovines were sampled according to Algerian regulations.

Tick Collection and Morphological Identification
A total of 1055 ticks were randomly collected from 112 bovines (with an average of nine ticks/bovine) between May 2015 and November 2017 in eight locations in the Kabylia region located in northeastern Algeria. The samples collected from each individual bovine were stored in 70% ethanol. Tick species (or genera) were determined using taxonomic keys developed by Walker et al. [49] and based on morphological characteristics observed under a stereomicroscope.

DNA Extraction
A random selection of one to three ticks (males and females) from each individual bovine (235 total ticks) were used for genomic DNA extraction. Prior to extraction, ticks were washed three times in sterile distilled water, dried and crushed individually using a sterile scalpel. DNA was then extracted from whole ticks using the NucleoSpin ® Tissue DNA extraction kit (Macherey-Nagel, Düren, Germany) following the manufacturer's instructions and stored at −20°C until use.

DNA Pre-Amplification
The Perfecta PreAmp SuperMix (Quanta Biosciences, Beverly, Massachusetts, USA) was used for DNA pre-amplification according to the manufacturer's instructions. First, all primers pair targeting TBPs were pooled, combining equal volumes with a final concentration of 0.2 µM each.
Then, reactions were carried out in a final volume of 5 µL containing 1 µL of 5× Perfecta Preamp, 1.25 µL of the pooled primer mixture, 1.5 µL of distilled water and 1.25 µL of tick DNA.
The PCR run conditions consisted of a first cycle of 95°C (2 min), followed by 14 cycles of amplification at 95°C (10 s) and 60°C (3 min). The pre-amplified products were diluted in ultra-pure water at 1:10 and kept at −20°C until use.

High-Throughput Microfluidic Real-Time PCR
The BioMark™ real-time PCR system (Fluidigm, San Francisco, USA) was used for the high-throughput microfluidic system, which can handle 48 real-time PCR reactions simultaneously in one single chip [6,14]. Real-time PCR reactions were performed using 6-carboxyfluorescein (FAM)-and black hole quencher (BHQ1)-labeled TaqMan probes with TaqMan Gene expression master mix in accordance with the manufacturer's instructions (Applied Biosystems, France). Amplification consisted of 2 min at 50°C, 10 min at 95°C, followed by 40 cycles of two-step amplification of 15 s at 95°C, and 1 min at 60°C.
We carried out the high-throughput microfluidic real-time PCR to screen the bacterial and parasitic species known to circulate in ticks. We thus simultaneously targeted 36 different microorganisms belonging to 10 genera (the list of pathogens is shown in Table 2 and the list of each primer set and probes used is given in Table 3). Moreover, two primer/probe sets targeting I. ricinus and R. sanguineus tick species were used as a positive control of tick species identification and one primer/probe set targeting tick species was used to control DNA extractions (primers/probes for other tick species were not available at that time). One negative control (ultra-pure water) and one positive control DNA of Escherichia coli were included in each chip. The results were acquired on the BioMarkTM real-time PCR system and analyzed using the Fluidigm real-time PCR analysis software to obtain crossing point (Cp) values.  Table 3. List of primers and probes used in this study for microfluidic real-time PCR [6,14,47].

Standard/Nested PCR and Sequencing
Samples were considered positive for a given microorganism if the Cp value was <30, and if they were positive for a given pathogen species and its corresponding genus and negative for all other species belonging to the same genus. Other positive samples were confirmed using nested PCR or conventional PCR with primers targeting genes or regions different from those of the BioMark™ system (for the primers used, see Table 4). The PCR products were sequenced by Eurofins Genomics (https://Cochin.eurofins.com (accessed on 1 October 2020)) then assembled using BioEdit software (Ibis Biosciences, Carlsbad, CA, USA). Our results were compared in online BLAST (http://www.ncbi.nlm.nih.gov/blast (accessed on 1 October 2020)). searches against sequences publicly available in GenBank (https://www.ncbi.nlm.nih.gov/ (accessed on 1 October 2020)).

Conclusions
We used high-throughput microfluidic real-time PCR to detect TBPs in cattle ticks in Algeria. We confirmed the presence of several bacteria and protozoan species in tickinfested cattle with a fast and highly sensitive molecular method. We detected pathogenic and non-pathogenic (e.g., Francisella-like endosymbionts) microorganisms in ticks feeding on cattle, with a high frequency of co-infections. Further studies on endosymbionts and their possible interactions with pathogens transmitted by ticks are now needed, particularly with regard to the highly predominant ticks Rhipicephalus and Hyalomma in Algeria. Finally, our results highlight the possible involvement of tick species other than those typically reported in the transmission of some pathogens of interest in Algeria; these atypical associations deserve to be investigated further.