Genotypic Analysis of Piroplasms and Associated Pathogens from Ticks Infesting Cattle in Korea

Tick-borne pathogens (TBPs) impose an important limitation to livestock production worldwide, especially in subtropical and tropical areas. Earlier studies in Korea have examined TBPs residing in ticks and animals; however, information on multiple TBPs in ticks infesting cattle is lacking. This study assessed the prevalence of TBPs in ticks parasitizing cattle. A total of 576 ticks, including 340 adults and 236 nymphs, were collected from cattle in Korea between 2014 and 2018. All ticks collected were identified as Haemaphysalis longicornis based on their morphological and molecular characteristics. Among piroplasms and other tick-associated pathogens, seven TBP genes, namely Theileria orientalis (5.0%), Anaplasma bovis (2.3%), Anaplasma capra (4.7%), Anaplasma phagocytophilum-like Anaplasma spp. (APL) clades A (1.9%) and B (0.5%), Ehrlichia canis (1.6%), and Candidatus Rickettsia longicornii (17.5%), were detected. Bartonella spp. and severe fever with thrombocytopenia syndrome virus were not found. To our knowledge, this is the first study to report the presence of the pathogens T. orientalis major piroplasm surface protein genotypes 3 and 7, A. capra, and APL in ticks from Korea. Cattle ticks may be maintenance hosts for many TBPs, and veterinary and medical clinicians should be aware of their high probability of infection and clinical complexity in humans.


Introduction
Many emergent tick-borne pathogens (TBPs) were circulating in animals and ticks long before their identification as causes of clinical diseases [1]. The global hazard of TBPs is increasing and fostering public health concerns, as novel pathogens have been continuously recognized during the past two decades [2]. The capability of ticks to attach to humans and transmit pathogens is affected by several factors, including human activities, geographical and climatic conditions, and tick abundance, biological stage, burden, and attachment duration [3].
TBPs impose an important limitation to livestock production across the world, especially in subtropical and tropical areas [4]. Much of the world's cattle population is influenced by ticks and TBPs, and the damage caused by them is extremely high [5]. Due to the significance of hard ticks (Acari: Ixodidae) in veterinary medicine and the financial cost of their control, the transmission of tick-borne diseases remains an issue for the cattle industry in subtropical and tropical regions, and it is a primary concern for numerous countries in these areas [6].

Tick Collection and Species Identification
In total, 576 ticks, including 236 nymphs and 340 adults, were collected from cattle in the northern (81 from Gyeonggi and 82 from Gangwon), central (35 from Chungbuk, 45 from Chungnam, and 77 from Gyeongbuk), and southern (75 from Jeonbuk, 88 from Jeonnam, and 93 from Gyeongnam) areas of Korea from 2014 to 2018. Four to fifteen ticks for each animal were collected by a simple random sampling method from 149 cattle (Hanwoo, native Korean cattle, Bos taurus coreanae) and then stored in tubes containing 70% ethanol. The collected ticks were primarily identified by their morphological features [15], with an additional classification by the molecular methods described below.

Molecular Detection of Ticks and TBPs
Genomic DNA was extracted from the ticks using a commercial DNeasy Blood & Tissue Kit (Qiagen, Melbourne, Australia), according to the manufacturer's instructions. The AccuPower HotStart PCR Premix Kit (Bioneer, Daejeon, Korea) was employed for the PCR amplification. Molecular identification of tick species was performed by amplifying the sequence of the mitochondrial cytochrome c oxidase subunit I (COI) gene using specific primers as described previously [16].
The ticks were then screened for several TBPs using primer sets specific to each pathogen. To detect piroplasm 18S rRNA, samples were first examined for infection by piroplasms by PCR using a commercial AccuPower Babesia and Theileria PCR Kit (Bioneer). Positive samples were then re-amplified by PCR using primers designed from the common sequence of the 18S rRNA gene of numerous piroplasm species [17], and major piroplasm surface protein (MPSP) genes of Theileria species were amplified by PCR [18]. Infection with rickettsiae was primarily tested by PCR via a commercial AccuPower Rickettsiales 3-Plex PCR Kit (Bioneer) for the detection of rickettsiae 16S rRNA. Additionally, positive samples were submitted to amplification for species identification. Positive samples of Anaplasma spp. were confirmed by amplifying 16S rRNA fragments using nested PCR (nPCR) [12,13], while positive samples of Rickettsia spp. were confirmed by PCR, which targeted the citrate synthase gene (gltA) [19]. nPCR was employed to amplify the internal transcribed spacer region sequence of Bartonella spp. [20] and the S segment of SFTSV [21].
All primers and amplification conditions used for detecting TBPs in ticks from cattle in the current study are described in Supplementary Table S1 [22]. Sequence alignment results were modified by BioEdit (v. 7.2.5) [23]. Phylogenetic analysis was conducted using MEGA (v. 6.0) [24], according to the maximum likelihood method with the Kimura two-parameter distance model. A similarity matrix was used to analyze the aligned sequences. The trees' stability was assessed by a bootstrap analysis using 1000 replicates.

Statistical Analysis
The two-sided Fisher's exact test was performed to analyze significant differences between pathogens for each tick stage, and a value of p < 0.05 was considered to indicate statistical significance. GraphPad Prism (v. 8.0; GraphPad Software Inc., La Jolla, CA, USA) was employed for the statistical analyses.

Identification of Ticks
In total, 576 ticks were collected in the study. Most ticks were partially fed. All of them, including 340 adults and 236 nymphs, were identified as Haemaphysalis longicornis by their morphological characteristics. Tick species were also molecularly identified using universal primers for the COI gene (expected size 710 bp) to avoid potential mistakes in the morphological identification. Both the morphological and molecular analyses identified all the ticks as H. longicornis. Furthermore, the nucleotide sequences from the representative ticks based on the developmental stage and collected region were assessed for the data analysis. The COI gene sequences obtained in this study shared close genetic relationships with H. longicornis (97.7-99.9% nucleotide identity). A phylogenetic tree was created according to the COI genes documented from several tick sequences deposited in GenBank ( Figure 1).

Molecular and Phylogenetic Analyses
Phylogenetic analyses showed that the 18S rRNA ( Figure  The five sequences of T. orientalis found in the present study shared a 97.4-98.8% identity with the 18S rRNA sequence. They also shared a 97.6-99.7% identity with the 18S rRNA sequences in previously reported T. orientalis isolates. The T. orientalis MPSP gene sequences were classified into five genotypes: types 1, 2, 3, 4, and 7. Among the 29 sequences, 11, 10, 3, 1, and 4 isolates were assigned to types 1, 2, 3, 4, and 7, respectively. The three representative sequences of types 1 and 2 found in the present study shared a 98.7-99.8% and 98.8-99.3% identity with the MPSP sequence, respectively. They also shared a 99.5-99.8% and 99.3-99.5% identity with the MPSP sequences in previously reported T. orientalis isolates, respectively. The three, one, and four sequences of types 3, 4, and 7 that we found each shared a 99.6-100%, 100%, and 97.4-99.9% identity with the MPSP sequence, respectively. Each of them also shared a 98.1-99.8%, 98.3-99.8%, and 97.9-99.3% identity with the MPSP sequences in previously reported T. orientalis isolates, respectively.
A., Anaplasma; APL: Anaplasma phagocytophilum-like Anaplasma spp.; E., Ehrlichia; R., Rickettsia; T., Theileria. Nymph A., Anaplasma; APL: Anaplasma phagocytophilum-like Anaplasma spp.; E., Ehrlichia; R., Rickettsia; T., Theileria.           The five sequences of T. orientalis found in the present study shared a 97.4-98.8% identity with the 18S rRNA sequence. They also shared a 97.6-99.7% identity with the 18S rRNA sequences in previously reported T. orientalis isolates. The T. orientalis MPSP gene sequences were classified into five genotypes: types 1, 2, 3, 4, and 7. Among the 29 sequences, 11, 10, 3, 1, and 4 isolates were assigned to types 1, 2, 3, 4, and 7, respectively. The three representative sequences of types 1 and 2 found in the present study shared a 98.7-99.8% and 98.8-99.3% identity with the MPSP sequence, respectively. They also shared a 99.5-99.8% and 99.3-99.5% identity with the MPSP sequences in previously reported T. orientalis isolates, respectively. The three, one, and four sequences of types 3, 4, and 7 that we found each shared a 99. 6  The three representative sequences of E. canis observed in our study shared a 100% identity with the 16S rRNA sequence and a 99.0-100% identity with previously reported E. canis isolates. Our three representative sequences of Candidatus R. longicornii shared a 100% identity with 16S rRNA and gltA sequences and a 100% and 98.9-99.2% identity with 16S rRNA and gltA sequences in previously reported isolates, respectively.

Discussion
In our study, only H. longicornis, including 340 adults and 236 nymphs, was found in cattle by both morphological and molecular methods. These findings are consistent with the result of a previous Korean study, which found H. longicornis (900/903, 99.7%) and Ixodes spp. (3/903, 0.3%) in cattle [10]. In another study [25], H. longicornis (15,020/19,821, 75.8%), H. flava (3889/19,821, 19.6%), and I. nipponensis (912/19,821, 4.6%) were identified from various habitats. H. longicornis is the most frequently identified species in Korea. The climate of Korea is steadily becoming subtropical due to global warming. The emergence of endemic TBPs might be associated with climate-driven changes to their geographic ecology and range. In the present study, TBPs were more prevalent in the southern area in the nymph and adult tick stages. This biogeoclimatic difference may clarify the observed differences in the prevalence of ticks and TBPs. These findings show that an additional geographical study is needed to fully understand the tick populations and clarify the distribution of TBPs in animals.
In the present study, the prevalence of TBPs was 1.4-times higher in the adults (38.2%) than in the nymphs (26.7%). This result is similar to that of a previous study [26] in which the overall infection rate was 2.7-times higher in adults compared with nymphs, most likely due to the transstadial accumulation in the mature ticks. In addition, the prevalence of multiple infections was higher in the adults than in the nymphs. In total, 1.3% of the collected H. longicornis nymphs and 3.8% of the adults were infected with more than one disease agent, constituting 4.8% and 10% of all the infected nymphs and adults, respectively. The COI sequences from the collected H. longicornis showed a 97.6-99.7% nucleotide identity with known COI sequences of H. longicornis (Figure 1). H. longicornis is typically collected from grasslands and herbaceous vegetation throughout Korea [25].
In this study, T. orientalis, A. bovis, A. capra, APL clades A and B, E. canis, and Candidatus R. longicornii were detected in cattle ticks by the molecular analysis. Theileriosis, one of the most important tick-borne hemoprotozoan diseases, can affect domestic animals, most frequently sheep and cattle in subtropical and tropical zones, and it causes great economic losses [27]. The members of the taxonomic group encompassing T. buffeli, T. sergenti, and T. orientalis are very similar, and the separate taxonomy of this group is controversial. Based on molecular studies, the three parasites are classified as one species, T. orientalis [28]. Recently in Korea, 18S rRNA genes were detected as T. orientalis in ticks (3.7%, 21/566 pools) from cattle [10] and T. orientalis in cattle (23.2%, 69/298) [29]. Here, 15 H. longicornis nymphs and 14 H. longicornis adults were positive for the T. orientalis 18S rRNA gene. The prevalence of T. orientalis in ticks herein (5.0%) was lower than that in a cattle study (23.2%) in Korea [29]. Based on the sequence analysis of the MPSP gene, T. orientalis consists of at least 11 different MPSP genotypes, including types 1-8 and N1-N3 [28]. Of the 11 MPSP genotypes, the Ikeda group consists of types 2 and 7, and the Chitose group consists of types 1, 3, 4, 5, 8, and N-3 [30]. Types N-1 and N-2 appear to have low sequence homology to each other [30]. Type 6 has been found in yaks and cattle, and it is classified as T. sinensis [31]. Recently in Korea, types 1, 2, 4, and 8 of the Theileria MPSP gene (2.7%, 15/556) were identified in ticks from grazing cattle [9], and types 1-3 and 7 (41.3%, 57/138) were identified in cattle [32]. In the present study, types 1-4 and 7 of the Theileria MPSP gene (5%, 29/576) were identified in cattle ticks. To our knowledge, this is the first study to report the presence of Theileria MPSP genotypes 3 and 7 in ticks in Korea. Of the five MPSP genotypes identified in this study, types 1 (37.9%, 11/29) and 2 (34.5%, 10/29) were the most commonly detected in ticks. In previous studies, type 2 in cattle [32] and types 2 and 4 in ticks from cattle [9] were also predominant. Types 1 (Ikeda) and 2 (Chitose) have been most commonly linked to clinical diseases. Of these, type 2 is more pathogenic [28], and it causes high parasitemia, severe anemia, and sometimes death. Therefore, it is imperative to conduct additional studies to identify the relationship between clinical signs and pathogenic types by determining the MPSP genotypes of T. orientalis that are related to animals and ticks.
The genus Anaplasma contains obligate intracellular Gram-negative bacteria belonging to the order Rickettsiales and the family Anaplasmataceae [33]. There are seven formally recognized species of Anaplasma (A. phagocytophilum, A. centrale, A. marginale, A. platys, A. bovis, A. ovis, and A. caudatum). A. capra represents a probable eighth species, and it has been argued as a new Anaplasma species whose name has not yet been formally recognized [34]. A. capra was recently identified from goats in China as a new tick-transmitted emerging zoonotic pathogen owing to its isolation from human blood (5.9%, 28/477) after tick bites [35]. It remains uncertain whether A. capra is pathogenic to both humans and animals, but if it is confirmed as such, it could pose a significant public health risk, the same as A. phagocytophilum [35]. In Korea, A. capra was formerly identified in cattle (0.4%, 5/1219) [12] and in Korean water deer (17.7%, 35/198) [36]. In the present study, one H. longicornis nymph and 26 H. longicornis adults tested positive for A. capra. The prevalence of A. capra in ticks herein (4.7%) was higher than that in a cattle study (0.4%) in Korea [12]. To our knowledge, this is the first study to report the presence of A. capra in ticks in Korea. A. capra is an emerging human pathogen in ticks parasitizing cattle and other animals in Korea. However, the vector capability of ticks for the transmission of A. capra is still unclear and needs additional evaluation for public health control.
A. phagocytophilum is the causative pathogen of granulocytic anaplasmosis in various species, such as humans, dogs, goats, horses, sheep, and cattle [37]. The rapid and accurate diagnosis of pathogenic and zoonotic diseases, such as human granulocytic anaplasmosis, is required for the risk estimation in TBP control programs [38]. Therefore, it is meaningful to differentiate between pathogenic A. phagocytophilum and APL species that do not cause clinical signs in infected animals and that are presently considered non-pathogenic [38]. APL clade A was detected in cattle (2.6%, 20/764) in Korea [13]. Several APL clades have also been reported in other countries: APL clade A was detected in cattle (2.0%, 1/50) and ticks (2.4%, 2/85) in Japan [39], APL clade B was found in ticks (0.8%, 3/388 pools) in China [40], and APL clades A and B were detected in cattle (1.9%, 7/367; 0.5%, 2/367), sheep (7.0%, 25/355; 5.4%, 19/355), and goats (13.3%, 32/241; 5.0%, 12/241) in Tunisia [38]. In the present study, APL clade A was detected in one H. longicornis nymph and 10 H. longicornis adults, and one H. longicornis nymph and two H. longicornis adults were positive for APL clade B. The prevalence of APL clade A in ticks herein (1.9%) was lower than that in a cattle study (2.6%) in Korea [13]. To our knowledge, this is the first study to report the presence of both APL clades A and B in ticks from Korea. Additional studies are required to provide more information on the molecular background and to trace the evolutionary tree of novel Anaplasma species. A. bovis, a monocytotropic species, has been reported in ruminants in numerous countries [41]. A. bovis was formerly detected in Korea in cattle (1.0%, 12/1219) [12], in ticks (7.5%, 20/266 pools) from Korean water deer [11], in H. longicornis ticks (2.5%, 1/40 pools) from native Korean goats [14], and in H. longicornis ticks (1.0%, 5/506 pools) [8]. In the present study, 13 H. longicornis adults were positive for A. bovis. The prevalence of A. bovis in ticks herein (2.3%) was higher than that in a cattle study (1.0%) in Korea [12]. A. bovis was the only species of TBPs that was not found in the nymph stage. Meanwhile, in our previous study [14], A. bovis was only detected in the nymph stage of H. longicornis ticks from goats. In another study [42], A. bovis was detected in all tick stages, including larvae, nymphs, and adults. Thus, this appears to be an issue of the number of ticks tested. If we were to collect more ticks, we would be more likely to detect A. bovis in nymphs.
Canine monocytic ehrlichiosis is a systemic infection in dogs caused by E. canis, which is transmitted by Rhipicephalus sanguineus sensu lato, known as the brown dog tick [43]. E. canis was previously identified in Korea in H. longicornis and Ixodes turdus (1.1%, 18/1638 pools) and small mammals (12.0%, 51/424) [7], in H. longicornis ticks (1.2%, 6/506 pools) [8], and in H. longicornis from cattle (22.3%, 126/566 pools) [10]. Moreover, reports have revealed human infections of E. canis in Venezuela [44] and a novel genotype of E. canis in a human from Costa Rica [45]. The presence of E. canis in human samples is likely associated with the high prevalence of this pathogen in ticks and dogs [45]. In our study, four H. longicornis nymphs and five H. longicornis adults tested positive for E. canis. Continuous monitoring for infected ticks and reservoir hosts is required to ensure the health and safety of animals and the public against the risks of TBP exposure.
Rickettsia spp. are emerging or re-emerging pathogens with public health importance [46]. Obligate intracellular bacteria belong to the spotted fever group (SFG) and cause tick-borne rickettsioses [47]. In the present study, one SFG rickettsia with the Candidatus status was identified in the ticks. A potentially new SFG rickettsia classified into the putative novel subgroup "Candidatus R. longicornii" has been detected in Korea, including in H. longicornis ticks (43.2%, 79/183) [48], in H. longicornis ticks (16.7%, 52/311 pools) [49], and in H. longicornis ticks (45%, 18/40 pools) from native Korean goats [14]. These pathogens were clustered together in a subgroup that represented a sister taxon separate from the known subgroups of SFG rickettsiae. Gene fragment sequences reported in GenBank for rickettsial isolates from H. longicornis in Korea, Japan, and China have uncertain taxonomic statuses and unidentified pathogenicity that are most likely correlated to Candidatus R. longicornii or to a very closely associated species [48]. In addition, the XY118 (KU853023) isolate was detected from a patient, representing its possible pathogenicity in humans. Further studies are required to determine the pathogenicity of this novel pathogen. Here, Candidatus R. longicornii was detected in 41 H. longicornis nymphs and 60 H. longicornis adults. Additional research is necessary to identify other novel Rickettsia spp. in ticks and animals in Korea.
The bootstrap support levels need to be evaluated cautiously. As a limitation in this study, some bootstrap values were low. This could be due to using small amplification fragments of genes for the phylogenetic analysis. Therefore, further studies are needed to analyze longer fragments of genes in a phylogenetic analysis for a better presentation of the data.

Conclusions
In the present study, seven TBPs were detected in cattle H. longicornis ticks from Korea, including human pathogens of A. capra, E. canis, and Candidatus R. longicornii. Overall, 33.5% of ticks harbored at least one TBP. In 1.3% of the nymphs and 3.8% of the adults, we found more than one TBP. Among them, Candidatus R. longicornii was the most prevalent. To our knowledge, this is the first study to report the presence of the pathogens T. orientalis MPSP genotypes 3 and 7, A. capra, and APL in ticks from Korea. Cattle ticks may be maintenance hosts for many TBPs, and veterinary and medical clinicians should be aware of their high probability of infection and clinical complexity in humans. Our results show that ticks parasitizing cattle could be possible maintenance hosts for TBPs, and because of the zoonotic pathogenic importance of TBPs, we need to increase the awareness of their wide distribution and adopt measures to prevent their spread. Moreover, this study shows that coinfections are represented and thus should be considered in the diagnosis of TBPs, and that additional studies in Korea are needed in the future.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-2607/8/5/728/s1, Table S1: Primers used for the detection of tick-borne pathogens in ticks from cattle in the present study.