Ehrlichia chaffeensis and Four Anaplasma Species With Veterinary and Public Health Significance Identified in Tibetan Sheep (Ovis aries) and Yaks (Bos grunniens) in Qinghai, China

Tick-borne diseases (TBDs) can cause serious economic losses and are very important to animal and public health. To date, research on TBDs has been limited in Qinghai-Tibet Plateau, China. This epidemiological investigation was conducted to evaluate the distribution and risk factors of Anaplasma spp. and Ehrlichia chaffeensis in livestock in Qinghai. A total of 566 blood samples, including 330 yaks (Bos grunniens) and 236 Tibetan sheep (Ovis aries) were screened. Results showed that A. bovis (33.3%, 110/330) and A. phagocytophilum (29.4%, 97/330) were most prevalent in yaks, followed by A. ovis (1.2%, 4/330), A. capra (0.6%, 2/330), and E. chaffeensis (0.6%, 2/330). While A. ovis (80.9%, 191/236) and A. bovis (5.1%, 12/236) infection was identified in Tibetan sheep. To our knowledge, it is the first time that A. capra and E. chaffeensis have been detected in yaks in China. Apart from that, we also found that co-infection of A. bovis and A. phagocytophilum is common in yaks (28.2%, 93/330). For triple co-infection, two yaks were infected with A. bovis, A. phagocytophilum, and A. capra, and two yaks were infected with A. bovis, A. phagocytophilum, and E. chaffeensis. Risk analysis shows that infection with A. bovis, A. phagocytophilum, and A. ovis was related to region and altitude. This study provides new data on the prevalence of Anaplasma spp. and E. chaffeensis in Qinghai, China, which may help to develop new strategies for active responding to these pathogens.


INTRODUCTION
Anaplasmosis and ehrlichiosis are important diseases caused by tick-borne pathogens, which result in additional economic losses to livestock (1,2). To date, seven Anaplasma species have been identified, including A. bovis, A. phagocytophilum, A. centrale, A. platys, A. marginale, A. ovis, and A. capra (3,4). A. bovis parasitizes monocytes and macrophages of ruminants and small mammals (5). A. phagocytophilum infects neutrophils of humans and animals, and causing human granulocytic anaplasmosis (HGA), tick-borne fever in ruminants, and canine and equine granulocytic anaplasmosis (5). A. centrale and A. marginale mainly infect erythrocytes of cattle, while A. ovis primarily infect small ruminant animals such as sheep and goats. (6). A. platys mainly infect canine platelets and cause cyclic thrombocytopenia in dogs (6). A. capra is an emerging pathogen, which can infect ruminants and humans (7). In addition, as a member of the Ehrlichia family, Ehrlichia chaffeensis can cause human monocytic ehrlichiosis (HME) (8), and ehrlichiosis in animals (9).
Over the past several decades, the Anaplasma and Ehrlichia infections are very common in many countries (3,(10)(11)(12). A. bovis is mainly distributed in Africa, Asia, and South America, and cattle are considered the primary hosts (6). Similarly, A. ovis is the leading cause of anaplasmosis in small ruminants, which is widely distributed around the world (13). Recently, A. phagocytophilum, A. capra, and E. chaffeensis have received much attention for their potential threats to public health (7,14). A. phagocytophilum has been detected in sheep, cattle, Capreolus pygargus, goats, and humans in different areas of China (15)(16)(17)(18). E. chaffeensis infections are very common in the United States, with an annual rate of 4.46 cases/1,000,000 population (19). For A. capra, it was initially isolated from goats and humans in China (7). Subsequently, it was found in many countries (20,21).
Qinghai is the source of the Yangtze River, the Yellow River, and the Lancang River, located in the northeast of Qinghai-Tibet Plateau and northwest of China with an average altitude of more than 3,000 meters. The complicated topographic features and changeable climate bless the region with advantageous conditions of rich natural resources. Tibetan sheep (Ovis aries) and yaks (Bos grunniens) are the main domestic animals in Qinghai and an important source of life and income for herders. Ixodid tick infestation in livestock is a common and severe problem, and more than 25 tick species in six genera have been reported in Qinghai (22,23). However, information about tick-borne diseases (TBDs) in the region has been limited. Therefore, to better understand the situation of TBDs in Qinghai, China, a molecular epidemiologic study was conducted investigating exposure to Anaplasma spp. and E. chaffeensis in domestic animals across the area.

Blood Sample Collection of Yaks and Tibetan Sheep
A total of 566 blood samples of yaks (n = 330) and Tibetan sheep (n = 236) were collected using random sampling from six sampling sites in Maqin (35 • (Figure 1). GL and YS are similar in altitude and climate, and both belong to the continental climate of the plateau. Except for about 400 Tibetan sheep in Maqin, the number of yaks and Tibetan sheep in other sampling sites is between 100 and 200. All animals adopt a free grazing system. Ticks and Melophagus ovinus and their bites can be seen in Tibetan sheep, while ticks are rarely found on yaks.

Extraction and Quantification of DNA
According to the manufacturer's operation manual, genomic DNA was extracted from 200 uL whole blood samples by the TIANamp Genomic DNA kit (TIANGEN biotech, Beijing). The concentration of the extracted DNA was detected by NanoDrop 2,000 (Thermo Fisher Scientific, USA) and then stored at −20 • C for pathogens detection.

Detection of Anaplasma spp. and E. chaffeensis
Conventional PCR or nested PCR was used to screen for Anaplasma spp. and E. chaffeensis in extracted DNA. Nested PCRs were employed to detect A. bovis, A. phagocytophilum, A. centrale, A. platys, and E. chaffeensis based on 16S rRNA gene. Conventional PCR based on the msp4 genes was employed to detect A. marginale and A. ovis, while 16S rRNA gene for detection of A. capra. PCR primers and cycling conditions used in this study, as shown in Table 1. The DNAs extracted from the whole blood of Tibetan sheep and yaks infected with A. bovis, A. phagocytophilum, A. ovis, A. capra, and E. chaffeensis that had been verified by sequencing, were used as a positive control for corresponding PCR reactions; double-distilled water was used as a negative control. The PCR products were detected by 1.5% agarose gel electrophoresis with M5 Hipure Next III Gelred (Mei5 Biotechnology Co., Ltd., Beijing, China) stained.

Sequencing and Phylogenetic Analysis
PCR products of all positive samples for Anaplasma spp. and E. chaffeensis randomly selected from each sampling site were sequenced by BGI (Beijing, China). The sequence obtained by BGI sequencing was submitted to NCBI for BLASTn search (https://blast.ncbi.nlm.nih.gov/Blast.cgi), and sequence alignment and analysis. The representative nucleotide sequences of this study have been deposited in the GenBank database. Phylogenetic trees were constructed using the neighbor-joining method executed with the p-distance model in MEGA X. Bootstrap values were assessed with 1,000 bootstrap replicates (28,29).

Statistical Analysis
The data were grouped into four variables according to animal species, gender, sampling sites, and the altitude of sampling sites. The chi-square test was used to calculate the difference of infection rate in SPSS 25.0 software in each group. When p < 0.05, the difference was significant.

Prevalence of Anaplasma spp. and E. chaffeensis in Tibetan Sheep and Yaks
This study identified four Anaplasma species and E. chaffeensis from Tibetan sheep and yaks ( Table 2). Of the 566 samples tested, 50% (283/566) were positive for at least one pathogen. The infection rates of A. bovis and A. ovis were 33.3% and 1.2% in yaks, 5.1% and 80.9% in Tibetan sheep. The infection rates of A. phagocytophilum, A. capra, and E. chaffeensis were 29.4%, 0.6%, and 0.6% in yaks, respectively. This is the first time that A. capra and E. chaffeensis have been detected in yaks in China. Interestingly, we noticed A. ovis infection in yaks and A. bovis in Tibetan sheep. The most common co-infection was A. bovis and A. phagocytophilum, with an infection rate of 28.2% (93/330) in yaks. For co-infection with three pathogens, the infection rate of A. bovis, A. phagocytophilum, and A. capra was 0.6% (2/330), and the infection rate of A. bovis, A. phagocytophilum, and E. chaffeensis was 0.6% (2/330) ( Table 2). No co-infections by two or more pathogens were detected in Tibetan sheep.

Sequencing and Phylogenetic Analysis
In the current study, 15 representative sequences were obtained and submitted to GenBank ( Table 3). We compared and analyzed the partial 16S rRNA gene sequences of A. bovis, A. phagocytophilum, A. capra, and E. chaffeensis obtained from blood samples of Tibetan sheep and yaks. BLASTn analysis of the 16S rRNA gene showed that the Anaplasma spp. obtained in this study had 99.04-100% identities to either of A. bovis, A. phagocytophilum, A. capra, and E. chaffeensis sequences, respectively. The E. chaffeensis sequences (MW048788, MW048789) from yaks were 99.44-100% identical to E. chaffeensis isolated from goats (KX505292) in China. Phylogenetic analysis of 16S rRNA gene sequences confirmed A. bovis, A. phagocytophilum, A. capra, and E. chaffeensis in this study (Figures 2A,B, 3A,B). Additionally, we analyzed the msp4 genomic region of three A. ovis (MZ231113-MZ231115) obtained in this study. The results showed that the three sequences were consistent with the homology of the Iranian A. ovis (MH790273). A. ovis were classified as A. ovis msp4 Genotypes II based on T 366 C 470 (25). Phylogenetic analysis of msp4 gene sequences confirmed the identity of A. ovis in this study (Figure 4).

Risk Factors of Tibetan Sheep and Yaks Infected With Anaplasma spp. and E. chaffeensis
These factors include animal species, gender, sampling sites, and altitude of sampling sites, which were used as variables for statistical analysis of the infection patterns of Anaplasma spp. and E. chaffeensis. The results indicate that the prevalence of Anaplasma spp. and E. chaffeensis in female animals was similar  to that of male animals (P >0.05). The infection rates of A. bovis, A. phagocytophilum, and A. ovis in yaks in GL and YS were 39.7 and 21.6% (P = 0.001), 34.6 and 19.8% (P = 0.006), 0 and 3.5% (P = 0.005), respectively. In addition, the infection rate of A. bovis and A. phagocytophilum below 4,000 m was significantly higher than those above 4,000 m (P = 0.000). In Tibetan sheep,  the infection rate of A. ovis above 4,000 m was higher than that below 4,000 m (P = 0.022) ( Table 4).

DISCUSSION
In the present study, Anaplasma spp. and E. chaffeensis were investigated in domestic animals in Qinghai, China. Four Anaplasma species (A. bovis, A. phagocytophilum, A. ovis, and A. capra) and E. chaffeensis were identified in Tibetan sheep and yaks. Among them, E. chaffeensis and A. capra were detected in yaks for the first time in China. The genus Anaplasma are widely distributed in domestic animals, wild animals, ticks, and other vectors (23,(30)(31)(32). This study found relatively high A. ovis infection rates of 76.2, 94.1, and 81.3% in Tibetan sheep in three sampling sites, Maqin, Dari, and Qumalai, respectively, which is higher than in sheep in Xinjiang (40.5%) (16) and Gansu (5.7%) (33), and Tibetan sheep in northeast Qinghai (58%) (34). An explanation for higher infection rates of A. ovis in this area could be the bites of ticks and other arthropods. Ticks and M. ovinus were found in Tibetan sheep in Maqin and Dari, and data on that M. ovinus carried A. ovis has been reported in our previous study (31). In addition, we carried out the comparative analysis and phylogenetic analysis of the msp4 gene sequence of A. ovis (25). The results showed that the A. ovis strains isolated from Tibetan sheep were identical to those isolated in M. ovinus in our previous study (31). Whereas, the A. ovis isolated from Dermacentor nuttalli in Qinghai by Han et al. (23) belongs to genotypes III, which is in the same clade as those obtained from sheep in Italy (Figure 4) (25). Genotypes II and III were also isolated from M. ovinus in Xinjiang by Zhao et al. (35). Remarkably, an A. ovis variant was reported in humans (36), indicating that this agent has zoonotic potential. Taken together, there are two A. ovis genotypes prevalent in domestic animals in northwest China, and arthropods (including M. ovinus and ticks) may be the main vectors of A. ovis.
A. phagocytophilum and A. bovis are frequently detected in ruminants around the world. This study confirms that both A. phagocytophilum and A. bovis can infect yaks. The infection rate of A. phagocytophilum in yaks (29.4%) in this study was higher than that reported in sheep (9.9%), dairy cattle (12%), and white yaks (5.3%) in other areas of China (1,13,37), and lower than that in C. pygargus (33.3%) from Heilongjiang China (17). Since the first case of HGA, caused by A. phagocytophilum, was reported in Anhui, China (38), HGA has been reported in the USA, Europe, Africa, and Asia (11,39,40). For A. bovis, the infection rate in yaks (33.3%) was higher than that in cattle (4.8%) and white yaks (6.2%) from China (16,37), cattle (1.0%) from South Korea (20). Recent studies have shown that climate, altitude, longitude, latitude, season, tick bites, contact with wild animals, and feeding methods are important factors affecting Anaplasma infection (41). Previous reports have shown that Haemaphysalis qinghaiensis, Dermacentor abaensis, D. nuttalli, and Dermacentor silvarum are common ectoparasites among grazing livestock in high altitude areas (2,800 to 4,300 m), and the risk of tick bites with Anaplasma spp. was related to altitude and tick species (23). Our results also showed that the risk of infection with Anaplasma spp. in Tibetan sheep and yaks is mainly related to altitude and sampling sites. Furthermore, all animals in this study adopted a free grazing system, which increased the risk of domestic animals being exposed to ticks.
A. capra is a novel Anaplasma species that emerged in recent years. The novel species was first found in goats and then in sheep (30), C. pygargus (17), dogs (42), and ticks (23) in China. In addition, A. capra has also been detected in goats, cattle, and Hydropotes inermis argyropus in South Korea (32,43), cattle in Malaysia (10), and Cervus elaphus and Rucervus duvaucelii in France (21). In 2015, it was isolated from the blood samples of patients with a history of tick bites in northeastern China (7). Subsequently, Peng et al. (44) confirmed the ability of A. capra to infect human erythrocytes, HL-60 and TF-1, and further confirmed its zoonotic characteristics. In this study, we detected A. capra DNA in yaks in China for the first time. In Qinghai, H. qinghaiensis is the most dominant tick species infected with A. capra, followed by D. abaensis and D. nuttalli (23). The above evidence suggests that A. capra is widely distributed and could infect a wide range of hosts.
Ehrlichia species include E. chaffeensis, E. canis, E. ewingii, E. equi, E. muris, and E. ruminantium. These species have been detected in many ticks in China, for instance, Amblyomma testudinarium, Haemaphysalis yeni, Haemaphysalis longicornis, Ixodes sinensis, D. silvarum, Rhipicephalus sanguineus, and Rhipicephalus microplus (45)(46)(47)(48). In previous studies, E. canis infection was detected in Cervus nippon in Gansu (49), and high infection rates of E. canis and E. chaffeensis were reported in dogs, cattle, sheep, goats, donkeys, and humans in Xinjiang (9,18,50). Ehrlichia species were also detected in birds and small mammals in other parts of China (51,52). In the current study, the prevalence rate of E. chaffeensis was 0.61%. We present the first report of Ehrlichia infection caused by E. chaffeensis in yaks in China. However, it is unclear which ticks are responsible for the pathogen. Therefore, further study is needed to determine the vector or reservoir host for this pathogen.
Moreover, mixed-infection is also an important issue that would need to be considered in livestock. The present study results illustrate that mixed infection of A. phagocytophilum and A. bovis are very common in yaks in Qinghai. Coinfection involving three Anaplasma species of A. bovis, A. phagocytophilum, and A. capra was also observed in two yaks in this study. In addition, we found that two yaks were coinfected with A. bovis, A. phagocytophilum and E. chaffeensis. Currently, A. phagocytophilum, A. capra, and E. chaffeensis have been recognized as causative agents of human infection. Mixedinfection of tick-borne pathogens has also been observed in animals in other countries and regions (1,30,34,53). Above all, co-infection of tick-borne pathogens emphasizes the need for differential diagnosis of these pathogens in animal hosts and humans to improve the prevention and control of TBDs.
Notably, all pathogens were detected from apparently healthy animals in this study, consistent with other studies (54)(55)(56). This indicates that the appearance of clinical symptoms is mainly dependent on the pathogenicity of these pathogens strains and the breed or species of the infected animals (54). Alternatively, these animals have previously been infected with these pathogens and developed immunity against these pathogens (56). Further research is necessary to confirm these speculations.  In conclusion, we investigated the epidemic situation of the TBDs in yaks and Tibetan sheep in Qinghai province, China, and confirmed that Tibetan sheep and yaks could be infected with A. bovis, A. phagocytophilum, A. ovis, A. capra, and E. chaffeensis. This is the first report of A. capra and E. chaffeensis infection in yaks in China. These pathogens could pose a significant threat to livestock and human health. Thus, future studies should focus more on systematically assessing these pathogens' threats to veterinary and public health.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

ETHICS STATEMENT
The Animal Ethics Committee of the Institute of Zoology, Chinese Academy of Sciences approved the procedures of collecting blood samples from Tibetan sheep and yaks, and obtained the livestock owner's consent. Written informed consent for participation was not obtained from the owners because all the samples in this study were collected by local veterinarians during the daily epidemic surveillance.

ACKNOWLEDGMENTS
We want to express our heartfelt thanks to all the herdsmen and all the Guoluo and Yushu Animal Disease Prevention and Control Center staff for their strong support for this study, enabling us to smoothly carry out the research.