Novel variants of the newly emerged Anaplasma capra from Korean water deer (Hydropotes inermis argyropus) in South Korea

Background Anaplasma spp. are tick-borne Gram-negative obligate intracellular bacteria that infect humans and a wide range of animals. Anaplasma capra has emerged as a human pathogen; however, little is known about the occurrence and genetic identity of this agent in wildlife. The present study aimed to determine the infection rate and genetic profile of this pathogen in wild animals in the Republic of Korea. Methods A total of 253 blood samples [198 from Korean water deer (Hydropotes inermis argyropus), 53 from raccoon dogs (Nyctereutes procyonoides) and one sample each from a leopard cat (Prionailurus bengalensis) and a roe deer (Capreolus pygargus)] were collected at Chungbuk Wildlife Center during the period 2015–2018. Genomic DNA was extracted from the samples and screened for presence of Anaplasma species by PCR/sequence analysis of 429 bp of the 16S rRNA gene marker. Anaplasma capra-positive isolates were genetically profiled by amplification of a longer fragment of 16S rRNA (rrs) as well as partial sequences of citrate synthase (gltA), heat-shock protein (groEL), major surface protein 2 (msp2) and major surface protein 4 (msp4). Generated sequences of each gene marker were aligned with homologous sequences in the database and phylogenetically analyzed. Results Anaplasma capra was detected in blood samples derived from Korean water deer, whereas samples from other animal species were negative. The overall infection rate in tested samples was 13.8% (35/253) and in the water deer the rate was 17.8% (35/198), distributed along the study period from 2015 to 2018. Genetic profiling and a phylogenetic analysis based on analyzed gene markers revealed the occurrence of two distinct strains, clustered in a single clade with counterpart sequences of A. capra in the database. Conclusions Anaplasma capra infection were detected in Korean water deer in the Republic of Korea, providing insight into the role of wildlife as a potential reservoir for animal and human anaplasmosis. However, further work is needed in order to evaluate the role of Korean water deer as a host/reservoir host of A. capra.


Background
The cosmopolitan genus Anaplasma includes six species of Gram-negative obligate intracellular bacteria that are transmitted by ticks to a wide range of animals, including humans [1][2][3][4][5], resulting in considerable economic losses in the livestock industry and serious public health concerns [6,7]. Anaplasma phagocytophilum, A. ovis and recently reported A. capra are human pathogens [8][9][10][11][12], whereas other species in the genus have no known zoonotic potential. However, A. platys may have zoonotic potential after frequent reports of human infection [13,14].
Although different Anaplasma species have been detected in wildlife [23][24][25][26][27][28][29], little is known about the prevalence and genetic identity of A. capra in these animals in Korea. Using molecular tools, the present study aimed at investigating the occurrence and characterizing the genetic profile of this pathogen in wildlife in the Republic of Korea.

Collection of samples
Chungbuk Wildlife Center is located in Cheongju city, Chungcheongbuk-do province in the Republic of Korea (36°38′13.99ʺN, 127°29′22.99ʺE). The center receives terrestrial and avian wild animals for purposes of treatment from sickness/injuries and/or rehabilitation. Blood samples are collected for diagnosis and treatment of wildlife referred to the Chungbuk Wildlife Center. Blood samples are archived in EDTA-treated tubes and stored at -80 °C. A total of 253 blood samples including 198 from Korean water deer (Hydropotes inermis argyropus), 53 from raccoon dogs (Nyctereutes procyonoides) and one sample each from a leopard cat (Prionailurus bengalensis) and a roe deer (Capreolus pygargus), collected from January 2015 to June 2018, were used.

DNA extraction and PCR amplification
Frozen blood samples were thawed at room temperature and genomic DNA was extracted from 200 µl of blood using a Magpurix ® Blood Kit and Magpurix ® 12s automated nucleic acid purification system (Zinexts Life Science Corp., Taipei, Taiwan), according to the manufacturer's recommendations. DNA preparations were tested for the presence of Anaplasma species by PCR/ sequence analysis of 429 bp of the 16S rRNA gene as described previously [30]. Anaplasma capra-positive isolates were genetically profiled by the amplification of a longer fragment of 16S rRNA (rrs) gene as well as partial sequences of citrate synthase (gltA), heat-shock protein (groEL), major surface protein 2 (msp2) and major surface protein 4 (msp4) genes as described previously (Table 1). Amplified fragments were electrophoresed on 1.2% gel loaded with EcoDye ™ stain (BIOFACT, Daejeon, Korea) and visualized using UV light.

DNA sequence analysis
The PCR products (rrs and groEL) or secondary PCR product (for other gene markers) were purified and sequenced, either directly or after cloning in the pGEM-T vector (Promega, Madison, WI, USA), in both directions. Generated sequences were assembled using ChromasPro v.2.1.8 (https ://techn elysi um.com.au/wp/chrom aspro /).

Results
The overall infection rate of A. capra in tested animals was 13.8% (35/253); however, samples from raccoon dogs (n = 53), leopard cat (n = 1) and roe deer (n = 1) were negative.  sheep, cattle, ticks and humans; however, they had striking genetic differences, suggesting that they are novel strains. Sequences of the rrs gene fragment of both strains showed an identity of ~ 99.5% with counterparts in database and clustered in the clade of A. capra from different hosts (Fig. 1). Both strains had single nucleotide polymorphisms (SNPs), resulting in four genotypes at this gene locus. Phylogenetic analysis revealed that three sequences designated A. centrale (GenBank: AB211164, AF283007 and GU064903) and two sequences designated Anaplasma spp. (GenBank: AB454075 and AB509223) clustered within the A. capra clade, even though other A. centrale sequences from different hosts and geographical regions formed separate clusters in the ML tree.
The gltA gene of the Cheongju strain shared a similarity of 99.5% (with two substitutions, A/G at position 456 and T/C at position 533) with gltA sequences KM206274, KJ700628 and MH029895 isolated from a human, goat and tick, respectively [12,23]. Sequences of the Chungbuk strain showed a similarity of 98-99% with KX685885, KX685886 and MF071308 of A. capra from ticks and sheep [13,19]. Both strains clustered with their homologous sequences in the A. capra clade (Fig. 2). groEL gene sequences derived from the Cheongju strain shared a similarity of 99% (one substitution) with their counterparts from humans (GenBank: KM206275), goats (GenBank: KJ700629), sheep (GenBank: KX417356) and ticks (GenBank: KR261633 and KR261635), whereas sequences from the Chungbuk strain shared a similarity of 91% with the reference sequences (Fig. 3). The msp2 sequences showed extensive intra-and inter-sequence variations, including multiple InDels and single nucleotide substitutions; however, all sequences remained clustered in the A. capra clade (Fig. 4). A hypervariable stretch was detected between positions 285 and 414 of the generated sequences (corresponding to positions 550 and 679 in the reference sequence KM206276 of A. capra from humans). The msp4 sequences were identical in the two strains and showed an identity of 100% with those from humans (GenBank: KM206277) and ticks (Gen-Bank: KR261637 and KR261640) (Fig. 5).

Discussion
Wild animals act as reservoirs for a wide range of pathogens [31][32][33]. The emergence of infectious disease agents of wildlife origin is a prominent challenge to public health and the livestock industry [34][35][36]. Anaplasma capra has recently been isolated from human patients in China with non-specific clinical manifestations, with potential progression to CNS complications, suggesting that this species could pose a substantial threat to public health [12,37]. We detected A. capra DNA in blood samples of 35 out of 198 KWD (17.7% infection rate) at the Chungbuk Wildlife Center, Korea. Epidemiological data for this pathogen in wildlife are lacking in Korea; however, our findings are similar to those obtained from wildlife (five takins, three Himalayan gorals, three Reeves's muntjacs, one forest musk deer and one wild boar) in China [24]. In addition, a low percentage of infection rate was reported cattle, sheep and goats in China, Sweden and Korea [15][16][17][18][19], indicating that A. capra has a broad host range. Occurrence of infection during the study period from 2015 to 2018 indicates the persistence of the infection in KWD, suggesting that the species may act as a reservoir for this pathogen. However, it is difficult to explain the negative results from raccoon dogs in the present study. This may be attributed to persistent infection making the pathogen below detectable level in the blood of these animals. In support of this view, A. capra has been reported to infect endothelial cells [12,15], making its detection in the blood possible in case of considerable bacteremia or released endothelial cells, resembling Rickettsia species [12]. Furthermore, the sample size and species and/or the age of animals may play a role in these findings. Further investigations are needed to clarify these points.
Our genetic profiling results indicated that the newly generated 16S rRNA gene sequences shared a homology of > 99.5% with sequences of A. capra strains from humans, sheep, goats, cattle and ticks [12,15,16,18,20,22], suggesting that they likely are within the same species of bacteria [38,39]. Clustering of sequences named A. centrale from deer (Cervus nippon nippon) (GenBank: AB211164) and cattle (GenBank: AF283007) in Japan and from ticks (Haemaphysalis longicornis) in Korea (Gen-Bank: GU064903), as well as Anaplasma spp. from deer (GenBank: AB454075; direct submission) and Japanese The clustering pattern of these sequences in the A. capra clade does not support their assignment as sister taxa and suggests that these isolates are in fact A. capra [15,25] and may need re-description, since these sequences were more related to Chungbuk strain in the A. capra clade.
Due to the extinction of natural predators, the KWD is thriving in Korea and has been designated as "harmful wildlife" by the Ministry of Environment in 1994 owing to harmful interactions with humans and their properties. This close interaction poses substantial threats to domestic animals and human health in Korea. This study was limited by analyzing samples from one geographical area and few animal species, which my lead to biases in the results. A large-scale study is underway to fully elucidate the host range of wildlife, vector ticks, pathogenicity and geographical distribution of this organism in Korea.

Conclusions
To our knowledge, the results presented herein provide the first evidence for the presence of A. capra in Korean water deer in Korea. As an emerging human pathogen, the detection of A. capra in deer provides insight into the role of wildlife as a potential reservoir for human anaplasmosis. Furthermore, the obtained results expand the known geographical and host range of the Anaplasma capra.