Background

Anaplasmosis is a tick-borne disease caused by a group of obligate intracellular bacteria of the genus Anaplasma. It has a worldwide distribution, particularly in tropical, subtropical and temperate regions, and is considered one of the most important pathogens in humans and animals. Anaplasma belongs to the family of Anaplasmataceae, order Rickettsiales. To date, seven Anaplasma spp. are most commonly reported in domestic animals: A. phagocytophilum, A. ovis, A. bovis, A. marginale, A. centrale, A. platys and A. capra [1].

A. phagocytophilum is a zoonotic pathogen that mainly infects human and animal neutrophils, It can cause an acute febrile illness known as human granulocytic anaplasmosis (HGA) [2]. The first human case of HGA was detected in the United States in 1990, and several cases have since been reported in countries in Europe and Asia. In China, the first case of HGA was detected in Anhui Province in 2006, and then the prevalence of HGA was found in several provinces and regions in China. Anaplasmosis generally presents with nonspecific symptoms such as fever, chills, malaise, headache, and myalgias. In severe cases, multiple organ function is impaired, and death can occurs [3]. The reservoir for A. phagocytophilum includes domestic and wild mammals, such as sheep, goat, cow, horse, deer, cat, dog and human [4,5,6,7,8,9].

A. ovis, which has the the same affinity to host cells as A. marginale and A. centrale, is a tick-borne obligate intraerythrocytic bacterium that commonly infects camels, sheep, goats, cattle and wild ruminants [6, 10]. Additionally, it was detected in humans in Cyprus in 2006. Clinical manifestations such as fever, fatigue, loss of appetite, reduced milk production miscarriage and Lower mortality rates are common in ruminants with ovine anaplasmosis infections [11]. Goats are more prone to outbreaks than sheep [12]. Moreover, for other anaplasmas, A. ovis infection may predispose to other microbial or parasitic infections, leading to the exacerbation of clinical symptoms and eventual death [13].

A. marginale, which mainly infects cattle, has been increasingly detected in other animal species, horse and asses [14]. It is the most pathogenic among several Anaplasma spp.for cattle. The pathogen has been recorded on several continents, including South America [15,16,17,18], Latin America [19, 20], Africa [21,22,23,24], Asia [18, 25,26,27,28], Australia [24] and Europe [29]. However, in China, the reports of marginal apocrysis are very rare in China. A.centrale is closely related to A. marginale and is considered a subspecies of A. marginale, causing only mild anaemia in animals [30]. It is used as a live vaccine against the A. marginale in Africa, Australia, Latin America, and Israel [13].

A. bovis is an obligate parasitic pathogen in monocytes. A. bovis infection causes anaplasmosis in cattle, presenting a variety of clinical symptoms, including fever, weight loss, and lower milk production. In the acute phase of the disease, it occasionally causes abortion and death [31]. However, most of infected animals are asymptomatic. Besides the aforementioned ruminants, DNA of A. bovis has also been detected in goats, sheep, dogs, wild cats, swine, as well as monkeys worldwide [32,33,34,35].

A. capra, an emerging zoonotic Anaplasma, was first discovered in asymptomatic goats in China [36]. Soon after, 28 out of 477 hospital patients with a history of tick bites in Heilongjiang, China, were infected with A. capra [1]. The disease is widely distributed in several provinces in China and mainly affects sheep, goats and humans. In Liaoning, Henan, and Heilongjiang provinces of China, sheep infected with A. capra were detected [37]. The disease was also detected in goats in seven Chinese provinces/autonomous region: Guizhou, Henan, Inner Mongolia, Shanxi, Xinjiang, Yunnan and Gansu [38]. A new species with zoonotic potential, known as A. capra, has also been discovered in various hosts across European and Asian countries. The presence of A. capra has also been documented in domestic ruminants in Türkiye [39]. In addition to sheep, goats and humans, the pathogen also appears to have a wide range of hosts around the world, including cattle [40] and water deer [41] in Korea, deer [42] in France, water buffalo [43] in Türkiye, and cattle [44] in Kyrgyzstan.

A. platys, the agent of infectious canine cyclic thrombocytopenia (ICCT), is a bacterium parasitized in the canine platelets [21]. Additionally, two clinical cases with single A. platys infection were recently confirmed in two women from Venezuela [45]. These findings further indicate that, in addition to A. phagocytophilum being recognized as zoonotic, A. ovis, A. capra and A. platys are also potential zoonotic pathogens, which can be easily ignored.

There have been many reports about Anaplasma spp. in the central and northern regions of China [46,47,48,49,50,51,52]. However, there have been few studies on Anaplasma infections in southern China. Considering the scarce data regarding the epidemiology of Anaplasma in ruminants in Hainan, China, we conducted a survey in Hainan island/province to obtain this needed information. The aim of this study is to determine the species of Anaplasma in ruminants, and to evaluate the prevalence of Anaplasma in Hainan.

Materials and methods

Study area

This study was conducted in Hainan province (also called Hainan Island), which is located in the South China sea (between 108° 37’ and 111° 03’ E longitude and 18° 10’ and 20° 10’ N latitude) (Fig. 1). This part of the island is renowned for its tropical climate, which is completely different from that of the Chinese mainland. It encompasses 35,400 km2 of land, with an average rainfall between 1000–2600 mm/year (occurring mostly from July to October) and an average annual temperature of 26.5 °C.

Fig. 1
figure 1

The geographical location of the study area (Hainan Province, China) and the distribution sites of different cities and counties. QZ: Qiongzhong, DF: Dongfang, WC: Wenchang, CJ: Changjing, DZ: Danzhou, DA: Dingan, WZS: Wuzhishan, QH: Qionghai, LS: Lingshui, BS: Baisha, CM: Chengmai, BT: Baoting, WN: Wanning, LD: Ledong, TC: Tunchang, LG: Lingao, SY: Shanya

Full size image

Blood sample collection

Between June 2018 and August 2020, whole blood samples were collected from the jugular vein of goats and cattle using EDTA Vacutainer tubes. A total of 731 goat samples were randomly collected from 18 counties in Hainan province, of which 122 were male and 609 were female. Secondly, blood samples were collected from a total of 176 individual cattle from 4 counties, of which 155 were female and 21 were male. The blood samples were stored at − 20 °C until molecular analyses. The sampling locations are shown in Fig. 1. Additionally, the feeding system, sex, and age were recorded. Age was measured according to dentition; goats younger than 1.5 years were considered as young and goats older than 1.5 years were considered to be adults. For cattle, those younger than 3 years were considered young, and those older than 3 years were considered adults. Stall-feeding livestock farming is a system in which animals were well maintained and fed in stall with limited access to land. All procedures to gather samples from animals were approved by Hainan University Institutional Animal Care and Use Committee (HNUAUCC-2019-0000A).

Extraction of whole blood DNA

Frozen EDTA blood samples were thawed and vortexed at room temperature. Using the Blood Genomic DNA Extraction Kit (Sangon Biotech, Shanghai, China), the blood genome extraction procedure was performed according to the manufacturer's instructions, 200 μL of whole blood was aspirated for DNA extraction, and 50 μL of purified DNA was obtained. DNA concentration (ng/μL) and purity were measured using a Nanodrop ND1000 spectrophotometer (Thermo Scientific©, USA) to measure DNA concentration (ng/μL) and purity.DNA samples with a concentartion less than 5 ng/μL were discarded, and whole blood DNA was re-extracted. The DNA was stored at -20 °C.

Nested PCR and single PCR amplification

The primer sequences used in this study are presented in Table 1. For A. phagocytophilum and A. bovis, nested PCR was carried out to amplify the 16S rRNA gene as described previously [53]. The first round of amplification primers were EE1 and EE2 [54]. During the second round, A. bovis specific primers AB1f and AB1r were used that generate a product of 551 bp, and the A. phagocytophilum specific primers SSAP2f and SSAP2r were used to generate a product of 641 bp [53, 55, 56]. For the A. ovis and A. marginale, the partial segment of MSP4 gene was amplified using primers MSP4f and MSP4r [56]. Based on previously published data, PCR was performed using primers groELf and groELr to amplify partial groEL gene of A. capra [57]. All primers mentioned above were synthesized by Shanghai Bioengineering Co. The first round PCR reactions were performed in a final volume of 25 μL containing 13μL Mix enzyme (Novozymes, Nanjing, China) 10 μL of nuclease free water, 0.5 μL of each primer and 1 μL of DNA. Thermal cycling system was as described in a previous report [58] for A. phagocytophilum and A. bovis. The final volume of the second PCR reaction was 50 μL containing 25 μL of LMaxase (Novozymes, Nanjing, China), 21 μL of nuclease free water, 1 μL of each primer and 2 μL of the product of the first PCR reaction. the thermal cycling system was as described in a previous report [55]. For A. ovis, A. marginale and A. capra, the final system of the reaction was 25 μL. thermal cycling reaction conditions were as in Table 2. The PCR products were electrophoresed in a 1.5% agarose gel. The electrophoresis was also performed at a constant voltage of 120 V. When the electrophoretic bands ran to the appropriate position, the electrophoresis instrument was turned off, the agarose gel was removed and placed under a gel imaging UV light source to observe the results and photographed.

Table 1 Primers used for the amplification of 16S rRNA, MSP4 and Gro EL
Full size table
Table 2 PCR amplification conditions of Anaplasma
Full size table

Sequence variance comparison and sequence homology analysis

All Anaplasma positive PCR products were sequenced by Shanghai Bioengineering Sequencing Company. The sequences of the bidirectional sequencing results were spliced using DNAMAN 8.0 gene analysis software and logged into NCBI for sequence alignment. In this study, representative sequences were selected by multiple sequence alignment using the Clustal W program in MegAlign 7.2 (DNAStar, Madison, WI) software, and homology analysis of the sequences obtained in the study with known sequences was performed using this software. The selected sequences obtained in the study were analyzed by sequence difference comparison using DNAMAN 8.0 gene analysis software.

Phylogenetic tree analysis

A phylogenetic tree was constructed using 16S rRNA, gro EL (coding for a heat shock protein) and msp4 (major surface protein 4) gene sequences. Strains from around the world were first obtained from GenBank and screened (i.e., duplicate sequenced, incomplete sequences were discarded and a sequence was selected in a subset of very similar sequences). Comparisons were performed by using the Clustal W in MegAlign software (DNAStar, Madison, WI). Then, the phylogenetic tree was constructed using Mega X software using the Kimura-2-parameter model in the Neighbor-joining (NJ) algorithm, and bootstrop values were obtained using 1000 replicates. Finally, the original tree exported by Mega X was stained and edited, and the software chosen for staining and editing was Figtree (http://tree.bio.ed.ac.uk/software/figtree/).

Statistical analysis

Chi-square (χ2) and logistic regression tests (SPSS software, version 23) were used to assess the association between the molecular prevalence of Anaplasma and risk factors (age, gender, and feeding habits). Results were considered statistically significant for P < 0.05.

Results

Infection of Anaplasma spp. in cattle and goats in Hainan

An epidemiological study on Anaplasma infection in cattle and goats in Hainan province was carried out in Hainan Province from June 2018 to August 2020. A total of 731 goats from the 18 sampling regions in Hainan province were examined for the presence of Anaplasma spp.. Out of these, the samples were tested positive for Anaplasma spp. in 15 counties and negative in 3 counties. The infection rates were higher in Wenchang county (52%) and Chengmai county (76%) compared to other counties. The overall infection rate of Anaplasma spp. in goat samples was 30.1%. Categorically, the infection rates of A. bovis, A. phagocytophilum, A. ovis and A. capra were 22.7% (166), 13.8% (101), 5.2% (38) and 3.4% (25), respectively. A. bovis had the highest positive rate, followed by A. phagocytophilum and A. ovis (Table 3). In cattle samples, the infection rates of A. bovis, A. phagocytophilum and A. marginale were 11.4% (20), 6.3 (11), 5.7(10), respectively, with an overall infection rate of Anaplasma spp. at 14.8% (26) (Table 4). A. bovis had the highest positive rate, followed by A. phagocytophilum. A. orvis was only found in goats, and A. marginale was only found in cattle (Table 2). In both goat and cattle samples, the infection rates of A.bovis and A. phagocytophilum were relatively high, while A. ovis and A. capra were detected only in goat samples, and A. marginale was found in cattle. A. capra, a recently reported species of Anaplasma, is a new zoonotic pathogen that has been detected only in Chengmai county (Table 3).

Table 3 Anaplasma positivity rates in goats from Hainan, China
Full size table
Table 4 Anaplasma positivity rates in cattle from Hainan, China
Full size table

Mixed infection of Anaplasma spp.

The rate of infection with two or more Anaplasma spp. was 11.8% (86/731) in the731 goatsincluded in this study. Co-infection with A. bovis and A. phagocytophilum was 8.2% (60/731), followed by A. bovis + A. ovis (0.5%, 4/731), A. bovis + A. capra (0.4%, 3/731), A. phagocytophilum + A.ovis (0.1%, 3/731), and A. bovis + A. capra + A. phagocytophilum (2.0%, 15/731) (Table 5). In cattle, co-infections with two or three species of Anaplasma spp. were observed, with a positive co-infection rate of 6.8% (12/176) (Table 5). Specifically, the infection rate of A. bovis + A. phagocytophilum was 2.0% (3/176), and the infection rate of A. bovis + A. phagocytophilum was 1.1% (2/176). The infection rate of A. phagocytophilum + A. marginale was 0.6% (1/176), and the infection rate of A. bovis + A. phagocytophilum + A. marginale was 3.4% (6/176) (Table 5).

Table 5 Co-infection rates in goats and cattle samples
Full size table

Prevalence distribution of Anaplasmosis by age, gender, and feeding mode in cattle and goats

According to the age, gender, and feeding habits, the infection of Anaplasma spp. in goats and cattle were investigated. A total of 37 male goats and 183 female goats were infected. The infection rate among male goats and female goats was 8.2% (37/122), 30.0% (183/609), respectively. There were differences in prevalence between the age groups and between the feeding habits. The infection rates were 20.0% (39/195), 30.2% (76/252), 36.8% (105/284) in goats younger than 1.5 years old, between 1.5 to 3 years old and older than 3 years old, respectively. Goats that were > 3 years old had the highest prevalence (36.8%) than those of other age groups (OR = 5.3, CI = 3.2–8.9, P < 0.001). The situation of Anaplasma spp. Infection differed between free-range and captive feeding modes of goats. The infection rate in free-range goats was 39.7% (115/390), significantly higher than that in captive goats (19.1%, 61/341) (OR = 4.3, CI = 3.0–6.3, P < 0.001).

For cattle, the infection rates in male and female were 14.3% (3/21) and 14.8% (23/155), respectively. The rates of Anaplasma spp. infections were 14.3% (9/63), 15.1% (13/86) and 11.1% (3/27) in cattle younger than 3 years, between 3 to 6 years, and older than 6 years, respectively. In the univariable statistical analysis, the feeding mode was associated with Anaplasma spp. infection, while no significant association was found between gender, age, and the prevalence of Anaplasma spp. (P > 0.05). Furthermore, the infection rate in free-range cattle was 73.7% (11/15), higher than that in the captive cattle (19.1%, 61/341) (OR = 28.4, CI = 7.7–104.9, P < 0.001) (Table 6).

Table 6 Anaplasma infection rates with respect to genders, ages, and feeding patterns
Full size table

Sequence and phylogenetic analysis

Phylogenetic and sequence analysis of A. phagocytophilum based on 16S rRNA gene

In this study, 112 strains of A. phagocytophilum 16S rRNA sequences (641 bp) were obtained, of which 9 sequences contained polymorphic sites. Among them, 4 strains were derived from goats (CMAP30, BXAP37, HRAP1, ZQAP10), and 5 strains were derived from cattle (NDAAP2, RYAP3, RYAP4, RYAP6, XYAP2). The sequences of these 9 strains were compared with the reference strains (KJ782381), revealing 2 ~ 11 base site differences (Table 7).

Table 7 16S rRNA gene of A.phagocytophilum and their sequence base site differences
Full size table

Homologous sequence analysis of A. phagocytophilum showed that the 9 A. phagocytophilum 16S rRNA sequences obtained in this study were compared with reference strains from China (KT944029, KJ782381, KU321298, KF569915, MG002405) and different countries’ reference strains, including South Africa (KU870667), Pakistan (MN216240), Japan (AB196720, AB196721), ranged from 97.4% to 100%. Meanwhile, the homology range with A. bovis (MH255938) was 95.6% ~ 98.8%, A. ovis (KJ459342) was 40.5% ~ 41.1%, and Rickettsia (JX885456) was 30.1% ~ 30.9%. A. platys (KU500907) has a homology range of 44.1% ~ 44.9%, and A. capra (MT052418) had a homology range of 44.1% ~ 44.7%. A. centrale (MH588233) had a homology range of 44.1% ~ 44.9% (Fig. 2).

Fig. 2
figure 2

Sequences homology analysis of A. phagocytophilum based on 16S rRNA gene

Full size image

The 9 A. phagocytophilum 16S rRNA sequences obtained in this study were compared with 27 known A. phagocytophilum 16S rRNA sequences from different regions (AB196720, AB19672, GQ175174, JN558812, JN558816, JN990105, KC246018, KF569915, KJ782381, KJ782386, KP062963, KP276588, KR002114, KT944029, KU321298, KU870667, KX083402, KX236051, KX236051, KX450278, LC060986, LC060987, MF992253, MG002405, MN097858, MN216240, NR044762), as well as Rickettsia (JX885456), A. ovis (KJ459342), A. platys (KU500900, KU500914, MH255941), A. bovis (MH255928, MH255938), A.centrale (MH588232, MH588233) and A. capra (MH762071, MT798602, MT052418) were used as outgroups to construct the phylogenetic tree. Phylogenetic tree analysis of A. phagocytophilum showed that the sequences from four goat strains (CMAP30, HRAP1, ZQAP10, BXAP37) and one bovine strain (NDAAP2) were grouped into one clade. The other clade from four bovine strains (XYAP2, RYAP3, RYAP4, and RYAP6) was isolated into a group. In this study, four types sequence derived from goat CMAP30, BXAP37, HRAP1 and ZQAP1 were compared with those from Japanese wild deer (AB196720), South African water giraffe (KU870667), Chinese dog (KX632051) and Zhejiang breed goat (JN558812), respectively. CMAP30 isolates had the longest genetic distance. In addition, the sequence of the bovine NDAAP2 strain in this study was similar to that of the bovine reference strain isolated from Pakistan (MN216240) (Fig. 3).

Fig. 3
figure 3

The phylogenetic analysis of the A. phagocytophilum 16S rRNA gene sequence (541 bp) by the neighbor-joining method. The number on each node corresponds to 1000 repeated guided analyses (only represents a percentage greater than 60%). The amplified sequences in this study are highlighted with colored fonts, different colored fonts indicate different animals (blue: goat;green: cow), and different branches are displayed with different colored backgrounds

Full size image

A. bovis sequence analysis and phylogenetic tree analysis

In this study, partial nucleotide sequences (551 bp) of 16S rRNA of 168 strains A. bovis were obtained. All the obtained A. bovis sequences were compared, and 13 A. bovis strains were screened out as having sequences representing 16S rRNA. Among them, 6 strains were derived from goats (CMAB16, YLAB7, CMAB24, LSAB11, HKAB10, BSAB89) and 7 strains were derived from cattle (RYAB3, RYAB6, RYAB7, RYAB9, XYAB2, DAAB145, DAAB155). The 13 A. bovis representative sequences were compared with the known reference strain (MN309843), and the results showed that there were differences in 1 to 4 base sites (Table 8).

Table 8 16S rRNA gene of A.bovis and the sequence base site differences
Full size table

A. bovis homology analysis showed that the 13 A. bovis 16S rRNA sequences obtained in this study were compared with the A. bovis sequences from China and abroad (KU509992, KU509996, KX450273, KY242455,LC432092, MH255927, MH255936, MH255939, MN044717, MN309842, MT036513, MK028574) were analyzed with homology in the range of 95.5% ~ 99.8%. Meanwhile, the homology with A. ovis (KC484562) from Russia, A. platys (KU500900) from Malaysia, and A. capra (MT798602) from Korea were 95.7% ~ 96.9%, 96.9% ~ 99.8%, 95.5% ~ 96.3% respectively. The homology with Rickettsia (JX885456) from other area of China was 44.1% ~ 44.5% (Fig. 4).

Fig. 4
figure 4

Sequences homology analysis of A. bovis based on 16S rRNA gene

Full size image

In this study, 13 partial and 18 intact gene sequences of A. bovis 16S rRNA isolates from different regions were known (KU509992, KU509996, KX450273, KY242455, MH255927, MH255934, MH255935, MH255936, MH255939, MH255940, MH255941, MK028572, MK028574, MK446832, MN044717, MN309842, MN309843, MT036513). 16S rRNA gene sequences from Rickettsia (JX885456), A. ovis (KC484562, MH795156), A. platys (KU500900, KU500914) and A. capra (LC432092, MH762071, MT052418, MT798602) were used as outgroups to construct the phylogenetic tree. A. bovis phylogenetic tree analysis showed that A. bovis sequences from Hainan were clustered into a group with isolates from other provinces of China and other Asian countries (including Malaysia, Russia, and Korea), supported by 94% nodes (Fig. 5). In addition, 4 strains from goats (MAB16, YLAB7, CMAB24, LSAB11) and 2 strains from cattle (RYAB3, RYAB7) formed two separate subgroups on the same branch with 70% and 60% support on the branch nodes, respectively (Fig. 5).

Fig. 5
figure 5

The phylogenetic analysis of the 16S rRNA gene sequence (551 bp) of A. bovis by the neighbor joining method. The number on each node corresponds to 1000 repeated guided analyses (only represents a percentage greater than 60%).The amplified sequences in this study are highlighted with colored fonts, different colored fonts indicate different animals (blue: goat; green: cow), and different branches are displayed with different colored backgrounds. In addition, indicate the host or medium, country of origin, and GenBank accession number. Nine gene sequences of other strains as outgroups

Full size image

Sequence and phylogenetic tree analysis of A. capra

In this study, we obtained partial sequences of 25 strains A. capra Gro EL genes (878 bp), and found that they have only one sequence type through multiple sequence alignment. Homology analysis of A. capra showed that 9 strains (CMAG14, CMAG24, CMAG34, CMAG35, CMAG38, CMAG39, CMAG40, CMAG42, CMAG43) were selected and compared with those from other areas of China (MH716420, MH714931, MG940875, MG869454, MG869415, MG869416, MG869387, MG869388, MG869389, KX987394, KX417341) and from Korea (LC432173, LC432182, LC43218, LC4321824, MT721150) have a high homology of 99.8% to 100%. Meanwhile, the homology with other types of reference strains of A. bovis (MH255905), A. ovis (MG778623), A. centrale (KY522999), A. marginale (JQ839013), Ehrlichia sp. (U96731) was 76.7%, 81.6%, 82%, 81.4% and 75%, respectively (Fig. 6).

Fig. 6
figure 6

Sequences homology analysis of A. capra based on Gro EL gene

Full size image

In this study, we obtained 9 strains A. capra Gro EL gene partial sequences and 17 strains A. capra GroEL gene reference sequences (LC432182, LC432184, MG869387, MG869389, MG869399, MG869415, MG869416, MG869454, MG940875, MH174931, MT721150, LC432183, LC432173, KX417341, KX417341, KX987393, MH716420) from different regions, and GroEL gene sequences of 2 strains A. centrale (KY522999, KY523000), 3 strains A. marginale (JQ839013, JQ839014, KY523034), 4 strains A. ovis (FJ460434, MG383905, MG778623), 2 strains A. platys (KU585953, KU585953) and 3 strains A. bovis (MH255905, MH255906, MH255907), were used as outgroups to construct phylogenetic trees.

The phylogenetic tree analysis of A. capra showed that 9 strains A. capra sequences and 17 strains reference strains all clustered in the same group in this study, and the support rate of branch nodes was as high as 100%. In addition, the genetic distance between CMAG43 sequence and other sequences was the longest (Fig. 7).

Fig. 7
figure 7

The phylogenetic analysis of the A. capra Gro EL gene sequence (878 bp) was performed using the neighbor-joining method. The number on each node corresponds to 1000 repeated guided analyses (only represents a percentage greater than 60%). The sequences amplified in this study are shown in red font. In addition, the GenBank accession number of he sequence used for phylogenetic analysis is also shown. Thirteen gene sequences of other strains as outgroups

Full size image

Sequence and phylogenetic tree analysis of A. ovis

In this study, 38 strains A. ovis MSP4 partial gene sequences (851 bp) were obtained from goats, and 10 different sequences (TCAO11, TCAO40, QHAO19, SZAO26, LDAO19, EMAO14, DZAO1, CJAO4, CJAO16, BXAO11) were screened. The reference strain sequence (LC141091) from goats was compared with the selected representative sequences of 10 strains of A. ovis, and the results showed that there were differences in 3–20 base positions (Table 9).

Table 9 MSP4 gene of A. ovis and the sequence base site differences
Full size table

The homology analysis results of A.ovis showed that the partial gene sequences of 10 strains A. ovis MSP4 obtained in this study were homologous to those of other Chinese strains (MG668814, HQ456350, MG283274), Turkish sheep (KY283958), Mongolian goats (LC141080, LC141091) and Sudan goats (KU497709), and the homology was 94.0% -99.0%. At the same time, the homology comparison with different species of Anaplasma, A. marginale (KX17990) from Algeria, A. phagocytophilum (KM205427) from Slovenia, A. capra (LC432231) from South Korea and A. centrale (KY305621) from South Africa were 34.4%—35.2%, 55.2%—56.4%, 66.3%—67.2% and 59.0%—60.0%, respectively (Fig. 8).

Fig. 8
figure 8

Sequences homology analysis of A. ovis based on MSP4 gene

Full size image

In this study, we obtained 10 strains A. ovis MSP4 gene partial sequences, and 18 strains A. ovis MSP4 gene reference sequences (GQ925816, HQ456350, KC432643, KU497709, KU497710, KU497712, KY283958, LC141077, LC141080, LC141081, LC141091, MG283274, KP608305, MG668814, MG564176, MN198191, MK358053, MH790274) from different regions, and MSP4 gene sequences of 3 strains A. centrale (KY305601, KY305620, KY305621), 2 strains A. marginale (KX179906, KX179906), 2 strains A. capra (MK83607, LC432231) and 1 strains A. phagocytophilum (EU008082), were used as outgroups to construct phylogenetic trees. The phylogenetic tree analysis of A. ovis showed that A. ovis, A. centrale, A. marginale, A. capra and A. phagocytophilum independently clustered into different clades.

In this study, 10 A. ovis sequences and 18 strains known reference sequences were clustered into the same group, and the branch node support rate was 82.5%. However, the genetic distance between CJAO4 and other sequences is far (Fig. 9).

Fig. 9
figure 9

The phylogenetic analysis of A. ovis MSP4 gene sequence (851 bp) by the neighbor joining method. The number on each node corresponds to 1000 repeated guided analyses (only represents a percentage greater than 60%). The amplified sequences in this study are shown in blue font. In addition, the GenBank accession number of the sequence used for phylogenetic analysis is also shown. 11 gene sequences of other strains as outgroups

Full size image

Analysis of A. marginale sequence and phylogenetic tree

Ten A. marginale MSP4 gene sequences (DAAM9, DAAM95, RYAM2, RYAM3, RYAM 4, RYAM5, RYAM6, RYAM7, RYAM9, XYAM4) were obtained in this study. The sequences of the 10 A. marginale MSP4 strains obtained in this study were compared with the known reference strain (MK809386), and the results showed that there were differences at 3 ~ 9 base positions. The results showed that there were differences in 3 ~ 9 base positions (Table 10).

Table 10 MSP4 of A. marginale and the sequencebase site differences
Full size table

In the present study, 10 A. marginale MSP4 partial gene sequences were analyzed for homology with reported domestic and international sequences (MK809386, KX989513, KX989516, AY665997, AF428082), and their homology ranged from 98.3.0% to 99.1%. The homology analysis with A. phagocytophilum (EU180058), A. centrale (KY305621) and A. ovis (KU499307) was also performed, and the homology ranged from 62.5% ~ 62.8%, 72.5% ~ 72.9%, and 89.7% ~ 90.5%, respectively (Fig. 10).

Fig. 10
figure 10

Sequences homology analysis of A. marginale based on MSP4 gene

Full size image

The partial sequences of the 10 A. marginale MSP4 genes obtained in this study were compared with the known sequences of 27 A. marginale MSP4 genes from different regions (AF428082, AF428086, AY283190, AY456002, AY665997, AY665999, AY851150, EF053264, EU283844, EU677383, JN564646, KX989516, KX989512, KX989513, MG676453, MG676455, MG676459, MH026093, MH172467, MH373246. MH939155, MK809379, MK809381, MK809384, MK809386, MK809387, MT268094), two strains of A. ovis (HM063433, KU497703), three strains of A. centrale (KY305601, KY305604, KY305621) and three strains of A. phagocytophilum (EU008082, EU180058, MF974857) were used as outgroups for MSP4 gene sequences to construct phylogenetic trees. Phylogenetic tree analysis based on the MSP4 gene showed that the MSP4 gene sequences of A. marginale, A. phagocytophilum, A. centrale and A. ovis were independent and clustered into different branches. The sequences of 10 A. marginale strains obtained in this study clustered into one taxon with the known reference sequences of 27 A. marginale strains, with 100% support of branching nodes, among which DAAM9 strains were genetically distant (Fig. 11).

Fig. 11
figure 11

The phylogenetic analysis of A. marginale MSP4 gene sequence (851 bp) by the neighbor joining method. The number on each node corresponds to 1000 repeated guided analyses (only represents a percentage greater than 60%). The amplified sequences in this study are shown in blue font. In addition, the GenBank accession number of the sequence used for 11 gene sequences of other strains as outgroup

Full size image

Discussion

It has been reported that Anaplasma spp. are widely distributed in China, and their DNA has been found in a variety of wild and domestic animals and ticks [37, 50, 59,60,61,62]. This study was conducted for molecular detection and phylogenetic analysis of Anaplasma pathogens based on 16S rRNA, Gro EL and MSP4 as marker genes in ruminants from Hainan province/island. This study reported the prevalence of anaplasmosis in ruminants in Hainan province/island.

The overall prevalence of anaplasmosis in cattle and goats in Hainan province was recorded as 27.1% (246/907), which was lower than the prevalence reported in small ruminants in Anhui province in central China 33.0% (67/203). The prevalence in goats and cattle was 30.1% (220/731) and 14.8% (26/176), respectively, which were higher than the prevalence reported in cows in Xinjiang in the north west of China (3.2%, 16/493), but lower than those reported in sheep in Heilongjiang in the north of China (30.2%, 103/341). The prevalence of anaplasmosis in ruminant was also reported in other Asian countries, including cattle in Iran, 15.5% [63], cattle in Malaysia, 60.7% [64], cattle in Mozambique, 86.3% [65], Cattle in Algeria, 42.2% [66], sheep, 35.6% and goats, 46% in northern Tunisia [67]. These differences may be due to different geographical areas, climatic conditions [68], different diagnostic methods used, seasons of sample collection, tick infestation intensity, and animal breeds with different susceptibility to pathogens.

In this study, A. bovis was more prevalent than other Anaplasma spp. which is consistent with previous studies from different parts of China. Previous studies have shown that the prevalence of A. bovis varied in other areas of China, ranging from 4.5% to 49.0% [58, 61, 69,70,71,72]. The feeding method was also significantly associated with the prevalences. In this study, the prevalence in ruminants in free-range farming was higher than that in intensive rearing system. This result was consistent with another study that reported animals browsing outdoor were at higher risk than the animals fed indoor [73]. It may be due to conditions controlled in large-scale farming environments that reduce the incidence of tick bites. Under free range conditions, there is an increased incidence of tick bites, which may lead to a higher rate of Anaplasma spp. infection. The prevalence of Anaplasma spp. infection was higher in old animals, > 3 years (36.8%, 105/204) than in younger animals, < 1.5 years of age (20%, 39/195) and those between 1.5 and 3 years of age (30.2%, 76/252). The difference in the prevalence in goats at different ages was possibly because older goats had a higher chance of being bitten by ticks.

A. phagocytophilum has been recognized as a zoonotic pathogen. In 1994, A. phagocytophilum was first identified as the pathogen of HGA in the United States. In China, the first suspected human case was reported in Anhui province in 2006. Recently, A. capra, a newly reported species originally found in goats; caused 28 infections in people reported in Heilongjiang province in China in 2015 [1, 36]. An Anaplasma sp. was first detected in a Cyprus patient with fever, hepatosplenomegaly, and lymph node disease in 2006 [74]. In addition, two cases of flat anplasmosis were detected in two women in Venezuela [45]. To date, besides A. phagocytophilum infections in humans, there have been relatively few reports of A. capra, A. ovis and A. platys human infections. However, they are all potential zoonotic pathogens, which are easily overlooked in disease prevention [75]. The study aimed to investigate the presence, prevalence, and genotypes of A. phagocytophilum, A. ovis, and A. capra in sheep from Kyrgyzstan. Polymerase chain reaction (PCR), restriction fragment length polymorphism (RFLP), and sequencing techniques targeting the 16S SSU rRNA, groEL, and gtlA genes were employed. The findings revealed, for the first time, the presence of A. phagocytophilum-like 1, A. ovis, and A. capra in sheep from Kyrgyzstan. The positivity rates for A. phagocytophilum-like 1, A. ovis, and A. capra genotype-1 were 6.9%, 22.5%, and 5.3%, respectively. It was also observed that A. capra exhibits two distinct genotypes, namely A. capra genotype-1 and A. capra genotype-2 [76].

In the past, anaplasmosis in ruminants was a blood borne disease that was easily overlooked, but in recent years, with the widespread increase in incidence trends, it will be increasingly valued. Clinical symptoms of cattle and sheep infected with Anaplasma include anemia, emaciation, jaundice, and other clinical symptoms, sometimes leading to the death of cattle and sheep, causing serious economic losses in pastoral areas.

A. Phagocytophilum is a zoonotic pathogen, and HGA poses a serious threat to human health. This study aims to investigate the 16S rRNA/Gro EL/MSP4 gene of this pathogen, which can help to trace its transmission and understand whether there are host differentiation and geographical differentiation characteristics of the Anaplasma pathogen in Hainan region, thus providing favorable reference materials for the prevention and control of this disease.

Conclusion

This study investigated the molecular epidemiology of anaplasmosis in ruminants in Hainan province/island, China. The total infection rate in goats was 30.1%. The infection rates of A. bovis, A. phagocytophilum, A. ovis and A. capra were 22.7%, 13.8%, 5.2% and 3.4%, respectively. The infection rate in goats with two or more species of Anaplasma was 11.8%. The rate of Anaplasma infection in cattle was 14.8%. The infection rates of A. bovis, A. phagocytophilum and A. marginale in cattle were 11.4%, 6.3% and 5.7%, respectively. The co-infection rate of Anaplasma in cattle was 11.8%. The results showed that Anaplasma was prevalent in ruminants in Hainan province, China and co-infection was common. In this study, 16S rRNA/ Gro EL/MSP4 gene phylogenetic tree analysis showed that A. phagocytophilum, A. bovis, A. ovis, A. capra, and A. marginale did not exhibit characteristics of geographical isolation and differentiation from isolates in different regions in Hainan. However, the phylogenetic tree of A. phagocytophilum revealed distinct branches associated with different hosts, suggesting host differentiation. Hainan province is located in tropical and subtropical regions, which are more suitable for the growth and reproduction of certain tick species. However, the specific tick species carrying Anaplasma spp. in Hainan are unknown. This study conducted a molecular epidemiological investigation on Anaplasma spp. in ruminants in Hainan, providing a favorable basis for understanding the spread, prevention, and treatment of these infections in the province.

In conclusion, anaplasmosis represents a potential threat to the ruminant husbandry in Hainan, and our studies have also shown that co-infection of Anaplasma spp. is common. The present study demonstrated that a significant proportion of cattle and goats infected with Anaplasma spp., even though none of the animals showed clinical symptoms. These animals act as carriers of the bacteria. It is important for farmers, local veterinarians, and the local government in Hainan to take preventive measures against anaplasmosis in domestic animals.