Molecular Characterization of Anaplasma spp. among Dairy, Cashmere, and Meat Goats in Shaanxi Province, Northwestern China

Simple Summary Anaplasmosis is an important tick-borne disease caused by Anaplasma spp., significantly threating public health safety and breeding industry. The present study reported the occurrence of Anaplasma infection among dairy, cashmere, and meat goats in Shaanxi province, northwestern China, with the total prevalence of 58.5% (298/509) in goats. Anaplasma phagocytophilum, A. bovis, and A. ovis were the dominant species in meat, dairy, and cashmere goats, respectively, with the absence of A. ovis in meat goats. Furthermore, the different influencing factors (production categories, species, regions, and ages) were analyzed, and statistically significant differences were found. Frequent occurrence of Anaplasma in this study indicated One Health-based intervention approaches were urgently needed to block the transmission between humans and animals. Abstract Anaplasma spp. are important tick-borne pathogens endangering the health of humans and various animals. Although several studies have reported Anaplasma infection in livestock in China, little is known about the impact of production categories on the occurrence of Anaplasma species. In the present study, PCR tools targeting the 16S rRNA and msp4 genes were applied to investigate the prevalence of Anaplasma spp. in 509 blood samples of dairy (n = 249), cashmere (n = 139), and meat (n = 121) goats from Shaanxi province. The prevalence of Anaplasma spp. was 58.5% (298/509) in goats, and significant differences (p < 0.001) were identified in the prevalence among production categories, with the highest in meat goats (84.3%, 102/121), followed by cashmere goats (58.3%, 81/139) and dairy goats (46.2%, 115/249). Significant differences (p < 0.001) in prevalence were also found among sampling sites and age groups. Meanwhile, the prevalence was 36.9% (188/509) for A. phagocytophilum, 36.1% (184/509) for A. bovis, and 11.0% (56/509) for A. ovis, and significant differences (p < 0.001) in prevalence of A. phagocytophilum, A. bovis and A. ovis were recognized among production categories and sampling sites. A. phagocytophilum, A. bovis and A. ovis were dominant species in meat, dairy, and cashmere goats, respectively, and A. ovis was absent in meat goats. Co-infections were found in 124 (24.4%) investigated samples. Goats aged < 2, 3–6, and 7–12 months, and goats from Qingjian and Zhenba were risk factors associated with the occurrence of Anaplasma. Phylogenetic analysis indicated separate clades for the distribution of A. phagocytophilum from different ruminant, reflecting potential host adaption within this species. This study reported the colonization occurrence of Anaplasma spp. among production categories in goats in Shaanxi province and enriched our knowledge on the transmission of Anaplasma spp. in goats in China. Considering the existence of zoonotic A. phagocytophilum in goats in this study and previous reports, interventions based on One Health are needed to be developed to control the transmission of Anaplasma spp. between humans and animals.


Sampling
To investigate the prevalence and species composition of Anaplasma in meat, cashmere, and dairy goats in Shaanxi province, we collected samples from representative farms in seven main breeding sampling sites, with 10~20% of animals sampled for each farm ( Figure 1). From April to July 2017, a total of 509 blood samples of dairy (n = 249), cashmere (n = 139), and meat (n = 121) goats were collected from 15 representative farms (Table 1). Due to the larger scale of dairy goats compared with other two production categories, we collected more farms and more samples from dairy goats (Table 1). Blood samples were collected from the jugular vein of each animal and placed into separate tubes containing 1.5% Ethylenediaminetetraacetic acid tripotassium (EDTA-3K) with basic information (e.g., sampling sites, breeds, and ages), immediately transported to the department of parasitology of Northwest A&F University under cool condition, and then kept at 4 • C for further analysis. department of parasitology of Northwest A&F University under cool condition, and then kept at 4 °C for further analysis.  Good: good animal husbandry practice with regular usage of drugs against ectoparasite; Medium: medium animal husbandry practice with random usage of drugs against ectoparasite; Poor: poor animal husbandry practice with little usage of drugs against ectoparasite.  Good: good animal husbandry practice with regular usage of drugs against ectoparasite; Medium: medium animal husbandry practice with random usage of drugs against ectoparasite; Poor: poor animal husbandry practice with little usage of drugs against ectoparasite.

Genomic DNA Extraction
The Blood Genomic DNA Isolation Kit (Sangon Biotech, Shanghai, China) was applied to extract genomic DNA (gDNA) samples from approximately 100 µL blood of each sample according to the procedures of the manufacturer, and the gDNA samples were kept at −20 • C until further analysis. Meanwhile, Anaplasma negative blood samples preserved in our lab were also applied for genomic DNA extraction to be used as negative control in PCR amplification.

PCR Amplification
The occurrence of A. phagocytophilum and A. bovis was identified using nested-PCR targeting a~551 bp fragment and a~641 bp fragment of the 16S rRNA gene as reported, respectively [25,26] (Table 2). Nested PCRs were carried out in a 25 µL reaction mixture containing 1 × Ex Taq Buffer (Mg 2+ free), 2 mM MgCl 2 , 0.2 mM dNTP Mixture, 1 U TaKaRa Ex Taq, 0.4 µM each primer, 1 µL gDNA for the primary PCR, or 1 µL primary PCR product for the secondary PCR under the following conditions for both primary and secondary PCRs: an initial denaturing at 94 • C for 5 min, followed by 35 cycles of 94 • C for 30 s, 55 • C for 1 min, and 72 • C for 1 min, and a final extension at 72 • C for 10 min. A. ovis DNA was identified by PCR amplification targeting a~867 bp fragment of the msp4 gene [27] ( Table 2). PCRs were conducted in a 25 µL reaction mixture containing 1 × Ex Taq Buffer (Mg 2+ free), 2 mM MgCl 2 , 0.2 mM dNTP Mixture, 1 U TaKaRa Ex Taq, 0.4 µM each primer, and 1 µL gDNA under the following conditions: an initial denaturing at 94 • C for 30 s, followed by 40 cycles of 94 • C for 30 s, 60 • C for 30 s, and 68 • C for 1 min, and a final extension at 68 • C for 10 min. A negative control without Anaplasma was used in each PCR amplification. Positive PCR products of the msp4 gene and secondary nested-PCR products of the 16S rRNA gene will appear a band of the expected size under a UV transilluminator after 1% agarose gel electrophoresis. A Good Laboratory Practice was followed to avoid contamination in each step during the whole experiment [28].

Sequencing and Sequence Analysis
All positive amplicons were sequenced at forward direction by Sangon Biotech (Shanghai, China) using an ABI PRISM 3730XL DNA Analyzer (Applied Biosystems, Bedford, MA, USA). The obtained sequences were identified to be A. phagocytophilum 16S rRNA gene, A. bovis 16S rRNA gene, or A. ovis msp4 gene by BLAST analysis within NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi; accessed on 15 April 2021). To assess the phylogenetic placement of Anaplasma species found in this study, phylogenetic trees were constructed by using the neighbor-joining (NJ) method with the Kimura 2-parameter model and the calculation of substitution rates with the bootstrap evaluation of 1000 replicates within software MEGA V6.0 [29].

Statistical Analysis
Data analysis was conducted using the package R V4.0.5 (https://www.r-project. org/; accessed on 7 July 2021) and RStudio V1.1.463 (http://rstudio.com/; accessed on 7 July 2021). A chi-square test was applied to analyze differences in prevalence of Anaplasma spp. by production categories, age groups, or sampling sites. Logistic regression was used to analyze the association between risk factors and the occurrence of Anaplasma spp. in dairy, cashmere, and meat goats. Initially, univariate analysis was used to assess the strength of association. Then, a multivariate model was built using variables with p ≤ 0.2, with p < 0.05 being recognized as significant in the final model. Odds ratio (OR) with 95% confidence intervals (CI) was analyzed for the identification of risk factors for the occurrence of Anaplasma spp. in goats.

Nucleotide Sequence Accession Numbers
Representative nucleotide sequences of A. phagocytophilum 16S rRNA gene, A. bovis 16S rRNA gene, and A. ovis msp4 gene in the present study were available in Gen-Bank™ under the accession numbers of MZ489423-MZ489426, MZ489427-MZ489430, and MZ502497-MZ502499, respectively.
To assess the genetic diversity of A. phagocytophilum in goats, all the positive samples for A. phagocytophilum at the 16S rRNA locus were sequenced, and a total of 184 sequences were obtained in the present study. Based on the BLAST analysis at NCBI and sequence alignment in MEGA V6.0, four types of A. phagocytophilum 16S rRNA sequences, namely LY05 (6), LY49 (3), LY75 (170), and LY89 (5), were identified in goats, with 99.2-99.6% sequence identity among these sequence types. Phylogenetic analysis indicated that four A. phagocytophilum 16S rRNA sequences in the present study and other reference sequences from goats were included in the same clade, while sequences from cattle and sheep formed other two separated clades (Figure 2).
Further, a total of 184 samples positive for A. bovis were successfully sequenced, and the obtained 184 sequences formed four sequence types, namely LA28 (4), LB03 (2), LY02 (172), and LY88 (6), with 98.7-99.7% sequence identities to each other. Phylogenetic analysis indicated that A. bovis samples from goats and cattle were included in the same clade
Further, a total of 184 samples positive for A. bovis were successfully sequenced, and the obtained 184 sequences formed four sequence types, namely LA28 (4), LB03 (2), LY02 (172), and LY88 (6), with 98.7-99.7% sequence identities to each other. Phylogenetic analysis indicated that A. bovis samples from goats and cattle were included in the same clade ( Figure 3).
Based on the PCR-sequencing of A. ovis msp4 gene, a total of three sequence types, namely MZ16 (30), MZ25 (18), and SDC23 (8), were recognized in the obtained 56 sequences, with 99.7-99.8% sequence similarity among these sequence types. Further phylogenetic analysis based on the msp4 gene indicated that MZ16, MZ25, SDC23, and reference sequences were included in the same clade, showing high sequence similarity among A. ovis from divergent areas in China (Figure 4).
Based on the PCR-sequencing of A. ovis msp4 gene, a total of three sequence types, namely MZ16 (30), MZ25 (18), and SDC23 (8), were recognized in the obtained 56 sequences, with 99.7-99.8% sequence similarity among these sequence types. Further phylogenetic analysis based on the msp4 gene indicated that MZ16, MZ25, SDC23, and reference sequences were included in the same clade, showing high sequence similarity among A. ovis from divergent areas in China (Figure 4).

Discussion
Anaplasma spp. are important zoonotic pathogens with worldwide distribution in humans and various animals [4,8,19,25,[30][31][32][33]. To understand the occurrence of Anaplasma spp. among production categories, the present study investigated the prevalence of A. phagocytophilum, A. bovis, and A. ovis in dairy, cashmere, and meat goats in seven sampling sites within five main breeding areas in Shaanxi, northwestern China by using PCR-sequencing tools based on the 16S rRNA gene and the msp4 gene. Of the 509 blood samples, a total of 298 (58.5%) samples were positive for Anaplasma spp. in goats in the present study, and statistically significant differences were found in the prevalence of Anaplasma spp. among production categories.
A. phagocytophilum is recognized as one emerging tick-borne pathogen of public health in humans [34]. Besides humans, a variety of animals, e.g., cattle, sheep, goats, deer, horses, cats, dogs, and rats are also susceptible to this pathogen, and infected hosts can show a series of symptoms, including high fever, anorexia, and weight loss [20,24,[35][36][37][38]. In the present study, the prevalence of A. phagocytophilum in goats was 36.9% (188/509) ( Table 2), which was consistent with that in goats in Gansu (38.5%) [39], but higher than that in other provinces in China (5.7-30.8%) [18][19][20][21]24]. Recently, one study reported a higher average prevalence of 71.5% for A. phagocytophilum in goats in Weinan city, Shaanxi [22]. The differences in the occurrence rates of A. phagocytophilum in goats were likely affected by multiple factors, including different detection methods, geographic regions, sampling sizes, and animal management practice [40,41].
Previously, several reports indicated the existence of differences in host infectivity among divergent A. phagocytophilum strains [42][43][44]. Phylogenetic analysis indicated that four A. phagocytophilum 16S rRNA sequences in the present study and other reference sequences from goats were included in the same clade, while sequences from cattle and sheep formed two separated clades, suggesting the possible host adaptation for the distribution of A. phagocytophilum in these domesticated animals ( Figure 2). Similar results have also been identified in A. phagocytophilum-positive samples from sheep and cattle in one

Discussion
Anaplasma spp. are important zoonotic pathogens with worldwide distribution in humans and various animals [4,8,19,25,[30][31][32][33]. To understand the occurrence of Anaplasma spp. among production categories, the present study investigated the prevalence of A. phagocytophilum, A. bovis, and A. ovis in dairy, cashmere, and meat goats in seven sampling sites within five main breeding areas in Shaanxi, northwestern China by using PCR-sequencing tools based on the 16S rRNA gene and the msp4 gene. Of the 509 blood samples, a total of 298 (58.5%) samples were positive for Anaplasma spp. in goats in the present study, and statistically significant differences were found in the prevalence of Anaplasma spp. among production categories.
A. phagocytophilum is recognized as one emerging tick-borne pathogen of public health in humans [34]. Besides humans, a variety of animals, e.g., cattle, sheep, goats, deer, horses, cats, dogs, and rats are also susceptible to this pathogen, and infected hosts can show a series of symptoms, including high fever, anorexia, and weight loss [20,24,[35][36][37][38]. In the present study, the prevalence of A. phagocytophilum in goats was 36.9% (188/509) ( Table 2), which was consistent with that in goats in Gansu (38.5%) [39], but higher than that in other provinces in China (5.7-30.8%) [18][19][20][21]24]. Recently, one study reported a higher average prevalence of 71.5% for A. phagocytophilum in goats in Weinan city, Shaanxi [22]. The differences in the occurrence rates of A. phagocytophilum in goats were likely affected by multiple factors, including different detection methods, geographic regions, sampling sizes, and animal management practice [40,41].
Previously, several reports indicated the existence of differences in host infectivity among divergent A. phagocytophilum strains [42][43][44]. Phylogenetic analysis indicated that four A. phagocytophilum 16S rRNA sequences in the present study and other reference sequences from goats were included in the same clade, while sequences from cattle and sheep formed two separated clades, suggesting the possible host adaptation for the distribution of A. phagocytophilum in these domesticated animals ( Figure 2). Similar results have also been identified in A. phagocytophilum-positive samples from sheep and cattle in one previous study, reflected by the formation of separate clades for this pathogen from sheep and cattle by phylogenetic analysis [19].
In the present study, the 16S rRNA gene was applied for the genetic characterization of A. phagocytophilum of goats and found variants among different domesticated animals by phylogenetic analysis. The conserved 16S rRNA gene has been widely applied to detect A. phagocytophilum infection in humans and ruminants, identifying variants within this pathogen. However, the variants of the 16S rRNA are controversial due to the low discriminatory power. Therefore, several molecular markers, such as groESL, ankA, and Msp2, were tested and likely showed better function in distinguishing variant of divergent pathogenicity or origins compared with 16S rRNA [45]. The drhm gene was previously recognized as a maker for the pathogenicity of A. phagocytophilum, but it was not useful for several European A. phagocytophilum strains [46]. To increase the resolution in the characterization of A. phagocytophilum, variable number tandem repeat (VNTR), multi-locus sequence typing (MLST), and other multi-locus sequence analysis were used, which could significantly contribute to the comprehensive understanding of A. phagocytophilum [45,47].
A. bovis can infect a variety of mammals, including cattle, sheep, goats, deer, cats, and dogs [18,25,[48][49][50][51]. In the present study, the prevalence of A. bovis in goats was 36.1% (184/509) ( Table 2), which was lower than that in Tunisia (72.0%) [8], but higher than that in other regions in China (12.3-20.3%) [18,21]. Previously, some studies reported lower prevalence of 9.5% for A. bovis in goats in Xi'an city, Shaanxi [52], and higher prevalence of 62.1% in Weinan city, Shaanxi [22]. Animal management practices with the good application of drugs and procedures against ectoparasite possibly contribute to the low prevalence of A. bovis in goats since the common occurrence of this pathogen in several ticks, including Rh. microplus, Ha. longicornis, and Ixode crenulatus [41,53]. Contrary to the distribution of A. phagocytophilum, A. bovis samples from goats and cattle were included in the same clade by phylogenetic analysis, suggesting the possible absence of host segregation for the distribution of A. bovis (Figure 3), which was in accordance with the previous report in sheep and cattle [19]. Nevertheless, potential geographic differences for the distribution of A. bovis were identified in goats in China [22].
The average prevalence of A. ovis (11%, 56/509) in goats in the present study was lower than that in France (52%) [54], Mongolia (71.3%) [55], and other regions in China, e.g., Guizhou (17.8%), Xinjiang (40.5%), Zhejiang (26.3%) [18,19], but higher than that in Heilongjiang (2.6%), Henan (8.7%), and Hubei (7.2%) in China [18,21]. One previous study reported lower prevalence of A. ovis (0.9%) in goats in Xi'an city, Shaanxi [56], and the other study found higher prevalence of A. ovis (25.2%) in goats in Weinan city, Shaanxi [22]. Sampling sites, breeds, animal management practices, as well as other multiple factors, may lead to the differences in the prevalence of A. ovis among studies [40,41]. Notably, the occurrence of ticks (e.g., Rhipicephalus bursat) may also influence the prevalence of A. ovis in goats [54]. Among those three production categories in the present study, A. ovis was absent in meat goats, which was likely caused by the specific species composition of ticks in the studied regions due to the possible biased distribution of Anaplasma spp. among divergent ticks [9,54]. As an important major surface protein of Anaplasma spp., the msp4 gene is likely to have higher evolution rate under the selective pressure of host immune system compared with other genes [27]. In the present study, limited variation was observed within the sequences of the msp4 gene among A. ovis isolates, which was in accordance with the previous reports, reflected by low genetic diversity of A. ovis isolates from Xinjiang and Shaanxi [19,22].
Notably, co-infections of two to three Anaplasma species were identified in the present study (24.4%, 124/509), which was also commonly found in previous reports [18,22,49]. Significant differences were found among co-infections of A. phagocytophilum and A. bovis, A. bovis and A. ovis, and A. phagocytophilum and A. ovis (χ 2 = 134.594, df = 2, p < 0.001), with the co-infection of A. phagocytophilum and A. bovis being the dominant one (83.1%, 103/124). Interesting, co-infection of A. phagocytophilum, A. bovis, and A. ovis was recognized in 6 samples from dairy and cashmere goats, which has also been reported in 6 samples from goats in a previous report [18].
Previous reports indicated that the age was an influence factor for the infection of Anaplasma [8,23]. In the present study, significant difference was found among ages, with the highest prevalence in young animals under 2 months, which may be caused by low immunity or few samples tested in these animals. However, a higher prevalence of Anaplasma species in old goats (≥2 years) than young goats (<2 years) was found in one study in Anhui, China [23]. The difference in prevalence could be old animals being exposed to tick infection for a longer time. More investigations with a wide age range and geographic region settings are needed to further understand the influence of ages on Anaplasma infection.

Conclusions
In the present study, Anaplasma spp. were found in goats, with a prevalence of 58.5%, and significant differences in prevalence were found in animals from different production categories and age groups. A. phagocytophilum, A. bovis, and A. ovis were dominant species in meat, dairy, and cashmere goats, respectively, and A. ovis was absent in meat goats. Meanwhile, co-infections with two or three pathogens were frequently recognized. Frequent occurrence of A. phagocytophilum, A. bovis, and A. ovis in goats in this study indicates that there possibly exist potential risks for the zoonotic transmission between humans and animals due to the appearance of those pathogens in humans in previous reports. These findings can provide baseline information for the understanding of the transmission and zoonotic potential of Anaplasma spp. in goats. Considering the public health risks to humans of zoonotic A. phagocytophilum being found in goats, interventions based on One Health, such as eliminating ticks, maintaining good farm husbandry practices, and keeping personal hygiene, are needed to reduce the transmission of this pathogen from goats to humans and other animals. Certainly, there exist several limitations in the present work, and several questions should be addressed in future studies. For example, did the distribution and species composition of ticks effect the prevalence of Anaplasma spp. in goats? Are tick species and/or burdens related to the distribution of Anaplasma species or not? Why were young children more susceptible to Anaplasma then elder animals? Answers to these questions will assist us to comprehensively illustrate the transmission of Anaplasma between humans and animals.