First Evidence of Ehrlichia minasensis Infection in Horses from Brazil

The genus Ehrlichia includes tick-borne bacterial pathogens affecting humans, domestic and wild mammals. Ehrlichia minasensis has been identified in different animal species and geographical locations, suggesting that this is a widely distributed and generalist Ehrlichia. In the present study, we evaluated Ehrlichial infection in 148 Equidae presented to the Medical Clinic Department of a Veterinary Hospital from a midwestern region of Brazil. Blood samples and ticks collected from the animals were tested by Polymerase Chain Reaction (PCR) for the presence of Ehrlichia spp. A multigenic approach including Anaplasmataceae-specific (i.e., 16S rRNA, groEL, gltA) and Ehrlichia-specific (i.e., dsb and trp36) genes was used for accurate bacteria identification. Sera samples were also collected and evaluated for the detection of anti-Ehrlichia antibodies by indirect fluorescent antibody test (IFA). Possible associations between molecular and serological diagnostics and clinical and hematological manifestations were tested using chi-squared or Fisher’s exact tests. Sequence analysis of the dsb fragment revealed that three horses (2.03%) were exposed to E. minasensis. Sixty-one (41.2%) Equidae (58 equines and three mules), were seropositive for Ehrlichia spp., with antibody titers ranging between 40 and 2560. Seropositivity to ehrlichial antigens was statistically associated with tick infestation, rural origin, hypoalbuminemia and hyperproteinemia (p ≤ 0.05). The present study reports the first evidence of natural infection by E. minasensis in horses from Brazil.


Introduction
Ehrlichia genus is composed of Gram-negative, obligatory intracellular bacteria transmitted by ticks. Several Ehrlichia bacteria are important pathogens, causing disease in humans and domestic and wild mammals. This genus belongs to the order Rickettsiales, family Anaplasmataceae, and has six recognized species: Ehrlichia canis, Ehrlichia chaffeensis, Ehrlichia ewingii, Ehrlichia muris, Ehrlichia ruminantium and Ehrlichia minasensis [1,2].
Despite studies on Ehrlichia infection in equine being scarce in Brazil, seropositivity to Ehrlichia antigens in horses was observed in the States of Paraná [20] and Mato Grosso [21], and the bacterial DNA was also detected in horses of Paraná [8]. Subsequent molecular surveys in Brazil suggest that the Brazilian isolate of Ehrlichia detected in horses may potentially belong to the same group as that detected in Nicaragua [5,22].
Given these antecedents and the lack of studies involving Equine ehrlichiosis in Brazil, this study aims to analyze the occurrence of Ehrlichia spp. and anti-ehrlichial antibodies in horses from the state of Mato Grosso, as well as to evaluate the possible association between bacterial infection and tick parasitism, clinical manifestations, and hematological changes.

Ethical Statement
The animal use procedures for this study were approved by the Animal Use Ethics Committee of the University of Cuiabá (UNIC), under the protocol number 009/2018.

Tick and Blood Sample Collection and Sera Preparation
One hundred and forty-eight equids (horses n = 142 and mules n = 6) with a history of tick parasitism, that attended the Medical Clinic Service of a Veterinary Hospital, Mato Grosso State, Brazil, between February 2018 and April 2018, were included in the study. The animals were located in ranches of the Cuiabá municipality (15 • 35 56" S, 56 • 5 42" W) (n = 84 horses), Poxoreu municipality (15 • 50 27" S, 54 • 23 39 W) (n = 14 horses) and the Poconé municipality (16 • 7 2" S 56 • 57 38" W) (n = 44 horses and six mules) Blood samples were collected in duplicate tubes: one with EDTA (BD Vacutainer ® , Franklin Lakes, NJ, USA), one for complete blood count and another for PCR testing (kept at −80 • C). The other tubes, without anticoagulant (BD Vacutainer ® , Franklin Lakes, NJ, USA), were used for obtaining serum samples. Blood samples without anticoagulant were kept at room temperature (25 • C) until visible clot formation. Subsequently, the samples were centrifuged at 1500× g for 5 min, and the obtained sera was used for biochemical and serological evaluation. Samples were kept at −20 • C until further use.
Ticks, when observed on the animals, were collected and kept in labeled isopropanol tubes until identification according to morphological taxonomic keys for different tick stages [23][24][25].
At the time of collection, an epidemiological questionnaire was applied in order to obtain variables related to the origin (urban or rural) and clinical information such as apathy, fever, mucosal pallor, bleeding, lymphadenemegaly and bloody diarrhea.

DNA Extraction and Molecular Detection of Anaplasmataceae Bacteria
The wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA) was used for DNA extraction from whole-blood and tick samples, following the manufacturer's instructions. Tick DNA was extracted according to Sangioni et al. [27]. In order to monitor the quality of DNA extraction in blood and ticks, samples were selected randomly to perform conventional PCR assays for the mammal endogenous gene glyceraldehyde-3-phosphate dehydrogenase (gapdh) [28] and tick mitochondrial 16S rRNA gene [29], respectively.
PCR reactions were prepared with GoTaq Green Master Mix (Promega, Madison, WI, USA) according to the manufacturer's instructions. The amplification protocol consisted of an initial denaturation step at 95 • C for 5 min, 35 cycles of denaturation (95 • C, 15 s), annealing (52 to 58 • C, 15 s) and extension (72 • C 1 min), followed by an additional extension step at 72 • C for 5 min. Amplification products were mixed with GelRed (Uniscience, São Paulo, SP, Brazil) at 1.5% for visualization, and run on agarose gels in horizontal electrophoresis for 20 min at 120 V. DNA of Ehrlichia canis (Cuiabá strain # 1) [35] and ultrapure water were added as positive and negative controls, respectively. Amplified fragments of the expected size were purified by ReliaPrep DNA Clean-Up and Concentration System ® Kit (Promega, Madison, WI, USA) according to the manufacturer's instructions. Products were then subjected to nucleotide sequencing reaction using the Big Dye Kit (Applied Biosystems, Austin, TX, USA) and 3500 Genetic Analyzer automate sequencer (Applied Biosystems, Austin, TX, USA) according to the manufacturer's instructions. The sequences obtained were edited in the Geneious software (Biomatters, Auckland, New Zealand). Sequence identity analysis was performed in GenBank using the Basic Local Alignment Search Tool (BLAST).

Serologic Analysis
Sera samples were tested by indirect fluorescent antibody test (IFA) using E. canis Cuiabá # 1 strain antigens. The bacterium was cultured in DH82 cells as previously described [21]. Dilutions of the sera samples at 1:40 to 1:5120 were prepared in PBS (pH 7.2) and applied to the IFA slides containing the antigen previously fixed using Acetone. Negative and positive control sera were included in each slide. Rabbit anti-equine IgG conjugated with FITIC (Sigma Diagnostics, St. Louis, Mo, USA) was added at dilution 1:800. Glycerin (pH 8.5) was added to each slide to coverslip. Seroreactions were visualized under an epifluorescence microscope (Scope.A1 Zeiss) using a 40× objective.

Statistical Analysis
For statistical analysis, the R software (R Development Core Team, 2010) was used. Statistical evaluation was based on the association between clinical findings, laboratory tests, and any positive results for PCR or IFA, as well as for both of them. Therefore, we considered the dichotomous variable named "result", which assumed that the value of 1 in horses was characterized as positive in IFA and PCR, and the value of 0 as negative.
To calculate the association measure used for qualitative variables, the Chi-square test (χ2) and Fisher's exact test were used when necessary, with a 95% confidence interval. For the numerical variables, the normal distribution of the data was verified by the Kolmogovov-Smirnov test, and the Mann-Whitney test was applied due to the non-normal distribution. Differences were considered significant at p ≤ 0.05.

Molecular Detection of Anaplasmataceae Bacteria
The gapdh and tick mitochondrial 16S rRNA gene were consistently amplified from DNA horse and tick samples. Amplicons for dsb gene were detected in blood samples of three horses (2.03%) and were negative for the 16S rRNA gene. Sequence analysis of the dsb fragments revealed a sequence 99-100% identical to E. minasensis sequences (GenBank accession numbers: MH500007, KX258450, KT970783, KT314243, KF621012). The dsb sequences generated in this study were deposited into GenBank and assigned the nucleotide accession numbers of MT232423, MT232424, MT232425. These three samples were negative for the groEL, trp36 and gltA genes. All ticks were also negative in the PCR assays.

Discussion
This paper reports, for the first time, the molecular detection of E. minasensis in horse blood in Brazil. Ehrlichia minasensis was first described in cattle from Canada [36], and later in R. microplus ticks and cattle from Brazil [7,10,36,37]. Additionally, E. minasensis has been the subject of studies on morphological and genetic characterization [38] and their distribution has been shown in recent reports in Corsica, France [14], Pakistan [15], Ethiopia [39], South Africa [13], Israel [40] and China [16].
Our result reinforces the hypothesis that this ehrlichial species is not restricted to cattle. Indeed, previous reports detected E. minasensis DNA in deer samples in Canada [11]. Horses can be infected while grazing on infested R. microplus pasture or when raised close to bovines [17]. Although no positive ticks for E. minasensis were detected, a previous study showed that populations of A. sculptum ticks exposed to naturally infected cattle, were negative by PCR. In contrast, R. microplus ticks were positive under the same conditions [12].
In this sense, even under a low frequency of R. microplus among Equidae in the present study, the transmission of E. minasensis by this tick species must be considered. In addition, the three positive PCR samples were from animals from a single farm, which suggests a source of infection with epidemiological relevance for horses in this site.
In this context, approximately 70% of seropositive equids were infested with at least one tick species, and these animals had 2.1 chances to obtain anti-Ehrlichia spp. antibodies (Table 4). Our result corroborates results in the USA [4], which found greater positivity in horses parasitized by ticks.
Seropositives Equidae were found in 41.2% of samples and titers of anti-Ehrlichia spp. antibodies ranging from 40 to 2560. The presence of serum immunoglobulins against ehrlichia was previous reported in horses from the Pantanal wetlands [21], but with lower frequency and antibody titers. Serological surveys of ehrlichial infection in equids in Brazil are scarce; another study reported the presence of anti-Ehrlichia spp. antibodies in horses from a rural settlement in the Parana state, Southern Brazil, with similar IFA antibody frequencies to the present study [20].
In line with the risk observed with tick's parasitism, animals from a rural environment had a 7.9 chance of being seropositive compared to animals living in urban environments. This association reinforces the possibility of transmission by R. microplus ticks, since the presence of cattle in urban areas is usually insignificant when compared to their presence in the rural environment.
Hematological variables, such as anemia, hyperproteinemia and hypoalbuminemia, were associated with seropositivity for Ehrlichia spp. This result agrees with the clinical and laboratory characteristics of ehrlichiosis, whether in dogs, cattle or humans, that were usually found to have a significant hyperproteinemia [10,41,42]. In this study, positive horses had 5.2 chances of having high plasma protein compared to negative horses (p < 0.05). Hyperproteinemia with hyperglobulinemia is a common abnormality found in dogs with canine monocytic ehrlichiosis, due to continuous prolonged antigen stimulation [42]. In this study, hyperglobulinemia (33/61) was observed in positive equid; however, statistical significancy was not detected.
The anemia variable can be explained by the inflammatory response when serum iron is unavailable, which compromises erythropoiesis. Significant decreases in iron levels were identified in horses with local and systemic inflammatory disease [43]. Although the albumin variable showed a statistical result (p < 0.05), the highest observed frequency was between normal albumin values, followed by hypoalbuminemia. Thus, in the present study, this result does not reflect any laboratory parameter for equine ehrlichiosis. However, hypoalbuminemia can be observed in several situations, such as chronic liver damage, a deficit of intestinal protein absorption due to helminths, and kidney disease [44].
Increased serum fibrinogen levels were observed in both PCR/IFA-positive animals, because the highest fibrinogen levels were more frequent among the positive horses. However, these increases were not statistically significant (p = 0.06/p = 0.11), but the results suggest the tendency of an increase in serum fibrinogen, observed in positive animals. Elevated fibrinogen, occurring in horses, is relevant in acute phases of infections [45,46], and this result denotes an inflammatory response related to the positivity to diagnosis ehrlichiosis seen in both assays in the present study. Additionally, hyperfibrinogenemia can be justified by animals presenting plasma hyperproteinemia [26].

Conclusions
The present study reports the first evidence of natural infection by E. minasensis in horses in Brazil. Additionally, seropositivity against Ehrlichia antigen was associated with parasitism by ticks, rural origin and anemia. Hyperproteinemia and hyperfibrinogenemia were detected in most positive animals, reflecting inflammatory response. In the absence of a statistically significant association between the positivity of animals and clinical signs, further studies of Ehrlichia spp. infection in horses should be performed to better understand their relevance in the natural history of this ehrlichiosis in Brazil.