MOLECULAR DETECTION OF EHRLICHIA CANIS IN THE PET-DOG POPULATION IN R. N. MACEDONIA

Canine monocytic ehrlichiosis (CME) is a widespread, tick-borne, canine disease, caused by an obligate intracellular bacterium, Ehrlichia canis . The main vector, a brown-dog tick, Rhipicephalus sanguineus , is widely distributed, especially in areas with tropic, subtropic, or Mediterranean climates (Central and South America, Eastern and Western Asia, Africa, Australia and Southern Europe). The study performed in 2012, by Stefanovska et al., determined a seroprevalence of 18.7% of E. canis among the Macedonian dog population. Up to date, the presence of E. canis , using molecular diagnostic methods, has not been investigated in Macedonia. Therefore, this study aimed to con ﬁ rm the presence of E. canis , in the pet-dog population on the territory of the city of Skopje, North Macedonia, using a highly sensitive multiplex Real-Time PCR method (qPCR). Whole blood samples from 80 dogs of different breeds and ages, with clinical symptoms of CME and positive serology result for the presence of antibodies against E.canis , were collected for analyses. Out of 80 dogs, 36 (45%) were found as positive. The present work reports the ﬁ rst molecular detection of E. canis in pet dogs on the territory of the city of Skopje, Macedonia.


INTRODUCTION
Canine monocytic ehrlichiosis (CME) is a widespread, tick-borne, canine disease, caused by an obligate intracellular bacterium, Ehrlichia canis. The other two closely related species, Ehrlichia ewingii, and Ehrlichia chaffeensis, are known to infect dogs, as well, causing similar but milder clinical diseases [1].
The main vector of E. canis, a brown-dog tick, Rhipicephalus sanguineus, is widely distributed, especially in the areas with tropic, subtropic or Mediterranean climates (Central and South America, Eastern and Western Asia, Africa, Australia, and Southern Europe) [2][3][4]. Besides causing the disease in dogs, E. canis has a zoonotic potential and can be a cause of human ehrlichiosis [5]. On the other hand, the epidemiological importance of E. ewinvii and E. chaffeensis seems to be limited only to North America, because the presence of their vectors, Amblyomma americanum and Dermacentor variabilis ticks, has not been reported outside this continent [6].
The infection with E. canis in dogs can be asymptomatic for months, or with the immediate development of clinical signs. The incubation period can vary from eight to twenty days. The disease is of multisystemic character and can be manifested as acute form with fever, depression, anorexia, splenomegaly, lymphadenopathy and hemorrhage predisposition, subclinical form with no apparent clinical signs of the disease, or chronic form with symptoms similar to those in the acute form, but more severe [2,7]. Different forms of the disease can hardly be distinguished in everyday clinical practice [8]. The main hematological alterations are moderate to severe thrombocytopenia, non-regenerative anemia, and depending on the disease stage, moderate to severe leucopenia [9].
Even though there is no perfect diagnostic test for E. canis, serological tests are most commonly used for the diagnosis of E. canis infections [10]. Antibodies against E.canis appear 7-35 days post infection and are not in correlation with the clinical manifestation and the duration of the disease [8]. Antibodies for E.canis can persist in the circulation for several months to years after treatment, and they cannot be used in the determination of active or past infection [11]. Therefore, the molecular diagnostic methods are considered as the most appropriate for the diagnosis of E. canis infection [12]. Polymerase chain reaction (PCR) is a very sensitive method for the detection of acute monocytic ehrlichiosis in dogs, and it is an irreplaceable diagnostic tool, especially in the fi rst two weeks of infection, prior to antibody development and the onset of clinical signs [13]. However, the use of PCR for the detection of E. canis in the subclinical and chronic form may have limited value due to the low concentration of the microorganism in the circulation, which can result in false-negative results [14,15].
Serological evidence for the presence of CME in dogs has been reported in almost all Balkan countries (Greece, Albania, Serbia, and Bulgaria) [22][23][24][25]. The study performed in 2012, by Stefanovska et al, determined the seroprevalence of 18.7% of E. canis among the Macedonian dog population [26]. However, published data on studies for molecular detection of Е. Canis are reported only from Greece [27].
Up to date, the presence of E. canis, using molecular diagnostic methods, has not been investigated in Macedonia. Therefore, this study aimed to confi rm the presence and determine the prevalence of E. canis, in the pet-dog population on the territory of the city of Skopje, using a highly sensitive multiplex Real-Time PCR method (qPCR).

Sample selection
The targeted group was pet dogs brought to the University Veterinary Clinic (UVC) at the Faculty of Veterinary Medicine -Skopje (FVMS), and several other private veterinary clinics on the territory of the city of Skopje, exhibiting clinical symptoms characteristic for tick-borne diseases (fever, anorexia, lethargy, and pale mucosa), and hematological alterations with the presence of thrombocytopenia and/or anemia. All dogs were screened for the presence of antibodies against E. canis using a commercially available immunochromatography based point-of-care (POC) test (SNAP® 4Dx®; IDEXX Laboratories, Inc. U.S.A). Finally, only the seropositive animals were selected for further investigation and detection of E. canis.
From May 2018 until September 2019, 80 whole blood samples from naturally infected pet dogs (47 males and 33 females), of different breeds and ages, were collected and analyzed using the qPCR method. Whole blood samples were collected in 3ml vacuum tubes sprayed with Ethylenediaminetetraacetic Acid (Vacuette, K3E K3EDTA, Greiner Bio-One), and stored at -20°C until DNA extraction.

DNA extraction
Total DNA was obtained from 200µL of whole blood, using an automated extraction system (SaMag-24, Sacace, Italy), and an appropriate DNA extraction kit (SaMag Blood DNA Extraction kit, REF SM001), according to the manufacturer's protocol. For optimization of the DNA input in the qPCR reaction, the quantity and purity of the extracted DNA was evaluated for each sample, using the UV-Vis spectrophotometer (NanoDrop 2000C, Thermo Scientifi c), and determination of the A 260/280 ratio.

Detection of Ehrlichia canis by qPCR
For detection of E. canis DNA, a highly sensitive, multiplex, Taqman-based qPCR protocol, previously described by Peleg et al., 2010, was used. [28] The protocol is designed to amplify a short sequence within the 16S rRNA of E. canis, and additionally, the canine beta actin gene (ACTB) as an internal control (IC). Even though this protocol detects E. ewingii and E. chaffeensis, as well, аs neither the presence of competent vectors nor the pathogens themselves have been confi rmed in Europe so far, we considered the positive amplifi cation of the 16S rRNA fragment of Erlichia genom to be specifi c for E. canis. Based on the internal verifi cation and determination of the limit of detection (LOD) of the qPCR protocol, the cut-off value for the positive result was set at < Ct 39 (data not presented).

RESULTS AND DISCUSSION
Eighty blood samples from pet dogs, living on the territory of the city of Skopje, were analyzed for the presence of E. canis using the qPCR method, of which 36 were found as positive, giving an overall prevalence of 45%. The prevalence of E. canis was calculated as 55.6% (20 of 47) among the male dogs, and 44.4% Unspecifi c clinical manifestation, together with the lack of sensitive diagnostic procedures impairs the overall diagnosis of CME [12]. Thus, the combination of clinical, serological and molecular diagnostic procedures is required for early detection of circulating E. canis. Through personal communication with the major veterinary clinics/ambulances in Macedonia, the commercial POC tests for detection of IgG antibodies against E. canis, was found to be the most commonly used tests for diagnosis of CME. Even though, the presence of E. canis has been indirectly confi rmed with previous serological studies [26], direct detection of the pathogen, using molecular diagnostic methods, has never been performed in Macedonia. This study presents the results of the fi rst molecular survey designed for confi rmation of the presence and determination of the prevalence of E. canis among pet-dogs, exhibiting clinical signs of a tick-borne disease on the territory of the city of Skopje. In addition to clinical signs, the presence of hematological alterations (thrombocytopenia and/ or anemia), and confi rmation of the presence of antibodies against E. canis, were used as a diagnostic approach to maximize the probability of detecting circulating E. canis using the qPCR method. From the 80 seropositive pet dogs included in this study, in 36, successful amplifi cation of the specifi c fragment within the 16S rRNA of E. canis, was obtained. The established prevalence of 45.0%, was similar to the prevalence observed in molecular studies conducted in several countries from different continents [16][17][18][19][20][21]. However, some studies report a rather low prevalence of E. canis (Malaysia -2%, Myanmar -0.75%, etc.), or even fail to detect the presence of E. canis in the studied dog population (Croatia) [10,29,30]. The established differences in the prevalence of E. canis in different studies are mainly infl uenced by the wide variety of approaches used for the study design. In general, studies reporting a low prevalence of E. canis mostly used a simple random sampling method of apparently healthy dogs, as a method for selection, while the majority of studies reporting higher prevalence used a more targeted approach selecting only dogs exhibiting clinical symptoms of tick-borne disease or only seropositive dogs, such was the case with our study. The established high prevalence of E. canis among pet dogs living on the territory of the city of Skopje, indicates a frequent exposure, as well as a low level or poor protection against ectoparasites of this population. No signifi cant difference (p> 0.05) was observed when comparing the prevalence on a gender level. However, the slightly higher prevalence observed in the male in comparison to female dogs (55.6%, and 44.4%, respectively), goes in line with the data from similar studies conducted in Portugal and Malaysia [21,10].
In 55% (44 of 80) of the seropositive dogs, we failed to detect circulating E. canis in the analyzed whole blood samples. The whole blood samples were chosen because of the noninvasive nature of this sampling technique and the established suitability of these samples for molecular detection of E. canis [31,32]. The reason for this is the ability of the bacteria to "hide" in the spleen and other organs, in the subclinical phase of the disease, while the level of antibodies is high and the level of circulating bacteria is very low (severe pancytopenia in the chronic phase) [31,33,34].
Based on the results from the study conducted by Waner et al. (2014), the sensitivity of the qPCR method dramatically decreases 17 days post infection (DPI) [35]. Thus, only 36.11% (13 of 36) of the confi rmed positive dogs in this study, with Ct values lower than 30, could be considered as being detected within 15 DPI (acute phase). The use of spleen aspirates for the confi rmation of E. canis, in seropositive/qPCR negative dogs, will contribute to increasing the level of sensitivity of the overall diagnostic approach, and identifi cation of the carrier status for CME in dogs with a subclinical or chronic form of the disease [15].
Besides the presence of clinical symptoms (fever, depression, anorexia, etc.), and serological confi rmation of the antibodies against E. canis (100%), thrombocytopenia was the most prevalent fi nding, present in 95% (76 of 80) of the dogs included in this study. This highlights the diagnostic value and signifi cance of this parameter for the general approach in CME diagnostic.
The study should be further continued with the determination of the genetic characteristics of Macedonian E. canis isolates, and the establishment of epidemiological relations with the closely related isolates deposited in the accessible, open-source, genetic databases.