Clinicopathological and molecular profiles of Babesia vogeli infection and Ehrlichia canis coinfection

Background and Aim: Canine babesiosis, a tick-borne parasitic disease, is caused by the hemoprotozoa, Babesia vogeli, and Babesia gibsoni. Infection with these parasites, which is endemic globally, leads to life-threatening immunosuppression in dogs. The merozoites invade the red blood cells (RBCs) of infected dogs. Ehrlichia canis, an intracellular bacterium that infects monocytes, is transmitted by the same tick species (Rhipicephalus sanguineus) during blood consumption and coinfection with B. vogeli and E. canis has been reported. Although the hematology and biochemistry of canine babesiosis have been studied, more studies are needed to develop a better understanding of the hematobiochemical and molecular profiles associated with cases of single infection and coinfection of canine babesiosis in Thailand. This study aimed to investigate the hematological, biochemical, and molecular profiles of B. vogeli infection and E. canis coinfection. Materials and Methods: The study included 33 B. vogeli–positive blood samples and 11 E. canis–coinfected blood samples. To exclude coinfection with Hepatozoon canis and Anaplasma platys, only dogs with B. vogeli infection and B. vogeli–E. canis coinfection were included in the study. A multiplex polymerase chain reaction (PCR) assay was conducted to detect B. vogeli, E. canis, and H. canis, and a conventional PCR assay was conducted for the detection of A. platys. Besides, the PCR assay and sequencing, comprehensive data analysis was conducted, including a microscopic blood parasite examination and hematological and biochemical data analysis. Results: The comparison of the hematobiochemical data between the B. vogeli–positive and E. canis coinfection groups identified that there were statistically significant differences in the RBC parameters, including RBC count, hemoglobin concentration, hematocrit, and RBC distribution width (p=0.001). Neither B. vogeli infection nor coinfection with E. canis was associated with the sex, breed, recorded clinical signs, geographic origin of the dog and also B. vogeli 18S rRNA gene sequencing results. Conclusion: Coinfection with E. canis increased the severity of babesiosis. The pathogenic mechanisms underlying this infection, such as destruction of RBCs, require further investigation. This study may enhance diagnosis, treatment, and prevention of canine babesiosis.

Babesia-infected dogs may be asymptomatic [8] or present with various clinical signs that range from mild to peracute and deadly. Clinical manifestations include anorexia, lethargy, fever, pale mucous membrane, jaundice, and renal disease [13][14][15], depending on the parasite and the host's, age, sex, and breed [12,16]. Natural and experimental in vivo infections with B. vogeli have shown subclinical signs. In immunosuppressed animals, such as splenectomized dogs, the pathogen causes severe acute infection, which is followed by fever, anorexia, malaise, regenerative anemia, thrombocytopenia, and increased white blood cell (WBC) count [6,17]. By contrast, many studies have reported that there are decreased numbers of WBC in dogs with canine babesiosis [18,19]. A retrospective canine babesiosis study in the Small Animal Available at www.veterinaryworld.org/Vol.13/July-2020/7.pdf Teaching Hospital of Chulalongkorn University in Thailand found that hypocytic hypochromic anemia and thrombocytopenia were the major clinical hematological findings in dogs with canine babesiosis [20]. However, other studies have shown the occurrence of macrocytic hyperchromic anemia in canine babesiosis [18,19]. Hematological profiles of canines with coinfections of Babesia-Ehrlichia have been studied in South Africa, but the prevalence of coinfection in South Africa is low and the main hemoprotozoan species is B. rossi [21]. Although coinfections have frequently been reported in Thailand in the literature describing the canine hemoparasite prevalence [9,10,22], information about the hematobiochemical patterns of the infection and the molecular diversity of B. vogeli and Ehrlichia canis coinfection remains lacking.
This study aimed to investigate 44 Babesiapositive blood samples, including samples that had a single infection with Babesia and samples that had Babesia-Ehrlichia concurrent infections that were confirmed by microscopic and molecular examinations. Comprehensive and hematobiochemical data were analyzed, and Babesia 18S rRNA and Ehrlichia 16S rRNA genes were sequenced for the identification of Babesia subspecies, as well as for the genetic variation of both hemoparasites.

Ethical approval
The study protocol (no. MUVS-2018-12-66) was approved by the Faculty of Veterinary Science-Animal Care and Use Committee (FVS-ACUC), Mahidol University, Thailand.

Comprehensive data collection
Retrospective comprehensive data, including signalment (age, sex, breed, and geographic origin), recorded clinical signs, and hematobiochemical profiles, from 44 dogs were gathered from medical records provided by Prasu-Arthorn Animal Teaching Hospital, Mahidol University, on the 1 st day of registration, between February and December 2019. A map showing the provinces in Thailand where the babesiosis cases were located was drawn using a template obtained from www.simplemaps.com (Figure-1).

Blood sample preparation
The blood samples (1 mL) were obtained from Prasu-Arthorn Animal Teaching Hospital, Faculty of Veterinary Science, Mahidol University, Salaya Campus, Nakhon Pathom, Thailand. The samples were drawn from the cephalic or saphenous veins of the dogs and preserved in ethylenediaminetetraacetic acid in plain tubes for hematobiochemical analysis. Thin buffy coat smears were prepared and stained using Giemsa solution and sent to the laboratory of the parasitology unit for routine microscopic detection. Thirty-three Babesia-positive samples and 11 samples that had Babesia-Ehrlichia coinfection were identified and further confirmed by molecular examination.  Available at www.veterinaryworld.org/Vol.13/July-2020/7.pdf The multiplex PCR amplifications were conducted using 5 μL of genomic DNA with the primers Ba103F, Ba721R, Ehr1401F, Ehr1780R, Hep001F, and Hep737R. The PCR reactions contained 5 μL total DNA template, 20 µL of the reagents which composed of 0.4 pmol of each primer, and 300 μM of each deoxyribonucleotide triphosphate (dNTP), four units of HotStarTaq DNA Polymerase (QIAGEN, Hilden, Germany), 1X PCR buffer (20 mM Tris-HCl pH 8.4, 50 mM KCl), 1.5 mM MgCl 2 , and RNase-free water to a total volume of 25 μL. The reactions were conducted in a S1000 Thermal Cycler (Bio-Rad, Hercules, CA, USA), with the following steps: 15 min at 95°C; 30 cycles of 45 s at 94°C, 45 s at 65°C, and 90 s at 72°C; and 10 min at 72°C. The PCR products were examined through gel electrophoresis on a 2.5% agarose gel stained with GelRed (Biotium, Hayward, CA, USA) and visualized under ultraviolet (UV) light (GeneGenius, Cambridge, UK). The amplicons (619 base pairs) were then purified using QIAquick PCR purification kits (QIAGEN Inc., Valencia, CA, USA), according to the manufacturer's protocol. Bidirectional sequencing of all PCR products was conducted using First BASE Laboratories (Selangor, Malaysia). The DNA sequences were then analyzed using an ABI 3730xl sequencer (Applied Biosystems, Foster City, CA, USA) with fluorescent dye terminator sequencing. The DNA sequencing was conducted using the Babesia-specific primers Ba103F and Ba721R.

Anaplasma platys conventional PCR
Conventional PCR for the detection of A. platys was used to exclude A. platys coinfection. The reactions were conducted using 5 μL of the total DNA as a template and 20 μL of 0.4 pmol of each primer (Ana45F: 5′GTCGAACGGATTTTTGTCGT3′ and Ana671R: 5′GCCACTGGTGTTCCTCCTAA3′) [24], 300 μM of each dNTP, four units of iTaq DNA Polymerase (iNtRON Biotechnology, Kyungki-Do, South Korea), 1X PCR buffer (20 mM Tris-HCl pH 8.4, 50 mM KCl), 1.5 mM MgCl 2 , and RNase-free water. The amplification was conducted in a T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA). The thermocycling steps consisted of one step for 10 min at 95°C followed by 35 cycles of 30 s at 94°C, 30 s at 55°C, and 45 s at 72°C, with a final extension step of 10 min at 72°C. Aliquots of the amplicons were detected using gel electrophoresis on 2.0% agarose gel stained with GelRed (Biotium, Hayward, CA, USA) and visualized under UV light (Syngene, Cambridge, UK).

E. canis 16S rRNA gene amplification and sequencing
For the detection of E. canis, conventional PCR was used to amplify the 16S rRNA genes. The reactions were conducted using 5 μL of the total DNA template and 20 μL of 0.4 pmol of each primer (Ecan 16S-94F: 5′GTGGCAGACGGGTGAGTAAT3′ and Ecan 16S-1102R: 5′GAGTGCCCAGCATTACCTGT3′), 300 μM of each dNTP, four units of iTaq DNA Polymerase (iNtRON Biotechnology, Kyungki-Do, South Korea), 1X PCR buffer (20 mM Tris-HCl pH 8.4, 50 mM KCl), 1.5 mM MgCl 2 , and RNase-free water. The amplification was conducted in a T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA). The thermocycling consisted of one step of 10 min at 95°C followed by 35 cycles of 30 s at 94°C, 30 s at 58°C, and 45 s at 72°C, with a final extension step of 10 min at 72°C. Aliquots of the amplicons were detected using gel electrophoresis with a 2.0% agarose gel stained with GelRed (Biotium, Hayward, CA, USA) and visualized under a UV light (Syngene, Cambridge, UK). The amplicons (1009 base pairs) were then purified using the QIAquick PCR purification kit (QIAGEN Inc., Valencia, CA, USA), according to the manufacturer's protocol. Bidirectional sequencing of all PCR products was conducted using First BASE Laboratories (Selangor, Malaysia). The DNA sequences were then analyzed using an ABI 3730xl sequencer (Applied Biosystems, Foster City, CA, USA) with fluorescent dye terminator sequencing. The DNA sequencing was conducted using E. canis-specific primers Ecan 16S-94F and Ecan 16S-1102R.

Bioinformatic and phylogenetic analysis
The B. vogeli 18S rRNA and E. canis 16S rRNA sequence results were analyzed using several programs. All sequencing results were compared with sequences available in the GenBank database using the Basic Local Alignment Search Tool (http://blast. ncbi.nlm.nih.gov/Blast.cgi). Multiple alignments of all nucleotide sequences were conducted using the ClustalW web-based tool (https://www.genome.jp/ tools-bin/clustalw) [25]. Phylogenetic trees were reconstructed using maximum likelihood analysis with bootstrapping (100 replications) in the advanced mode of the Phylogeny.fr web server (http://www. phylogene.fr/) [26]. All sequences were compared with published sequences in the GenBank database that originated from other geographic locations globally. H. canis and Trypanosoma evansi were included as the outgroup for the 18S rRNA and A. platys was included as the outgroup for the 16S rRNA phylogenetic tree, respectively.
As shown in Table-1, the sex (male and female), breed (pure and mixed), age, clinical signs, and geographic origin of the dogs were not significantly different between the B. vogeli and the coinfected group (p=0.79, 0.48, 0.57, 0.57, and 0.89, respectively).

Analysis of hematobiochemical data
The RBC count, hemoglobin concentration, hematocrit, and RBC distribution width (RDW) were significantly different between the B. vogeli and the coinfected group (p=0.001, 0.001, 0.001, and 0.005, respectively). The RBC count, hemoglobin concentration, and hematocrit levels in the coinfected group were lower than the reference value ranges, whereas the RDW was higher. The platelet count was not  (Table-2). Plasma protein was not significantly different (p=0.357), with both groups exhibiting hyperproteinemia (plasma protein more than 7.5 g/L). All 33 (100%) dogs in the B. vogeli-infected group and 8/11  Available at www.veterinaryworld.org/Vol.13/July-2020/7.pdf (72.73%) dogs in the coinfected group displayed hyperproteinemia (Table-2).

Phylogenetic tree analysis
Phylogenetic analysis revealed that our sequences were closely related to the sequences from Thailand available in GenBank. The nucleotide sequences obtained from bidirectional sequencing of the 18S rRNA sequences of B. vogeli with coinfection of E. canis and the single B. vogeli infection groups (MT674935 and MT674936) showed that both nucleotide sequences were identical to those previously reported in Thailand (Chiangmai); B. vogeli 18S rRNA sequences (JF825145). The 16S rRNA sequences from E. canis in the coinfection group (MN660040) were identical to previously reported E. canis 16S rRNA sequences (EF139458) from samples from Thailand (Bangkok) (Figure-2).

Discussion
R. sanguineus is a common hard tick in Thailand (Southeast Asia) that can carry various canine hemoparasites, including E. canis, B. vogeli, H. canis, A. platys, and Mycoplasma spp., which infect both domestic and stray dogs [12,22,[27][28][29]. Occurrences of B. vogeli and E. canis coinfection and sole B. vogeli infections confirmed by molecular examination have been reported in Thailand [9,22]. Studies on tickborne hemoparasite prevalence found that the ratio of coinfections of B. vogeli and E. canis to single B. vogeli infections was 2:59 (1:29.5) in Khon Kaen Province [22] and 2:17 (1:8.5) in Mahasarakham Province [9]. In the present study, to have a reliable sample comparison, we gathered 11:33 (1:3) ratio of samples of these infections from small animal hospital cases. The dogs of both groups were mainly from the Bangkok Metropolis and bordering areas, such as the Nakhon Pathom, Samut Sakhon, and Nonthaburi provinces. By contrast, a study conducted in South Africa, where the single Babesia pathogen was B. rossi, found that the ratio was 4:191 (1:48) [21]. To the best of our knowledge, the present study is the first to investigate the hematobiochemical parameters of B. vogeli infections and B. vogeli and E. canis coinfections.
In a hematobiochemical study, Niwetpathomwat et al. [20] reported decreases in the hemoglobin concentration, mean corpuscular hemoglobin (MCH), MHC concentration (MCHC), and mean platelet counts in registered babesiosis cases in the Chulalongkorn University's Small Animal Teaching Hospital in Bangkok. Moreover, a study investigating cases of B. vogeli and E. canis coinfection in Costa Rica identified that the main clinical signs of young coinfected dogs were anemia, lethargy, and fever [30]. A 3-month-old coinfected puppy in the Philippines had lack of appetite, a pale mucous membrane, and fever, with thrombocytopenia as the most important abnormal hematological finding [31]. However, these results are not consistent with the findings of our study. In our study, in the group infected with B. vogeli, the only similar result was a decrease in the platelet count; the hemoglobin concentration, median MCH, and median MCHC values were all within the reference ranges. However, the study by Niwetpathomwat et al. [20] did not confirm the single B. vogeli infection in the canine babesiosis samples by molecular examination. Importantly, in our investigation, the RBC count, hemoglobin concentration, and median hematocrit levels in the cases with coinfection were significantly lower than in the cases with single B. vogeli infection. The previous studies demonstrated that infections of E. canis, which were confirmed using molecular techniques, led to a significant reduction in RBC count and hematocrit levels [32]. This study also demonstrated that anemic dogs infected with Mycoplasma spp. that had hematocrit levels of less than 15% had a seven-fold risk of coinfection with E. canis, as compared with a single infection of Mycoplasma spp. [28]. Coinfection with E. canis may lead to increased infection severity. Available at www.veterinaryworld.org/Vol.13/July-2020/7.pdf The significant increase in RDW in our coinfection group may be associated with increased destruction of RBCs, which may lead to regenerative anemia, including immune-mediated hemolytic anemia. RDW is related to heterogeneous erythrocyte populations in the blood circulation, whose main population is reticulocytes, rather than mature RBCs [33]. The mechanism underlying canine babesiosis has been proposed to involve intravascular and extravascular hemolysis with immune-mediated hemolytic anemia [34]. Various types of anemia, such as normocytic normochromic anemia, caused by B. canis and B. gibsoni infections [35], and hypocytic hypochromic anemia, caused by B. vogeli infection [20], have been reported. Our results indicate that coinfection with E. canis may cause macrocytic and/or microcytic hypochromic anemia, leading to the destruction of erythrocytes through an immune-mediated mechanism [36][37][38] resulting in the elevation of RDW.
The significance of lymphocytosis in single B. vogeli infections observed in this investigation is consistent with the previous studies conducted in Egypt [39], Italy [40], and Indonesia [41], although one study in Thailand did not describe lymphocytosis [20].
Niwetpathomwat et al. [20] reported that the levels of alkaline phosphatase (ALP) enzyme increased in cases of babesiosis. In the present study, we did not measure the levels of ALP and aspartate aminotransferase (AST), a liver enzyme, which are related to infections with B. canis and B. gibsoni [3,42]. There are three main ALP isoenzymes in canine serum [43] and the elevation of canine ALP is also associated with hepatobiliary, hepatic, and bone diseases [44][45][46]. In azotemic dogs infected with B. canis, the AST/ ALT ratio decreased [47], and this ratio was not significantly different from those observed in B. vogeli infections [12].
In our study, the 44 bidirectional sequences of B. vogeli 18S rRNA displayed 100% identity and showed conservation with the various B. vogeli 18S rRNAs available in the GenBank database, including those from Chiangmai Province (the northern part of Thailand, JF825145) [8], Bangkok Metropolis (the central part of Thailand, KF621061-KF621074), and Khon Kaen Province (the north-eastern part of Thailand, KF621075-KF621081) [48]. By contrast, the sequences (around 200 base pairs) obtained from the Songkhla Province (the southern part of Thailand, KU765196 and KU765197) [49] had various genetic variations when aligned with our data. The genetic variation and genotyping of B. vogeli in Thailand should be further investigated using an immunodominant protein gene [50] with high levels of nucleotide diversity, such as an apical membrane antigen 1 [51].

Conclusion
Coinfection with E. canis increases the severity of babesiosis. Its pathogenic mechanisms, such as RBC destruction, should be further investigated. This study may contribute to improve the diagnosis, treatment, and prevention of the disease.