Molecular Detection of Borrelia burgdorferi Sensu Lato and Anaplasma phagocytophilum in Ticks Collected from Dogs in Urban Areas of North-Eastern Poland

From 2016 to 2018, ticks were collected from 272 dogs admitted to veterinary clinics in the city of Olsztyn (north-eastern Poland). Among 522 collected ticks, 423 were identified as Ixodes ricinus (413 females and 10 males) and 99 as Dermacentor reticulatus (62 females and 37 males). Non-engorged (86 individuals) and engorged (436 individuals) ticks were screened for the presence of Borrelia burgdorferi sensu lato and Anaplasma phagocytophilum DNA. Borrelia and A. phagocytophilum species detection was determined based on the sequence of the fla B and 16S RNA genes, respectively. DNA of B. burgdorferi s.l. was identified in 31.6% (165/522, 95% CI: 27.6–35.8%) of ticks (I. ricinus 151/423, 35.7%, 95% CI: 31.1–40.4%; D. reticulates 14/99, 14.1%, 95% CI: 7.9–22.6%). A. phagocytophilum was identified in 0.96% (5/522, 95% CI: 0.3–2.2%) of specimens. All positive samples were engorged I. ricinus females (5/402, 1.2%, 95% CI: 0.4–2.9%). In 85.4% (141/165, 95% CI: 79.1–90.4%) of Borrelia infected ticks, the DNA of one genospecies was revealed. The DNA of at least two different genospecies was detected in 14.5% of specimens (24/165, 95% CI: 9.5–20.8). The coexistence of B. burgdorferii s.l. and A. phagocytophilum was not detected.


Introduction
Along with mosquitoes, ticks are the most widespread vectors of pathogenic microorganisms (viruses, bacteria, and protozoa) for humans and domestic animals worldwide [1][2][3]. Ixodes ricinus and Dermacentor reticulatus are the most important tick species in Poland, but the role of I. ricinus in pathogen transmission is dominant [4]. One of the most frequently diagnosed zoonotic tick-borne diseases is Lyme borreliosis (LB). Worldwide, the main vectors of the B. burgdorferi s.l. are I. ricinus, I. scapularis, and I. persulcatus. In Central Europe, I. ricinus represents the main health risk to humans and many other vertebrate species as a vector of multiple pathogens, including Borrelia spirochetes [5]. The pathogens causing LB are spirochetes included in the Borrelia burgdorferi complex, which now comprises ca 20 Borrelia species. Nine of them have been detected in European I. ricinus ticks. The most s.l. is still pending, although the specific DNA of this pathogen has been detected in these ticks [40,41]. However, there is still no evidence of its role as a vector of spirochetes [4].
Based on the "One Health" theory, tick-borne diseases are associated with close relationships among ticks, humans, and companion animals. Additionally, the monitoring of tick-borne pathogens in ticks attached to animals is important to determine disease distribution and possible transmission to humans [42]. Considering the information above, the aim of this study was the detection of B. burgdorferi s.l. and A. phagocytophilum in non-engorged and engorged I. ricinus and D. reticulatus ticks collected from dogs in the urban areas of north-eastern Poland, a region endemic for tick-borne diseases.

Tick Collection
Vets from three veterinary clinics in Olsztyn collected non-engorged and engorged ticks from dogs between spring 2016 and autumn 2018. A total of 522 adult ticks were identified: 81% I. ricinus (423/522-413 females and 10 males) and 19% D. reticulatus (99/522-62 females and 37 males). All of the ticks were removed from a total of 272 dogs (on average there were 1.92 ticks per dog). These ticks were stored at −4 • C for further analysis.

Molecular Identification of Pathogens
Overall, 522 adult ticks were analyzed (

BLASTn Data Analysis
Comparative analysis with data registered in the GenBank database using BLASTn showed that all obtained sequences (Aph1-5: KY319143, KY828226, MK530241-MK530243) belonged to one haplotype and revealed 100% homology to the sequence of A. phagocytophilum first described in a human patient from Wisconsin, USA (U02521) and the A. phagocytophilum sequence from a tick removed from human skin in north-eastern Poland (DQ006828).

Discussion
Tick infestation and the risk of tick-borne diseases are commonly connected with forested and rural areas. However, many reports indicate that ticks are well adapted to urban and suburban environments [24,26,43,44]. Ticks that inhabit urban localities originate from tick populations persisting in wild natural habitats around cities and towns, and environmental conditions in both localities, urban and natural, are suitable and promote the development of a tick population [45,46]. In our opinion, in green areas located within the administrative boundaries of large cites, not only are residents exposed to the ticks and the pathogens transmitted by them, but so are their pets, such as domestic dogs and cats.
Lyme disease also has veterinary importance, affecting dogs, cattle, horses, and cats. The most common clinical signs in domestic animals are lameness, loss of appetite, weight loss, and kidney disease [47]. The infestation rates of domestic animals with pathogen-infected ticks are poorly documented in Europe, including in Poland. Even fewer publications report on the proportions of pathogen-infected ticks removed from animal or human skin. As early as the 1990s, in Germany (North Baden), 22% of I. ricinus and I. hexagonus removed from domestic animals were infected by B. burgdorferi, while in Lower Saxony, fewer than 10% of all human skin-attached ticks were Borrelia-positive. The PCR method was used in both studies using 23S rRNA and 16S rRNA genes, respectively [48,49]. In the latest research from Germany (2013-2017), the overall Borrelia infection rate of I. ricinus ticks attached to human skin by real-time PCR was 20.02% [50]. In 2016 in Great Britain, molecular analysis of the ospA gene confirmed that 1.8% of ticks collected from cat skin were Borrelia-positive, and the most frequent species was B. garinii [51]. In 2018 Geurden et al. [43] analyzed the prevalence of Borrelia spp. and A. phagocytophilum by the real-time PCR method. Ticks were collected from dogs and cats in Hungary, France, Italy, Belgium, and Germany. Borrelia spp. were mainly identified in I. ricinus collected from cats (18%) and to a lesser extent in dog-sourced ticks (1%); while A. phagocytophilum was also found in 17% of ticks. In Poland, a study was conducted on ticks collected in 2013 from domestic dogs and cats in the Wroclaw Agglomeration (south-west Poland). Using nPCR, the authors revealed that 22.5% and 21.3% of the ticks were Borrelia-positive, respectively. A comparable level of Borrelia infection between I. ricinus from pets and vegetation indicates that domestic animals may participate in the circulation of these pathogens, and that they do not have zooprophylactic competence [28]. Moreover, it is very well documented that there is a possibility of cross-infection (including B. burgdorferi and A. phagocytophilum) when multiple ticks, infected and non-infected specimens, are co-feeding on one animal [52].
We assumed that adult feeding ticks are characterized by the highest degree of infection due to the transstadial transmission of pathogens. It is known that engorged ticks cause problems in molecular testing. Substances present in mammalian blood can inhibit the PCR amplification. Beichel et al. [48] pointed out that ticks larger than 4 mm inhibit the PCR reaction. Therefore, in our study, only the anterior parts of engorged ticks were analyzed. We have shown that from 165 cases of Borrelia spp., as many as 153 cases were from engorged ticks and only 12 came from non-engorged ticks. The study of Scott et al. [8] showed that 36% of I. scapularis adults collected from 41 mammalian hosts (dogs, cats, humans) were positive for the Lyme disease bacterium. Actually, 35.7% of the examined I. ricinus and 14.1% D. reticulatus collected in the urban area of Olsztyn tested positive for the DNA of Borrelia spp. spirochetes. The dominant species of B. garinii was detected in over 80% of cases. In some contrast to our results, Kubiak et al. [44] showed that the overall infection rate of questing I. ricinus with Borrelia spirochetes was 27.4% in Olsztyn, and the dominant genospecies was B. afzelii (93.1%). It is noteworthy that they collected ticks by flagging while we specifically studied the ticks that were actively parasitizing dogs. Perhaps dogs in Olsztyn are mainly a reservoir of B. garinii, but this should be confirmed by a city-wide dog blood test for the presence of DNA from the Borrelia genospecies. This may be due to the different mechanism of transmission of the spirochete genospecies. Hovius et al. [38] observed a higher infestation of B. burgdorferi s.s. in non-engorged (either questing or attached) I. ricinus (12%) compared to semi-engorged ticks (2%), explaining this phenomenon by loss of infectivity during the start of the feeding phase. According to De la Fuente et al. [53], pathogen transmission by ticks requires many often unexplored tick-pathogen interactions, from the migration of these pathogens from the gut to their secretion in tick saliva. It is possible that B. garinii is not transmitted as quickly as the other two genospecies during tick feeding.
HGA, a zoonotic acute febrile disease, is difficult to recognize because its symptoms are rather nonspecific. European studies warn that the infection, although largely unrecognized, may be widespread in most of Europe, because the pathogen incidence in tick-carriers ranges from moderate to high, with the median prevalence of A. phagocytophilum in European I. ricinus ticks at approximately 3% [18,54,55]. In our study, only 5 out of 402 (1.2%) engorged I. ricinus females showed the presence of A. phagocytophilum DNA. The prevalence of A. phagocytophilum was 6% in the DNA of adult I. ricinus ticks collected from dogs in Latvia, whereas in I. persulcatus and D. reticulatus, the pathogen was not identified [56]. Even though similar studies are scarce, our data did seem rather low for the country. This is because in the urban areas of Wroclaw, 21.3% I. ricinus and 8.1% I. hexagonus ticks isolated from dogs and cats were A. phagocytophilum-positive [29]. This was higher than our rates of tick infection with A. phagocytophilum from 8.7% to 16.0%. These were also recorded in the north-eastern regions of Poland at the beginning of the century, and an extremely rapid increase was observed over the next few years of research, in some locations from 6.6% to a stunning 73.3% [37]. It is also important to note that in Poland, A. phagocytophilum more frequently infects ticks in urban areas than in natural forests [57]. Similar findings were also reported by Grzeszczuk and Stańczak [58] who published extremely disturbing data that 23.7% of A. phagocytophilum-positive I. ricinus ticks were removed from the skin of Bialystok residents (north-eastern Poland).
Nucleotide analysis of the partial sequence of the Anaplasma 16S rRNA gene showed that all amplified sequences belong to one haplotype. BLASTn analysis revealed a 100% similarity with the sequence of A. phagocytophilum from the first described human patient in Wisconsin (USA) and also with A. phagocytophilum from ticks removed from human skin in north-eastern Poland [58,59]. These results indicate that the Aph1-5 haplotype detected in I. ricinus collected from dogs in Olsztyn represents A. phagocytophilum, and that HGA in urban dogs may be significant in the future.
We also addressed the coexistence of B. burgdorferii s.l. and A. phagocytophilum in ticks, because in Pomerania (Pomorze, northern Poland) 5% of examined I. ricinus ticks contained both pathogens. Moreover, research in the urban and suburban forests in the Tri-City area of Gdańsk, Gdynia, and Sopot (northern Poland) showed the simultaneous presence of both pathogens in 8.3% of adult ticks. It is possible that B. burgdorferii s.l. and A. phagocytophilum consolidate in the same foci and often co-infect the same tick vector, thus increasing the risk of a mixed infection [60,61]. At the present time in Olsztyn, we have not observed such a dependence, most likely due to the low incidence (0.96%) of A. phagocytophilum. However, the epidemiological situation should be closely monitored, especially in the areas considered as endemic for tick-borne diseases.

Study Area and Tick Collection
For the analysis we used ticks collected from domestic dogs visiting veterinary clinics in the Olsztyn agglomeration (53 • 47 N 20 • 29 E; 88.33 km 2 ; 173,070 inhabitants) between 2016 and 2018. Olsztyn is the capital of the Warmia-Masuria Province in north-eastern Poland. Within the city there are a large number of parks, squares, and forested recreational areas, which occupy a quarter of the city. These areas also provide an excellent habitat for ticks. Ticks were collected from the dogs by veterinarians and were placed in tubes with 70% ethanol. In the laboratory, the species and sex of the ticks were identified, then, individuals were weighed, measured, and separated into non-engorged and engorged categories. To the non-engorged I. ricinus female group, those weighing on average 2.13 ± 1.15 mg and measuring 3.63 ± 0.50 mm were included; while the engorged group weighed on average 104.26 ± 112.29 mg and measured 7.84 ± 2.36 mm. Non-engorged I. ricinus males weighed and measured 0.94 ± 0.2 mg and 2.7 ± 0.48 mm accordingly, while non-engorged D. reticulatus included individuals weighing on average 4.94 ± 0.98 mg and 4.69 ± 1.45 mg, and measuring 4.64 ± 0.49 mm and 4.38 ± 0.46 mm, for females and males, respectively. Those classified as D. reticulatus engorged females were the largest of the analyzed ticks. They weighed and measured on average 220.29 ± 140.65 mg and 11.2 ± 2.68 mm.

DNA Extraction
To start, the ticks preserved in 70% ethanol were dried. Fully-engorged ticks were bisected, only the anterior end was used. Anterior parts or entire ticks were then crushed using a sterile mortar, moved to 2 mL tubes filled with lysis buffer (A&A Biotechnology, Gdynia, Poland) and were incubated for 2 h at 50 • C. Total DNA was extracted according to the manufacturer's tissue protocol (Micro AX Tissue Gravity, A&A Biotechnology, Gdynia, Poland) and stored at −70 • C.

PCR Conditions
The Borrelia region of the fla B gene (422 bp) was amplified using the primer BFL1/BFL2 [62] ( Table 2) and PCR was run under the following thermal cycle conditions: an initial activation step of 2 min at 94 • C was followed by 40 cycles of 30 s at 94 • C, 1 min at 58 • C, and 1 min at 72 • C. Finally, an extension step of 1 min was performed at 72 • C. For A. phagocytophilum, the primer sets EHR521 and EHR747 [63] (Table 2)

PCR-RFLP Analysis
PCR-RFLP analysis was carried out to identify three genospecies containing the B. burgdorferi s.l. complex: B. garinii, B. afzelii, and B. burgdorferi s.s. The amplified DNA was digested with the TasI endonuclease (Fast Digest Tsp 509I, Thermo Fisher Scientific, Waltham, MA, USA) to obtain the restriction patterns of the different genospecies of the spirochetes. For each positive sample, 10 µL of amplified DNA were digested in a Mastercycler Nexus (Eppendorf, Hamburg, Germany) at 65 • C for 15 min. The reaction mixture contained 17 µL of nuclease-free water, 2 µL of 10 × Fast-Digest buffer, and 1 µL of TasI enzyme. PCR-RFLP products were separated by electrophoresis in a 3% agarose gel and stained with Midori Green dye. DNA fragments subjected to restriction analysis were visualized in the G-BOX Syngene transilluminator. Membership in a given genospecies was determined by TasI restriction fragment sizes (bp) according to restriction patterns (

Statistical Analysis
All ticks were analyzed individually, and the prevalence was expressed as a percentage. Statistical analysis of the results was carried out using the two-sided Fisher's exact test (Prism 7 program, GraphPad Software, San Diego, CA, USA). Prevalence of pathogens was calculated with 95% confidence intervals (95% CI). In order to check whether there was a prevalence of Borrelia in both tested tick species or non-engorged and engorged ticks and a relationship between variables (the occurrence of B. garinii and the other two genospecies), the Chi-square test (χ 2 ) was used. Values of p < 0.05 were considered statistically significant.

Conclusions
This study indicates that domestic dogs in north-eastern Poland are at risk of infection from Borrelia species and A. phagocytophilum. The results of this study demonstrate the potential danger from ticks feeding on dogs. Ticks represent a serious risk of LB for companion animals and for city residents. There is a high percentage of Borrelia-positive ticks in the city of Olsztyn (north-eastern Poland); A. phagocytophilum-positive ticks were identified less frequently. However, veterinarians and physicians should be aware of anaplasmosis among domestic animals and among patients with a tick-bite history.