Long-Term Study of Borrelia and Babesia Species Distribution in Ixodes Ricinus and Dermacentor Recticulatus Ticks Removed From Humans in Poland, 2016-2019


 Monitoring changes in the prevalence of different Borrelia genospecies/ species in ticks might be an important indicator of risk assessment and of differences in pathogenicity in humans. Furthermore, the evaluation of pathogens in feeding ticks represents the risk of human exposure better than studies on questing ticks. The objective of our study was to assess the prevalence and distribution of Borrelia and Babesia species in ticks removed from humans, in a larger sample collected for several months during four years of studies. We confirmed high Borrelia prevalence, including B. miyamotoi, in ticks removed from humans as well as the shift in Borrelia genospecies/ species frequency of occurrence during the four-year study. Despite the fact that Babesia prevalence was relatively low, the majority of tested isolates are considered to be pathogenic for humans. The results of our study have also shown that Borrelia and Babesia coinfections in ticks are more common in Borrelia-infected ticks. Even if the overall risk of developing Lyme borreliosis after a tick bite in Europe is rather low, the knowledge of prevalence and distribution of Borrelia and Babesia species in ticks might be an important indicator of both tick-borne disease risk and pathogenicity assessment.


Introduction
With 85,000 cases reported annually in Europe, Lyme borreliosis (LB) is the most common vector-borne disease in temperate zones of the northern hemisphere [1]. The estimated incidence of LB in Poland increased dramatically from 20.3 per 100,000 inhabitants in 2007 to 53.6 per 100,000 inhabitants in 2019 (an estimated average increased from 7,735 cases per year in 2007 to 20,614 cases per year in 2019) (National Institute of Public Health -National Institute of Hygiene, Epidemiological reports, www.pzh.gov.pl). However, the reliability of LB incidence data is uncertain due to diagnostic problems and limited reporting [2]. At least ve species of Borrelia -Borrelia burgdorferi sensu stricto, Borrelia garinii, Borrelia afzelii, Borrelia spielmani and Borrelia bavariensis -are known to be pathogenic to humans and each genospecies is believed to be associated with different clinical manifestations. The heterogeneity among B. burgdorferi s.l. genospecies seems to be the main factor causing the regional differences in the clinical expression of human Lyme borreliosis [3]. Borrelia burgdorferi sensu stricto is particularly arthritogenic, B. afzelii primarily causes skin infections, and B. garinii is especially neurotropic. Infection usually begins with an expanding skin lesion, known as erythema migrans which, if left untreated, can be followed by early disseminated infection, particularly neurological abnormalities, and by late infection, especially arthritis or acrodermatitis chronica atrophicans (ACA). Recently, Borrelia miyamotoi has been identi ed as a human pathogen causing relapsing fever in Europe, and little is known about its local impact on human health. Borrelia miyamotoi disease (BMD) has also been con rmed in an immunocompetent patient, and BMD concurrent with Lyme disease has also been described [4].
In Europe, including Poland, other tick-borne diseases such as babesiosis are reported sporadically. About 60 con rmed cases of human babesiosis caused mainly by Babesia divergens have been described so far [5]. Non-speci c clinical symptoms of babesiosis, such as fever, u-like disease, headache, chills, sweats and myalgia, as well as diagnostic which can be a signi cant problem for people with impaired immune system whose percentage in society is constantly increasing [16].
The I. ricinus ticks are competent vectors for many species of pathogenic viruses, bacteria and protozoa. An important problem in the epidemiology of tick-borne diseases is co-infection, i.e. simultaneous, multi-species infections, especially di cult to diagnose in humans [17]. Co-infection in humans and animals might enhance disease severity and may have signi cant consequences in terms of tick-borne disease treatment and diagnosis. For instance, co-infected Lyme disease patients harboured more in uenza-like symptoms than those with Lyme disease alone [7]. In the case of concurrent babesiosis and Lyme disease, co-infected patients experienced a greater number of symptoms for a longer duration than those with Lyme disease alone.
The knowledge of Borrelia prevalence and genospecies distribution is crucial to understand epidemiology as well as the prevention and diagnosis of LB. There is a limited number of studies on particular species prevalence in ticks removed from humans, mainly providing information only on B. burgdorferi (s.l.) complex. In Poland, most of the previously conducted research concerned questing ticks or ticks collected from animals [18][19][20][21][22][23][24][25]. However, the evaluation of pathogens in feeding ticks represents the risk of human exposure better than studies on questing ticks. The aim of our study was to assess the prevalence and distribution of Borrelia and Babesia species in ticks removed from humans in Poland, in a larger sample collected for several months during four years of studies.

Ixodes ricnus Ticks
During four years of study, 1890 I. ricinus ticks were collected from humans: 54 (2.9%) larvae, 1,298 (68.7%) nymphs, 524 (27.7%) females and 14 (0.7%) males. Most of them were collected in 2018-2019 (n = 762 and n = 775, respectively), whereas in 2016-2017 only 335 ticks were tested (n = 126 and n = 227, respectively). The main peak of tick activity was observed in June and the second one in October; however, the mean number of ticks collected in October was almost four times lower (Fig. 1). The number of ticks in each stadium (larvae, nymphs and adults) removed from humans has varied signi cantly between months of study (month x number of I. ricinus tick in each stadium: χ 2 16 = 85.5, p < 0.000). Overall, the median number of larvae collected by month was 6, with a minimum of 2 larva (in May), a maximum of 18 (in July), and no larvae were collected in March-April and November. The proportion of nymphs over the total number of ticks during a particular month of study increased from 67.4% (62/92) Table 2). The relapsing fever spirochete B. miyamotoi was identi ed in 8.4% (95% CI: 5.4-12.3%) of analyzed ticks. Comparison of genospecies/ species distribution in diagnostic ticks removed from humans with those from questing ticks in our previous study [22] revealed that ticks removed from humans were by far more frequently infected with B. myiamotoi (p = 0.003), whereas questing ticks were more commonly infected with B. garinii (p = 0.0001). Detailed results are shown in Fig. 3.

Borrelia Genospecies/ Species Identi cation in D. reticulatus Ticks
Genospecies differentiation of Borrelia infected ticks was successful in 6 out of 8 positive tick samples (75%). All Borrelia isolates were identi ed on the basis of RFLP-PCR analysis as B. afzelii. Species typing was performed on the basis of sequencing of 18S rRNA gene fragment (~ 540 bp product); all positive PCR samples were sequenced. Alignment and BLAST-NCBI analyses revealed the presence of three Babesia species. Nine out of 15 isolates (60%) have shown high similarity level (> 99.5%) to B. microti strain Jena isolated originally from human patients in Germany (EF413181). The nucleotide sequences of ve isolates (33.3%) were identical to B. venatorum isolate from I. ricinus in France (FJ215873). One isolate was identi ed as B. canis with a similarity level of > 99% to another Polish isolate (JN107810).

Discussion
Analysis of available data revealed the high socio-economic impact of Lyme borreliosis on public health systems as well as on quality of life for infected patients [29,30]. In this study, we con rmed high Borrelia prevalence in ticks removed from humans as well as the shift in Borrelia genospecies/ species frequency of occurrence during the four-year study. The results of our study have also shown that Borrelia and Babesia coinfections in ticks are more common in Borrelia-infected ticks.
The ticks removed from humans in Poland were almost exclusively I. ricinus (97%), the most widespread and abundant ticks species in humans in Europe (European Centre for Disease Control & Prevention, 2019). Only a few specimens of D. reticulatus were collected (3%). While almost the whole of Europe is an endemic region for I. ricinus, the geographical range of D. reticulatus in Europe is discontinuous with two main macroregions, and the spreading of D. reticulatus is believed to be associated with the loss of forest area [31]. This tick species appeared to show bimodal activity pattern with the highest density in March-May and September-November, whereas no ticks were collected in summer, which is typical for this tick species [32]. Dermacentor reticulatus ticks were also removed from patients in Germany, Belgium and Poland [33][34][35]; however, the frequency of occurrence of this species does not exceed a few percent.
For I. ricinus, we observed the peak of activity in June which is congruent with the results of our previous study on questing ticks [22] and other studies on seasonality of I. ricinus bites on humans [11,12]. The predominance of nymphs of up to 73% in dependence of month of study was similar to other European studies on ticks collected from humans [33][34][35][36][37][38]. The activity of larvae was the highest in August and September; however, only 54 specimens in total were removed from humans. It is worth noting that the highest number of tick bites occurred during the summer period when people are more likely to be exposed to ticks by spending time outdoors, not only in natural areas. Our previous analysis of the frequency of occurrence of Borrelia spirochetes in ticks collected from areas with varying degrees of anthropopression has shown that although the population density of ticks in natural areas was signi cantly higher, the prevalence of Borrelia infection in I. ricinus ticks collected from natural and urban areas was similar (12% vs. 11%) [22].
To observe a long-term trend, Borrelia spirochetes prevalence as well as species/ genospecies distribution in ticks removed from humans were compared in the course of four years. Surprisingly, between 2016 and 2019, annual Borrelia prevalence in ticks decreased signi cantly from 38-25%. At the same time, the number or Lyme borreliosis cases in Poland decreased slightly from 21,220 in 2016 to 20,614 in 2019 (National Institute of Public Health -National Institute of Hygiene, Epidemiological reports, www.pzh.gov.pl). Similar uctuations in Borrelia prevalence in I. ricinus collected from humans were observed in Germany and Romania [33,37,38]; however, the differences were not so signi cant. Our previous studies have shown that annual Borrelia occurrence in questing I. ricinus ticks in Poland varied from 8-15% between 2013 and 2014 [22]. These inter-annual uctuations in Borrelia prevalence may be due to climatic or other ecological factors affecting tick density or the abundance and, as a result, the availability of reservoir hosts, such as rodents or birds. It has been proven that the relative abundance of the white-footed mouse is positively associated with nymphal infection prevalence value which is regarded as the most important indicator of Lyme borreliosis risk [39].
Overall in Europe, including Poland, the Borrelia prevalence in ticks removed from humans range from 5-29% [33,34,35,37,38,40,41,42,43,44]. In our study, the Borrelia prevalence has differed signi cantly between I. ricinus ticks removed from humans (25%) and questing ticks (11%, [22]). Some results suggest that the abundance of spirochaetes in questing Ixodes ticks may be low (below 300 copies of bacteria) and, therefore, often undetectable, while blood repletion or simply the increased ambient temperature triggers bacteria growth and rises detectability, but possibly only within a short period (around 72 h after changing the conditions) [45,46]. The knowledge of this phenomenon is still limited, and, in consequence, the number of infected Borrelia ticks removed from the host (human) may be higher than it has been evaluated in questing ticks, which could translate into higher risk of tick-borne infections.
Since each tick stadium has only one blood meal from different hosts and the probability of acquiring pathogens increase with every blood meal, the highest prevalence of infection is noted in adults ticks. It is believed that transovarial transmission of Borrelia is rare or non-existent and larval ticks are not important vectors of Lyme borreliosis [47]. Richter et al. [48] have suggested that questing larvae in nature may have acquired Borrelia spirochetes from an interrupted host contact. In our study, we con rmed Borrelia infection in 9% of removed larvae; however, only 54 of them were collected. Detection of the spirochetes in larvae was previously noticed at low prevalence in questing ticks [49] as well as in ticks removed from humans [34,37,38], which strengthens the evidence for transovarial transmission of Borrelia under eld conditions. Nonetheless, Faulde et al. [50] did not con rm the case of acquired Lyme borreliosis following the bite of an infected I. ricinus larva. Hence, the hypothesis of Borrelia transmission from larvae to human need further experimental studies.
Borrelia infection rate in D. reticulatus ticks does not exceed 13%; however, only 63 ticks were tested. The previous studies have shown that Borrelia prevalence in questing D. reticulatus ticks is signi cantly lower [51][52][53]. Nevertheless, the infection rates in engorged D. reticulatus ticks collected from dogs is similar to the results noted in this study [25].
Since different Borrelia species/genospecies are involved in distinct clinical manifestations, it is important to know accurate numbers for the prevalence of a particular species with regard to risk assessment. In our study, the species identi cation by sequencing or RFLP analysis was successful in 52.4% of Borrelia-positive I. ricinus ticks. The Borrelia species / genospecies differentiation revealed that B. afzelii was the most frequent species within four years of study with the prevalence ranging between 58% and 78%. The obtained results are comparable to data on questing and engorged ticks from other European countries (reviewed in 54,23,33,[35][36][37]. Borrelia garinii is believed to be the second dominant genospecies in I. ricinus ticks, followed by B. afzelii [55]. However, in our study, the second most frequent species were B. burgdorferi (10.8%), B. garinii (8.8%) and B. miyamotoi (8.4%). Borrelia valaisiana constituted only 5% of analyzed samples, while B. spielmanii and B. lusitaniae were even less common (1.2% vs. 0.4%, respectively). Similar Borrelia genospecies/ species distribution was noted in questing I. ricinus ticks in our previous studies (Fig. 3, [22]). The low frequency of B. spielmanii and B. lusitaniae could be explained by relatively low abundance of the competent reservoir host for those species, mainly dormice and lizards [56,57]. Coipan et al. [58] have also shown that the infection peak in seasonal dynamics in questing ticks is different for different pathogens, including B. afzelii and non-B. afzelii spirochetes, suggesting that they were acquired from the distinct vertebrate hosts. However, we have not con rmed signi cant differences in Borrelia genospecies/ species distribution between the month of study what might be the result of limited number of non-B. afzelii isolates.
Interestingly, in our study I. ricinus ticks removed from humans were more frequently infected with B. miyamotoi than questing ticks (8.4% vs. 2.2%, p = 0.003) [22], whereas the latter were signi cantly more often infected with B. garinii (8.8% vs. 21.3%; p < 0.0001). Nevertheless, the questing ticks were collected between 2012 and 2015 from selected natural areas of North-Eastern Poland and urban areas of Central Poland, whereas ticks were removed from habitants of multiple regions of the country and were delivered to laboratory between 2016 and 2019. Therefore, the differences in Borrelia prevalence in questing and engorged ticks might be the result of speci c eco-epidemiological conditions within the habitats affecting the availability and abundance of reservoir hosts for ticks as well as for Borrelia spirochetes. We have also observed that B. afzelii prevalence was noted more often in ticks removed from humans than in questing ticks (63% vs. 57%, p = 0.060). Similar results were obtained by Springer et al. [37] and Waindok et al. [33]. Coipan et al. [58] have shown that B. afzelii and B. bavariensis were signi cantly more frequent in human cases than in questing ticks, which is related with the fact that both are mammal-associated Borrelia species. Rodents are mainly reservoir hosts for B. afzelii as well as for I. ricinus larvae and nymphs; therefore, this phenomena might be also the result of spatial overlap between habitats of rodents with human activity areas and where the risk of tick bites is signi cant [37]. Nevertheless, no B. bavariensis isolates were observed in this study. It is likely due to using the single restriction enzyme DdeI which is not able to distinguish the recently described B. bavariensis from B. garinii [19]. However, the sequence analysis Borrelia isolates from 2016-2017 did not con rm the presence of B. bavariensis species.
Monitoring changes in the prevalence of different Borrelia genospecies/ species in ticks might be an important indicator of risk assessment and of differences in pathogenicity in humans [59]. The statistical analysis in our study has shown considerable annual variation in the frequency of non-B. afzelii genospecies/ species occurrence. Similar year-to-year variations were shown in I. ricinus ticks removed from humans in Germany [37,38] and in questing ticks collected in Europe [9,22,60]. It is well-known that the distribution and prevalence of Borrelia spp. in ticks show signi cant temporal and spatial variations. Surprisingly, in our study, the annual prevalence of B. miyamotoi was relatively high (up to 15.8% in 2016) compared to other European studies in questing as well as feeding ticks where the prevalence usually did not exceeded 5% [19,22,23,35,[61][62][63][64][65]. In contrast, Springer et al. [37] have con rmed B. miyamotoi infection in 7.4% of I.ricinus ticks removed from humans. Breuner et al. [66] have shown that single I. scapularis nymphs effectively transmit B. miyamotoi while feeding and transmission can occur within the rst 24 h of nymphal attachment. Additionally, probably due to the overlap of endemic areas for B. miyamotoi with B. burgdorferi s.l. complex, co-infections of B. miyamotoi with other spirochete species in I. ricinus ticks and humans have been observed [22,23,37,38]. Taken together, this data indicates that the risk of B. myiamotoi infection in Poland should not be underestimated. So far, only one case of human B. miyamotoi infection has been diagnosed [67]. However, Fiecek et al. [67] suggested that in case of the patients who do not meet the criteria for neuroboreliosis (presence of B. burgdorferi antibodies only in serum, no antibodies in PMR), B. miyamotoi disease should be considered. According to the National Institute of Public Health -National Institute of Hygiene in Poland (epidemiological reports), in 2013 only 14% of all reported cases with neurological symptoms (n = 1267) met the clinical and laboratory criteria of neuroborreliosis (detection of antibodies in PMR) [67].
Co-infections in ticks are frequently reported. This is likely due to a large variety of animals from which they can ingest blood, exposing the ticks to any pathogens currently infecting the hosts, including bacteria, parasites and viruses. In the present study, we have also investigated the occurrence of Borrelia coinfection. We have con rmed that 2% of tested I. ricinus ticks carried two Borrelia species and triple infections were detected only in 0.7% of ticks. The observed rate of coinfection prevalence was signi cantly lower than in feeding I. ricinus ticks in other European studies [33,37]. The mechanism by which Borrelia co-exists with other microbial pathogens within the tick, including different Borrelia species, remains unexplored. Furthermore, the extent to which different Borrelia species or strain engage in interactions or how multi-species/strain infections might in uence spirochete loads in ticks and, consequently, on transmission to humans and pathogenicity is yet to be discovered. Competition between strains of B. burgdorferi s.l. in the vertebrate host has been shown in eld studies [68] and experimental infections [69]. Field studies on I. ricinus population have found in coinfected questing nymphs that the spirochete load per strain decreased with increasing strain richness, and this result provides indirect evidence for competition [70]. Nonetheless, the low prevalence of coinfection with different Borrelia species has suggested that the risk of this type coinfection in humans in Poland is rather negligible.
In Europe, the majority of human babesiosis cases are caused by Babesia divergens [5]. However, in Poland so far only B. microti infections in humans have been noted [71][72][73][74]. Additionally, the molecular studies of questing I. ricinus ticks in Poland have shown that the B. microti species occurred signi cantly more often than B. divergens [75][76][77]. In the current study, we have con rmed the occurrence of three Babesia species, out two of them (B. microti and B. venatorum), are considered to be pathogenic for humans. Nonetheless, the Babesia prevalence in I. ricinus removed from humans is rather low (1.3%) and similar to other European studies on engorged as well as questing I. ricinus ticks [9,35,65,78].
The recent studies concentrating on Babesia microti and B. burgdorferi infections in rodents and ticks have indicated that coinfection with these pathogens is common in vectors and enzootic hosts with a greater probability of coinfection than predicted by chance, and they have suggested that co-infection provides a survival advantage for both pathogens [17].
Alekseev et al. [79] went one step further and put forward that B. microti infection can only survive in I. persulcatus in combination with Borrelia spp. Serological studies indicate that coinfection with B. microti and B. burgdorferi is also common in humans [80]. In endemic regions in the United States, almost 40% of Lyme disease patients reported concurrent babesiosis, while up to 25% of babesiosis patients also had Lyme disease (reviewed in [17]). Co-infection in humans and animals might enhance disease severity and may have signi cant consequences in terms of tick-borne disease treatment and diagnosis. Moreover, babesiosis and borreliosis can present with similar clinical manifestations [17]. In our study, Babesia-positive I. ricinus ticks were signi cantly more often observed among Borrelia-positive ticks (2.7%) than among ticks non-infected with Borrelia (0.8%). Therefore, our results seem to con rm the presence of positive interaction among these two pathogens; however, the molecular mechanism of these facilitation remain still unclear.
In conclusion, our study con rmed relatively high Borrelia prevalence in ticks removed from humans with signi cant annual variation of spirochete genospecies/ species. In spite of low D. reticulatus abundance, the prevalence of B. afzelii in this tick species is signi cant. Although B. afzelii constitutes the majority of detected isolates, the risk of B. miyamotoi disease in humans should not be underestimated. Analysis of Babesia prevalence suggests that risk of human babesiosis is rather negligible, which is consistent with babesiosis cases reported in Poland. Even if the overall risk of developing Lyme borreliosis after a tick bite in Europe is 4% [81], the knowledge of prevalence and distribution of Borrelia and Babesia species in ticks might be an important indicator of both tick-borne disease risk assessment and varying pathogenicity in humans.

Ethics approval and consent to participate
Written informed consent was obtained from all individual participants included in the study. We con rmed that all

Tick Collection and Identi cation
The ticks were delivered directly or by post to Diagnostic Laboratory of Parasitic Diseases and Zoonotic Infections AmerLab Ltd up to 5 days after removal from skin. Only ticks attached to skin were collected. The ticks were removed from habitants of multiple regions of the country and were collected from March to November in 2016-2019. Ticks were morphologically identi ed in terms of species and developmental stage. Specimens that could not be identi ed due to extensive damage induced by the removal from the skin were not included in the study.

DNA Extraction and PCR Analysis
Individual larvae, nymph and adult ticks were sterilized to avoid contamination and then homogenised. Genomic DNA from ticks was isolated with Genomic Tissue Spin-Up kit (AA Biotechnology) or DNeasy Blood & Tissue Kits (Qiagen) according to the manufacturer's protocol. Genomic DNA was used for molecular screening for spirochetes by ampli cation agellin gene ( aB) marker, with published primers [26]. Initial PCR conditions were modi ed as follows: initial denaturation in 95°C for 5 min, 35 cycles of denaturation in 95°C for 30 s, 30 s of primers annealing in 52°C, and elongation in 72°C for 80 s with the nal elongation in 72°C for 7 min. Nested PCR was performed with minor modi cation: denaturation in 95°C for 20 s and annealing in 55°C for 20 s, elongation in 72°C for 60 s. For B. miyamotoi detection among positive samples, speci c primers for aB marker were used [22]. Babesia spp. were detected and identi ed using GR2 and GF2 primers targeting the fragment of 18S rDNA. The primers and thermal pro les used in this study were previously described [27].
Negative controls were performed in the absence of template DNA. PCR products were visualized on 1.5% agarose gels stained with Midori Green Stain (Nippon Genetics Europe, Düren, Germany).

Borrelia and Babesia species identi cation
A Borrelia-positive samples from ticks collected in 2016-2017 and Babesia-positive samples from ticks collected in 2016-2018 were sequenced by a private company (Genomed S.A., Poland) in both directions. Obtained nucleotide sequences were analyzed using BLAST NCBI and MEGA v. 7.0 software [28] for sequence alignment and species typing.
Restriction fragment length polymorphism (RFLP) was used to differentiate Borrelia-positive isolates at the genospecies level obtained in 2018-2019. Positive amplicons after nested-PCR were digested with the restriction enzyme HpyF3I (Thermo Fisher Scienti c, USA), which recognizes the 5'C↓TNAG3' sequence [26]. The digestion was performed according to the producer's protocol in 37°C for 2 h. The enzyme was heat-inactivated at 65°C for 15 min. The digestion products were separated on 2% agarose gel, visualized and archived in the GelDoc-It imaging system (USA). The obtained restriction patterns enabled the recognition of the species of B. burgdorferi complex and B. miyamotoi.

Statistical Analysis
Statistical analysis was performed using IBM SPSS Statistics v. 25.0 software. Prevalence of Borrelia and Babesia infection (percentage of ticks infected) was analyzed by Maximum Likelihood techniques based on log-linear analysis of contingency tables (HILOGLINEAR). For analysis of the prevalence of Borrelia and Babesia in ticks, we tted the prevalence of pathogens as a binary factor (infected = 1, uninfected = 0) and then year (4 levels: 2016-2019 for Borrelia and 3 levels: 2016-2018 for Babesia), month (March-November), and tick stadium (larvae, nymphs, adults). P-values < 0.05 were considered statistically signi cant.

Declarations
Con ict of interest: The authors declare that they have no con ict of interest.