Occurrence and Identification of Ixodes ricinus Borne Pathogens in Northeastern Italy

In Europe, Ixodes ricinus is the main vector for tick-borne pathogens (TBPs), the most common tick species in Italy, particularly represented in pre-alpine and hilly northern areas. From 2011 to 2017, ticks were collected by dragging in Belluno province (northeast Italy) and analyzed by molecular techniques for TBP detection. Several species of Rickettsia spp. and Borrelia spp. Anaplaspa phagocitophilum, Neoerlichia mikurensis and Babesia venatorum, were found to be circulating in the study area carried by I. ricinus (n = 2668, all stages). Overall, 39.1% of screened pools were positive for at least one TBP, with a prevalence of 12.25% and 29.2% in immature stages and adults, respectively. Pathogens were detected in 85% of the monitored municipalities, moreover the presence of TBPs varied from one to seven different pathogens in the same year. The annual TBPs prevalence fluctuations observed in each municipality highlights the necessity of performing continuous tick surveillance. In conclusion, the observation of TBPs in ticks remains an efficient strategy for monitoring the circulation of tick-borne diseases (TBDs) in a specific area.


Introduction
Pathogens transmitted by ticks (TBPs) are responsible for the majority of the vectorborne diseases in temperate North America, Europe, and Asia [1]. The economic impact of tick-borne diseases (TBDs) is significant and increases every year but unfortunately the combined public health impact of TBDs remains mostly unquantified [1,2].
Given the fact that TBDs could severely concern both human and animal health, surveillance programs for TBDs and TBPs, based on harmonized One Health approaches, have been implemented in several European countries in recent years [3,4].
In Europe, Ixodes ricinus (Acari: Ixodidae) is considered one of the primary vectors of multiple pathogens that affect human and animal health [5]. These pathogens (i.e., virus, bacteria, and protozoa) circulate in enzootic cycles, alternating between ticks and suitable animal hosts. In addition, infected I. ricinus can transmit TBPs to humans, causing several TBDs [6,7].
The most frequently diagnosed zoonoses transmitted by I. ricinus are Lyme borreliosis (LB) and tick-borne encephalitis (TBE) [8][9][10]. Lyme borreliosis (LB) is caused by three species of spirochetes consisting of the B. burgdorferi sensu lato (s.l) complex (Borrelia burgdorferi sensu stricto (s.s), B. afzelii and B. garinii). Tick-borne encephalitis virus (TBEv, Flaviviridae) is described as the agent of the human TBE disease. This investigation began after several tick bite advisories from forestry workers and outdoor workers (professional hazard) and the cooperation with Carabinieri Corps and Forestry Team (Belluno department) carried out the monitoring activities and tick sampling during the years 2011-2017.
The aim of this study was to estimate the occurrence of TBPs in I. ricinus ticks collected in Belluno province.

Ticks Surveillance
The study area involved 20 municipalities in Belluno province, in which 39 monitored sites were found positive for tick presence, corresponding to 95.12% of total sites investigated during the seven years of study.
During the sampling period, 187 dragging transects of 100 m 2 each were performed (total 18,700 m 2 ) within 39 sites positive for ticks monitored 1 to 38 times. In total, 2668 ticks were collected, all belonging to I. ricinus. Nymphs (2062) and larvae (331) were the most frequently collected development stages, followed by adults (147 males and 128 females, total 275) and immature ticks, with an adult ratio of 8.7:1 (Tables 1 and A1). Collected specimens were pooled according to their life stage, sex, date, and sampling site, resulting in a total of 596 pools (25,296 and 275 pools for larvae, nymphs, and adults, respectively).
A considerable variability in tick numbers per collection (1 to 126 ticks/collection, median = 5) was seen throughout the study period, in addition to two peaks in activity of I. ricinus recorded in spring (May-June) and autumn (October-November). The peak of tick presence was characterized from May (n total = 801, 30.0%) to June (n total = 657, 24.6%): 60.9% of nymphs and 73.8% of adults were collected, while the majority of larvae (81.3%) were collected in autumn (September, October).
Mean observed densities of I. ricinus per 100 m 2 were 11.1, 1.8, and 1.5 per nymphs, larvae, and adults, respectively. The maximum number of nymphs per sampling was 160 specimens, followed by larvae (110) and adults (42).

Tick-Borne Pathogen Detection
After molecular analysis of 2668 host-seeking ticks (all stages) belonging to 596 pools, 233 pools (39.1% of total tested) were positive for at least one TBP, ranging from one to three different pathogens. No tick was found to be infected with the TBE virus, although human TBE cases were reported in the same area during the surveillance period [46]. Tick density was higher in late spring (May-June), whereas the pool positivity rate (PPR) was higher in  (Table 2). Co-infected adults were collected from April to June, while multiple TBP presence in pooled nymphs was observed in late spring and autumn ( Figure 1). June, while multiple TBP presence in pooled nymphs was observed in late spring and autumn ( Figure 1).  The highest TBP prevalence was detected in adult ticks (29.2) and was higher in females (32.0) compared to males (23.1). All tick stages were found to be infected with at least one TBP and R. helvetica and N. mikurensis were detected in each stage ( Table 3).
Despite adult ticks being more likely to be infected than nymphs with an overall prevalence of 29.2 (versus prevalence of 11.2 in nymphs) and higher prevalence for each TBP, in the latter a higher TBP diversity was detected, given that B. garinii and Ba. venatorum were detected only in nymphs. The highest TBP prevalence was detected in adult ticks (29.2) and was higher in females (32.0) compared to males (23.1). All tick stages were found to be infected with at least one TBP and R. helvetica and N. mikurensis were detected in each stage ( Table 3).
Despite adult ticks being more likely to be infected than nymphs with an overall prevalence of 29.2 (versus prevalence of 11.2 in nymphs) and higher prevalence for each TBP, in the latter a higher TBP diversity was detected, given that B. garinii and Ba. venatorum were detected only in nymphs.
Molecular investigation revealed the presence of coinfection in adults (1.8%) and multiple TBP detection in pooled nymphs (16.9%). Five adult ticks (1.8% of total) sampled in four different sites and periods were coinfected with two pathogen species (see Table 4 for details). Eight nymph pools (composed of 10 individuals each) coming from four different sites located in three different municipalities during six samplings (2011, 2012, 2014, and 2015) were found to be positive for three TBPs, while 42 nymph pools (from 1 to 13 specimens) collected during 30 different samplings in eight municipalities during the whole studied period were positive with two TBPs. In one site (Col della Feda), two nymph pools collected the same day were positive at four different TBPs (i.e., R. helvetica, B. garinii, B. burgdoferi s.s., and A. phagocytophilum). All the TBPs, except Ba. venatorum, were detected in different combinations: R. helvetica and A. phagocytophilum were the most frequent combination (12 pools), followed by R. helvetica and B. afzelii (9 pools) and R. helvetica and N. mikurensis (5 pools) ( Table 5).
All the 20 municipalities monitored tested positive for tick presence and among them, 17 (85%) tested positive for at least one TPBs during the study period ( Figure 2). In three municipalities (Sedico, Sospirolo, Ponte nelle Alpi) the circulation of more than six TBPs was demonstrated. The highest diversity in TBPs was detected in the Sospirolo municipality, with eight different pathogens circulating in the whole study period and seven TBPs co-circulating in the same year. Table 5. Multiple detection of TBPs in pooled I. ricinus nymphs (combination of three and two TBPs in the first and second box, respectively).

TBPs Combination N Pool
Rickettsia helvetica + Borrelia garinii + Anaplasma phagocytophilum   Despite Sedico, Sospirolo, and Ponte nelle Alpi municipalities registering the biggest diversity in TBP species occurring per year, the highest TBP prevalence was reached in La Valle Agordina (100.0%), Ospitale di Cadore (50.0%), and Alpago (44.4%) (see Table 6). In municipalities monitored for several years, notable prevalence fluctuations can be observed (e.g., Limana registered a minimum TBP prevalence of 6.7 in 2016 and a maximum TBP prevalence of 100.0 in 2012) (see Table A1 for geographical details). In general, the mean prevalence of TBPs in Belluno municipalities during the study period was 21.7%.
During a retrospective study (unpublished data), B. miyamotoi presence was investigated in a subsample of stored ticks collected in Belluno province during 2011-2017. Of the 343 (57.5%) analyzed pools, 15 were positive for B. miyamotoi (details are available in the Appendix A, Table A3) but unfortunately the data about the statistical prevalence of this emerging pathogen were unavailable.

Discussion
In this study, the presence and occurrence of endemic and emerging TBPs in ticks collected in northeastern Italy was deeply investigated, achieving epidemiological and biological TBPs and TBDs knowledge. Our data highlight the co-circulation of nine different pathogens in the study area, with an overall TBP prevalence of 12.5% in ticks in immature stages and of 29.2% in adults. As expected, the overall TBP prevalence was higher in adults than nymphs [51].
At a complex level, the most prevalent pathogen reported in our study was Rickettsia spp. (13.7%; 5.0% in immature stages and 8.7% in adults) followed by Borrelia spp. (12.6%; 2.6% in nymphs and 10.0% in adults).
Rickettsiae are known to be vertically transmitted; in fact, we found positivity in larvae with a prevalence of 1.0%. The Rickettsiae species detected in our study (R. helvetica and R. monacensis) are the main TBPs transmitted by I. ricinus circulating in Europe [52], and our prevalence appears to be lower than that previously reported in Italy, which ranged from 24.8% (northern Italy) to 18.4% (Central Italy) [41,53,54]. Rickettsia helvetica has been detected in I. ricinus ticks in at least 24 European countries [55], with a wide range of prevalence from 0.5% (Island of the Baltic Sea) to 66% in the Netherlands [3,56]. In northern Italy, R. monacensis was detected in the past with a higher prevalence (3.7-4.5%) than that reported of our study (0.3% in nymphs, −1.8% in adults) [57,58]. This could be explained when considering that Rickettsia spp. prevalence could be influenced by several factors, including the season, year of sampling, environment, and type of tick hosts [5].
The B. burgdorferi s.l. prevalence detected was lower compared to a previous study in northeastern Italy (4.7% in nymphs and 17.6% in adults) [38] and to other similar studies carried out in northwestern Italy (10.3-10.5%) [53,59], in the Alpine area (6.3%) in the province of Trento [60], in the Po River Valley (18.0%), and in Central Italy (20.0%) [61].
In our study, we found four genospecies (B. burgdorferi s.s., B. afzelii, B. garinii, and B. valaisiana). The most prevalent was B. burgdorferi s.s., followed by B. afzelii, which was different from a study conducted in northwestern Italy in which B. garinii and B. afzelii were the genospecies most frequently identified in questing ticks [53]. In Central Europe, B. afzelii, together with B. garinii, is one the most common genospecies with the highest prevalence rate [62]. Despite the first detection of B. miyamotoi in north Italy dating back to 2016 [38], during a retrospective study (unpublished data) B. miyamotoi presence was detected in Belluno province in 2011 and had been continuously present until 2017, with the exception of 2014. Considering that B. miyamotoi seems to be well established in the Belluno area, more attention should be paid in cases of non-specific febrile syndrome.
The most prevalent pathogen, at a species level, was A. phagocytophilum. The prevalence was higher in adults (7.6%) than in nymphs (2.9%) and was higher than the overall prevalence found previously in questing I. ricinus ticks (1.8%) in northern Italy [30]. In Europe, the infection rates range from 0.8% to 14.0% [63,64], with a greater risk of infections in eastern Europe than in western Europe [30]. The low infection rate is due to inability of A. phagocytophilum to be transmitted transovarially, and larvae can only be found infected after a meal on a bacteriemic host [22,65]. Accordingly, in our study no positivity for this pathogen was found in larvae.
Neoehrlichia mikurensis was detected in all stages with a prevalence of 0.3%, 1.6%, and 2.9% in larvae, nymphs, and adults, respectively; these results are lower compared to findings in adult ticks (10.5%) in a previous study conducted in the same area [58] but higher than data coming from northwestern Italy (2.0% in nymph and adult I. ricinus collected in a peri-urban park) [41]. In Europe, the prevalence of N. mikurensis ranged from 0.2% (Poland) [66] to 6.4% (Switzerland) [67] and in general, the median N. mikurensis infection rate was greater in western Europe than in Eastern Europe [8]. Despite this pathogen being discovered very recently, it is present in the Belluno area, confirming its high suitability for TBPs.
Few studies have been carried out in Italy to evaluate the distribution and prevalence of the zoonotic Ba. venatorum. In our study, Ba. venatorum was found only in nymphs with a prevalence of 0.1%, lower than the prevalence found previously in nymphs collected in northeastern Italy (2.4%) [29] and in the bordering province of Trento (3.8%) [30]. Transovarial transmission has been demonstrated [68] but in this study we did not find the pathogen in larvae, possibly because of the very low prevalence in this tick life stage. Babesia microti is associated with human infection [69] and it is considered a potential risk in north Italy [30], whereas it was not detected in our study.
Compared to central and eastern Europe, Italy is considered a low-risk region for TBEv. The virus seems to be restricted to areas of the northeastern part of the country [46], where it is historically endemic with a prevalence ranging from 0.2% to 2.5% [58,70,71], within the range registered in European endemic areas (0.1-5.0%) [72,73]. In our study TBE was undetected, probably because this agent has a typical distribution and presentation which is particularly scattered and circumscribed in foci; these characteristics reasonably allowed the pathogen to escape from the sampling [73]. To improve the tick positivity for TBEv detection, it would be useful to provide integrated human cases and tick monitoring surveillance [74].
In our study, co-infections in single adult ticks were reported in a similar percentage (1.8%) as in a previous study conducted in Italian humans (1.2%) [75]. The presence of multiple pathogens in ticks could cause co-transmission of pathogens to humans, affecting the medical diagnosis, severity of the disease, and prognosis [76,77]. In particular, in our study the dual co-infection between Borrelia spp. and Rickettsia spp. occurred, which is confirmed to be one of the most frequent co-infections as reported in Romania, Switzerland and Belgium [76,78,79]. Co-infection of Borrelia spp. with Babesia spp., suspected to enhance the severity of LB, was not observed in our study [80,81].
The majority of the ticks collected by dragging were nymphs (77.3%); this is probably a result of the dragging technique used [82,83]. Nymph and adult ticks are the most important stages in the transmission of pathogens to humans. Nymphs are the most dangerous because their small size and their activity period (spring and summer), which is the highest human outdoor activity time in temperate climates [84]. The I. ricinus density trend data (seasonality and distribution) reported in our study confirms what was already described in central Europe [11,85,86], although tick density in the Belluno area was lower than in other European areas [87]. The two peaks in tick abundance (May-June and October-November) correspond to the peak of TBP detection. Similarly, co-infection and multiple TBP detection in pools analyzed were recorded. Adult ticks showed a higher prevalence for each TBP compared to immature stages, where the higher TBP diversity was detected in nymphs.
All the monitored municipalities tested positive for tick presence and different species of TBPs were detected. Our results highlight the considerable variation of annual prevalence of TBPs in each municipality and detected fewer common species (e.g., Ba. venatorum). Regular monitoring is necessary to determine TBP prevalence of emerging or new TBPs. Tick surveillance should continue in order to detect abnormal prevalence peaks due to an imbalance between climatic conditions and reservoir availability [11,88].
The high diversity (up to nine species), proximity (four different TBPs in ticks collected within less than 100 m 2 ), frequency of detection (seven different species/sites/years), and prevalence of TBPs characterize the Belluno province as a high-risk area for TBP transmission. Tick surveillance is necessary to assess TBP prevalence and occurrence and it allows for the identification of the different genospecies circulating in a investigated area. These data would improve prospective surveillance of TBDs in the focus area by providing more insight into the ecological and epidemiological features of TBDs [9].
Understanding and mapping I. ricinus' spread and prevalence is pivotal to assess the risk of TBPs spreading. Furthermore, the assessment of TBDs temporal and spatial trends would be significantly improved by social communication of messages about risk related to TBDs to policymakers, stakeholders, and the citizens [9,89].

Study Area and Sampling Method
Tick collection was conducted in 20 municipalities in the Belluno province, Veneto Region (northeastern Italy). Sampling activity was performed in sites with human frequency for both work and leisure activities, such as parks, hiking trails, start of climbing routes, and peri-urban recreational areas (Table A1).
In order to cover the two peaks in tick activity (spring and autumn), sampling activity was carried out each year from 2011 to 2017 and from April to November [46,86].
The technique used by forest rangers and the local health unit for the collection of questing ticks was 'tick dragging', using a 1 m 2 white flannel cloth dragged across the top of the vegetation or forest floor in a designated transept (100 m) and regularly checking for the presence of ticks [58].
Collected ticks were identified based on morphological keys [90,91], and were pooled and stored at -80 • C until molecular analysis for the pathogen identification.

Molecular Analysis
Single adult ticks, pooled nymphs (maximum 13 specimens per pool), and larvae (maximum 22 specimens per pool) were homogenized in 600 µL of Phosphate Saline Buffer (PBS) with a 5 mm bead (Qiagen) using the instrument TissueLyser II (Qiagen). Then, 150 µL of supernatant was used for the nucleotide extraction using the All Prep DNA/RNA Mini Kit (Qiagen, Valencia, CA, USA), according to the manufacturer's instructions, and then kept frozen at −80 • C. RNA was screened for the detection of TBE virus using a TaqMan Real-Time PCR, as described elsewhere [92]. DNA was screened for the detection of B. burgdorferi s.l. and N. mikurensis using traditional PCR, as described elsewhere [93,94], and by using in-house SYBR Green Real-Time PCR (rtPCR) for the detection of A. phagocytophilum, Rickettsia spp., and Babesia spp. The target genes, primers used, and related references are listed in Table A2. Each reaction was carried out in a total volume of 20 µL, containing 10 µL of Quanti-Fast SYBR Green PCR Master mix 2× (Qiagen GmbH, Hilden, Germany), sense and reverse primers (concentration reported in Table 1), and 3 µL of DNA. Amplifications were performed in a StepOnePlus™ instrument (Applied Biosystems, Foster City, CA, USA). The thermal profile consisted of 5 min of activation at 95 • C, followed by 40 cycles at 95 • C for 15 s (denaturation), specific annealing temperature (Table 1) for 30 s (annealing), and 60 • C for 30 s (extension). Following amplification, a melting curve analysis was performed by slowly raising the temperature of the thermal chamber from 60 • C to 95 • C to distinguish between specific and non-specific amplification products. To ensure the effectiveness of DNA and RNA extraction, a PCR targeting the 18S rRNA gene internal control and a TaqMan Real-Time PCR targeting the 16S rRNA gene were applied, respectively [92,95]. Negative (sterile water) and positive controls (DNA or RNA of pathogens) were included in each run. PCR products were examined on 7% acrilamide gels stained with SYBR Gold®Nucleic Acid Gel Stain 1× (Thermo Fisher Scientific) and visualized on a Molecular Imager®Gel DocTM XR System (Biorad). DNA amplification products were directly sequenced for species identification. The amplicons were purified and sequenced in both directions with the same primers used for PCR and qPCR, using a 16-capillary ABI PRISM 3130xl Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). Sequences were aligned and compared with those available in GenBank by Basic Local Alignment Search Tool (BLAST-http://blast.ncbi.nlm.nih.gov/Blast.cgi; accessed on 5 January 2018).
Phylogenetic analysis related to Borrelia burgdorferi genospecies and the Rickettsia species identified in this study were performed using the representative sequences, from 2012 to 2017, based on years and site. Phylogenetic trees (Figures A1 and A2) can be inserted into Appendix A.
The DNA of a pool subsample (343/596 pools, 57.5%) was screened for the presence of Borrelia miyamotoi using a Traditional PCR targeting~900 bp of the glpQ gene [17].

Statistical Analysis
Tick density was calculated as average number of ticks per 100 m 2 [96]. Pool positivity rate (PPR), corresponding to the number of positive pools/total pool of ticks examined, was calculated (pool positive for more than one TBPs count as 1). Estimate prevalence (hereafter named Prevalence) for variable pool size together with uncertainly intervals (hereafter named confidence interval, CI) for each TBP and for TBP in general in each municipality (both annually and mean) were calculated through PoolTestR package [97]. Estimation of prevalence based on presence/absence tests on pooled samples was obtained using the script codes provided by PoolTestR package (see the package description for details). All data cleaning and preparation and graphics were conducted using R statistical software version 4.1.0 [98] and the package Tidyverse [99]. Maps and spatial data manipulation were carried out using ESRI ArcMap (ArcGIS Desktop: Release 10.5.1. Redlands, CA, USA: Environmental Systems Research Institute. Copyright© 1999-2017).

Conclusions
Continuous tick surveillance and TBP screening is of pivotal importance and should be maintained to collect updated data on TBP prevalence, occurrence of different genospecies circulating in the province, and to estimate the risk related to TBPs in the population. These aspects have been highlighted through this study, confirming that Belluno province is highly endemic to TBPs, with co-circulation of up to nine different TBPs. Furthermore, these data should be used to carry out risk communication campaigns aimed at sharing knowledge about different tick species present in the area, as well as the pathogens they can transmit and preventive measures against infectious tick bites. Furthermore, these data should be used to carry out risk communication campaigns aimed at implementing preventive measures against infectious tick bites. In conclusion, a multidisciplinary 'One Health' approach is crucial to perform eco-epidemiological research and surveillance specifically focused on the occurrence of ticks, their infection with pathogenic microorganisms, as well as on the presence of tick maintenance and reservoir vertebrate hosts.     Figure A1. Phylogenetic tree based on Fla gene partial sequences of Borrelia burgdorferi sensu lato. Sequence dataset was analyzed using MEGA v6.0 [103]; the tree was constructed using the neighbor-joining (NJ) method and bootstrap analysis (1000 replicates) based on the ClustalW    Figure A1. Phylogenetic tree based on Fla gene partial sequences of Borrelia burgdorferi sensu lato. Sequence dataset was analyzed using MEGA v6.0 [103]; the tree was constructed using the neighborjoining (NJ) method and bootstrap analysis (1000 replicates) based on the ClustalW algorithm. Significant bootstrapping values (>70%) are shown on the nodes. The tree shows the representative sequences of this study based on years and sites. • Reference sequences from Genbank. algorithm. Significant bootstrapping values (>70%) are shown on the nodes. The tree shows the representative sequences of this study based on years and sites. • Reference sequences from Genbank. Figure A2. Phylogenetic tree based on rOmpB gene partial sequences of Rickettsia spp. Sequence dataset was analyzed using MEGA v6.0 [103]; the tree was constructed using the neighbor-joining (NJ) method and bootstrap analysis (1000  Figure A2. Phylogenetic tree based on rOmpB gene partial sequences of Rickettsia spp. Sequence dataset was analyzed using MEGA v6.0 [103]; the tree was constructed using the neighbor-joining (NJ) method and bootstrap analysis (1000 replicates) based on the ClustalW algorithm. Significant bootstrapping values (>70%) are shown on the nodes. Rickettsia conorii was used as outgroup. The tree shows the representative sequences of this study based on years and sites. • Reference sequences from Genbank.