Distribution of Ixodes scapularis in Northwestern Ontario: Results from Active and Passive Surveillance Activities in the Northwestern Health Unit Catchment Area

The range of Ixodes scapularis is expanding in Ontario, increasing the risk of Lyme disease. As an effective public health response requires accurate information on disease distribution and areas of risk, this study aims to establish the geographic distribution of I. scapularis and its associated pathogen, B. burgdorferi, in northwestern Ontario. We assessed five years of active and passive tick surveillance data in northwestern Ontario. Between 2013 and 2017, 251 I. scapularis were submitted through passive surveillance. The submission rate increased over time, and the proportion infected with B. burgdorferi was 13.5%. Active tick surveillance from 2014 to 2016 found few I. scapularis specimens. In 2017, 102 I. scapularis were found in 10 locations around the city of Kenora; 60% were infected with B. burgdorferi, eight tested positive for A. phagocytophilum, and one for POWV. I. scapularis ticks were found in 14 locations within the Northwestern Health Unit area, with seven locations containing B. burgdorferi-positive ticks. We found abundant I. scapularis populations in the southern portion of northwestern Ontario and northward expansion is expected. It is recommended that I. scapularis populations continue to be monitored and mitigation strategies should be established for rural northern communities.


Introduction
Lyme disease, caused by the bacteria, Borrelia burgdorferi sensu stricto, is the most commonly reported vector-borne disease in North America [1,2]. The bacteria is transmitted in eastern and central North America by the blacklegged tick (sometimes called the deer tick), Ixodes scapularis [3]. Lyme disease typically manifests as an influenza-like illness with erythema migrans. However, if left untreated, the bacteria may spread to other tissues and organs, causing severe manifestations involving the skin, nervous system, joints, or heart [4].
The range of I. scapularis and consequently the risk of Lyme disease is expanding in Canada [5,6]. The number of documented risk areas has increased from one location in Southern Ontario in the

Passive Tick Surveillance
NWHU encourages members of the public within the catchment area to submit tick specimens found on people to any of NWHU's 12 offices throughout the catchment area. Additionally, physicians encourage their patients who have a blacklegged tick to submit it to NWHU. Ticks from animals are not accepted. Locations of ticks are based on the municipality where the submitter resides, and are not specifically geocoded to the exact location of suspected tick exposure. Any I. scapularis that are submitted are sent to the Public Health Ontario Lab (PHOL) for species identification. The PHOL then forwards I. scapularis ticks on to the National Microbiology Lab (NML) in Winnipeg for pathogen detection. Tick specimens submitted to the NML were shipped there on freezer packs and held at 4 C prior to DNA extraction and subsequent testing. Test results are shared by NWHU with the client. Details of the passive tick surveillance system in Ontario are described elsewhere [24].

Active Tick Surveillance
Active tick surveillance activities at NWHU are done by public health inspectors (PHIs) who travel throughout the catchment area to conduct tick dragging per Public Health Ontario's tick dragging standard operating procedure [25]. Active surveillance activities only document I. scapularis collected. After dragging the area, any ticks that are captured are sent to the NML for testing and tick specimens were held under the same storage condition as ticks from passive surveillance. Public health inspectors drag for ticks at various locations in the NWHU catchment area in the spring (May and June) and the fall (September and October) each year.
Prior to 2017, dragging locations were selected based on passive surveillance, anecdotal reports, or perceptions of PHIs of areas that might harbor I. scapularis. Based on previous years of passive surveillance data, Kenora and the surrounding area was the focus of active surveillance in 2017. Also for this year, the NWHU adopted a more systematic and methodological approach to selecting

Passive Tick Surveillance
NWHU encourages members of the public within the catchment area to submit tick specimens found on people to any of NWHU's 12 offices throughout the catchment area. Additionally, physicians encourage their patients who have a blacklegged tick to submit it to NWHU. Ticks from animals are not accepted. Locations of ticks are based on the municipality where the submitter resides, and are not specifically geocoded to the exact location of suspected tick exposure. Any I. scapularis that are submitted are sent to the Public Health Ontario Lab (PHOL) for species identification. The PHOL then forwards I. scapularis ticks on to the National Microbiology Lab (NML) in Winnipeg for pathogen detection. Tick specimens submitted to the NML were shipped there on freezer packs and held at 4 C prior to DNA extraction and subsequent testing. Test results are shared by NWHU with the client. Details of the passive tick surveillance system in Ontario are described elsewhere [24].

Active Tick Surveillance
Active tick surveillance activities at NWHU are done by public health inspectors (PHIs) who travel throughout the catchment area to conduct tick dragging per Public Health Ontario's tick dragging standard operating procedure [25]. Active surveillance activities only document I. scapularis collected. After dragging the area, any ticks that are captured are sent to the NML for testing and tick specimens were held under the same storage condition as ticks from passive surveillance. Public health inspectors drag for ticks at various locations in the NWHU catchment area in the spring (May and June) and the fall (September and October) each year.
Prior to 2017, dragging locations were selected based on passive surveillance, anecdotal reports, or perceptions of PHIs of areas that might harbor I. scapularis. Based on previous years of passive surveillance data, Kenora and the surrounding area was the focus of active surveillance in 2017.
Also for this year, the NWHU adopted a more systematic and methodological approach to selecting dragging locations. Using shapefiles obtained from the Ministry of Natural Resources, we mapped out the forest composition of our catchment area using QGIS Version 2.18.13 (QGIS Development Team, Open Source Geospatial Foundation Project) Potential sites to conduct active surveillance in were identified by the presence and quantity of broad-leaf trees (e.g., oak, aspen, poplar), which is a habitat that has been shown to have a positive association with the presence of I. scapularis [26]. Forest areas located near accessible roads where the most common type of tree is a broad-leaf were identified as areas to be dragged.

Pathogen Detection in Blacklegged Ticks
Blacklegged ticks submitted through passive tick surveillance were tested for the presence of Borrelia burgdorferi, Borrelia miyamotoi, Anaplasma phagocytophilum, and Babesia microti by real-time PCR as previously described [27]. Ticks submitted in 2017 were also screened for Borrelia mayonii [28]. Briefly, QIAGEN DNeasy 96 tissue kits (QIAGEN Inc., Mississauga, ON, Canada) were used for DNA extraction. A duplex screening assay was chosen to screen the samples for Borrelia spp. using the 23S rRNA real-time polymerase chain reaction (PCR) assay, and A. phagocytophilum using the msp2 real-time PCR assay [29]. Analysis for B. microti was conducted using the methods described by Nakajima et al. [30] for the detection of the CCT eta gene. All Borrelia spp.-positive samples were subsequently tested for B. burgdorferi using a confirmatory ospA real-time PCR assay, and B. miyamotoi using an IGS real-time PCR assay. B. miyamotoi-positive samples were further verified using the glpQ real-time PCR assay [27] while B. mayonii positive samples were confirmed using Light cycler melt curve analysis targeting the oppA2 gene as described by Pritt et al. [28].
To facilitate the detection of RNA viruses in blacklegged ticks, RNA was extracted from individual host-seeking ticks collected by drag sampling in the active tick surveillance program using QIAGEN RNeasy 96 kits (QIAGEN Inc., Mississauga, ON, Canada). Eluate, collected during the RNA extraction of individual ticks, contained both RNA and DNA, and was screened for all the DNA-based pathogens described above. In addition, RNA was also tested for evidence of infection with Powassan virus (POWV) using real-time Reverse Transcriptase (RT)-PCR protocols and primers designed against the NS5 gene [31]. The lineage of POWV (i.e., lineage 1 or 2) infecting positive ticks was determined using another set of primers against the NS5 gene as described by Dupuis et al. [31]. To account for possible contamination during extraction and the PCR/RT-PCR procedures, water or blank controls were included in all extractions and PCR/RT-PCR runs, respectively. All of these controls were negative during the course of this study. With the exception of A. phagocytophilum, each targeted pathogen was confirmed in ticks using a screening and confirmatory set of primers and probes.

Data Variables and Sources
Data from I. scapularis specimens collected through the passive tick surveillance program between 2013 and 2017 are stored in a secure, internal SharePoint-based database. Variables collected in this database include: the dates of tick collection; date of submission; municipality of collection; tick species submitted; and pathogen infection status.
Data from the active tick surveillance system collected between 2014 and 2017 was collated into one file and used for our analyses. This data file contains the following variables: location of drag sampling; date performed; number; life stage and species of ticks collected; and pathogen infection status.
The relationship between forest composition and I. scapularis presence was examined by comparing the average forest composition of areas where I. scapularis were found and areas where they were not found. Average forest composition was calculated by determining the average prevalence of each tree species across all sites where I. scapularis were found and all sites where they were not found between 2014 and 2017; four forest sections per dragging site were included in these calculations.

Statistical Analysis
Descriptive statistics (numbers, proportions) were used to report results. Linear trends were examined using the Chi-square test for linear trend. The level of significance was set at p ≤ 0.05 with 95% confidence intervals. Locations of active tick surveillance dragging sites were geocoded and mapped using QGIS Version 2.18.13. Passive tick surveillance results were geocoded and mapped based on the municipality where the submitter resides. All other analyses were completed using Stata software, version 15 (StataCorp, College Station, TX, USA).

Passive Tick Surveillance
From 2013 through 2017, a total of 908 ticks were submitted to NWHU by members of the public. The proportional representation of the different developmental stages was 76.9% female, 15.6% male, and 7.5% nymph. The most commonly-submitted ticks were Dermacentor variabilis, followed by I. scapularis (Table 1). A total of 251 I. scapularis were submitted. There was a general increasing trend in the submission rate of I. scapularis submitted each year (chi square for linear trend: 68.417, p < 0 .001, Table 2). The proportion of submitted I. scapularis infected with B. burgdorferi for all five years was 13.5%. No trend over time was observed (Chi square for linear trend: 0.144, p = 0.7043, Table 2). The geographic distribution of I. scapularis submissions covered a large geographic area, ranging from Red Lake in the northwest to Atikokan in the southeast, and B. burgdorferi was detected in blacklegged ticks from almost every site (Figure 2). In the Kenora area in 2017, 9.7% of ticks submitted through passive surveillance were positive for B. burgdorferi. Two I. scapularis tested positive for A. phagocytophilum, both in 2016, submitted by residents of Fort Frances and were negative for B. burgdorferi.

Active Tick Surveillance
Active tick surveillance activities from 2014 to 2016 yielded a small number of I. scapularis specimens. Four locations were dragged in 2014 yielding a total of four I. scapularis that were captured and tested. One tick tested positive for B. burgdorferi. Six locations were dragged in 2015 yielding six I. scapularis (zero positive). Four locations were dragged in 2016 yielding three I. scapularis (zero positive). In 2017, ten locations around the city of Kenora were dragged yielding a total of 102 I. scapularis. Sixty-one tested positive, equaling a percent positivity of 59.8% (95% CI: 49.6%, 69.4%; Table 3). In addition, eight I. scapularis tested positive for A. phagocytophilum and one for POWV (deer tick virus, lineage 2), all in 2017. Six out of the eight ticks that were positive for A. phagocytophilum were also positive for B. burgdorferi (two females and four males; all found in the Kenora area). The I. scapularis that tested positive for POWV was also positive for B. burgdorferi.
All I. scapularis captured were adults, with the exception of one nymph that was captured in 2017, which was negative for B. burgdorferi. I. scapularis were found in 14 unique locations within the NWHU catchment area, with seven of these locations containing B. burgdorferi-positive (Figure 3).
No relationship was found between forest composition and I. scapularis presence (Table A1).

Active Tick Surveillance
Active tick surveillance activities from 2014 to 2016 yielded a small number of I. scapularis specimens. Four locations were dragged in 2014 yielding a total of four I. scapularis that were captured and tested. One tick tested positive for B. burgdorferi. Six locations were dragged in 2015 yielding six I. scapularis (zero positive). Four locations were dragged in 2016 yielding three I. scapularis (zero positive). In 2017, ten locations around the city of Kenora were dragged yielding a total of 102 I. scapularis. Sixty-one tested positive, equaling a percent positivity of 59.8% (95% CI: 49.6%, 69.4%; Table 3). In addition, eight I. scapularis tested positive for A. phagocytophilum and one for POWV (deer tick virus, lineage 2), all in 2017. Six out of the eight ticks that were positive for A. phagocytophilum were also positive for B. burgdorferi (two females and four males; all found in the Kenora area).
The I. scapularis that tested positive for POWV was also positive for B. burgdorferi.
All I. scapularis captured were adults, with the exception of one nymph that was captured in 2017, which was negative for B. burgdorferi. I. scapularis were found in 14 unique locations within the NWHU catchment area, with seven of these locations containing B. burgdorferi-positive (Figure 3).
No relationship was found between forest composition and I. scapularis presence (Table A1).

Discussion
Active tick surveillance showed the presence of I. scapularis positive for B. burgdorferi around Kenora and Morson. As active tick surveillance has been shown to have low sensitivity but good specificity, the capture of an I. scapularis by drag sampling indicates an emerging, self-sustaining, reproducing population is present in that location [32]. Adult ticks found by active tick surveillance in these areas had a prevalence of B. burgdorferi of about 60%. This is similar to established endemic areas across Michigan (61%) [33]. In addition, an average of two (range: 0-6) human cases of Lyme disease per year between 2011 and 2017 were reported among residents of the NWHU catchment area [34], and previous studies have found an established I. scapularis population in the region [20].

Discussion
Active tick surveillance showed the presence of I. scapularis positive for B. burgdorferi around Kenora and Morson. As active tick surveillance has been shown to have low sensitivity but good specificity, the capture of an I. scapularis by drag sampling indicates an emerging, self-sustaining, reproducing population is present in that location [32]. Adult ticks found by active tick surveillance in these areas had a prevalence of B. burgdorferi of about 60%. This is similar to established endemic areas across Michigan (61%) [33]. In addition, an average of two (range: 0-6) human cases of Lyme disease per year between 2011 and 2017 were reported among residents of the NWHU catchment area [34], and previous studies have found an established I. scapularis population in the region [20]. This indicates that I. scapularis are likely established in the southern portion of northwestern Ontario. On average, over the study period, the percent positivity for passive tick surveillance was 13.5%. Although there was some fluctuation, no significant trend was observed. This was similar to other areas with emerging I. scapularis populations, such as Manitoba (9.7%), Quebec (13.0%), and Nova Scotia (15.1%) [35]. However, this is lower than long-established areas, such as Long Point, Ontario (67%) [36]. In Kenora in 2017, 10% of ticks submitted through passive surveillance were positive for B. burgdorferi. We speculate that the high percent positivity due to active surveillance in 2017 (60%) could be due to the presence of particular areas within the region that have a large I. scapularis population with a high proportion of B. burgdorferi. The presence of one of these areas of high endemicity has been previously described in the Kenora area [20].
Failure to detect I. scapularis during active tick surveillance along the southern border of the health unit catchment area is likely due to poor selection criteria for sampling sites. The southern border areas are characterized by a higher percentage of agricultural land, making some locations sampled less suitable for ticks. The methodological change in 2017 to sample areas with a high percentage of broad-leaf tree cover (which yielded a much greater prevalence of ticks) was not done along the southern border. Rather, specific sites for active tick surveillance prior to 2017 were chosen based on where members of the public tend to visit (e.g., trails, parks). Furthermore, because of the relatively lower sensitivity of drag sampling, the absence of ticks does not indicate they are not present, but rather they could be present at a density below which they could be detected by active tick surveillance [32]. Passive tick surveillance data also indicate widespread occurrence of I. scapularis across NWHU. With an average population density of 0.5 persons/km 2 , it is expected that passive tick surveillance would yield very few submissions [22,30]. However, many ticks were detected across NWHU via passive tick surveillance, leading to the assumption that I. scapularis are prominent across the region where the habitat is suitable.
Passive tick surveillance data indicate that the number of I. scapularis in northwestern Ontario is increasing and expanding. This trend has been previously reported in the Rainy River and Kenora areas of northwestern Ontario [21], and in other areas in Canada [6,7,24]. Northward expansion of I. scapularis has been observed in Minnesota, with areas of risk reaching the US-Canada border [37]. The increase could be due to the awareness of Lyme disease increasing among Canadians, with government entities producing education campaigns for the public and health care professions [38], leading to increased submission of ticks. Moreover, favorable changes in climate allowing more ticks to survive and establish a population is also a factor in this expansion [6]. By 2080, the average temperatures in northwestern Ontario are expected to increase by 7.0 • C in the northern portion and 6.1 • C in the south [39]. As the climate warms, tick dispersal by birds is expected to occur earlier and cover larger distances [14]. In addition, more adventitious ticks will be able to complete their life cycle and establish self-sustaining populations [14,32]. As I. scapularis populations expand across Canada, the proportion carrying B. burgdorferi may also increase, raising the risk of Lyme disease among Canadians and pushing this risk further north [24]. Northward expansion to rural and remote communities is a public health concern because, although population densities are low, these areas have limited access to public health and medical services for tick species detection and Lyme disease testing.
We did not find a significant difference in forest composition between areas where I. scapularis were found and those were they were not. This is likely explained by the fact that the selection of dragging locations was not originally designed for research purposes. As such, a control group of locations where we would expect to find no I. scapularis was not included and as a result most of our dragging locations had relatively similar forest compositions.
Lyme disease and the prevalence and patterns of I. scapularis in northwestern Ontario remain issues of public health importance. This indicates the ongoing need for disease and vector surveillance, public education on the disease and prevention measures, and support for health care providers as it relates to diagnosis and treatment. Passive surveillance will continue to be important to direct the health unit towards possible newly emerging I. scapularis populations as our climate continues to warm, while active surveillance will be important in the determination of specific risk areas for Lyme disease. Climate change adaptation and vulnerability assessments will be an important tool for health units to assess risks, such as Lyme disease, and establish targeted adaptation strategies. This includes establishing or adapting existing mitigation strategies for rural northern communities. For example, NWHU recently launched a mobile application, which allows individuals to submit photos of ticks to the health unit for identification, broadening our services to areas without public health offices.
There are some limitations associated with this study. Passive tick surveillance locations are logged based on where the submitter resides, and, therefore, may not capture the exact location of exposure. Active tick surveillance methodology has changed over time, limiting any trend information over time. In addition, the vast geographic area of the health unit could not be systematically covered by active surveillance due to time and resources available. Therefore, only a limited number of sites were surveyed across the region. While we were unable to detect a correlation between forest composition and the presence and abundance of I. scapularis ticks, a similar analysis should be completed comparing I. scapularis populations across areas of differing forest consumption.

Conclusions
Our study indicates that populations of I. scapularis are well entrenched in the populated regions of NWHU and will likely continue to expand north as the climate continues to change. Vigilance by the public and health care professionals is required to minimize the impact of tick-borne diseases in this region of Ontario.

Conflicts of Interest:
The authors declare no conflict of interest.