Borrelia burgdorferi and Borrelia miyamotoi are tick-vectored zoonotic pathogens maintained in wildlife species. Tick populations are establishing in new areas globally in response to climate change and other factors. New Brunswick is a Canadian maritime province at the advancing front of tick population establishment and has seen increasing numbers of ticks carrying B. burgdorferi, and more recently B. miyamotoi. Further, it is part of a region of Atlantic Canada with wildlife species composition differing from much of continental North America and little information exists as to the presence and frequency of infection of Borrelia spp. in wildlife in this region. We used a citizen science approach to collect a wide range of animals including migratory birds, medium-sized mammals, and small mammals. In total we tested 339 animals representing 20 species for the presence of B. burgdorferi and B. miyamotoi. We have developed new nested PCR primers and a protocol with excellent specificity for detecting both of these Borrelia species, both single and double infections, in tissues and organs of various wildlife species. The positive animals were primarily small non-migratory mammals, approximately twice as many were infected with B. burgdorferi than B. miyamotoi and one animal was found infected with both. In addition to established reservoir species, the jumping mouse (Napaeozapus insignis) was found frequently infected; this species had the highest infection prevalence for both B. burgdorferi and B. miyamotoi and has not previously been identified as an important carrier for either Borrelia species. Comprehensive testing of tissues found that all instances of B. burgdorferi infection were limited to one tissue within the host, whereas two of the five B. miyamotoi infections were diffuse and found in multiple systems. In the one coinfected specimen, two fetuses were also recovered and found infected with B. miyamotoi. This presumptive transplacental transmission suggests that vertical transmission in mammals is possible. This finding implies that B. miyamotoi could rapidly spread into wildlife populations, as well as having potential human health implications.
Citation: Zinck CB, Lloyd VK (2022) Borrelia burgdorferi and Borrelia miyamotoi in Atlantic Canadian wildlife. PLoS ONE 17(1): e0262229. https://doi.org/10.1371/journal.pone.0262229
Editor: Brian Stevenson, University of Kentucky College of Medicine, UNITED STATES
Received: October 26, 2021; Accepted: December 20, 2021; Published: January 21, 2022
Copyright: © 2022 Zinck, Lloyd. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study was funded by a Venture grant from the Canadian Lyme Disease Foundation to CBZ (2018; https://canlyme.com) and a Discovery Grant (RGPIN/4426-2015) from the Natural Science and Engineering Research Council of Canada (https://www.nserc-crsng.gc.ca/index_eng.asp) to VKL.The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Borrelia burgdorferi and B. miyamotoi are zoonotic spirochete bacteria found throughout the world. B. burgdorferi is the principal causative agent of Lyme disease in North America, whereas B. miyamotoi causes Borrelia miyamotoi disease, a relapsing fever-like disease not unlike Lyme disease . These two bacteria are vectored by Ixodid ticks, primarily Ixodes scapularis and Ixodes pacificus in North America [2–5]. Both Borrelia species propagate through a series of reciprocal vector and host transmissions. Larval and nymphal ticks acquire the bacteria by feeding on previously tick-infected animals, they maintain the bacteria through molting (transtadial transmission) to then transmit the infection to future hosts, which can thus receive pathogens collected from all prior hosts [5, 6]. In addition, B. miyamotoi is able to be transmitted transovarially in ticks to allow the production of ticks infected from hatching [7, 8].
Ixodes species ticks have been expanding their range globally in response to climate change and natural and anthropomorphic factors [9–12]. New Brunswick, Canada is an Atlantic Canadian province bordered by the Atlantic Ocean and, on land, by Maine, USA, Quebec, Canada, and Nova Scotia, Canada and is on the emerging front of tick population establishment . Endemic regions are regions where I. scapularis populations are known to be established and both ticks and wildlife are found infected with B. burgdorferi. It can be challenging to document newly established endemic regions. At the time of this study, New Brunswick had two identified tick endemic regions in the southwest portion of the province, however, tick recoveries are reported throughout the province and B. burgdorferi infection of both I. scapularis and dogs are elevated in the southern third of the province, relative to the provincial average. This broad region encompasses the areas of this study [10, 13]. No endemic or risk areas in Maritime Canada have been identified for B. miyamotoi as it is understudied due to its relative rarity in ticks in this region, relative to B. burgdorferi. The presence of B. burgdorferi reservoir competent wildlife species is well documented in Maine and Quebec [3, 14, 15] and many of these species are also present in New Brunswick. Additionally, New Brunswick has some Atlantic species of non-migratory birds and mammals not abundant in continental North America . The potential for a variety of Borrelia species and strains to establish in the New Brunswick wildlife is further enhanced by the intersection of both trans-Atlantic and North American bird migration flyways over the province, which provides a potential source of ticks infected with non-local Borrelia [17–22].
The primary reservoir of B. burgdorferi in North America is the white-footed mouse (Peromyscus leucopus) and the closely related deer mouse (Peromyscus maniculatus), with overlapping ranges covering central and eastern North America [8, 23–25]. A number of other B. burgdorferi reservoir species have been identified: the eastern chipmunk (Tamias striatus), eastern and western grey squirrels (Sciuris griseus, Sciuris carolinensis), and some shrew species (Blarina brevicauda, Sorex cinereus) [14, 23, 26–28]. These are animals that have been shown to competently maintain and transmit the infection to ticks, maintaining B. burgdorferi between tick generations. B. miyamotoi has been studied less than B. burgdorferi [6, 29]. Some European B. miyamotoi reservoir species have been identified [29–32] and in North America, the white-footed mouse has been identified as a B. miyamotoi reservoir [8, 25]. The deer mouse has not been studied for B. miyamotoi and is the dominant wild mouse species in Atlantic Canada. The impact of B. miyamotoi on wildlife is unknown, B. miyamotoi has a higher blood spirochete load in mammals than B. burgdorferi suggesting they inhabit different niches in the host, so may have different effects on host health . While B. miyamotoi is capable of vertical transmission within ticks [7, 8], vertical transmission has not been previously shown in mammalian hosts. However vertical transmission in mammals is well documented for another relapsing fever Borrelia, B. duttonii [8, 33, 34].
The Borrelia transmission cycle requires both infected ticks and animals so to assess the risk of Borrelia exposure in a region, both need to be monitored [35, 36]. To investigate the incidence of B. burgdorferi and B. miyamotoi in Atlantic Canada, we used a citizen science approach to survey a wide range of animals including migratory birds, medium-sized mammals, and small mammals. This design was employed to yield a wide variety of species as opposed to a targeted approach as no previous information existed on which animals, if any, are acting as reservoirs for Borrelia spp. in New Brunswick. To determine the infection status of an animal, multiple tissues and organs were tested. An important consideration for an animal to be a reservoir is that there be a negligible loss of fitness from carrying the bacteria as this ensures its ability to transmit the bacteria between ticks. In human and animal models, late-stage infections among symptomatic hosts are caused by the dissemination of B. burgdorferi throughout the body [37–39]. In contrast animals with robust immune responses or antibiotic treatment show infections localised to a single organ [40, 41]. Determining the distribution of Borrelia spp. within the tissues of an infected animal can inform whether this animal is potentially an unaffected carrier or if they could be suffering a loss of fitness.
Materials and methods
To maximize the number and diversity of species collected two non-targeted collection approaches were used. In the summer of 2016 (May-August) medium-sized animals were collected twice-weekly through roadkill collection. All roadside specimen recovery locations were recorded using GPS, and UTM coordinates (S2 Table and S2 Fig). Small animals such as rodents, squirrels, and smaller birds, were collected by using a public participatory approach where, primarily, cat-killed specimens were provided by the public from May 2016 to December 2017. These collection approaches provided animals and tissues that were either frozen (small mammal community donations) or of varying degrees of freshness (medium sized mammals, roadside collections), which precluded culturing from the sample tissues. Mammals and birds were identified morphologically by Gay Hansen, curator of the Mount Allison university ornithology collection; specimens too damaged for morphological identification but from which tissues could still be obtained were also tested and are listed as “unknown species”. Locations for these animals were noted as the civic addresses of the donors or locations nearest to where they were found.
Roadkill dissections were performed on the roadside. Animals were surface sterilised with 70% ethanol prior to dissection and samples were held at approximately 4°C for the duration of the sampling trip until frozen at -20°C prior to DNA extraction. Small animals were stored whole at -20°C prior to dissection and then dissected under sterile conditions in the laboratory. All laboratory dissections were performed under sterile conditions in a fume hood. Animals were surface sterilised with 70% ethanol prior to dissection, and all implements were autoclaved and stored in 70% ethanol between uses and scalpel blades were discarded after single use. Dissections were performed wearing approved personal protective equipment. Tissue samples were placed in sterile microcentrifuge tubes and all specimens were stored in separate sealed plastic bags. An initial dissection was performed to remove liver and kidney samples, if present, as well as bladder, muscle, and skin samples, if present. Skin samples were taken from the ear and tail, if applicable, for mammalian specimens and from the neck and breast for avian specimens. Positive samples identified in the initial screening of kidney and liver, as well as representative negatives were dissected to obtain samples of the brain, lungs, heart, spleen, and gonads. In one instance, fetuses that were present were also collected and dissected from their amniotic sacs to obtain tissue exclusively from the fetus. Blood was not considered due to the difficulty in obtaining samples from deceased animals in variable condition. Representative negatives were species matched negative animals independent of location. In 5 cases, animals initially thought to be negative because their kidney and liver were negative but were later found to have other infected tissues. In these cases, additional species matched negatives were selected. All specimens were thoroughly inspected for ticks, but only rabbit ticks (Haemaphysalis leporispalustris) were found on two snow shoe hares. As ticks rapidly detach from dead hosts , ticks would be unlikely to be found on less infested dead hosts.
Approval for the use of tissues obtained from necropsy was given by the Animal Care Coordinator of Mount Allison University, protocol number NEC 2016–01. Approval for collection of all specimens excluding migratory birds was given through the government of New Brunswick, Fish and Wildlife branch, permit number SP16-006. Approval for collection of migratory birds was given through the Canadian Wildlife Service, permit number SS2025.
For each animal, target tissue samples were removed and stored at -20°C prior to DNA extraction. Whole DNA was extracted using the AquaGenomicTM (Multi target Pharmaceuticals) tissue protocol. For all samples this procedure was performed under sterile conditions in a sterilized fume hood. Tissue samples were placed in 100μL AquaGenomic solution with proteinase K (5μg) for a 90-minute incubation at 55°C followed by a 10-minute deactivation step at 95°C. The samples were then hand vortexed for approximately 30 seconds then centrifuged at 12000Xg in a desktop microcentrifuge for 5 minutes. The supernatant was removed and mixed with an equal volume of isopropanol to precipitate the DNA, mixed and re-centrifugation at 5 minutes to collect the precipitated DNA. The pellet was rinsed with 70% ethanol and left to air dry overnight at room temperature in sterile conditions, before being resuspended in 50μL nuclease-free water (ThermoFisher) and stored at -20°C.
Detection of Borrelia spp. used nested-PCR (nPCR) to enhance specificity. Outer primers (Rrs and Rrl; Table 1) designed to detect multiple Borrelia species were as described by DiBernardo et al. . These primers target the 16-23s intergenic spacer, a conserved region of the Borrelia genus, and detect multiple Lyme and relapsing fever Borrelia species . Inner primers specific for B. burgdorferi and B. miyamotoi were then designed and used (Burg23S and Miya23S; Table 1). All nPCR reactions were set up in a surface- and UV-sterilized laminar hood and performed in 0.5mL tubes in a Labnet MultiGene OptiMax Thermal Cycler with 12.5μL GoTaqGreen polymerase (Promega), 8.5μL of nuclease free water, 1μL of the forward and reverse primers (Sigma), and 2μL of sample DNA. Negative controls, pre and post-reaction, in which animal DNA was replaced by nuclease free water were performed and animal dissection, DNA extraction, PCR reactions and gel electrophoresis were all performed in different work spaces with independent air flow. The amplification program for the initial reaction was 4 min at 94°C, 35 cycles of 94°C 1 min, 50°C 1 min, 72°C 1 min, followed by 72°C for 10 min . The second round of amplification was performed as described above: 5 min at 95°C, 40 cycles of 95°C 30 sec, annealing temperature (Table 1) 30 sec, 72°C 30 sec, followed by 72°C for 10 min. PCR products were size-separated by agarose gel electrophoresis (1.2% agarose in sodium borate buffer). The liver and kidney were targeted initially from each specimen for B. burgdorferi and B. miyamotoi. A positive from either tissue meant that the animal was considered positive. All positive amplicons were confirmed by Sanger sequencing through Genome Quebec, McGill University, Canada.
Sensitivity and specificity testing of B. burgdorferi detection
All amplicons produced during testing were sequenced to assess primer specificity. Both primer sets created in this project were sequence-validated for use in each of the species tested. While not every species had a representative positive, there were no false positives found through sequencing, indicating that the primers do not cross-react with any of the mammalian or avian host species’ DNA. Amplification sensitivity was assessed by spiking negative animal DNA samples with DNA isolated from pure B. burgdorferi B31 culture (ATCC #35210). The spiked DNA was assessed by nPCR as described above. DNA concentration was measured using a NanoDrop 1000 Spectrophotometer (ThermoFisher). The number of genomes present was calculated based on the total genome size, including plasmids, of B. burgdorferi strain B31 chromosome on Genbank (Accession: NC_001318). No sensitivity testing was performed for the B. miyamotoi detection as no commercial culture or whole genome DNA was available.
Sequence and phylogenetic analysis
To confirm sequences, FinchTV software (FinchTV 1.4.0, Geospiza, Inc.; Seattle, WA, USA; http://www.geospiza.com) was used to manually compare opposite direction reads and resolve any discrepancies through reference to the original chromatogram. An nBLAST analysis of the NCBI Genbank database  nucleotide collection (nr/nt) with no exclusions or organismal selection was performed selecting for highly similar sequences (megablast), most recently on October 15, 2018. Only sequences confirmed in both directions were used to identify sequences with the closest match to the amplified product (S1 Table). Amplified sequences from all Borrelia-positive samples have been deposited in the Genbank database; accession numbers, sequences and nBLAST results are indicated in S1 Table. Maximum-likelihood phylogenetic trees were constructed following the protocol established by Hall . Both trees (Fig 1A and 1B) included outgroups and reference sequences for comparison retrieved from Genbank. To build the trees, first model selection was run using Mega X , with partial deletion and a 95% site coverage cutoff. For the B. miyamotoi tree, two sequences (MH796099 & MH796103) from different tissues of the same organism were omitted, as the overall analysis is constrained by the shortest sequence included. Both trees (Fig 1A and 1B) were based on the Tamura-Nei model  and bootstrapped for 1000 replicates . The B. burgdorferi tree was rooted on B. bissettii strain SCCH8 (Genbank accession #JF791735.1), while the B. miyamotoi tree was rooted on B. burgdorferi strain B31 (Genbank Accession #NC_001318.1). Both trees are shown with any branch appearing in less than 50% of the bootstrap replicates collapsed.
B. burgdorferi (A) and B. miyamotoi (B) aligned sequences were run through Clustal Omega to produce phylogenetic trees. Additional sequences retrieved from Genbank were included in the trees. All sequences have their accession number, and where they come from the same specimen are indicated with their sample number (e.g. C006). Both trees are after 1000 bootstrap iterations and have any branches produced in less than 50% of the iterations removed. The B. burgdorferi tree (A) is rooted on B. bissettii (strain SCCH8) and the B. miyamotoi tree (B) is rooted on B. burgdorferi (strain B31).
Spatial trend analysis was done using ArcMap (ESRI 2015). Samples were plotted with their location of recovery and infection status for both B. burgdorferi and B. miyamotoi. A Getis-Ord spatial clustering analysis was done for both Borrelia species . Determination of any species predictive of Borrelia infection was done by logistic regression using Rstudio (2016) and the package ROCR  to determine if there were species with a higher likelihood of infection than others.
B. burgdorferi and B. miyamotoi in wildlife species
The goal of this study was to identify Borrelia infected species in New Brunswick, assess the distribution of Borrelia in their tissues as an indirect way to gauge their ability to act as pathogen reservoirs and, lastly, to determine if they were present outside of known endemic areas. As such, collection methods with as little species bias as possible were required. Use of vehicle- and cat-killed specimens was successful in obtaining a plentiful and diverse collection of species (Table 2).
The primers developed to detect B. burgdorferi and B. miyamotoi for this study were specific; amplicons of the expected size yielded Borrelia sequences and there were no false positives found through sequencing, indicating that the primers do not cross-react with any of the mammalian or avian host species’ DNA (S1 Table). Host-derived amplicons of smaller sizes were observed, but readily distinguishable from Borrelia-derived sequences by size (S2 Fig). While Borrelia was not found in every host species, sensitivity tests were performed on the Borrelia burgdorferi specific inner primers using negative host animal DNA spiked with B. burgdorferi B31 DNA. Reliable detection occurred with as few as 100 genomes B. burgdorferi added, or 50 genomes/μL of sample DNA. The ratio of B. burgdorferi genomes to wood thrush (Hylocichla mustelina, avian test sample) genomes at reliable detection was 1:3,090. The ratio of B. burgdorferi genomes to meadow vole (Microtus pennsylvanicus, mammalian test sample) at reliable detection was 1:13,200. This indicates that this protocol is able to detect few Borrelia genomes despite the overwhelming presence of host DNA. Further, coinfection of B. miyamotoi and B. burgdorferi was able to be simultaneously detected, although the presence of multiple B. burgdorferi strains could confound sequencing.
Over the two-year study period, a total of 339 animals were obtained and tested. Every animal had a minimum of two tissues tested by nPCR for both B. burgdorferi and B. miyamotoi. B. burgdorferi and B. miyamotoi were found in a wide range of mammalian host species: meadow vole, deer mouse, eastern grey squirrel, jumping mouse, shrew, porcupine, and American crow. Of these 339 animals, 13 had sequence-confirmed B. burgdorferi infections, 6 had sequence-confirmed B. miyamotoi infection, and one animal (jumping mouse) was infected with both Borrelia species (Table 2).
Sequence analysis of both B. burgdorferi and B. miyamotoi amplicons showed that all differed slightly and clustered with each other more closely than with B. burgdorferi strain B31 (Fig 1), as would be expected for wild-derived samples. There was no obvious clustering by host species or host tissue, which is not unexpected for a pathogen transmitted by a non-species-specific vector. The 16-23S intergenic spacer is a non-coding region which allows for it to have highly variable sections. As such conclusions cannot be made as to the presence of different strains from these sequences.
The most frequently B. burgdorferi-infected species were porcupine (2/21), crow (1/11), and jumping mouse (2/21), all with approximately 9.5% of individual specimens with sequence-confirmed B. burgdorferi. For B. miyamotoi, the jumping mouse was the species with the highest Borrelia detection at 3/21 (14.2%). To determine if specific species were more likely to have an infection, we used logistic regression, separately for B. burgdorferi and B. miyamotoi. The variable numbers obtained for each species, as well as the low representation of some species reduced the power of statistical analysis. For B. burgdorferi, a logistic regression model was created with a McFadden r2 of 0.098, an accuracy based on ROC curve of 0.734, and no significant species found. The model for B. miyamotoi was similar with a McFadden r2 of 0.292, an accuracy based on ROC curve of 0.899, and no significant species found. A targeted approach directed at specific species would allow more robust statistical analysis to identify a single predictive species, however, this was not the goal of this study.
Having no single significant predictive species allows all the samples to be used in the spatial analysis to determine if infected animals were clustered, potentially representing an isolated endemic area, or non-clustered, representing many isolated or a large region of bacterial penetration into wildlife. To test spatial clustering of B. burgdorferi and B. miyamotoi positive samples, a Getis-ord General G was calculated. This is a measure of spatial autocorrelation to determine whether high or low values are clustered . In this context, it is measuring whether positive samples are clustered or randomly distributed. This was calculated using a fixed-Euclidean distance as a conservative measure. Spatial analysis was done separately for B. burgdorferi and B. miyamotoi. B. burgdorferi had a p-value of 0.426. This distribution of positives does not differ from random so there are no clusters. For B. miyamotoi, a p-value of 0.088 was found. While not significant at a threshold of 0.05, when taken with a positive z-score (z-score = 0.796109) suggests a trend towards clustering positives. In both cases, resolution of this analysis is limited by the non-random sample collection; medium-sized animals were recovered from roadways and smaller samples originated primarily from the southeastern region of New Brunswick near the university. Nevertheless, for both B. burgdorferi and B. miyamotoi, the lack of significant clustering means that infected animals were not restricted to single geographic areas and were, instead, widely dispersed.
Tissue distribution of Borrelia
Fourteen animals, representing 5 species (Table 3), identified as having B. burgdorferi and/or B. miyamotoi based on testing of liver and kidney, along with negative specimens of the same species, then had additional target tissues tested to assess in what tissues and organs the spirochetes could be found. For both B. burgdorferi and B. miyamotoi, the liver was the most common tissue found infected, and was the most common single organ to be found infected in an animal. For both B. burgdorferi and B. miyamotoi, one animal was found where the infection was exclusively detected in the brain; a shrew infected with B. burgdorferi and a jumping mouse infected with B. miyamotoi. All identified B. burgdorferi infections were in a single tissue/organ whereas there were two instances of diffuse infection for B. miyamotoi. One of these animals was a female, pregnant, jumping mouse (Napaeozapus insignis) co-infected with B. burgdorferi and B. miyamotoi. For this animal, in addition to the tissues listed in Table 3, two late development fetuses (Fig 2) were removed and tested separately for the presence of both bacteria. B. miyamotoi was found in the placentae and in both fetuses.
(A) Still in the amniotic sac and with attached placenta (left; thin line on right is a dissecting pin). (B) Removed from the amniotic sac and separated from placenta. Both fetuses have separation between their digits, and resolved tails, indicated with arrows.
Each animal was tested for B. burgdorferi and B. miyamotoi with positives denoted as “B” or “M”, respectively. All specimens were from New Brunswick or Cumberland County, Nova Scotia, which directly borders New Brunswick. The year of collection is given in the second column.
This study reports B. burgdorferi and B. miyamotoi in multiple members of five non-migratory small mammal species within New Brunswick, including the jumping mouse, a species not previously identified as an important reservoir species. We found that B. miyamotoi, unlike B. burgdorferi, was found disseminated into multiple organs and we documented the first evidence of trans-placental transmission of B. miyamotoi in mammals. In order to maximize number and diversity of animal species collected, we used a public participatory or citizen/community science approach. An important advantage of this approach is that it allows collection of a large number of specimens with limited cost. By collecting road and cat-killed specimens, we also avoided the intentional sacrifice of animals. However, such non-targeted collection methods have disadvantages, in this case resulting in geographically uneven collections and uneven distributions of samples among species. These approaches also could have been biased towards recovery of slower, ill, or reckless individuals.
B. burgdorferi is expected to be found in a variety of wildlife species. However, the Atlantic region is home to some species not abundant in the rest of Canada, and equally, lacks species common in central and southern parts of North America  and there has been no investigation of which species would be suitable for ongoing surveillance in the Canadian Maritimes. The most common North American reservoir species is the white-footed mouse (Peromyscus leucopus), but this species is rare in New Brunswick. A close relative, the deer mouse (Peromyscus maniculatus) is abundant, however, and both the deer mouse and white-footed mouse have similar reservoir capabilities [23, 31, 52, 53]. Site studies in American locations with high incidences of Lyme disease but only small populations of white-footed mice have shown the deer mouse , and the meadow vole (Microtus pennsylvanicus)  to be the principal reservoirs of B. burgdorferi. Other species found to be reservoirs are the eastern chipmunks, eastern grey squirrels, and short-tailed shrews [14, 36]. These species are abundant in New Brunswick. One species not previously identified as either a reservoir species or one having a high infection prevalence is the jumping mouse, Napaeozapus insignis. While Anderson et al.  did note that infected jumping mice have been found, in our study this species had the highest prevalence of infection for both B. burgdorferi and B. miyamotoi singly, as well as a double infection.
Borrelia miyamotoi was first discovered in Japan in 1995, and in North America in 2001 and since detected in humans, ticks, and wildlife worldwide [8, 31, 55]. Currently B. miyamotoi reservoir species are believed to be primarily the same as B. burgdorferi . Our findings support this distribution as the only animal in which we identified B. miyamotoi that had not been previously described as a reservoir species was, again, the jumping mouse Napaeozapus insignis. B. miyamotoi has been documented in ticks found across Canada, with an approximately 1% infection prevalence in New Brunswick as of 2012 . Our identification of B. miyamotoi in ~2% of a variety of wildlife species from different regions of New Brunswick in 2016/2017 indicates that B. miyamotoi, along with B. burgdorferi, is established in wildlife in the province. While it is not currently as prevalent as B. burgdorferi, proliferation in the wildlife, particularly if vertical transmission is possible, will lead to an increase in the likelihood of human and companion animal infections.
While most of the species found in this study have been shown to be B. burgdorferi reservoirs [8, 32, 36], as our specimens were collected primarily as road or cat kill, this study could not directly test the ability of any given species to transmit infection to a feeding tick and so cannot show formally that these species are acting as reservoirs. This question can, however, be addressed indirectly by looking at tissue distribution of the Borrelia infections.
Tissue distribution of Borrelia
In our study, the liver and kidney were targeted for initial screening. The liver was chosen as the primary organ for direct bacterial detection due to it being a common site of disseminated B. burgdorferi infection in experimental models, as well as a common refuge tissue for immune evasion in long-term infections . It has also been used as a source organ for direct culturing of live B. burgdorferi [36, 56]. The kidney is similarly an organ often used for the culturing of live B. burgdorferi spirochetes [26, 36, 56, 58]. In human and animal models, the dissemination of B. burgdorferi is thought to be dependent on time since infection. At the onset of infection, the bacteria localise to the skin and tissue surrounding the site of the tick bite. The bacteria then disperse systemically through the circulatory and lymphatic systems before they begin to extensively replicate and colonise organs [37–39]. This is followed by the late disseminated stage, where bacteria are detectable in numerous systems and reflect failure of the host immune system to control the B. burgdorferi infection. In either host species with robust immune responses or late-stage infection hosts treated with antibiotics, B. burgdorferi persistence in refuge tissues is commonly observed [40, 41]. This persistence is thought to entail B. burgdorferi evading the immune response and antibiotic treatment by adopting less metabolically active states in one or few organs. In mice models of infection, the liver was a common refuge tissue for B. burgdorferi where the bacteria evaded Kupffer cells, and thus the host immune response . However, this same study also reported the highest levels of B. burgdorferi being found in the liver within 24 hours of injection into the bloodstream . Since blood is filtered through the liver it is a likely organ for initial infection when B. burgdorferi leaves the circulatory system.
Of the ten B. burgdorferi positive animals found in this study, eight had an infection only detectable in the liver. This means that these animals were either killed within 24 hrs of the B. burgdorferi reaching the dissemination phase from a natural infection via tick bite, or, more likely, they are in a persistent late infection stage in which B. burgdorferi are confined to the refuge tissue to evade the host immune response. The other two infected animals also showed only one infected tissue, a meadow vole (Microtus pennsylvanicus) with the infection present in a kidney, and a shrew (Sorex maritimensis) with infection confined to the brain (Table 3). The single-organ distribution of infected animals in this study, as well as the narrow window of localized distribution early in infection suggests that most of the animals recovered were likely at the long-term persistent infection stage; the host immune response was keeping the bacteria from complete dissemination through the body, but not able, or yet able, to eliminate it completely.
In humans, lab mice (Mus musculus), rhesus macaques, dogs, and horses, B. burgdorferi and B. miyamotoi infection causes a range of symptoms, from erythema migrans to arthritis and carditis [4, 40, 57, 59]. The functional effects of infection can be obtained or inferred from patient reports of symptoms, supplemented with serological findings and sometimes culturing or direct molecular detection in blood or cerebrospinal fluid [4, 60]. Assessment by biopsy or autopsy is rarely performed . In animal disease models, effects of infection are monitored based on health cues such as weight and activity level, and with B. burgdorferi the swelling of joints is a common measurement in mice [24, 37, 40]. Terminal tissue testing can be used to correlate the presence of infection in the tissues with physiological and histological signs of disease. Correlating this information with the location of infection, inferences can be made as to the effects of infection on the biologically relevant host animals in the wild, although such extrapolations are not straightforward. In human and non-human models such as Mus musculus, even after antibiotic intervention to remove the majority of the B. burgdorferi infection, symptoms can persist, which has been attributed to a number of causes, including persistent infection, physical damage arising from infection, or autoimmune response to bacterial proteins [59, 62, 63].
In contrast, the white-footed and deer mouse can both maintain a persistent infection and mount an immune response without obvious effects [24, 25, 64]. For other wildlife species, it is not known if infection impairs health, as in Mus musculus, or has no observable effect as reported for Peromyscus spp.; capturing and experimentally infecting a large number of species of wild animals is impractical. Additionally, with wild animals, it is impossible to be certain that signs of disease are caused by the Borrelia infection or arise from some other infection or pre-existing condition. Thus, cautious extrapolations from the location of infection is the most practical way of inferring physiological and functional effect of infection on wildlife health. The distribution of Borrelia spp. within an infected animal can be used to infer whether the infection is active, suggesting an ineffective immune response, or contained in a single organ. If an animal is negatively affected by the bacterial infection, it will reduce their effectiveness as a reservoir host by increasing their mortality. If they are affected, this could lead to the number of presumed late-stage infections being over represented in this study as they are more likely to be killed. If they are unaffected, they could function as ideal reservoirs, similar to the white-footed mouse, and be more capable of transmitting the infection to ticks.
Unlike B. burgdorferi, there is little pre-existing information on B. miyamotoi infection progression in wildlife or animal models. B. miyamotoi is known to cause recurrent spirochetemia, and serious febrile illness and arthritic symptoms in humans and lab mice, but its effects on wildlife is not well described . B. miyamotoi is thought to share the same reservoirs as B. burgdorferi, however, the bacteria are present in greater concentrations in the mammalian blood than B. burgdorferi, similar to many relapsing fever group Borrelia [6, 31, 56]. In this study, seven animals were found infected with B. miyamotoi, of these five had the infection present in a single tissue (Table 3). These animals include three jumping mice, having infection solely in the brain, liver, and kidneys. Additionally, a deer mouse and an eastern grey squirrel also had the infection only present in the liver. If B. miyamotoi exhibits the similar immune evasion strategies to B. burgdorferi , this would suggest that these animals also represent persistent late disseminated infections associated with a host immune response preventing fully disseminated infection, but not having eliminated infection. Confirmation would require controlled infections to be studied in captive populations of each host species. In addition to animals with single-organ infection, a jumping mouse and meadow vole showed systemic infections of B. miyamotoi. Whether this occurred because of the stage of infection or because the specific host was unable to control the infection is a question of ecological significance that requires further study. In the single B. miyamotoi-infected meadow vole identified, the bacteria was detected in the liver, bladder, brain, and lungs. This would suggest that B. miyamotoi was able to cross the blood brain barrier as well as establish in less vascularised tissue such as the bladder. In this study, B. miyamotoi was found in the brains of three infected animals, showing a capability to cross the blood-brain barrier, as seen in human cases . The infected jumping mouse with a systemic B. miyamotoi infection was also co-infected with B. burgdorferi. In this specimen B. miyamotoi was found throughout the animal, but B. burgdorferi was only present in the liver. In addition to the maternal tissues, two late-development fetuses, and their placentae, were extracted and tested separately for the presence of both bacteria. B. miyamotoi was found in the placentae and in both fetuses of this jumping mouse. Finding B. miyamotoi in two fetuses shows that it can also cross the transplacental barrier, passing from the infected mother to its offspring. It is unknown whether this infection would have led to a viable birth, as the effects of the systemic infection are unknown. Vertical transmission has been well documented with the relapsing fever Borrelia duttonii where it can lead to spontaneous abortion or infant death after birth in humans [33, 34]. The body-wide infection of the mother suggests that B. miyamotoi is successfully propagating and has not been sequestered by the host immune response. Whether this systemic infection reduces host fitness is necessary work in understanding B. miyamotoi and how it will spread through and affect wildlife. If B. miyamotoi vertical transmission is common, there will be enhanced spread of this pathogen in wildlife populations. Additionally, if transplacental transmission is a property of B. miyamotoi infections, there are veterinary and human medical implications.
Surveillance tools for B. burgdorferi and B. miyamotoi detection in wildlife
New Brunswick reports a low incidence of human Lyme disease compared to the bordering provinces and states. The low number of human cases of Lyme disease might suggest that B. burgdorferi is not present in the local wildlife, with adventitious ticks accounting for the observed tick and human infections. However, in our study, we detected multiple infected wildlife species. Further, all infected animals detected in this study were from regions far removed from the identified endemic sites, although within zones designated as I. scapularis and human Lyme disease risk areas . Both B. burgdorferi and B. miyamotoi are known to be present in ticks across New Brunswick [43, 66]. As well canine infections are prevalent through the province [10, 67]. Our findings of multiple members of reservoir-competent species infected with B. burgdorferi and B. miyamotoi demonstrate that Borrelia is established in New Brunswick wildlife populations and emphasizes the value of broader risk areas rather than discrete endemic areas in public health messaging.
As the risk of Lyme disease for human and companion animals arises from the presence of both ticks and infected wildlife reservoirs to infect the ticks, surveillance efforts need to monitor both components of Lyme disease risk. The primers described here allow for a rapid effective surveillance from necropsy tissue of both mammalian and avian wildlife species for B. burgdorferi or B. miyamotoi. These primers did not produce spurious amplification of the targeted sizes in any of the animals tested, and in the one instance of coinfection, they successfully amplified DNA from both bacterial species, with no cross-reaction. Further, this approach takes advantage of a public participatory citizen science model and so allows inexpensive and high-density sampling of a region. It directly detects bacterial DNA so, as opposed to serology, distinguishes between current and previous infections. However, this method is invasive and thus necessarily limited to necropsy tissue, prohibiting further study of individual animals. While effective for broad surveillance and initial investigations in an area, species-specific surveillance, using the candidate species identified in other works and this one, would be good targets for future surveillance to hone risk predictions.
This study identifies the meadow vole, deer mouse, eastern grey squirrel, jumping mouse, shrew, porcupine, and American crow as candidate species to be studied for their role in carrying Borrelia spp. in Atlantic Canada. As well, a new species of interest, the jumping mouse, has been identified. As a preliminary step in determining if these species are able to function as Borrelia reservoirs and how these Borrelia might affect host health, we tested multiple organs and tissues from infected individuals. All ten of the B. burgdorferi-infected animals tested had the bacteria present in a single tissue, either the liver, kidney, or brain. Five of the seven B. miyamotoi-infected animals also only had a single positive tissue, either the liver, kidney, or brain. This single-site of infection suggests late stage of infection where the bacteria persists in refuge tissues and that the bacteria are not rapidly cleared could indicate that the species could act as reservoir species. In addition, two of the seven B. miyamotoi-infected animals showed widely disseminated infections, including transplacental transmission of B. miyamotoi in one jumping mouse. This discovery raises the question of whether B. miyamotoi is capable of vertical transmission, a question with both ecological and veterinary and human health implications. Our findings clearly demonstrate that the B. burgdorferi and B. miyamotoi are broadly present in non-migratory wildlife in this region of Atlantic Canada suggesting that these Borrelia species and pose a risk to humans and companion animal health. The methods developed here are a resource for the detection of B. burgdorferi and B. miyamotoi in mammalian tissues.
S1 Fig. Sample locations and identified positives from New Brunswick for 2016 and 2017.
The Tantramar region of southeastern New Brunswick (main) and Southern New Brunswick (inset) with infections for both B. burgdorferi and B. miyamotoi indicated by blue and red symbols, respectively. Black symbols indicate negative samples.
S2 Fig. Representative gel products showing the positive amplicons.
Amplification of B. burgdorferi (top) and B. miyamotoi (bottom) sequences, indicated by the red arrows in wildlife hosts. Also visible are lower weight amplicons in other samples. None of these amplicons yielded usable sequences. Every amplicon at the correct size was sequence confirmed as B. burgdorferi or B. miyamotoi.
S1 Table. Top 10 nucleotide-BLAST results for each identified positive sequence.
All sequences are represented by their accession number. They were assessed against the NCBI Genbank nucleotide collection, specifying for highly similar sequences (megablast). Each positive tissue was sequenced resulting in some animals having multiple sequence entries.
S2 Table. Location data for all samples.
Roadkill, denoted with an “R” sample ID, have their UTM coordinates (zone 20N) reported. The donated samples, denoted with a “C” sample ID, have their community or county of origin listed to protect the privacy of donors. The infection status of each animal is indicated by a 1, meaning detection of that Borrelia species, or 0 indicating it was negative.
We thank veterinary clinics and members of the public throughout New Brunswick for assistance with specimen collection. We thank the late Gay Hansen for her assistance with specimen identification. We thank Cheryl Waterlot, and Jessica Thomas-Ebbett for additional assistance with specimen acquisition and J. Sperling, R. Kaufman and members of the Lloyd lab for discussion.
- 1. Telford SR, Goethert HK, Molloy P, Berardi V, Chowdri HR, Gugliotta JL, et al. Borrelia miyamotoi disease (BMD): Neither Lyme disease nor relapsing fever. Clin Lab Med. 2015 Dec;35(4):867–82. pmid:26593262
- 2. Hue F, Ghalyanchi Langeroudi A, Barbour AG. Chromosome sequence of Borrelia miyamotoi, an uncultivable tick-borne agent of human infection. Genome Announc. 2013 Sep 12;1(5):e00713–13. pmid:24029760
- 3. Ogden NH, Margos G, Aanensen DM, Drebot MA, Feil EJ, Hanincová K, et al. Investigation of genotypes of Borrelia burgdorferi in Ixodes scapularis ticks collected during surveillance in Canada. Applied and Environmental Microbiology. 2011;77(10):3244–54. pmid:21421790
- 4. Rudenko N, Golovchenko M, Grubhoffer L, Oliver JH. Borrelia carolinensis sp. nov., a novel species of the Borrelia burgdorferi sensu lato complex isolated from rodents and a tick from the south-eastern USA. International Journal of Systematic and Evolutionary Microbiology. 2011 Feb;61(2):381–3. pmid:20305062
- 5. Sperling JLH, Sperling FAH. Lyme borreliosis in Canada: Biological diversity and diagnostic complexity from an entomological perspective. Canadian Entomologist. 2009;141(6):521–49.
- 6. Wagemakers A, Staarink PJ, Sprong H, Hovius JWR. Borrelia miyamotoi: A widespread tick-borne relapsing fever spirochete. Trends in Parasitology. 2015;31(6):260–9. pmid:25892254
- 7. Rollend L, Fish D, Childs JE. Transovarial transmission of Borrelia spirochetes by Ixodes scapularis: A summary of the literature and recent observations. Ticks and Tick-borne Diseases. 2013;4(1–2):46–51. pmid:23238242
- 8. Scoles GA, Papero M, Beati L, Fish D. A relapsing fever group spirochete transmitted by Ixodes scapularis ticks. Vector-Borne and Zoonotic Diseases. 2001;1(1):21–34. pmid:12653133
- 9. Léger E, Vourc’h G, Vial L, Chevillon C, McCoy KD. Changing distributions of ticks: causes and consequences. Exp Appl Acarol. 2013 Feb;59(1–2):219–44. pmid:23015121
- 10. Lieske DJ, Lloyd VK. Combining public participatory surveillance and occupancy modelling to predict the distributional response of Ixodes scapularis to climate change. Ticks and Tick-borne Diseases. 2018;9(3):695–706. pmid:29478885
- 11. Talbot B, Slatculescu A, Thickstun CR, Koffi JK, Leighton PA, McKay R, et al. Landscape determinants of density of blacklegged ticks, vectors of Lyme disease, at the northern edge of their distribution in Canada. Sci Rep. 2019 Nov 13;9(1):16652. pmid:31723147
- 12. Lindgren E, Tälleklint L, Polfeldt T. Impact of climatic change on the northern latitude limit and population density of the disease-transmitting European tick Ixodes ricinus. Environ Health Perspect. 2000 Feb;108(2):119–23. pmid:10656851
Public Health. Office of the Chief Medical Officer of Health [Internet]. 2016 [cited 2018 Feb 10]. Available from: http://www2.gnb.ca/content/gnb/en/departments/ocmoh/cdc/content/vectorborne_andzoonotic/
- 14. LoGiudice K, Ostfeld RS, Schmidt KA, Keesing F. The ecology of infectious disease: Effects of host diversity and community composition on Lyme disease risk. PNAS. 2003 Jan 21;100(2):567–71. pmid:12525705
- 15. Ogden NH, Trudel L, Artsob H, Barker IK, Beauchamp G, Charron DF, et al. Ixodes scapularis ticks collected by passive surveillance in Canada: analysis of geographic distribution and infection with Lyme borreliosis agent Borrelia burgdorferi. J. Med Entol. 2006;43(3):600–9. pmid:16739422
New Brunswick Department of Natural resources. Department of Natural Resources: General Status of Wildlife in New Brunswick [Internet]. [cited 2018 Feb 10]. Available from: https://www1.gnb.ca/0078/WildlifeStatus/search-e.asp
- 17. Anderson JF, Johnson RC, Magnarelli LA, Hyde FW. Involvement of birds in the epidemiology of the Lyme disease agent Borrelia burgdorferi. Infection and Immunity. 1986;51(2):394–6. pmid:3943893
- 18. Hasle G. Transport of ixodid ticks and tick-borne pathogens by migratory birds. Frontiers in Cellular and Infection Microbiology. 2013;3(September):1–6. pmid:24058903
- 19. Dusek RJ, Hallgrimsson GT, Ip HS, Jónsson JE, Sreevatsan S, Nashold SW, et al. North Atlantic migratory bird flyways provide routes for intercontinental movement of avian influenza viruses. PLoS ONE. 2014;9(3). pmid:24647410
- 20. Madhav NK, Brownstein JS, Tsao JI, Fish D. A dispersal model for the range expansion of blacklegged tick (Acari: Ixodidae). Journal of Medical Entomology. 2009;41(5):842–52.
- 21. Scott JD, Fernando K, Banerjee SN, Durden LA, Byrne SK, Banerjee M, et al. Birds disperse Ixodid (Acari: Ixodidae) and Borrelia burgdorferi -infected ticks in Canada. Journal of Medical Entomology. 2001;38(4):493–500. pmid:11476328
- 22. Ogden NH, Lindsay LR, Hanincová K, Barker IK, Bigras-Poulin M, Charron DF, et al. Role of migratory birds in introduction and range expansion of Ixodes scapularis ticks and of Borrelia burgdorferi and Anaplasma phagocytophilum in Canada. Applied and Environmental Microbiology. 2008;74(6):1780–90. pmid:18245258
- 23. Salkeld DJ, Leonhard S, Girard YA, Hahn N, Mun J, Padgett KA, et al. Identifying the reservoir hosts of the Lyme disease spirochete Borrelia burgdorferi in California: the role of the western gray squirrel (Sciurus griseus). The American Journal of Tropical Medicine and Hygiene. 2008;79(4):535–40. pmid:18840740
- 24. Schwanz LE, Voordouw MJ, Brisson D, Ostfeld RS. Borrelia burgdorferi has minimal impact on the Lyme disease reservoir host Peromyscus leucopus. Vector-Borne and Zoonotic Diseases. 2011;11(2):117–24. pmid:20569016
- 25. Barbour AG. Infection resistance and tolerance in Peromyscus spp., natural reservoirs of microbes that are virulent for humans. Seminars in Cell and Developmental Biology. 2017;61:115–22. pmid:27381345
- 26. Anderson JF, Johnson RC, Magnarelli LA, Hyde FW. Identification of endemic foci of Lyme disease: isolation of Borrelia burgdorferi from feral rodents and ticks (Dermacentor variabilis). J. Clin Microbiol. 1985;22(1):36–8. pmid:3926816
- 27. Brisson D, Dykhuizen DE. ospC diversity in Borrelia burgdorferi: Different hosts are different niches. Genetics. 2004;168(2):713–22. pmid:15514047
- 28. Brisson D, Dykhuizen DE, Ostfeld RS. Conspicuous impacts of inconspicuous hosts on the Lyme disease epidemic. Proceedings of the Royal Society B: Biological Sciences. 2008;275(1631):227–35. pmid:18029304
- 29. Szekeres S, Coipan EC, Rigó K, Majoros G, Jahfari S, Sprong H, et al. Eco-epidemiology of Borrelia miyamotoi and Lyme borreliosis spirochetes in a popular hunting and recreational forest area in Hungary. Parasites and Vectors. 2015;8(1):1–8.
- 30. Burri C, Schumann O, Schumann C, Gern L. Are Apodemus spp. mice and Myodes glareolus reservoirs for Borrelia miyamotoi, Candidatus Neoehrlichia mikurensis, Rickettsia helvetica, R. monacensis and Anaplasma phagocytophilum? Ticks Tick-borne Diseases. 2014;5(3):245–51. pmid:24582511
- 31. Barbour AG, Bunikis J, Travinsky B, Hoen AG, Diuk-Wasser MA, Fish D, et al. Niche partitioning of Borrelia burgdorferi and Borrelia miyamotoi in the same tick vector and mammalian reservoir species. Am J Trop Med Hyg. 2009 Dec;81(6):1120–31. pmid:19996447
- 32. Ruyts SC, Frazer-Mendelewska E, Van Den Berge K, Verheyen K, Sprong H. Molecular detection of tick-borne pathogens Borrelia afzelii, Borrelia miyamotoi and Anaplasma phagocytophilum in Eurasian red squirrels (Sciurus vulgaris). European Journal of Wildlife Research. 2017;63(3):9–12.
- 33. Fuchs PC, Oyama AA. Neonatal relapsing fever due to transplacental transmission of Borrelia. JAMA. 1969 Apr 28;208(4):690–2. pmid:5818572
- 34. Larsson C, Andersson M, Guo BP, Nordstrand A, Hägerstrand I, Carlsson S, et al. Complications of pregnancy and transplacental transmission of relapsing‐fever borreliosis. The Journal of Infectious Diseases. 2006;194(10):1367–74. pmid:17054065
- 35. Ogden NH, Bigras-Poulin M, O’Callaghan CJ, Barker IK, Kurtenbach K, Lindsay LR, et al. Vector seasonality, host infection dynamics and fitness of pathogens transmitted by the tick Ixodes scapularis. Parasitology. 2007;134(2):209–27.
- 36. Vuong HB, Canham CD, Fonseca DM, Brisson D, Morin PJ, Smouse PE, et al. Occurrence and transmission efficiencies of Borrelia burgdorferi ospC types in avian and mammalian wildlife. Infection, Genetics and Evolution. 2014;27:594–600. pmid:24382473
- 37. Preac Mursic V, Patsouris E, Wilske B, Reinhardt S, Gross B, Mehraein P. Persistence of Borrelia burgdorferi and histopathological alterations in experimentally infected animals. A comparison with histopathological findings in human Lyme disease. Infection. 1990 Dec;18(6):332–41. pmid:2076905
- 38. Elsner RA, Hastey CJ, Olsen KJ, Baumgarth N. Suppression of long-lived humoral immunity following Borrelia burgdorferi infection. PLoS Pathogens. 2015;11(7):1–19. pmid:26136236
- 39. Shih C-M, Pollack RJ, Telford SR, Spielman A. Delayed dissemination of Lyme disease spirochetes from the site of deposition in the skin of mice. Journal of Infectious Diseases. 1992 Oct;166(4):827–31. pmid:1527418
- 40. Cadavid D, O’Neill T, Schaefer H, Pachner AR. Localization of Borrelia burgdorferi in the nervous system and other organs in a nonhuman primate model of Lyme disease. Laboratory Investigation. 2000;80(7):1043–54. pmid:10908149
- 41. Lee WY, Moriarty TJ, Wong CHY, Zhou H, Strieter RM, Van Rooijen N, et al. An intravascular immune response to Borrelia burgdorferi involves Kupffer cells and iNKT cells. Nature Immunology. 2010;11(4):295–302. pmid:20228796
- 42. Tsunoda T. Detachment of hard ticks (Acari: Ixodidae) from hunted sika deer (Cervus nippon). Experimental and Applied Acarology. 2014;63(4):545–50. pmid:24659518
- 43. Dibernardo A, Cote T, Ogden NH, Lindsay LR. The prevalence of Borrelia miyamotoi infection, and co-infections with other Borrelia spp. in Ixodes scapularis ticks collected in Canada. Parasites and Vectors. 2014;7(1):1–8.
- 44. Zinck CB, Priest JM, Shutler D, Boudreau M, Lloyd VK. Detection of Borrelia spp., Ehrlichia canis, Anaplasma phagocytophilum, and Dirofilaria immitis in Eastern Coyotes (Canis latrans) in Nova Scotia, Canada. J Wildl Dis. 2021 Jul 1;57(3):678–682. pmid:33956091.
- 45. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. Journal of Molecular Biology. 1990;215(3):403–10. pmid:2231712
- 46. Hall BG. Building phylogenetic trees from molecular data with MEGA. Mol Biol Evol. 2013;30(5):1229–35. pmid:23486614
- 47. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Molecular Biology and Evolution. 2018;35(6):1547–9. pmid:29722887
- 48. Tamura K, Nei M. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol. 1993;10(3):512–26. pmid:8336541
- 49. Felsenstein J. Confidence limits on phylogenies: an approach using the bootstrap. Evolution. 1985 Jul;39(4):783–91. pmid:28561359
- 50. Getis A, Ord JK. The Analysis of Spatial Association by Use of Distance Statistics. Geographical Analysis. 1992;24(3):189–206.
- 51. Sing T, Sander O, Beerenwinkel N, Lengauer T. ROCR: Visualizing classifier performance in R. Bioinformatics. 2005;21(20):3940–1. pmid:16096348
- 52. Fiset J, Tessier N, Millien V, Lapointe FJ. Phylogeographic structure of the white-footed mouse and the deer mouse, two Lyme disease reservoir hosts in Québec. PLoS ONE. 2015;10(12):1–18. pmid:26633555
- 53. Rand PW, Lacombe EH, Smith RP Jr, Rich SM, Kilpatrick WC, Dragoni CA, et al. Competence of Peromyscus maniculatus (Rodentia: Cricetidae) as a reservoir host for Borrelia burgdorferi (Spirochaetares: Spirochaetaceae) in the wild. Journal of Medical Entomology. 1993 May 1;30(3):614–8. pmid:8510121
- 54. Markowski D, Ginsberg HS, Hyland KE, Hu R. Reservoir competence of the Meadow Vole (Rodentia: Cricetidae) for the Lyme disease spirochete Borrelia burgdorferi. Journal of Medical Entomology. 1998 Sep 1;35(5):804–8. pmid:9775612
- 55. Fukunaga M, Takahashi Y, Tsuruta Y, Matsushita O, Ralph D, McClelland M, et al. Genetic and phenotypic analysis of Borrelia miyamotoi sp. nov., isolated from the Ixodid tick Ixodes persulcatus, the vector for Lyme disease in Japan. International Journal of Systematic Bacteriology. 2009;45(4):804–10.
- 56. Krause PJ, Fish D, Narasimhan S, Barbour AG. Borrelia miyamotoi infection in nature and in humans. Clinical Microbiology and Infection. 2015;21(7):631–9. pmid:25700888
- 57. Summers BA, Straubinger AF, Jacobson RH, Chang YF, Appel MJG, Straubinger RK. Histopathological studies of experimental Lyme disease in the dog. Journal of Comparative Pathology. 2005;133(1):1–13. pmid:15904927
- 58. Burgess EC, Windberg LA. Borrelia Sp. infection in coyotes, black-tailed jack rabbits and desert cottontails in southern Texas. Journal of Wildlife Diseases. 1989 Jan 1;25(1):47–51. pmid:2644452
- 59. Embers ME, Barthold SW, Borda JT, Bowers L, Doyle L, Hodzic E, et al. Persistence of Borrelia burgdorferi in rhesus macaques following antibiotic treatment of disseminated infection. PLoS ONE. 2012;7(1):1–12. pmid:22253822
- 60. Platonov AE, Karan LS, Kolyasnikova NM, Makhneva NA, Toporkova MG, Maleev VV, et al. Humans infected with relapsing fever spirochete Borrelia miyamotoi, Russia. Emerging Infectious Diseases. 2011;17(10):1816–23. pmid:22000350
- 61. Tavora F, Burke A, Li L, Franks TJ, Virmani R. Postmortem confirmation of Lyme carditis with polymerase chain reaction. Cardiovascular Pathology. 2008;17(2):103–7. pmid:18329555
- 62. Rudenko N, Golovchenko M, Vancova M, Clark K, Grubhoffer L, Oliver JH. Isolation of live Borrelia burgdorferi sensu lato spirochaetes from patients with undefined disorders and symptoms not typical for Lyme borreliosis. Clinical Microbiology and Infection. 2016;22(3):267.e9-267.e15.
- 63. Steere AC, Gross D, Meyer AL, Huber BT. Autoimmune mechanisms in antibiotic treatment-resistant Lyme arthritis. Journal of Autoimmunity. 2001;16(3):263–8. pmid:11334491
- 64. Brown RN, Lane RS. Natural and experimental Borrelia burgdorferi infections in woodrats and deer mice from California. Journal of Wildlife Diseases. 1994 Jul 1;30(3):389–98. pmid:7933283
- 65. Stone BL, Brissette CA. Host immune evasion by Lyme and relapsing fever borreliae: Findings to lead future studies for Borrelia miyamotoi. Frontiers in Immunology. 2017;8(JAN). pmid:28154563
- 66. Gasmi S, Ogden NH, Lindsay LR, Burns S, Fleming S, Badcock J, et al. Surveillance for Lyme disease in Canada: 2009–2015. Can Comm Dis Report. 2017;43(10):194–9. pmid:29770045
- 67. Bjurman NK, Bradet G, Lloyd VK. Lyme disease risk in dogs in New Brunswick. Canadian Veterinary Journal. 2016;57(9):981–4. pmid:27587892