Abstract
The genetic diversity of Borrelia burgdorferi sensu stricto, the agent of Lyme disease in North America, has consequences for the performance of serological diagnostic tests and disease severity. To investigate B. burgdorferi diversity in Canada, where Lyme disease is emerging, bacterial DNA in 309 infected adult Ixodes scapularis ticks collected in surveillance was characterized by multilocus sequence typing (MLST) and analysis of outer surface protein C gene (ospC) alleles. Six ticks carried Borrelia miyamotoi, and one tick carried the novel species Borrelia kurtenbachii. 142 ticks carried B. burgdorferi sequence types (STs) previously described from the United States. Fifty-eight ticks carried B. burgdorferi of 1 of 19 novel or undescribed STs, which were single-, double-, or triple-locus variants of STs first described in the United States. Clonal complexes with founder STs from the United States were identified. Seventeen ospC alleles were identified in 309 B. burgdorferi-infected ticks. Positive and negative associations in the occurrence of different alleles in the same tick supported a hypothesis of multiple-niche polymorphism for B. burgdorferi in North America. Geographic analysis of STs and ospC alleles were consistent with south-to-north dispersion of infected ticks from U.S. sources on migratory birds. These observations suggest that the genetic diversity of B. burgdorferi in eastern and central Canada corresponds to that in the United States, but there was evidence for founder events skewing the diversity in emerging tick populations. Further studies are needed to investigate the significance of these observations for the performance of diagnostic tests and clinical presentation of Lyme disease in Canada.
INTRODUCTION
Lyme disease risk is currently emerging in eastern and central Canada due to northern expansion of the range of the tick vector Ixodes scapularis (30, 31). Studies in one part of the zone of emergence in Canada suggest that founder populations of the agent of Lyme disease, Borrelia burgdorferi, comprise genotypes introduced from the northeastern United States, most likely by migratory birds (29, 31). It has been hypothesized that the agent of Lyme disease in North America, B. burgdorferi, is currently undergoing a period of adaptive radiation (23). The resulting genetic diversity of the bacterium could have consequences for disease severity and clinical symptoms observed in infected animals and humans (18, 33, 36, 37, 38), as well as for the performance of some diagnostic tests (17, 39).
In this study, we have investigated diversity by two methods: housekeeping genes located on the main linear chromosome (by multilocus sequence typing [MLST] [24, 25]), which are under purifying selection and evolve nearly neutrally, and the plasmid-borne outer surface protein C (ospC) gene, which is under balancing selection, and consequently, analyses of MLST and ospC alleles are not correlated on large geographic scales (24). The balancing selection of ospC is thought to be driven by host immune responses, because it is expressed early during B. burgdorferi infection of vertebrate hosts (32, 35). However, some researchers have raised the hypothesis that multiple-niche polymorphism and fitness variation of B. burgdorferi carrying different ospC alleles in diverse reservoir host communities explains the observed evidence for balancing selection (4). This may mean that geographic variations in host community assemblages drive geographic variation in the frequency of ospC alleles.
We have investigated two hypotheses. The first is that B. burgdorferi in ticks collected in Canada in passive surveillance are being dispersed from the United States into Canada by migratory birds and thus will carry the same MLST types that are found in the United States, with geographic patterns similar to those seen in the United States (15). Second, ospC may be a more direct indicator of clinically and diagnostically significant diversity of B. burgdorferi than MLST. If ticks carrying B. burgdorferi in passive surveillance in Canada are mostly dispersed from the United States into Canada by migratory birds, ospC diversity should also be the same in Canada as in the United States (29).
These investigations have practical importance in helping us to understand the pattern of emergence of tick and B. burgdorferi populations in Canada, whether patterns in Canada are different from those in the United States, and whether the ongoing range expansion of the tick vector may have implications for the bacterial populations with an effect on clinical symptoms and diagnosis in affected patients.
MATERIALS AND METHODS
Samples used in this study.
I. scapularis ticks were collected from companion animals and humans by veterinary clinics and medical clinics from Alberta to Newfoundland in Canada between 2005 and 2007 as part of the national passive surveillance system (27, 31). The ticks were tested at the National Microbiology Laboratory of the Public Health Agency of Canada for B. burgdorferi infection. DNA was purified using a Qiagen DNeasy 96 Tissue kit (Qiagen Inc., Mississauga, ON, Canada) optimized for recovery of low-copy-number DNA from ticks. Real-time PCR targeting the 23S rRNA locus and ospA was used to screen the ticks for Borrelia infection, as described previously (7, 27). Only data from ticks that had tested positive by this method were used in the genetic and statistical analyses described below.
ospC analysis.
The ospC alleles carried by B. burgdorferi in infected ticks were identified by reverse line blot (RLB) hybridization as previously described (29). Briefly, a 522-bp region of the ospC gene of B. burgdorferi was amplified by a seminested PCR using the external primers OC6 (+) and OC623 (−) and internal primers OC6 (+Fluo) and OC602 (−), which target conserved regions, as previously described (4). The amplicons were probed with ospC type-specific probes by RLB as previously described (4, 32).
In some samples, identified by RLB as carrying single-allele (i.e., unmixed) infections, ospC was amplified by nested PCR using primers and conditions described by Bunikis and coauthors (6). Amplicons were sequenced in both directions in order to identify the ospC allele carried by B. burgdorferi in that tick. The sequences were then aligned and compared with reference sequences downloaded from GenBank using the ClustalW algorithm implemented in MEGA version 3.1 (20). The reference sequences were as follows: type A, AF029860; type B, AF029861; type C, AF029862; type D, AF029863; type E, AF029864; type F, AF029865; type G, AF029867; type H, AF029868; type I, AF029869; type J, AF029870; type K, AF029871; type L, L42899; type M, U01892; type N, L42897; type O, X84778; type T, AF065143; and type U, AF065144. The criterion for inclusion within an ospC type was ≥99% similarity, and the criterion for exclusion from an ospC type was ≤90% similarity (32).
MLST and phylogenetic analyses.
Most tick samples shown by RLB to be infected with B. burgdorferi carrying single ospC alleles (suggesting that the ticks were less likely to carry B. burgdorferi of mixed STs) were used for MLST as described previously (24, 25). Briefly, eight housekeeping genes were amplified by nested PCR and sequenced by the Genomics Core Facility at the National Microbiology Laboratory of the Public Health Agency of Canada. Sequences of individual genes were compared with sequences present in the Borrelia MLST database at http://borrelia.mlst.net to obtain allele numbers and to determine the allelic profile and sequence types (STs) of samples. New alleles were compared with each other using the nonredundant database (NRDB) software available on the MLST website (http://www.mlst.net), and consecutive numbers were given to each distinct new allele. All sequence data for housekeeping genes are available at the MLST website hosted at Imperial College London, London, United Kingdom (http://borrelia.mlst.net). To ensure that Borrelia DNA in tick samples was capable of being amplified and to identify false-negative results, all tick samples subjected to MLST analysis were also tested by a Borrelia-specific PCR targeting the 16S-23S intergenic spacer (IGS) region (6).
Allelic profiles were analyzed using eBurst (9) and goeBurst (global optimal eBurst) (11). eBurst is based on a simple model of clonal expansion and divergence and provides a convenient method to establish relationships of descent for bacterial populations (http://www.mlst.net [9]). An implemented bootstrap procedure can give statistical confidence to the assignment of clonal complex founders. goeBurst, a further development of eBurst, has some additional features, such as a global optimization procedure (instead of local optimization) and an extended set of tie break rules, and allows better graphical representation of clonal complexes, including double-locus variants (DLV) and triple-locus variants (TLV) (11). Both algorithms are tailored for the use of MLST data and cluster STs as disjointed tree collections based on a set of hierarchical rules related to the number of single-locus variants (SLV), DLV (eBurst), and TLV (goeBurst). To obtain bootstrap support for the founder strains of clonal complexes for B. burgdorferi, we included all STs found in Canada and added 78 previously described STs from the northeastern (NE) and Midwestern (MW) United States (15) to an eBurst analysis. The minimum number of identical loci for group definition was set to 5, and the minimum count of SLV for subgroup definition was set to 0. The same samples were used in goeBurst to obtain a graphical display of clonal complexes.
For phylogenetic analyses, the sequences of all eight housekeeping genes were concatenated, and trees were generated using the maximum-likelihood method available on the PhyML website (http://www.atgc-montpellier.fr/phyml) (12). The settings were as follows: GTR was chosen as a substitution model, and the gamma parameter, invariant sites, and DNA models were estimated. For tree improvements, two methods, nearest-neighbor interchange (NNI) and subtree pruning and regrafting (SPR), were employed (16). Approximate likelihood ratios (aLRT) were calculated using the Shimodaira-Hasegawa (SH)-like procedure (2).
Statistical analysis.
We investigated the occurrence of spatial clustering in identified MLST STs and ospC alleles using cluster analysis in SaTScan version 8.0 and a Bernoulli model (19) with a temporal precision of 1 year. The maximal spatial cluster size was set at 50% of the population, and the latitudes and longitudes for each submitted tick were those obtained from Natural Resources Canada (Geographical Names of Canada [http://gnss.nrcan.gc.ca/gnss-srt/searchName.jsp?language=en]) for the town or village of origin identified on the submission. A number of outcomes were investigated by this method. First the region of origin (MW or NE United States) of STs that were identical to STs already identified in studies in the United States (15, 24) was investigated to see if STs already identified in the United States have a similar geographic distribution in Canada. Second, the clustering of STs newly identified in this study or present in the MLST database but not yet further described was investigated to see if there was a geographic pattern to STs that to date are unique to Canada. Third, the spatial clustering of each ospC allele was investigated to provide initial investigation of possible geographic variations in ospC allele frequencies.
Multiple pairwise comparisons among ospC alleles were carried out using generalized linear models with logi link function in STATA/SE for Windows (Statacorp LP, TX) to investigate whether there were significant differences in the likelihoods that different alleles occurred together in mixed infections. In each case, multivariable models were created for each allele with all the other alleles considered explanatory variables. These models were reduced to the most parsimonious model by stepwise elimination of explanatory variables. Because these multiple comparisons are prone to type I errors, correspondence analysis was conducted on the data set in SAS version 9.2 (Cary, NC) to see if associations observed in pairwise comparisons were supported when accounting for the full variance-covariance matrix of the data. The level of significance was a P value of <0.05 throughout.
RESULTS
MLST typing.
Of the tick samples suitable for MLST, 180 were randomly selected for analysis. Eighteen of them showed mixed infections after sequencing of one or more of the housekeeping genes, and PCR products were not obtained for one or more genes in a further 20 cases. Six of the latter samples, three from Nova Scotia and three from Ontario, produced 16S-23S IGS amplicons that yielded products of 500 to 600 bp rather than the 804 to 812 bp expected for B. burgdorferi (6). The sequenced products from two of these ticks were qualitatively sufficient to be compared to sequences available in GenBank and showed the highest similarity to Borrelia miyamotoi (GenBank accession number AY531879.1). Therefore, we suggest that these ticks carried relapsing-fever-like spirochetes, likely to be B. miyamotoi by PCR of the 16S-23S intergenic spacer region. Usable sequences of all eight (MLST) housekeeping genes of B. burgdorferi sensu lato were amplified from 136 samples, and 30 previously published samples from Quebec were included in the study (31). The 166 samples for which sequences were obtained originated from the Atlantic provinces (New Brunswick, Nova Scotia, Prince Edward Island, and Newfoundland; n = 61), Ontario (n = 67), Quebec (n = 31, including those analyzed in a previous study [31]), Manitoba (n = 6), and Alberta (n = 1) (Table 1).
Table 1.
Sequence types and allelic profiles of B. burgdorferi in ticks collected in surveillance in Canada from 2005 to 2007
| NML ID no.a | P/Tb | STc | Allele no. |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|
| clpA | clpX | nifS | pepX | pyrG | recG | rplB | uvr | |||
| AB07-3 | AB | 300 | 8 | 1 | 1 | 14 | 2 | 16 | 1 | 10 |
| MB07-20 | MB | 301 | 12 | 1 | 1 | 7 | 112 | 6 | 1 | 10 |
| MB07-21 | MB | 12 | 3 | 3 | 2 | 4 | 3 | 4 | 4 | 4 |
| MB07-22 | MB | 302 | 5 | 5 | 4 | 114 | 1 | 15 | 1 | 6 |
| MB07-25 | MB | 29 | 18 | 12 | 1 | 11 | 2 | 15 | 1 | 2 |
| UMB06-10 | MB | 12 | 3 | 3 | 2 | 4 | 3 | 4 | 4 | 4 |
| UMB07-5 | MB | 221 | 3 | 3 | 2 | 2 | 3 | 4 | 4 | 4 |
| NB06-101 | NB | 4 | 8 | 1 | 1 | 1 | 4 | 6 | 1 | 7 |
| NB06-127 | NB | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| NB06-129 | NB | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NB06-48 | NB | 7 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 8 |
| NB06-5 | NB | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NB06-85 | NB | 18 | 7 | 6 | 6 | 1 | 1 | 5 | 5 | 5 |
| NB07-119 | NB | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| NB07-121 | NB | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| NB07-137 | NB | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| NB07-17 | NB | 4 | 8 | 1 | 1 | 1 | 4 | 6 | 1 | 7 |
| NB07-50 | NB | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| NB07-87 | NB | 305 | 5 | 5 | 4 | 5 | 5 | 5 | 94 | 6 |
| NB07-95 | NB | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NF06-4 | NFL | 19 | 4 | 4 | 3 | 3 | 3 | 3 | 3 | 3 |
| NS05-104 | NS | 306 | 5 | 5 | 4 | 5 | 5 | 6 | 1 | 6 |
| NS06-10 | NS | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NS06-100 | NS | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| NS06-113c | NS | 18 | 7 | 6 | 6 | 1 | 1 | 5 | 5 | 5 |
| NS06-129 | NS | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NS06-151 | NS | 7 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 8 |
| NS06-196a | NS | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| NS06-2 | NS | 16 | 2 | 2 | 1 | 2 | 2 | 2 | 2 | 2 |
| NS06-233 | NS | 7 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 8 |
| NS06-234 | NS | 307 | 9 | 1 | 1 | 7 | 1 | 1 | 1 | 10 |
| NS06-235 | NS | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NS06-240c | NS | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| NS07-121 | NS | 281 | 127 | 88 | 87 | 107 | 105 | 108 | 88 | 98 |
| NS07-125 | NS | 59 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 19 |
| NS07-128 | NS | 38 | 21 | 1 | 15 | 8 | 1 | 18 | 4 | 7 |
| NS07-166 | NS | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| NS07-279 | NS | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| NS07-28 | NS | 308 | 4 | 91 | 1 | 1 | 1 | 6 | 1 | 7 |
| NS07-290 | NS | 7 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 8 |
| NS07-2a | NS | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| NS07-68 | NS | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| NS07-75 | NS | 37 | 7 | 6 | 12 | 1 | 1 | 5 | 5 | 5 |
| NS07-20 | NS | 309 | 7 | 6 | 12 | 115 | 1 | 114 | 5 | 5 |
| NS07-96 | NS | 18 | 7 | 6 | 6 | 1 | 1 | 5 | 5 | 5 |
| ON06-102 | ON | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| ON06-107 | ON | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| ON06-108 | ON | 38 | 21 | 1 | 15 | 8 | 1 | 18 | 4 | 7 |
| ON06-187c | ON | 19 | 4 | 4 | 3 | 3 | 3 | 3 | 3 | 3 |
| ON06-188 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON06-190 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON06-194 | ON | 226 | 8 | 1 | 15 | 94 | 2 | 20 | 1 | 7 |
| ON06-231 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON06-255a | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON06-255b | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON06-260a | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON06-262 | ON | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| ON06-277 | ON | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| ON06-292c | ON | 311 | 8 | 1 | 1 | 7 | 1 | 1 | 1 | 10 |
| ON06-293 | ON | 34 | 8 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| ON06-299 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON06-30 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON06-336b | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON06-371b | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON06-422 | ON | 29 | 18 | 12 | 1 | 11 | 2 | 15 | 1 | 2 |
| ON06-66 | ON | 18 | 7 | 6 | 6 | 1 | 1 | 5 | 5 | 5 |
| ON06-67a | ON | 57 | 25 | 4 | 3 | 16 | 3 | 21 | 1 | 2 |
| ON06-90a | ON | 50 | 8 | 1 | 1 | 17 | 1 | 6 | 1 | 10 |
| ON07-1001 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-1073 | ON | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| ON07-1115 | ON | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| ON07-1151 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-1194 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-1218 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-1231 | ON | 37 | 7 | 6 | 12 | 1 | 1 | 5 | 5 | 5 |
| ON07-124b | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-128 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-1289 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-139a | ON | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| ON07-14 | ON | 222 | 5 | 5 | 4 | 5 | 1 | 15 | 1 | 6 |
| ON07-147 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-15 | ON | 222 | 5 | 5 | 4 | 5 | 1 | 15 | 1 | 6 |
| ON07-150 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-157b | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-158a | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-219 | ON | 313 | 10 | 5 | 4 | 116 | 1 | 15 | 1 | 6 |
| ON07-223 | ON | 55 | 23 | 1 | 17 | 20 | 2 | 1 | 1 | 10 |
| ON07-285 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-289 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-292a | ON | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| ON07-336 | ON | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| ON07-351 | ON | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| ON07-361 | ON | 4 | 8 | 1 | 1 | 1 | 4 | 6 | 1 | 7 |
| ON07-399 | ON | 29 | 18 | 12 | 1 | 11 | 2 | 15 | 1 | 2 |
| ON07-424 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-430 | ON | 307 | 9 | 1 | 1 | 7 | 1 | 1 | 1 | 10 |
| ON07-441 | ON | 314 | 14 | 1 | 11 | 1 | 113 | 1 | 1 | 10 |
| ON07-471 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-472 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-507 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-520 | ON | 19 | 4 | 4 | 3 | 3 | 3 | 3 | 3 | 3 |
| ON07-576 | ON | 222 | 5 | 5 | 4 | 5 | 1 | 15 | 1 | 6 |
| ON07-590 | ON | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| ON07-600 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-625a | ON | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| ON07-727 | ON | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| ON07-766b | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-791b | ON | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| ON07-795 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-860 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| ON07-861 | ON | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ON07-873 | ON | 315 | 8 | 1 | 1 | 17 | 1 | 6 | 95 | 10 |
| ON07-88 | ON | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| PEI05-15 | PEI | 316 | 10 | 1 | 4 | 6 | 1 | 6 | 1 | 6 |
| PEI05-20 | PEI | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| PEI05-30 | PEI | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| PEI06-20 | PEI | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| PEI06-23 | PEI | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| PEI06-27 | PEI | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| PEI06-34 | PEI | 18 | 7 | 6 | 6 | 1 | 1 | 5 | 5 | 5 |
| PEI06-37 | PEI | 221 | 3 | 3 | 2 | 2 | 3 | 4 | 4 | 4 |
| PEI06-48 | PEI | 19 | 4 | 4 | 3 | 3 | 3 | 3 | 3 | 3 |
| PEI06-49 | PEI | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| PEI06-55 | PEI | 59 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 19 |
| PEI06-56 | PEI | 19 | 4 | 4 | 3 | 3 | 3 | 3 | 3 | 3 |
| PEI06-67 | PEI | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| PEI06-84 | PEI | 317 | 128 | 3 | 2 | 4 | 3 | 4 | 4 | 4 |
| PEI06-85 | PEI | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| PEI07-10 | PEI | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| PEI07-24 | PEI | 7 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 8 |
| PEI07-33b | PEI | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| PEI07-48 | PEI | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| PEI07-522 | PEI | 318 | 21 | 1 | 15 | 8 | 1 | 1 | 4 | 7 |
| PEI07-74 | PEI | 19 | 4 | 4 | 3 | 3 | 3 | 3 | 3 | 3 |
| PEI07-77 | PEI | 9 | 10 | 5 | 4 | 6 | 1 | 6 | 1 | 6 |
| QC05-156 | QC | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| QC07-344 | QC | 38 | 21 | 1 | 15 | 8 | 1 | 18 | 4 | 7 |
| QC07-362 | QC | 34 | 8 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| QC07-399 | QC | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| QC07-402 | QC | 238 | 10 | 5 | 75 | 6 | 1 | 15 | 1 | 6 |
| QC07-484 | QC | 29 | 18 | 12 | 1 | 11 | 2 | 15 | 1 | 2 |
| QC07-493 | QC | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| QC07-565b | QC | 4 | 8 | 1 | 1 | 1 | 4 | 6 | 1 | 7 |
| QC07-595 | QC | 36 | 10 | 5 | 4 | 6 | 1 | 15 | 1 | 6 |
| QC07-603 | QC | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| QC07-650 | QC | 16 | 2 | 2 | 1 | 2 | 2 | 2 | 2 | 2 |
| QC07-723 | QC | 11 | 5 | 7 | 5 | 1 | 6 | 1 | 4 | 9 |
| QC07-755 | QC | 7 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 8 |
| QC07-765 | QC | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| QC07-776 | QC | 12 | 3 | 3 | 2 | 4 | 3 | 4 | 4 | 4 |
| QC07-785 | QC | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| QC07-815 | QC | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| QC07-819 | QC | 8 | 5 | 5 | 4 | 5 | 5 | 5 | 1 | 6 |
| QC07-83 | QC | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| QC07-84 | QC | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| QC07-851 | QC | 59 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 19 |
| QC07-908 | QC | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| QC07-951 | QC | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| QC07-1008 | QC | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| QC07-1048 | QC | 59 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 19 |
| QC07-1054 | QC | 3 | 4 | 1 | 1 | 1 | 1 | 6 | 1 | 7 |
| QC07-161-5 | QC | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| QC07-175-3 | QC | 59 | 6 | 1 | 5 | 1 | 1 | 7 | 1 | 19 |
| QC07-181-1 | QC | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| QC07-181-4 | QC | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| QC07-182-4 | QC | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
| QC07-182-6 | QC | 14 | 9 | 1 | 1 | 7 | 1 | 6 | 1 | 10 |
All ticks were feeding adults except for those in italics, which were questing adults. NML, National Microbiology Laboratory.
P/T, province or territory. AB, Alberta; MB, Manitoba; NB, New Brunswick; NS, Nova Scotia; NFL, Newfoundland and Labrador; PEI, Prince Edward Island; ON, Ontario; QC, Quebec.
STs in boldface and underlined are new to this study, although data from all but one of the ticks collected in Quebec have been presented previously (31). The STs in boldface were present in the MLST database but had not been further analyzed.
A total of 146 ticks carried B. burgdorferi STs that had previously been described from the NE and MW United States (Table 1) (15). Two STs dominated the sample set (Table 2): ST1 (n = 35) and ST3 (n = 26). The next highest frequencies were found for ST14 (n = 11) and ST9 (n = 11), and all of these STs had been found in the NE United States previously.
Table 2.
Frequencies of B. burgdorferi STs in the sample of ticks collected in surveillance in Canada
| ST | Frequency | Geographic regiona | New ST | Geographic region of nearest ancestor for new STs |
|---|---|---|---|---|
| 1 | 35 | NE | No | |
| 3 | 26 | NE | No | |
| 4 | 4 | NE | No | |
| 7 | 6 | NE | No | |
| 8 | 7 | NE | No | |
| 9 | 11 | NE | No | |
| 11 | 1 | NE | No | |
| 12 | 3 | NE + MW | No | |
| 14 | 11 | NE | No | |
| 16 | 2 | NE | No | |
| 18 | 5 | NE | No | |
| 19 | 6 | NE | No | |
| 29 | 4 | NE + MW | No | |
| 34 | 2 | NE | No | |
| 36 | 7 | NE | No | |
| 37 | 2 | NE | No | |
| 38 | 3 | NE | No | |
| 50 | 1 | NE | No | |
| 55 | 1 | MW | No | |
| 57 | 1 | NE | No | |
| 59 | 5 | NE | No | |
| 221 | 2 | NE + MW | NE + MW | |
| 222 | 3 | NE + MW | NE | |
| 226 | 1 | NE | MW | |
| 238 | 1 | NE | Yes | NE |
| 300 | 1 | MW | Yes | MW |
| 301 | 1 | MW | Yes | MW |
| 302 | 1 | MW | Yes | NE |
| 305 | 1 | NE | Yes | NE |
| 306 | 1 | NE | Yes | NE |
| 307 | 2 | NE | Yes | NE |
| 308 | 1 | NE | Yes | NE |
| 309 | 1 | NE | Yes | NE |
| 311 | 1 | NE | Yes | NE |
| 313 | 1 | NE | Yes | NE |
| 314 | 1 | NE | Yes | MW |
| 315 | 1 | NE | Yes | NE |
| 316 | 1 | NE | Yes | NE |
| 317 | 1 | NE | Yes | NE + MW |
| 318 | 1 | NE | Yes | NE |
NE, STs already identified in NE United States; MW, STs already identified in MW United States; New ST, STs identified for the first time in this study. Also shown for STs new to this study is the geographic region of occurrence of the nearest ancestor.
Samples carrying STs that had been described only from NE regions of the United States were not found west of 80°W. Samples carrying STs that had been described only from MW regions of the United States were not found east of this longitude, except in the case of two ticks that came from Quebec. Ticks carrying MW STs formed a significant cluster (relative risk [RR] = 56.67; P = 0.001) centered on 49.77°N and 97.32°W, with a radius of 1,511 km and encompassing ticks collected in Alberta, Manitoba, and western Ontario to Simcoe (Fig. 1).
Fig. 1.
Geographic distribution of sites of collection of ticks carrying the different MLST STs of B. burgdorferi identified in the study. The triangles indicate the locations of ticks carrying those STs already identified in the NE United States (blue triangles) and those STs already identified in the MW United States (green triangles) and ticks carrying STs identified for the first time in this study (red triangles). The black ellipse indicates the geographic locations of ticks that formed a spatial cluster that carried new STs, and the red ellipse indicates the geographic locations of ticks that formed a spatial cluster that carried STs already identified in the MW United States.
In total, 19 new STs were identified, 9 of which showed new allelic profiles, while the other 10 contained one or more new alleles (Table 1). Three STs were present in the MLST database but have not yet been further described (i.e., ST221, ST222, and ST226), so we consider them new to the present study. The ratio of new STs to previously described STs was 3/6 (50%) in Manitoba, 8/67 (12%) in Ontario, 1/31 (3.2%) in Quebec, and 8/61 (13%) in the Atlantic region. ST222 was found in Ontario east and west of 80°W and ST221 in Manitoba and on Prince Edward Island (Table 1), and ST301 has been found in questing ticks sampled in Wisconsin (unpublished data). A significant cluster of new STs (RR = 5.15; P = 0.005) was identified centered on 53.54°N and 113.49°W, which encompassed ticks collected in Alberta, Manitoba, and Ontario, except for ticks from southeastern Ontario along the St. Lawrence valley (Fig. 1). Within this cluster, 12/27 (44.4%) ticks carried new STs, while outside the cluster, 12/139 (8.6%) ticks carried a new ST. In logistic regression analysis, occurrence within the cluster explained the occurrence of new STs better than the simpler explanatory variable of longitude (χ2 = 5.2; degree of freedom [df] = 1; P < 0.01).
Several clonal complexes (CC) were identified in goeBurst (Fig. 2); some had been described previously by Hoen and colleagues (15) (i.e., CC34, CC37, and CC19), but there were also new ones (CC36 and CC12) through connection of new STs with potential founder strains originating in the NE United States (Fig. 2). However, the bootstrap values for founder assignments were 64% or less (data not shown), which likely reflects the low number of SLV descending from each founder. This is consistent with the goeBurst data, in which higher levels of tie break rule (indicated by different colors in Fig. 2) refer to lower levels of confidence in the link.
Fig. 2.
Population snapshot of 244 samples of B. burgdorferi found in Canada (166 samples) and the Unites States (78 samples) as determined by goeBurst analysis. Each colored dot corresponds to an ST: the dark-green dots are STs previously found in the upper MW United States, the dark-blue dots are STs previously found in the NE United States, the light-green STs are found in Canada west of 79°W, and the light-blue STs are found in Canada east of 79°W, while the orange dots indicate STs found in both the eastern and middle western regions of Canada or the United States. The size of the dot corresponds approximately to the sample size. The lines link related STs (clonal complexes). Potential founders of clonal complexes are circled in black. Colored lines connecting STs indicate descending order of certainty: black lines are inferred without tie break rules, blue lines are inferred using tie break rule 1 (number of SLV), and green lines are inferred using tie break rule 2 (number of DLV). The light-gray lines correspond to DLV. TLV are indicated by dashed lines. The majority of the new STs found in the Canadian sample set are linked to previously defined STs.
Nevertheless, most of the new STs identified in the data set analyzed here were single-locus (ST300 to ST302, ST305 to ST308, ST311, ST313, and ST315 to ST318), double-locus (ST309, ST226, and ST222), or triple-locus (ST314) variants of previously described STs (Fig. 3). Most strains linked in the same CC also clustered in the same clade in the phylogenetic tree. The exceptions were ST52, ST53, and ST301, which were assigned to CC34 by goeBurst but fell into a different clade in the phylogenetic tree (Fig. 3). ST281 was genetically divergent from B. burgdorferi, showing new alleles for all eight housekeeping genes (Table 1), and phylogenetic analysis revealed that this ST is more closely related to Borrelia bissettii than to B. burgdorferi and belongs, in fact, to a new species within the B. burgdorferi species complex (Fig. 3) (26).
Fig. 3.
ST tree of Borrelia samples collected in the United States and Canada. The tree was constructed using concatenated sequences of eight housekeeping genes for maximum-likelihood analysis (PhyML). The outgroup branch is not according to scale, as indicated by slashes. The scale bar indicates 1% divergence.
ospC analysis.
Of the 400 tick samples available to the study, ospC alleles were obtained from 309. All 17 known alleles were represented in the sample, but alleles A, K, and N, which have all been associated with disseminated Lyme disease (33), were dominant (Fig. 4). A spatial cluster of ticks carrying allele I was observed in Nova Scotia centered on 45.09°N and 64.37°W, with a 67.51-km radius (Fig. 5). A spatial cluster of ticks carrying allele L was observed that extended from Manitoba into western Ontario, centered on 48.72°N and 94.57°W, with a 1,282-km radius (Fig. 5). There was no significant spatial clustering of ticks carrying other alleles.
Fig. 4.
Frequency distribution of ospC alleles identified in ticks in this study.
Fig. 5.
Geographic distribution of sites of collection of ticks carrying different ospC alleles identified in the study. A significant spatial cluster of ticks infected with B. burgdorferi carrying allele L is indicated by the black ellipse, and a significant spatial cluster of ticks infected with B. burgdorferi carrying allele I is indicated by the red circle.
More than one ospC allele was identified in 108 ticks (60 with 2 alleles, 36 with 3 alleles, 10 with 4 alleles, 1 with 5 alleles, 1 with 6 allele, and 2 with 7 alleles). In pairwise analyses, ticks infected with B. burgdorferi carrying allele F were significantly more likely to be coinfected with B. burgdorferi carrying allele M (odds ratio [OR] = 2.99; 95% confidence interval [95% CI], 0.1 to 0.77; P = 0.047), and coinfections in the same tick by B. burgdorferi carrying alleles G and H, J and T, and L and O were more likely to occur than by chance (OR = 3.29, 10.60, and 9.09; 95% CI, 1.3 to 8.1, 1.0 to 107, and 1.7 to 498; P = 0.010, 0.046, and 0.011, respectively). Coinfections by B. burgdorferi carrying alleles A and F were significantly less likely to occur in the same tick (OR = 0.26; 95% CI, 0.1 to 0.8; P = 0.015). No significant positive or negative three-way (or more complex) relationships were found. These findings are summarized in Table 3. Correspondence analysis supported the significance of these pairwise comparisons when the full variance-covariance matrix of the data was accounted for: all positively associated pairs occurred in the same parameter space, while the negatively associated pair (A and F) occupied very separated parameter spaces (Fig. 6).
Table 3.
Numbers of ticks infected with B. burgdorferi that carried pairs of ospC alleles for each possible pairing
| Allele | Total no. of infected ticks carrying allele | No. of ticks with pairinga | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B | C | D | E | F | G | H | I | J | K | L | M | N | O | T | U | ||
| A | 96 | 8 | 2 | 4 | 5 | 4− | 11 | 5 | 3 | 0 | 21 | 3 | 5 | 16 | 2 | 4 | 4 |
| B | 33 | 1 | 0 | 1 | 4 | 2 | 2 | 2 | 1 | 11 | 0 | 0 | 3 | 0 | 2 | 0 | |
| C | 9 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | ||
| D | 14 | 1 | 0 | 2 | 2 | 0 | 1 | 2 | 1 | 0 | 2 | 1 | 0 | 1 | |||
| E | 22 | 2 | 3 | 1 | 2 | 1 | 4 | 1 | 2 | 2 | 2 | 1 | 1 | ||||
| F | 34 | 3 | 2 | 3 | 0 | 8 | 1 | 5+ | 2 | 1 | 0 | 0 | |||||
| G | 40 | 8+ | 4 | 0 | 12 | 1 | 4 | 4 | 2 | 2 | 3 | ||||||
| H | 27 | 3 | 0 | 4 | 0 | 2 | 2 | 0 | 0 | 0 | |||||||
| I | 14 | 0 | 4 | 0 | 1 | 0 | 0 | 1 | 1 | ||||||||
| J | 5 | 0 | 0 | 0 | 1 | 0 | 1+ | 0 | |||||||||
| K | 81 | 0 | 8 | 14 | 3 | 2 | 0 | ||||||||||
| L | 10 | 0 | 1 | 2+ | 0 | 0 | |||||||||||
| M | 20 | 5 | 1 | 0 | 0 | ||||||||||||
| N | 56 | 1 | 2 | 1 | |||||||||||||
| O | 10 | 0 | 0 | ||||||||||||||
| T | 8 | 0 | |||||||||||||||
| U | 13 | ||||||||||||||||
Significant associations are in boldface; positive associations are indicated by a superscript + and negative associations by a superscript minus sign.
Fig. 6.
Graphical representation of the results of correspondence analysis of ospC allele pairwise comparisons showing that positively associated pairs (indicated by ellipses) occur in the same parameter space while a negatively associated pair (indicated by the double-headed arrow) occurs in different parameter spaces.
DISCUSSION
In this study, we identified a wide diversity of B. burgdorferi STs, as well as the presence of other tick-borne Borrelia spp. These findings have significance for the diagnosis of Lyme disease, and the possible occurrence of disease caused by other tick-borne bacteria, in eastern and central Canada.
First, this study determined that, as well as carrying B. burgdorferi, I. scapularis ticks collected in Canada also carry relapsing-fever-like spirochetes, most likely B. miyamotoi. The implications for public health are currently uncertain, because it is not known whether this species is pathogenic to humans. The prevalence of ticks infected with these spirochetes in nature is likely to be higher than revealed in our study: the screening PCRs would not have been positive if ticks had been infected only with B. miyamotoi. We identified only ticks coinfected with B. miyamotoi and B. burgdorferi, not ticks that were infected with B. miyamotoi alone. Studies on questing I. scapularis in the United States have shown an approximate 1:10 ratio of B. miyamotoi to B. burgdorferi (3), while in our study the ratio was 1:24.
Second, one tick carried an entirely new Borrelia sp. (ST281), which was most closely related to B. bissetii, and we have proposed that it be named Borrelia kurtenbachii (26). Strains belonging to this bacterial species found in the United States, e.g., 25015, have been shown to be mildly pathogenic in mice (1, 10). However, determining that ticks collected in Canada can carry this Borrelia sp. and that ticks in Canada can carry B. miyamotoi raises the possibility (which needs further study) that infections in humans for which we currently have no diagnostic methods could be occurring in Canada. Therefore, the possible presence of these species needs to be taken into consideration in diagnosis of human and animal patients.
Third, there was a high degree of diversity of B. burgdorferi, with 40 STs identified, 16 of which were entirely new while 3 were present in the MLST database but have not been further analyzed. Variations in clinical symptoms of affected humans, as well as differences in performance of serodiagnostic tests, are seen among European species of the B. burgdorferi sensu lato complex (34). These species frequently have different natural reservoir hosts, mostly due to their capacity to resist the alternative pathways of complement of different host species (21, 22, 23). It is unknown what confers the ability among Lyme disease spirochetes to cause disease in humans and to what extent ecological differences directly drive clinical differences. Although precise determinants of pathogenicity are unknown, differences in pathogenicity among different B. burgdorferi genotypes from the United States have been demonstrated (18, 33, 36, 37, 38), as well as the fact that the performance of some serodiagnostic tests varies among genetic variants (17, 39). While B. burgdorferi remains a generalist in North America (13), some studies suggest evidence of early signs of multiple-niche polymorphism (4), and we have identified a number of possible drivers for adaptive radiation of B. burgdorferi in North America (23). These include climate effects on tick seasonality (28), for which there may be evidence from the geographic variation in the occurrence of ribosomal sequence types (RSTs) in the United States (15).
Studies on differential pathogenicity (i.e., the ability or lack of ability to cause disseminated Lyme disease) of B. burgdorferi genotypes or different abilities of infected individuals to produce antibodies that are detectable in standard serological assays, have mainly focused on differences among RSTs (14, 38). RSTs can be directly assayed from IGS types or inferred from ospC alleles because of linkage disequilibrium among these loci, which has been observed (at least in the northeastern United States [6, 14]). The wide range of ospC alleles found in our study suggests that variations in pathogenicity and sensitivity to serological tests can be expected in Canadian patients, since it is also observed in the United States (8, 37, 38, 39). To date, MLST typing has not been used to identify genotypes of B. burgdorferi that differ in the ability to cause disseminated disease in humans or to elicit immune responses detectable by commercial test kits. However, the utility of MLST analysis in this regard may be worth investigating, as studies in Europe suggest that MLST is able to detect phenotypic differences among B. burgdorferi sensu lato spirochetes in terms of identifying ecotypes that vary in the ability to infect different reservoir hosts (25).
For those STs common to the United States and Canada, the geographic (at least longitudinal) patterns of their occurrence were similar in the two countries. While a number of new STs were identified, the goeBurst analysis clustered many of the new STs as SLV to the previously described CC34, −37, −19, and −7 with founders in the United States (15), suggesting a close relationship of the populations. Two new complexes, CC36 and CC12, were formed due to STs being SLV or DLV of North American ST36, -8, -9, and -12. For individual complexes, the founder assignment and direction of descent could not be established with certainty due to a paucity of SLV associated with complex founders. This does not invalidate the CCs, it simply means that the direction of descent, i.e., founder/subfounder assignment, may change when more SLV are added. Further sampling and additions to the B. burgdorferi MLST database are likely to lead to denser “forests” and better resolution of clonal complexes, which would improve the inference of founders and descendants. Such enhanced data sets will likely aid resolution of the phylogeography of B. burgdorferi in North America, which appears complex. This complexity may be due to repeated expansions and contractions of the geographic range of B. burgdorferi (perhaps associated with glacial-interglacial cycles), which result in observations on long- and short-term evolution of B. burgdorferi in North America that at first sight appear to conflict (5, 15). Nevertheless, in our study, the likely ancestor identified in all the clonal complexes was an ST found in the United States. For the most part, the complexes followed the geographic pattern expected from (15), i.e., NE STs being ancestors of MW STs and STs new to this study having ancestors that originated from the same longitude in the United States (Fig. 3). Therefore, the goeBurst analysis was consistent with the general pattern of south-north transportation of B. burgdorferi in ticks suggested by cluster analysis. Earlier studies of B. burgdorferi diversity, which analyzed outer surface protein sequence variations, identified latitudinal conformity of B. burgdorferi in the eastern United States (32). This was interpreted as balancing selection predominating over migration as a driver of diversity. Our study, using MLST of housekeeping genes with nearly neutral variation and employing ticks collected across a wide longitudinal range, suggests that south-north migration (in the case of Canada) may indeed be an explanation for latitudinal conformity.
In a previous study by Hoen et al. (15) that had similar sample sizes from the NE (n = 41) and MW (n = 37) United States, 37 unique STs were identified, 20 of which were found in the NE and 17 in the MW. The finding of more new STs in western parts of Canada in the present data set may be related to differences in sampling efforts in the United States and Canada. In the United States, field-collected ticks (mostly nymphs) were analyzed, while in Canada, the samples were mostly adult ticks obtained by passive surveillance. Many of these ticks, and the B. burgdorferi isolates they carry, were likely carried into Canada from the United States by migratory birds (29). However, some ticks are likely to have come from endemic populations in Canada. Therefore, an alternative explanation would be that some Borrelia populations in Canada, for example, at Long Point, Lake Erie (30), are refugial populations that have been isolated for long enough to accumulate mutations and are now themselves spreading their geographic ranges. The fact that a circular cluster, rather than simply accounting for a longitudinal cline, explained the spatial distribution of new STs in Canada may support this hypothesis.
Geographic variation in the occurrence of ospC alleles was in part similar to that seen in the United States. For example ticks carrying allele L occurred mostly in the western range of I. scapularis collected in passive surveillance, while allele I occurred only in eastern Canada, similar to findings in the United States (5). Nevertheless, spatial clustering of ticks carrying ospC alleles I and L (which was particularly spatially constrained for allele I) could indicate spatial variation in ospC allele frequencies associated with founder events that are likely to be occurring at present in Canada (31).
The spatial pattern of some ospC alleles and the occurrence of positive and negative associations between some alleles in the same tick may support suggestions of multiple-niche polymorphism. Coinfections of adult ticks with two alleles occurred more frequently than by chance in some cases. This could suggest that immature ticks fed on hosts that were coinfected with B. burgdorferi carrying ospC alleles that were associated with specificity for similar host species. We speculate that variations among species in mounting an effective immune response to different ospC alleles could be one mechanism for host specialization (17). Negative associations among ospC allele coinfections could also indicate differing host specificities of ospC alleles by indicating particularly cross-protecting immune responses among some ospC alleles.
Our study shows that there is a wide range of Borrelia species in ticks collected in Canada, and until proved otherwise, they must be considered a potential risk to public health. The study was based on ticks collected in passive surveillance, and the proportion of ticks that originated from locations in Canada where ticks are endemic versus those that may have been dispersed from the United States by migratory birds is mostly unknown. Assuming that the ticks came from both sources, our study shows that B. burgdorferi isolates in ticks collected in Canada have a wide genetic diversity. The diversity observed mostly reflected that found in the United States (in terms of degree and geographic location), where multiple long- and short-term evolutionary processes are in play. Our study does, however, suggest the possibility that in Canada long-term evolutionary processes (e.g., the possible occurrence of refugial populations following glacial contractions of the B. burgdorferi geographic range) and short-term evolutionary processes (e.g., founder events) shape a population structure of B. burgdorferi that is distinct from that in the United States. This means that the genetic diversity of B. burgdorferi in eastern and central Canada could be clinically or diagnostically significant. More prospective field studies in existing and emerging areas where Lyme disease is endemic, combined with studies on the diversity of B. burgdorferi in Canadian Lyme disease patients, are needed to elucidate the processes of evolution of B. burgdorferi in Canada and their significance for public health.
ACKNOWLEDGMENTS
We acknowledge the important contribution of the late Klaus Kurtenbach to this study, as well as colleagues in provincial and territorial public health organizations involved in the collection of ticks for the study. We thank Benjamin Delisle for assistance in testing ticks for ospC and Brynn Kaplen and Shaun Tylor for assistance with sequencing of housekeeping genes. We also acknowledge the expert technical assistance of Brynn Kaplen and Erika Landry in the sequencing of bacterial genomic DNA.
This work was funded by the Public Health Agency of Canada, while the contributions of I.S. and K.H. were supported by NIH grant AR041511.
Footnotes
Published ahead of print on 18 March 2011.
REFERENCES
- 1. Anderson J. F., Barthold S. W., Magnarelli L. A. 1990. Infectious but nonpathogenic isolate of Borrelia burgdorferi. J. Clin. Microbiol. 28:2693–2699 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Anisimova M., Gascuel O. 2006. Approximate likelihood-ratio test for branches: a fast, accurate, and powerful alternative. Syst. Biol. 55:539–552 [DOI] [PubMed] [Google Scholar]
- 3. Barbour A. G., et al. 2009. Niche partitioning of Borrelia burgdorferi and Borrelia miyamotoi in the same tick vector and mammalian reservoir species. Am. J. Trop. Med. Hyg. 81:1120–1131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Brisson D., Dykhuizen D. E. 2004. ospC diversity in Borrelia burgdorferi: different hosts are different niches. Genetics 168:713–722 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Brisson D., Vandermause M. F., Meece J. K., Reed K. D., Dykhuizen D. E. 2010. Evolution of northeastern and midwestern Borrelia burgdorferi, United States. Emerg. Infect. Dis. 16:911–917 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bunikis J., et al. 2004. Sequence typing reveals extensive strain diversity of the Lyme borreliosis agents Borrelia burgdorferi in North America and Borrelia afzelii in Europe. Microbiology 150:1741–1755 [DOI] [PubMed] [Google Scholar]
- 7. Courtney J. W., Kostelnik L. M., Zeidner N. S., Massung R. F. 2004. Multiplex real-time PCR for detection of Anaplasma phagocytophilum and Borrelia burgdorferi. J. Clin. Microbiol. 42:3164–3168 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Dykhuizen D. E., et al. 2008. The propensity of different Borrelia burgdorferi sensu stricto genotypes to cause disseminated infections in humans. Am. J. Trop. Med. Hyg. 78:806–810 [PMC free article] [PubMed] [Google Scholar]
- 9. Feil E. J., Li B. C., Aanensen D. M., Hanage W. P., Spratt B. G. 2004. eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J. Bacteriol. 186:1518–1530 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Fikrig E., et al. 1992. Borrelia burgdorferi strain 25015: characterization of outer surface protein A and vaccination against infection. J. Immunol. 148:2256–2260 [PubMed] [Google Scholar]
- 11. Francisco A. P., Bugalho M., Ramirez M., Carrico J. A. 2009. Global optimal eBURST analysis of multilocus typing data using a graphic matroid approach. BMC Bioinformatics 10:152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Guindon S., Gascuel O. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol. 52:696–704 [DOI] [PubMed] [Google Scholar]
- 13. Hanincová K., Kurtenbach K., Diuk-Wasser M., Brei B., Fish D. 2006. Epidemic spread of Lyme borreliosis, northeastern United States. Emerg. Infect. Dis. 12:604–611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hanincová K., Liveris D., Sandigursky S., Wormser G. P., Schwartz I. 2008. Borrelia burgdorferi sensu stricto is clonal in patients with early Lyme borreliosis. Appl. Environ. Microbiol. 74:5008–5014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hoen A. G., et al. 2009. Phylogeography of Borrelia burgdorferi in the eastern United States reflects multiple independent Lyme disease emergence events. Proc. Natl. Acad. Sci. U. S. A. 106:15013–15018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Hordijk W., Gascuel O. 2005. Improving the efficiency of SPR moves in phylogenetic tree search methods based on maximum likelihood. Bioinformatics 21:4338–4347 [DOI] [PubMed] [Google Scholar]
- 17. Ivanova L., et al. 2009. Comprehensive seroprofiling of sixteen B. burgdorferi OspC: implications for Lyme disease diagnostics design. Clin. Immunol. 132:393–400 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Jones K. L., et al. 2006. Borrelia burgdorferi genetic markers and disseminated disease in patients with early Lyme disease. J. Clin. Microbiol. 44:4407–4413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Kulldorff M. 1997. A spatial scan statistic. Comm. Stat. Theor. Method 26:1481–1496 [Google Scholar]
- 20. Kumar S., Tamura K., Nei M. 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief. Bioinform. 5:150–163 [DOI] [PubMed] [Google Scholar]
- 21. Kurtenbach K., Sewell H.-S., Ogden N. H., Randolph S. E., Nuttall P. A. 1998. Serum complement sensitivity is a key factor in Lyme disease ecology. Infect. Immun. 66:1248–1251 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kurtenbach K., et al. 2002. Differential survival of Lyme borreliosis spirochetes in ticks that feed on birds. Infect. Immun. 70:5893–5895 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Kurtenbach K., et al. 2006. Key processes in the evolutionary ecology of Lyme borreliosis. Nat. Rev. Microbiol. 4:660–669 [DOI] [PubMed] [Google Scholar]
- 24. Margos G., et al. 2008. MLST of housekeeping genes captures geographic population structure and suggests a European origin of Borrelia burgdorferi. Proc. Natl. Acad. Sci. U. S. A. 105:8730–8735 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Margos G., et al. 2009. A new Borrelia species defined by multilocus sequence analysis of housekeeping genes. Appl. Environ. Microbiol. 75:5410–5416 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Margos G., et al. 2010. Multilocus sequence analysis of Borrelia bissettii strains from North America reveals a new Borrelia species, Borrelia kurtenbachii. Ticks Tick Borne Dis. 1:151–158 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Ogden N. H., et al. 2006. Ixodes scapularis ticks collected by passive surveillance in Canada: analysis of geographic distribution and infection with the Lyme borreliosis agent Borrelia burgdorferi. J. Med. Entomol. 43:600–609 [DOI] [PubMed] [Google Scholar]
- 28. Ogden N. H., et al. 2007. Tick seasonality, host infection dynamics and fitness of Ixodes scapularis-borne pathogens. Parasitology 134:209–227 [DOI] [PubMed] [Google Scholar]
- 29. Ogden N. H., et al. 2008. The role of migratory birds in introduction and range expansion of Ixodes scapularis ticks, and Borrelia burgdorferi and Anaplasma phagocytophilum in Canada. Appl. Environ. Microbiol. 74:1780–1790 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Ogden N. H., Lindsay R. L., Sockett P. N., Morshed M., Artsob H. 2009. Emergence of Lyme disease in Canada. CMAJ 180:1221–1224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Ogden N. H., et al. 2010. Active and passive surveillance, and phylogenetic analysis of Borrelia burgdorferi elucidate the process of Lyme disease risk emergence in Canada. Environ. Health Perspect. 118:909–914 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Qiu W. G., Dykhuizen D. F., Acosta M. S., Luft B. J. 2002. Geographic uniformity of the Lyme disease spirochete (Borrelia burgdorferi) and its shared history with tick vector (Ixodes scapularis) in the Northeastern United States. Genetics 160:833–849 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Seinost G., et al. 1999. Four clones of Borrelia burgdorferi sensu stricto cause invasive infection in humans. Infect. Immun. 67:3518–3524 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Strle F., Stanek G. 2009. Clinical manifestations and diagnosis of Lyme Borreliosis. Curr. Probl. Dermatol. 37:51–110 [DOI] [PubMed] [Google Scholar]
- 35. Tilly K., et al. 2006. Borrelia burgdorferi OspC protein required exclusively in a crucial early stage of mammalian infection. Infect. Immun. 74:3554–3564 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Wang G., et al. 2002. Disease severity in a murine model of Lyme borreliosis is associated with the genotype of the infecting Borrelia burgdorferi sensu stricto strain. J. Infect. Dis. 186:782–791 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Wormser G. P., et al. 1999. Association of specific subtypes of Borrelia burgdorferi with hematogenous dissemination in early Lyme disease. J. Infect. Dis. 180:720–725 [DOI] [PubMed] [Google Scholar]
- 38. Wormser G. P., et al. 2008. Borrelia burgdorferi genotype predicts the capacity for hematogenous dissemination during early Lyme disease. J. Infect. Dis. 198:1358–1364 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Wormser G. P., et al. 2008. Effect of Borrelia burgdorferi genotype on the sensitivity of C6 and 2-tier testing in North American patients with culture-confirmed Lyme disease. Clin. Infect. Dis. 47:910–914 [DOI] [PMC free article] [PubMed] [Google Scholar]