Evidence for Host-Genotype Associations of Borrelia burgdorferi Sensu Stricto

RESEARCH ARTICLE Evidence for Host-Genotype Associations of Borrelia burgdorferi Sensu Stricto Samir Mechai1, Gabriele Margos2,3, Edward J. Feil4, Nicole Barairo5, L. Robbin Lindsay5, Pascal Michel1,6, Nicholas H. Ogden1,6* 1 Groupe de Recherche en Épidémiologie des Zoonoses et Santé Publique, Faculté de Médecine Vétérinaire, Université de Montréal, Saint-Hyacinthe, Québec, Canada, 2 National Reference Centre for Borrelia, Oberschleissheim, Germany, 3 Bavarian Health and Food Safety Authority, Oberschleissheim, Germany, 4 The Milner Centre for Evolution, Department of Biology and Biochemistry, University of Bath, Claverton Down, Bath, United Kingdom, 5 National Microbiology Laboratory, Public Health Agency of Canada, Winnipeg, Manitoba, Canada, 6 National Microbiology Laboratory, Public Health Agency of Canada, Saint-Hyacinthe, Québec, Canada * a11111 OPEN ACCESS Citation: Mechai S, Margos G, Feil EJ, Barairo N, Lindsay LR, Michel P, et al. (2016) Evidence for HostGenotype Associations of Borrelia burgdorferi Sensu Stricto. PLoS ONE 11(2): e0149345. doi:10.1371/ journal.pone.0149345 Editor: Brian Stevenson, University of Kentucky College of Medicine, UNITED STATES Received: November 5, 2015 Accepted: January 29, 2016 Published: February 22, 2016 Copyright: © 2016 Mechai et al. 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 Statement: All relevant data are within the paper and its Supporting Information files. Funding: The study was funded by the Public Health Agency of Canada. Competing Interests: The authors have declared that no competing interests exist. Abstract Different genotypes of the agent of Lyme disease in North America, Borrelia burgdorferi sensu stricto, show varying degrees of pathogenicity in humans. This variation in pathogenicity correlates with phylogeny and we have hypothesized that the different phylogenetic lineages in North America reflect adaptation to different host species. In this study, evidence for host species associations of B. burgdorferi genotypes was investigated using 41 B. burgdorferi-positive samples from five mammal species and 50 samples from host-seeking ticks collected during the course of field studies in four regions of Canada: Manitoba, northwestern Ontario, Quebec, and the Maritimes. The B. burgdorferi genotypes in the samples were characterized using three established molecular markers (multi-locus sequence typing [MLST], 16S-23S rrs-rrlA intergenic spacer, and outer surface protein C sequence [ospC] major groups). Correspondence analysis and generalized linear mixed effect models revealed significant associations between B. burgdorferi genotypes and host species (in particular chipmunks, and white-footed mice and deer mice), supporting the hypotheses that host adaptation contributes to the phylogenetic structure and possibly the observed variation in pathogenicity in humans. Introduction In North America, Borrelia burgdorferi sensu stricto (hereafter termed B. burgdorferi for simplicity) is a member of the bacterial genospecies complex B. burgdorferi sensu lato (s.l.) that is associated with Lyme disease [1]. In Eurasia, five genospecies of the B. burgdorferi s.l. complex are associated with Lyme disease [1]: B. burgdorferi, B. afzelii, B. garinii, B. bavariensis and B. spielmanii and the two main tick vectors are Ixodes ricinus (in Europe) and I. persulcatus (in Asia) [2, 3]). In North America, B. burgdorferi is mostly transmitted by two tick species: I. scapularis in the regions encompassing northeastern USA and southeastern Canada, and the PLOS ONE | DOI:10.1371/journal.pone.0149345 February 22, 2016 1 / 25 Host-Genotype Associations of Borrelia burgdorferi upper Midwest USA and south central Canada, and I. pacificus in the western coastal states of the USA and in British Columbia, Canada. In Eurasia, the different B. burgdorferi s.l. genospecies are associated with different types of clinical disease [4]. Arthritis is associated with B. burgdorferi infection; neuroborreliosis with B. garinii and B. bavariensis infection, and chronic dermatological manifestations with B. afzelii [5–7]. Most of the clinical features seen in Europe are also seen in North America, and these include those of early Lyme disease (Erythema migrans: EM), early disseminated Lyme disease (neuroborreliosis including facial palsy, meningitis and peripheral radiculoneuropathy, and atrioventricular block) and late disseminated Lyme disease (including Lyme arthritis) [2, 8, 9]. In North America there is evidence that different genotypes of B. burgdorferi show different levels of pathogenicity in humans, specifically whether or not the bacterium disseminates systemically from the early phase infection in the skin where the infective tick bit the patient [10–14]. In Europe, B. burgdorferi s.l. genospecies are frequently specialized for transmission by different host species [15]: B. afzelii and B. bavariensis are rodent host specialists [15, 16], B. garinii is a bird specialist [8] and B. lusitaniae may be a lizard specialist [17]. In North America, B. burgdorferi is considered a host generalist [4, 18], although more stable suitable environments associated with expanding woodland habitats, increased abundance of tick vectors and reservoir hosts [19] are thought to be creating conditions favourable for adaptive radiation and multiple niche polymorphism [4]. Most parasites show some degree of host preference [20–22], which is a critical pre-adaptation for host specialization if conditions for transmission are suitable [23], as they may increasingly be for B. burgdorferi in North America. There is some evidence of host associations for B. burgdorferi in the form of unequal frequencies of B. burgdorferi genotypes in samples collected in the field from different sources [24–26], and differential infection and transmission efficiency among different host-genotype pairings [25, 27]. Such associations are of public health interest as they may be linked to the capacity of the different genotypes to show different pathogenicity in humans and varying capacity to stimulate antibodies detectable in current serological tests, while the existence of host associations may allow prediction of regions and habitats where different genotypes are more likely to occur [28]. The clade structure of the B. burgdorferi phylogenetic tree obtained using concatenated housekeeping genes of a multilocus sequence typing (MLST) method does not seem to be based on geographic isolation of genotypes [29, 30]. It has been hypothesized that the clades were associated with introductions and/or population expansions after bottlenecks possibly associated with glacial-interglacial periods [31], although ecological isolation driven by host species associations may also explain the (...truncated)