Delineating A. phagocytophilum Ecotypes | CDC EID

EID Journal Home > Volume 15, Number 12–December 2009

Volume 15, Number 12–December 2009
Research
Delineating Anaplasma phagocytophilum Ecotypes in Coexisting, Discrete Enzootic Cycles
Kevin J. Bown, Xavier Lambin, Nicholas H. Ogden, Michael Begon, Gill Telford, Zerai Woldehiwet, and Richard J. Birtles
Author affiliations: University of Liverpool, Liverpool, UK (K.J. Bown, M. Begon, Z. Woldehiwet, R.J. Birtles); University of Aberdeen, Aberdeen, UK (X. Lambin, G. Telford); and Public Health Agency of Canada, Saint-Hyacinthe, Quebec, Canada (N.H. Ogden)

Suggested citation for this article

Abstract
The emerging tick-borne pathogen Anaplasma phagocytophilum is under increasing scrutiny for the existence of subpopulations that are adapted to different natural cycles. Here, we characterized the diversity of A. phagocytophilum genotypes circulating in a natural system that includes multiple hosts and at least 2 tick species, Ixodes ricinus and the small mammal specialist I. trianguliceps. We encountered numerous genotypes, but only 1 in rodents, with the remainder limited to deer and host-seeking I. ricinus ticks. The absence of the rodent-associated genotype from host-seeking I. ricinus ticks was notable because we demonstrated that rodents fed a large proportion of the I. ricinus larval population and that these larvae were abundant when infections caused by the rodent-associated genotype were prevalent. These observations are consistent with the conclusion that genotypically distinct subpopulations of A. phagocytophilum are restricted to coexisting but separate enzootic cycles and suggest that this restriction may result from specific vector compatibility.

The tick-transmitted bacterium Anaplasma phagocytophilum is the causative agent of granulocytic anaplasmosis, an infection of medical and veterinary importance that is widely encountered across the temperate zones of the Northern Hemisphere (1–3). Although considerable effort has been put into determining the natural diversity of A. phagocytophilum (4–6), our understanding of ecologic and evolutionary processes that underlie this diversity remains far from complete. A. phagocytophilum has been detected in the blood of a wide range of mammals and in several different Ixodes species, which suggests that it is a generalist parasite with the capacity to exploit multiple hosts and vectors (1–3,5–9). However, evidence for the existence of subpopulations that are adapted to specific reservoir host species has recently been forthcoming (7,9,10), and these subpopulations appear to possess differing potential to be a public health threat (7,9,10). This phenomenon has also been described within another tick-borne generalist species–complex, Borrelia burgdorferi sensu lato (11) and, more recently, within the 1 generalist member of this complex, B. burgdorferi sensu strictu (12,13). As yet, no evidence has shown that subpopulations of either A. phagocytophilum or B. burgdorferi have adapted to different Ixodes species as vectors.

Knowledge of the existence of host- or vector-adapted subpopulations is important given the public health consequences of multivector transmission by these pathogens. For example, we and other researchers (14–19) have hypothesized that pathogen populations maintained in efficient tick-rodent cycles by nidicolous (nest-living and host-specialist) ticks, such as I. trianguliceps in the United Kingdom and I. spinipalpis and I. minor in the United States, could spill over into the human population due to the co-occurrence of sympatric exophilic (and human-biting) tick species such as I. ricinus in the United Kingdom and I. pacificus and I. scapularis in the United States.

The purpose of this study was to characterize the diversity of A. phagocytophilum strains circulating in a natural multihost, multivector system and to determine whether the observed diversity had any ecologic basis. We obtained compelling evidence to support the proposition that different subpopulations of A. phagocytophilum exploit different tick and mammal species and, as a result, occur in discrete enzootic cycles even though both vectors and hosts are sympatric.

Materials and Methods
Natural Multihost, Multivector Study System
Kielder Forest is a managed plantation forest dominated by Sitka and Norway spruce that covers 620 km2 in northern England (55°13´N, 2°33´W). The harvesting of timber generates clearcut areas of 5–12 ha, which are subsequently replanted and progress through grassland and the thicket stage after 12–15 years. The most abundant mammal species encountered within clearcut areas is the field vole (Microtus agrestis), which exhibits multiannual population cycles in which densities can reach >700/ha (20). Roe deer (Capreolus capreolus) are also abundant at an estimated density of 3 deer/km2 across the forest and are frequent visitors to clearcut areas (21). The presence of I. ricinus and I. trianguliceps ticks in clearcut areas has been documented (18,19).

Monitoring of Host and Vector Populations
Protocols for the handling and sampling of wild rodents were approved by the University of Liverpool Committee on Research Ethics and were conducted in compliance with the terms and conditions of licenses awarded under the UK Government Animals (Scientific Procedures) Act, 1986. Voles were trapped at 4-week intervals from March 2004 through November 2005 (excepting December 2004 and February 2005) at 4 principal study sites that were 3.5 km–12 km apart. Each animal captured was processed as described previously and a blood sample was taken from the tail tip (19). Voles were examined for ticks, with all larvae being removed and stored in 70% ethanol for identification (22,23) before releasing the animal at the point of capture. Nymph and adult ticks were not removed to minimize any effect on the transmission of tick-borne infections, which were being studied as part of an extensive longitudinal program. Host-seeking I. ricinus nymphs and adults were collected at monthly intervals from the principal study sites from March 2004 through September 2005 as previously described (19) and from 17 additional sites widely distributed across the Kielder Forest District. Collected ticks were stored and identified as described above. Roe deer blood samples were collected from January 2004 through July 2006 from animals culled throughout the forest and stored in EDTA-containing tubes at –20°C.

Host Bloodmeal Source Identification
The relative importance of different species as hosts for I. ricinus larvae was determined as previously described (24). Probes for the following taxa were used: Myodes spp., Apodemus spp., Microtus agrestis, Sciurus spp., Sorex araneus, Meles meles, and C. capreolus, together with a generic "bird" probe (24).

Monitoring of A. phagocytophilum Genotypes
Crude nucleic acid extracts were prepared from blood samples and host-seeking I. ricinus nymphs as previously described (11). The presence of A. phagocytophilum DNA in each extract was assessed by a real-time PCR (25).

Genotyping of A. phagocytophilum strains exploited sequence variation at 3 genetic loci, 16S rDNA, msp4, and DOV1. Fragments of msp4 and 16SrDNA were amplified and analyzed as described (18,25). DOV1 is a noncoding region of ≈275 bp lying immediately downstream of a previously described variable number tandem repeat (VNTR) locus (6). Amplification of this locus was achieved by using seminested PCR. The first round of amplification contained 10 ρmol of each of the primers DOV1f (5´-GAT CTA TGA ATT GCY RGT GTT ATA-3´) and DOV1r1 (5´-ACA TTG TCA ATT TCT CAC CAT-3´), 12.5 μL of 2× Master Mix (Abgene, Epsom, UK), 1 μL of nucleic acid extract and sterile H2O to a final volume of 25 μL, which was subjected to a thermal program of 95°C for 3 min, then 35 cycles of 95°C for 10 s, 58°C for 10 s, and 72°C for 50 s, then a final extension step of 72°C for 5 min. The second round of amplification incorporated 1 μL of first-round product into a reaction containing 10 ρmol of each of the primers DOV1f and DOV1r2 (5´-CAA YRC ACR YAC ATT TCT AGG A-3´), 22.5 μL of Reddymix (ABgene), made up to a final volume of a 25 μL with sterile H2O. This reaction mix was subjected to the same thermal program as used for the first round of amplification. DOV1 amplicons were sequenced directly by using the second round primers. DNA sequences from all 3 loci studied were verified, assembled and aligned by using Chromas Pro (Technelysium Pty Ltd, Tewantin, Queensland, Australia) and MEGA4 software (26). Sequence similarity calculations and phylogenetic inferences were conducted by using MEGA4 software (26).

Results
Monitoring of Host and Vector Populations
Figure 1 (please, see the full-text)

Figure 1. Prevalence of Anaplasma phagocytophilum infection in field voles (A) and of infestation of Ixodes ricinus tick larvae (black line)...


Figure 2 (please, see the full-text)

Figure 2. Number of Anaplasma phagocytophilum–infected (black bars) and uninfected (gray bars) animals encountered among Kielder Forest District roe deer sampled during January 2004–July 2006.


Figure 3 (please, see the full-text)

Figure 3. Phylogenetic tree inferred from alignment of Anaplasma phagocytophilum msp4 sequences obtained in this study or available from GenBank...


Figure 4 (please, see the full-text)

Figure 4. Phylogenetic tree inferred from alignment of Anaplasma phagocytophilum DOV1 sequence types obtained in this study...

A total of 2,926 blood samples from 1,503 voles at the 4 study sites was obtained. Similar numbers of voles were encountered at each site and the population size at all sites fluctuated in a broadly synchronous manner, in keeping with the well-documented seasonal and multiannual population cycles (27). A. phagocytophilum DNA was detected in 183 (6.3%) of the blood samples, representing 157 (10.4%) of individual animals tested. Except for the bacterium being seemingly absent from 1 site in 2004, the seasonal variation in prevalence of infection was similar at all sites, with infections disappearing over winter, before reappearing in the spring and persisting until late autumn (Figure 1, panel A).

Of the 3,378 ticks that were recorded on the surveyed voles, 83.6% (2,823) were larvae, 13.4% (454) were nymphs, and 2.9% (101) adults. Approximately equal numbers of I. ricinus (1,618, 57.3%) and I. trianguliceps (1,205, 42.7%) were identified among the larvae, the seasonal dynamics of which are shown in Figure 1, panel B. I. ricinus larvae were most abundant in late spring/early summer, whereas I. trianguliceps larvae were most abundant in late autumn. The dramatic spike in the number of I. ricinus larvae recorded in May 2005 resulted from a small number of voles at one of our principal study sites having an extremely high number of larvae. Although nymph and adult ticks were not removed from voles (so could not be identified to species), their numbers were recorded. Of relevance to this study, virtually no nymphs or adults were observed on voles between November and April (Figure 1, panel B). The absence of the life stages that are capable transmitters of A. phagocytophilum underlies the disappearance of infections in voles during winter.

Blood samples were collected from 279 roe deer and A. phagocytophilum DNA was detected in 132 (47.3%) of these samples. Infections were detected throughout the year, with infection prevalence consistently high during the late spring/early summer of the years surveyed (Figure 2).

In total, 4,984 nymphs, 680 adult males, and 656 adult female host-seeking I. ricinus ticks were collected by dragging. The seasonal dynamics of both life stages have been presented elsewhere (19). A. phagocytophilum DNA was detected in 30 of 4,256 nymphs tested (0.7%), 9 of 263 adult females (3.4%) and 8 of 321 adult males (2.5%). Infected nymphs were encountered at 10 different sites. Infected host-seeking nymphs were collected during the same dragging session on only 8 occasions, suggesting that, for the most part, infected nymphs had fed on different animals.

abrir aquí para acceder al documento CDC EID completo del cual se reproduce un 40%:
Delineating A. phagocytophilum Ecotypes | CDC EID