Geographic and Temporal Variations in Population Dynamics of Ixodes ricinus and Associated Borrelia Infections in The Netherlands

Abstract

In a countrywide investigation of the ecological factors that contribute to Lyme borreliosis risk, a longitudinal study on population dynamics of the sheep tick Ixodes ricinus and their infections with Borrelia burgdorferi sensu lato (s.l.) was undertaken at 24 sites in The Netherlands from July 2006 to December 2007. Study sites were mature forests, dune vegetations, or new forests on land reclaimed from the sea. Ticks were sampled monthly and nymphal ticks were investigated for the presence of Borrelia spp. I. ricinus was the only tick species found. Ticks were found in all sites, but with significant spatial and temporal variations in density between sites. Peak densities were found in July and August, with lowest tick numbers collected in December and January. In some sites, questing activities of I. ricinus nymphs and adults were observed in the winter months. Mean monthly Borrelia infections in nymphs varied from 0% to 29.0% (range: 0%–60%), and several sites had significantly higher mean nymphal Borrelia infections than others. Four genospecies of Borrelia burgdorferi s.l. were found, with B. afzelii being dominant at most sites. Borrelia infection rates in nymphal ticks collected in July, September, and November 2006 were significantly higher (23.7%, p < 0.01) than those in the corresponding months of 2007 (9.9%). The diversity in Borrelia genospecies between sites was significantly different (p < 0.001). Habitat structure (tree cover) was an effective discriminant parameter in the determination of Borrelia infection risk, as measured by the proportion of nymphal ticks infected with B. burgdorferi s.l. Thickness of the litter layer and moss cover were positively related to nymphal and adult tick densities. The study shows that Borrelia-infected ticks are present in many forest and dune areas in The Netherlands and suggests that in such biotopes, which are used for a wide variety of recreational activities, the infection risk is high.

Introduction

L yme borreliosis has become a vector-borne disease of great significance in the northern hemisphere since its discovery in 1976. This disease, caused by spirochetes of the Borrelia burgdorferi complex, is transmitted by hard ticks of the Ixodes ricinus complex. In Europe, I. ricinus L. is the dominant vector and its ecology has been reviewed by Gray (1998). The species is found predominantly in forested habitats, where microclimate and host composition are suitable for survival of the four life stages—egg, larva, nymph, and adult—the last three requiring a single blood meal from a vertebrate host (mammalian or avian). Should the host be a reservoir of Borrelia bacteria, it can infect the ticks, which remain infected throughout successive life stages. Borrelia burgdorferi sensu lato (s.l.) is a species complex with at least 15 genospecies, of which 7 are present in western Europe (Hubalek and Halouzka 1997, Vennestrom et al. 2008). The mean infection rate of I. ricinus nymphs with B. burgdorferi s.l. in Europe is ∼10%, but varies from 0% to over 30% (Rauter and Hartung 2005). Borrelia infection rates in ticks can exhibit great temporal and spatial fluctuations (Rauter and Hartung 2005), which are ascribed to variations in environmental conditions such as animal host density and climate (Gray 1998). In addition to Borrelia, I. ricinus may also be infected with several other human pathogens such as Rickettsia spp., Anaplasma (Ehrlichia) spp., Babesia spp., and Bartonella spp. as well as, in some regions, tick-borne encephalitis (TBE) virus. In particular, coinfections of Borrelia and Rickettsia spp. are common (Schouls et al. 1999, Christova et al. 2003, Skarphedinsson et al. 2007).

Although the ecology of Lyme borreliosis is reasonably well understood, the factors that determine the observed large variations in tick density, Borrelia infections, and other determinants for Lyme borreliosis are not (Maetzel et al. 2005, Killilea et al. 2008). For example, a strong correlation between Lyme borreliosis risk and forested areas has been reported (De Mik et al. 1997, Estrada-Peña 2001, Jaenson et al. 2009), but risk level varies considerably between forests (Killilea et al. 2008). The abundance of deer is thought to contribute to risk variation, but factors such as human behavior, litter layer, abundance of small animals, and micro- as well as macroclimate may also affect risk level. In The Netherlands, Lyme borreliosis has developed into an important disease in the last decade, with an estimated 17,000 recorded cases in 2005 (Hofhuis et al. 2006). The relationship between tick activity and seasonal and regional variations in Borrelia infection rates of ticks is poorly understood. The present study was undertaken to examine the spatial and temporal dynamics of I. ricinus across The Netherlands and to establish their relationship to Borrelia infection rates of ticks collected at selected study sites. The effects of habitat characteristics on the spatial variation in tick densities and Borrelia infections in ticks are discussed.

Materials and Methods

Description of study sites

Twenty-four sites, distributed across The Netherlands, were selected for tick collections (Fig. 1). Criteria for selection were (1) likelihood of finding I. ricinus ticks and (2) the availability of a team of volunteers to make monthly collections. Within each location, two marked transects of 100 m² were monitored consistently. Geographic coordinates of each transect were established with Global Positioning Device (GPS), and a description of the habitat was made and classified according to Braun-Blanquet (1964) and Weeda et al. (2005) (Tables 1 and 2). Ten of the 24 sites, evenly distributed across The Netherlands and representing the different habitat classes present among the 24 sites, were chosen for detailed analysis of Borrelia infections of nymphal ticks.

FIG. 1.

Map of The Netherlands, showing the location of the 24 study sites where Ixodes ricinus was collected. Encircled sites represent the sentinel sites referred to in the study.

Table 1.

Plant Communities (Associations) of All 48 Transects in the 24 Sites, After Braun-Blanquet (1964) and Weeda et al. (2005)

No.	Location	Coordinates transect 1	Vegetation type transect 1	Coordinates transect 2	Vegetation type transect 2
1	Bellingwedde	53°04′26N, 7°07′36E	Betulo Quercetum roboris	53°04′26N, 7°07′37E	Betulo Quercetum roboris
2	Schiermonnikoog	53°29′34N, 6°09′46E	Fago Quercetum	53°29′44N, 6°09′55E	Cladonio Pinetum sylvestris
3	Appelscha	52°55′33N, 6°20′34E	Eurhynchiumpraelongum Pseudosderopodium-purum-[Vaccinio-Piceetea]	52°55′35N, 6°20′45E	Prunus serotina [Dicrano-Pinion]
4^a	Gieten	53°00′54N, 6°45′18E	Betulo Quercetum roboris deschampsietosum	53°00′56N, 6°45′12E	Fago Quercetum holcetosum
5^a	Ruinen	52°47′07N, 6°18′22E	Leucobryo-Pinetum deschampsietosum	52°47′07N, 6°18′22E	Betulo Quercetum roboris vaccinietosum
6	Hoogeveen	52°47′08N, 6°18′28E	Deschampsio Fagetum typicum	52°45′05N, 6°31′26E	Betulo Quercutum roboris vaccinietosum
7	Nijverdal	52°20′18N, 6°25′27E	Leucobryo Pinetum vaccinietosum	52°20′18N, 6°25′27E	Cladonio Pinetum sylvestris dicranetosum polyseti
8^a	Haaksbergen	52°09′50N, 6°47′54E	Betulo Quercetum roboris molinietosum	52°09′50N, 6°47′54E	Betulo Quercetum roboris cladonietosum
9	Montferland	51°53′29N, 6°13′10E	Leucobryo-pinetum vaccinietosum	51°55′42N, 6°13′23E	Quercion roboris
10	Apeldoorn	52°12′27N, 5°53′31E	Leucobryo Pinetum vaccinietosum	52°12′54N, 5°54′26E	Betulo Quercetum roboris vaccinietosum
11^a	Ede	52°01′41N, 5°41′49E	Betulo Quercetum roboris cladonietosum	52°01′41N, 5°41′49E	Betulo Quercetum roboris cladonietosum
12^a	Hoog Baarlo	52°06′20N, 5°52′29E	Betulo Quercetum roboris vaccinietosum	52°06′20N, 5°52′29E	Betulo Quercetum roboris vaccinietosum
13	Bilthoven	52°09′22N, 5°13′40E	Leucobryo-Pinetum deschampsietosum	52°09′27N, 5°13′47E	Betulo Quercetum roboris deschampsietosum
14	Dronten	52°31′38N, 5°41′06E	Fraxino-Ulmetum typicum	52°31′31N, 5°41′09E	Fraxino-Ulmetum typicum
15	Texel	53°04′11N, 4°44′26E	Fago Quercetum holcetosum	53°04′11N, 4°44′26E	Betulo Quercetum roboris dryopteridetosum
16	Bergen aan Zee	52°38′40N, 4°38′05E	Crataego Betuletum pubescentis	52°38′48N, 4°39′15E	Carex arenaria Calamagrostis epigejos-[Dicrano-Pinion]
17^a	Twiske	52°26′52N, 4°53′35E	Artemisio-salicetum albae agrostietosum stoloniferae	52°26′52N, 4°53′35E	Lychnido-Hypericetum Tetrapteri orchietosum morionis
18^a	Kwade Hoek	51°50′25N, 3°58′58E	Rhamno Crataegetum	51°50′25N, 3°58′58E	Rhamno Crataegetum
19^a	Wassenaar	52°09′32N, 4°21′41E	Carex arenaria - Calamagrostis epigejos-[Dicrano-Pinion]	52°09′23N, 4°21′38E	Hippophao Ligustratum typicum
20	Schouwen	51°42′37N, 3°44′22E	Fago Quercetum convallarietosum	51°42′34N, 3°44′32E	Fago Quercetum holcetosum
21^a	Veldhoven	51°25′19N, 5°20′02E	Betulo Quercetum roboris molinietosum	51°25′19N, 5°20′02E	Betulo Quercetum roboris deschamsitosum
22	Cuijck	51°42′14N, 5°48′51E	Fago Quercetum molinietosum	51°42′15N, 5°48′57E	Carici-Elongatae Alnetum rubetosum idaei
23^a	Eijsden	50°47′23N, 5°44′14E	Stellaria Carpinetum polystichetosum	50°48′54N, 5°44′45E	Stellario Carpinetum allietosum
24	Vaals	50°47′38N, 5°57′19E	Stellario - Carpinetum allietosum	50°45′48N, 5°58′57E	Luzulo Luzuloides Fagetum

No. corresponds to the sites in Figure 1. Coordinates are given in degrees, latitude, and longitude.

Ticks from 10 selected sentinel sites were inspected for B. burgdorferi sensu lato infections bimonthly during the study period.

Table 2.

Habitat Characteristics of All 48 Transects in the 24 Sites

		Transect 1					Transect 2
No.	Location	Litter	Moss	Herb	Brush	Tree	Litter	Moss	Herb	Brush	Tree
1	Bellingwedde	1	8	6	8	1.0	1	6	1	8	0.5
2	Schiermonnikoog	1	8	1	1	1.0	1	3	1	8	5.0
3	Appelscha	8	7	3	8	8.0	8	4	8	7	10.0
4	Gieten	1	8	3	8	3.0	1	3	6	8	5.0
5	Ruinen	3	2	2	6	7.5	6	7	5	7	17.5
6	Hoogeveen	1	1	1	8	20.0	8	5	6	4	10.0
7	Nijverdal	6	7	1	7	5.0	5	8	5	1	0.0
8	Haaksbergen	4	8	5	8	4.0	7	8	3	6	1.0
9	Montferland	8	8	6	7	5.0	1	8	5	7	2.0
10	Apeldoorn	3	8	4	7	15.0	3	8	4	7	15.0
11	Ede	8	8	4	4	14.0	8	6	4	4	18.7
12	Hoog Baarlo	2	8	3	8	3.6	7	8	3	5	2.8
13	Bilthoven	3	5	3	5	6.0	5	3	4	6	6.5
14	Dronten	1	8	4	8	0.0	1	6	4	8	0.0
15	Texel	1	3	6	6	2.0	1	6	5	7	3.0
16	Bergen aan Zee	1	8	8	7	1.0	1	8	1	8	2.0
17	Twiske	1	8	1	1	0.0	1	1	1	1	0.0
18	Kwade Hoek	1	7	4	1	0.0	1	4	8	1	0.0
19	Wassenaar	2	6	7	8	1.0	6	7	5	1	1.0
20	Schouwen	2	7	4	8	5.0	2	4	5	8	8.0
21	Veldhoven	4	8	4	7	9.0	7	7	4	3	6.0
22	Cuijck	3	7	7	7	12.5	6	7	8	4	10.0
23	Eijsden	6	8	3	8	0.0	2	8	4	8	0.0
24	Vaals	2	8	6	8	0.0	1	8	4	8	10.0

No. corresponds to the sites in Figure 1. Litter is thickness of litter layer (in cm); the other variables are coverage on a 1–10 scale of the moss layer, herbal layer, brush layer, and tree layer, respectively (Braun-Blanquet 1964, Weeda et al. 2005).

Tick collections

Ticks were collected once per month between July 2006 and December 2007. Teams of experienced volunteers, recruited through the Association for Environmental Education (IVN) and familiar with ecological investigations, were given a training session by the researchers at the start of the first sampling period. They were visited by one of the authors after 6 months to verify the quality of the sampling procedure. At each of the 24 sites (Table 1), tick collections were made by dragging a white cotton cloth (1 m²) over the two marked transects (Daniels et al. 2000). The cloth was inspected for the presence of ticks at intervals of 25 m (Tälleklint and Jaenson 1996). Larvae, nymphs, and adult ticks were collected using forceps, placed in 70% ethanol, and dispatched by mail to the research team in Wageningen, where all ticks were recounted and identified to species and life stage.

Determination of Borrelia infections in ticks

At 2-month intervals, starting in July 2006, nymphal ticks from 10 of the 24 sites (see Description of study sites section) were examined for the presence of Borrelia infections. Detection was done using PCR followed by the reverse line blot procedure, modified after Schouls et al. (1999) for identification of the Borrelia species (Van Overbeek et al. 2008). If fewer than 30 nymphs were collected, all ticks were examined; if more than 30 nymphs were found, a random sample of 30 ticks was selected for Borrelia analysis. On two occasions, in May and September 2007, nymphal ticks collected from all (n = 24) sites were examined for Borrelia infections. DNA extraction was performed as described by Schouls et al. (1999) and 60 μL of each extract was stored at −80°C until further analyses.

Data analysis

Statistical analyses were performed using Genstat software (release 12.1). Generalized linear models (GLM) were used to analyze temporal and spatial variations in tick densities. Because tick densities are counts, a Poisson distribution with a logarithm link function was chosen. Tick densities were used as response variates, and temporal units (e.g., year and month) and location were used as explanatory variables in the regression models. Analyses of spatial and temporal variations in nymphal infection with B. burgdorferi s.l. were performed with a GLM with a binomial distribution and a logit link function, and infections were linked to the total number of nymphs analyzed in each observation. Post hoc t-probabilities were calculated to test pairwise differences between sites, months, and years. Relationships among habitat characteristics, tick densities, and B. burgdorferi s.l. infection prevalence were calculated using GLM as described above, but here using the sum of a given variable over the 18-month study period as response variables and the vegetation characteristics given in Table 2 as explanatory variables. In all GLM analyses, overdispersion was taken into account where appropriate. Differences in composition of the dominant species B. afzelii, B. garinii, B. burgdorferi sensu stricto, and B. valaisiana between sites were analyzed using log-linear regression on a contingency table that was derived from Table 3, including ticks with coinfections. The presence of a significant interaction between location and species indicates that species composition differs between sites.

Table 3.

Borrelia Species Distribution in I. ricinus Collected in 10 Sentinel Locations from July 2006 to December 2007

Location	Number of ticks examined	Average % of ticks infected	B. afzelii	B. garinii	B. sensu lato only	B. sensu stricto	B. valaisiana	B. ruski (B. afzelii-like)	No. with coinfections	Coinfection (no.)	Significance of Borrelia diversity
Ede	329	15.5	43	3	3	1	4	0	3	Bg+Bs (1); Bg+Bv (2)	a
Eijsden	105	10.5	8	0	0	1	2	0	0		a
Gieten	135	19.3	14	6	3	2	1	1	2	Bg+Ba (1); Bg+Br (1)	b
Haaksbergen	174	7.5	4	1	2	5	1	0	0		b
Hoog Baarlo^a	193	6.7	4	3	1	1	4	0	0		b
Kwade Hoek	119	38.7	41	1	1	1	7	0	5	Bv+Ba (4); Bg+Ba (1)	a
Ruinen	102	9.8	8	1	0	1	0	0	0		a
Twiske	167	13.2	14	0	1	3	3	0	0		a
Veldhoven	273	8.8	23	0	1	0	0	0	0		a
Wassenaar	47	4.3	2	0	0	0	0	0	0		a
Total examined and infected	1644		161	15	12	15	22	1	10
% infected of all sites combined		13.1
Frequency (%) of Borrelia genospecies			71.2	6.6	5.3	6.6	9.7	0.4	4.4

Different letters in the last column indicate significant differences in Borrelia species diversity (log-linear regression: p < 0.001).

Thirteen infections in 2006; no infections in 2007.

Stepwise removal and pairwise comparison of sites was used to reveal the sites that differed significantly. In all analyses, effects were considered to be significant at a level of p ≤ 0.05.

Results

Population dynamics

I. ricinus was found in all sites studied. In Dronten (site 14), the species was found on only 6 of 18 sampling dates and in very low numbers. In Ede (site 11) and Hoog Baarlo (site 12), ticks were found on all sampling occasions. There was a strong seasonal variation in tick density, with a single peak in the summer, and reduced activity of larvae and adults in the winter (Fig. 2). At this time of the year, however, activity of nymphs was still observed, with a mean density of 1.4 (January) and 2.5 (February) per 200 m², respectively. Comparing the months July–December 2006 and 2007, significantly more larvae and adults were collected in 2007 than in 2006 (larvae: p = 0.001; adults: p = 0.002). Comparing the same months, there was a significantly higher nymphal density in 2007 than in 2006 (p = 0.01). Considering month-to-month variations, larval densities were significantly different in all months (p = 0.01) except for November 2006 and 2007, September 2006 and October 2007, and July 2007 and September 2007. Adult densities were significantly higher in September 2006, July 2007, and August 2007 than in all other months (p = 0.013, p < 0.001, and p < 0.001, respectively).

FIG. 2.

Mean monthly number of Ixodes ricinus collected per 200 m² of habitat in 24 sites in The Netherlands from July 2007 to December 2007. L, larvae; N, nymphs; A, adults (males + females). Error bars represent standard errors of the mean.

During the 18 months of the study, and considering the 24 study sites, there was a significant effect of the thickness of the litter layer and moss cover on numbers of nymphs and adults per location (nymphs: p = 0.002 and p = 0.01; and adults: p = 0.027 and p = 0.026, respectively). Tree, shrub, and herbal coverage, separately, did not have an effect on the number of ticks at each site.

Borrelia burgdorferi s.l. infection rates

The average infection rate with Borrelia burgdorferi s.l. of all nymphal ticks combined in May 2007 was 9.1% (n = 421), and in September 2007, 10.3% (n = 275). The range of Borrelia infections in May was 3.3%–60.0% and in September 3.9%–50.0%. In 8 of the 24 sites, no infected ticks were detected (Fig. 3). In seven sites, nymphs contained Borrelia only in one of these 2 months, whereas nymphs from nine sites had Borrelia infections in both months. Nymphs from four sites had Borrelia-positive ticks in May, but none in September. Three sites had only Borrelia-positive nymphs in September. The mean Borrelia infection rate of all sites was similar in May and September 2007 (p = 0.885). Infection rates in these months differed between sites, although these were at the border of significance (p = 0.051; Fig. 3). Further, there was no relationship between Borrelia infection rate and tick density (p = 0.264; Fig. 4).

FIG. 3.

Borrelia infections of nymphal Ixodes ricinus collected in 24 study sites in May and September 2007. Black bars represent infection percentage in May 2007; gray bars represent infection percentage in September 2007. Nymphal densities are represented by black and gray diamonds for May and September 2007, respectively.

FIG. 4.

Relationship between Borrelia infection rates of nymphal ticks and tick density. Data points represent monthly nymphal densities on 100 m² transects, where at least one nymph was collected and analyzed (n = 215 data points).

Considering the 2 months in which Borrelia infections were examined in nymphal ticks from all 24 study sites, there was a highly significant negative effect of tree cover on Borrelia infection rate of nymphal I. ricinus (p < 0.001); the more open the forest, the more ticks were infected. A similar effect was observed on the pooled data of the 10 sentinel sites over the 18-month study period.

Considering the 10 sentinel sites only, where nymphal ticks had been examined for Borrelia infections at 8-week intervals from July 2006 to December 2007, the mean Borrelia infection rate over 18 months was 13.7% (n = 1644), which differed significantly between sites (p < 0.001; Table 3). There was a significant difference between infection rate for corresponding months in 2006 and 2007 (Fig. 5; p < 0.001). No Borrelia infections could be detected in the nymphal ticks collected in January 2007 (n = 22). In March and May 2007, Borrelia infections rates were 1.3% and 6.1%, respectively. Remarkably, the Borrelia infection rate of nymphal ticks from Hoog Baarlo in 2006 was 16.0% (n = 58), whereas in 2007 it was 0% (n = 112).

FIG. 5.

Bimonthly mean number of nymphs collected per 200 m² of habitat and the corresponding Borrelia infection rate (%) from 10 sentinel sites in The Netherlands between July 2006 and December 2007. Error bars represent standard errors of the mean.

In the 10 sentinel sites, four Borrelia genospecies were identified (Table 3). B. afzelii was the dominant species in the complex (71.2%, p < 0.001), and B. ruski [a B. afzelii-like isolate (Alekseev et al. 2001, Wielinga et al. 2006)], the most uncommon, was found on one occasion only. Twelve Borrelia infections (5.3%) could not be identified at species level. The distribution of Borrelia genospecies over the study sites was heterogeneous, but differed significantly between sites (p < 0.001; Table 3). For example, Borrelia burgdorferi sensu stricto was the main genospecies in Haaksbergen (site 8), whereas B. valaisiana was more abundant in Kwade Hoek (site 18) than in other sites, and B. garinii prevailed in Gieten and Hoog Baarlo. Infections with two Borrelia genospecies were found in 4.4% of all ticks examined (Table 3).

Discussion

The present study provides, for the first time in The Netherlands, a countrywide and seasonal overview of I. ricinus populations and their corresponding B. burgdorferi s.l. infections. I. ricinus was found at all 24 study sites, but with a large spatial and temporal variation. This is not surprising, because I. ricinus depends on several factors that determine habitat suitability. Vegetation structure, micro- as well as macroclimate, and host composition and abundance can affect not only tick density, but also the proportion of ticks infected with Borrelia (Gray et al. 1998, 2009, Eisen et al. 2006). European studies having extensive geographical coverage are scarce, but have taken place in the United Kingdom (Pietzsch et al. 2005, Scharlemann et al. 2008), Switzerland (Jouda et al. 2004), and Sweden (Jaenson et al. 2009). The latter authors concluded that the distribution of I. ricinus in Sweden corresponds to the distribution of deciduous trees other than birch or aspen and that proliferation of ticks is aided by availability of deciduous vegetation that is favorable to mammals and birds, which serve as hosts for the ticks. From these and our own study, it became evident that I. ricinus has a wide distribution within the northern temperate climate zone and that most forest ecosystems below 1600 m in altitude are suitable for I. ricinus (Estrada-Peña et al. 2006).

There was a strong seasonal variation in questing tick abundance, with a single peak in early June and lowest abundance in January. This is not surprising, as summer climate is favorable for I. ricinus and historic winter temperatures in western Europe were generally too low for tick activity. However, the relatively high numbers of ticks caught between December 2006 and February 2007 (Fig. 2) challenge this paradigm. The winter of 2007 was relatively warm (www.knmi.nl/, accessed January 22, 2010) so that a portion of the tick population remained active. Dautel et al. (2008) similarly reported unusually high tick activity in Germany throughout January/February and ascribed this to the extremely mild winter of 2006/2007. It has been suggested that climate change has contributed to this shift in seasonal dynamics (Lindgren and Gustafson 2001, Gray et al. 2009).

There is some evidence that the abundance of ticks has increased in the United Kingdom (Scharlemann et al. 2008) and in The Netherlands (Hofhuis et al. 2006) in recent years, but because reliable historical data on tick abundance and distribution are missing, we can only suggest that, based on the growing trend in tick bites on humans in The Netherlands and on red grouse in the United Kingdom, tick populations may be escalating. Possible reasons for this are expansion of nature reserves, increased abundance of wildlife, the continuing reduction in the use of pesticides in agriculture and forestry (Horne and Fielding 2002), and climate change (Gray et al. 2009).

In the selection of our study sites, we gave preference to areas characterized as coniferous/deciduous woodland and/or nature reserve, as these were considered to be prime biotopes for I. ricinus in similar ecozones of Europe (Estrada-Peña et al. 2006, Jaenson et al. 2009). In our study, a large variation in tick abundance between the 24 sites was found, with consistent associations between tick density and sites. Diuk-Wasser et al. (2006) reported similar phenomena from the United States, where abundance of the black-legged tick I. scapularis appeared associated with specific sites, having different host abundances. In Europe, similar associations between sites and tick density were also reported (e.g., Jouda et al. 2004). In our study, habitat characteristics varied considerably between the 24 sites. These habitat differences can facilitate different microclimate conditions and variable host abundance and diversity. Thickness of litter layer and cover of the moss layer were the only measured parameters that were positively correlated with tick density. This confirms that these are essential for tick survival because they serve as refuge for ticks during periods of unsuitable climatic conditions (Randolph and Storey 1999) and shelter tick eggs and blood-fed molting ticks during the largest part of their life. In addition, macroclimatic conditions in The Netherlands could explain differences in tick phenology between inland and coastal areas (Smit et al. 2003). However, these differences are considered marginal in explaining differences in tick population size (Gray, 1991) and were not included in our study.

Apart from habitat characteristics, host composition and abundance may also affect tick populations. In western Europe, I. ricinus has been found to feed on a wide variety of mammals and birds (Matuschka et al. 1991, Humair et al. 2007). In particular, small rodents (mice and voles) and birds are important hosts for larval stages. There are no published data on small mammals for our study sites, although several recent studies in The Netherlands showed that mice and voles are abundant in many locations (Jagers op Akkerhuis et al. 2003, Smit et al. 2003, Gassner et al. 2008). Similarly, roe deer are present at most study sites (S. van Wieren, pers. comm.) and are considered to be a primary host for nymphal and adult I. ricinus (Tälleklint and Jaenson 1997, Pichon et al. 2006, Vor et al. 2010). We conclude that, overall, none of the study sites were unsuitable for I. ricinus and that the large variations in abundance between the sites were most likely due to variation in the composition of the vegetation and abundance and species composition of host animals.

The observed differences in Borrelia infections, with variations from 3% to 60% on some monthly sampling occasions, fall within data reported from elsewhere in western Europe (Rauter and Hartung 2005). Although there was no significant correlation between Borrelia infection rate and study site for the 24 sites in May and September 2007, the 10 sentinel sites, where ticks were analyzed more frequently, can be grouped into classes with medium-to-high Borrelia infections (e.g., Kwade Hoek, Gieten, Ede) and those with low-to-medium Borrelia infections (e.g., Hoog Baarlo, Haaksbergen, Wassenaar). A high proportion of the rodent hosts in Kwade Hoek was infected with Borrelia spp. (F. Gassner, unpublished data), which may explain why on some occasions more than 50% of nymphal ticks collected at this site were infected. On several sampling occasions, B. burgdorferi s.l. was not found in ticks. This could have been due to the fact that the number of ticks sampled may have been too low to find infections. Also, annual fluctuations in Borrelia infection rates of ticks are common (Mejlon and Jaenson 1993, Eisen et al. 2004, Wielinga et al. 2006) and may explain the absence of B. burgdorferi s.l. from these sites.

Four genospecies of B. burgdorferi s.l. were present, with B. afzelii being the most abundant. This suggests that I. ricinus in The Netherlands frequents small rodents, which are the most important reservoir host of B. afzelii (Hanincova et al. 2003, Piesman and Gern 2004, Richter et al. 2004). In Kwade Hoek, a relatively high frequency of B. valaisiana was found, which suggests that, in this area, birds may be an important host for I. ricinus next to the available rodents (Hanincova et al. 2003, Taragel'ova et al. 2008). At several other sites, the Borrelia species diversity was different from the national trend (Table 3). The reasons for these differences in species diversity may be related with the fact that several different mammalian and avian host species are involved in the Borrelia life cycle (Kurtenbach et al. 1998, Gern, 2008). The 12 Borrelia isolates that could not be further identified to genospecies level may possibly comprise European Borrelia species that could not be detected by the set of probes used in our reverse line blot, such as B. spielmanii and B. bavariensis. Coinfections of Borrelia spp. in I. ricinus were infrequent and associated with study sites that had the highest Borrelia infection rates.

The present study supports the reported high incidence of early symptoms of Lyme borreliosis in The Netherlands (Hofhuis et al. 2006) by the fact that relatively high abundances of Borrelia-infected ticks were found in many study sites. The 24 sites investigated in this study are representative of many natural biotopes in the country and are all accessible to the general public. With Borrelia infection rates in ticks varying from 3% to 60%, ticks that attach to the body are potential sources of Borrelia infections. We could not identify a singular factor that serves to predict high-risk areas for Borrelia transmission, although tick abundance was in many cases determined by habitat type, especially thickness of the leaf litter and moss cover, and infection rates were associated with tree cover. The results from this study suggest that more accurate predictions of tick abundance and the associated risk of Lyme borreliosis, including the level of human exposure to infected ticks, are needed to improve the quality of public health risk monitoring programs. In particular, the factors that cause strong annual variation in Borrelia infection rate of I. ricinus deserve specific attention.

Footnotes

Acknowledgments

The authors are very grateful to the volunteers who have dedicated much time and effort to monthly collections of ticks. Aldo Bergsma is thanked for making available . The authors thank Peter Wielinga for his advice on Borrelia identifications. Part of this research was funded by the program “Ruimte voor Geoinformatie” and by Wageningen University and Research Centre. Marcel Dicke and Françoise Kaminker are thanked for their comments on a later version of the manuscript.

Disclosure Statement

No competing financial interests exist.

References

Alekseev

, Dubinina

, van de Pol

, Schouls

. Identification of Ehrlichia spp. and Borrelia burgdorferi in Ixodes ticks in the Baltic regions of Russia. J Clin Microbiol, 2001; 39:2237–2242.

Braun-Blanquet

. Pflanzensoziologie: Grundzuege der Vegetationskunde. Berlin: Springer Verlag, 1964.

Christova

, van de Pol

, Yazar

, Velo

. et al. Identification of Borrelia burgdorferi sensu lato, Anaplasma and Ehrlichia species, and spotted fever group Rickettsiae in ticks from southeastern Europe. Eur J Clin Microbiol, 2003; 22:535–542.

Daniels

, Falco

, Fish

. Estimating population size and drag sampling efficiency for the blacklegged tick (Acari: Ixodidae) J Med Entomol, 2000; 37:357–363.

Dautel

, Dippela

, Kaemmer

, Werkhausen

. et al. Winter activity of Ixodes ricinus in a Berlin forest. Int J Med Microbiol, 2008; 298:50–54.

De Mik

, van Pelt

, Docters-van Leeuwen

, van der Veen

. et al. The geographical distribution of tick bites and Erythema migrans in general practice in The Netherlands. Int J Epidemiol, 1997; 26:451–457.

Diuk-Wasser

, Gatewood

, Cortinas

, Yaremych-Hamer

. et al. Spatiotemporal patterns of host-seeking Ixodes scapularis nymphs (Acari: Ixodidae) in the United States. J Med Entomol, 2006; 43:166–176.

Eisen

, Eisen

, Chang

, Mun

. et al. Acarologic risk of exposure to Borrelia burgdorferi spirochaetes: long-term evaluations in north-western California, with implications for Lyme borreliosis risk-assessment models. Med Vet Entomol, 2004; 18:38–49.

Eisen

, Eisen

, Lane

. Geographical distribution patterns and habitat suitability models for presence of host-seeking ixodid ticks in dense woodlands of Mendocino County, California. J Med Entomol, 2006; 43:415–427.

10.

Estrada-Peña

, Venzal

, Sanchez Acedo

. The tick Ixodes ricinus: distribution and climate preferences in the western Palaearctic. Med Vet Entomol, 2006; 20:189–197.

11.

Estrada-Peña

. Distribution, abundance, and habitat preferences of Ixodes ricinus (Acari: Ixodidae) in northern Spain. J Med Entomol, 2001; 38:361–370.

12.

Gassner

, Verbaarschot

, Smallegange

, Spitzen

. et al. Variations in Ixodes ricinus density and Borrelia infections associated with cattle introduced into a woodland in The Netherlands. Appl Environ Microbiol, 2008; 74:7138–7144.

13.

Gern

. Borrelia burgdorferi sensu lato, the agent of Lyme borreliosis: life in the wilds. Parasite, 2008; 15:244–247.

14.

Gray

, Dautel

, Estrada-Peña

, Kahl

. et al. Effects of climate change on ticks and tick-borne diseases in Europe. Interdiscip Perspect Infect Dis, 2009; 2009:1–12.

15.

Gray

, Kahl

, Robertson

, Daniel

. et al. Lyme borreliosis habitat assessment. Zentralbl Bakteriol, 1998; 287:211–228.

16.

Gray

. Review: The ecology of ticks transmitting Lyme borreliosis. Exp Appl Acarol, 1998; 22:249–258.

17.

Gray

. The development and seasonal activity of the tick Ixodes ricinus: a vector of Lyme borreliosis. Rev Med Vet Entomol, 1991; 79:323–333.

18.

Hanincová

, Schäfer

, Etti

, Sewell

. et al. Association of Borrelia afzelii with rodents in Europe. Parasitology, 2003; 126:11–20.

19.

Hanincová

, Taragelová

, Koci

, Schäfer

. et al. Association of Borrelia garinii and B. valaisiana with songbirds in Slovakia. Appl Environ Microbiol, 2003; 69:2825–2830.

20.

Hofhuis

, van der Giessen

, Borgsteede

, Wielinga

. et al. Lyme borreliosis in The Netherlands: strong increase in GP consultations and hospital admissions in past 10 years. Euro Surveill, 2006; 11:E0606222.

21.

Horne

, Fielding

. Recovery of the peregrine falcon Falco peregrinus in Cumbria, UK, 1966–99. Bird Study, 2002; 49:229–236.

22.

Hubálek

, Halouzka

. Distribution of Borrelia burgdorferi sensu lato genomic groups in Europe, a review. Eur J Epidemiol, 1997; 13:951–957.

23.

Humair

, Douet

, Cadenas

, Schouls

. et al. Molecular identification of bloodmeal source in Ixodes ricinus ticks using 12S rDNA as a genetic marker. J Med Entomol, 2007; 44:869–880.

24.

Jaenson

TGT

, Eisen

, Comstedt

, Mejlon

. et al. Risk indicators for the tick Ixodes ricinus and Borrelia burgdorferi sensu lato in Sweden. Med Vet Entomol, 2009; 23:226–237.

25.

Jagers op Akkerhuis

, Dimmers

, Kats

, van der Hoef

. et al. Observations on Lyme disease, tick abundance and the diversity of small mammals in different habitats in The Netherlands and implications for the “inflation” and “dilution”: effect approaches. Proc Exp Appl Entomol, 2003; 14:31–38.

26.

Jouda

, Perret

, Gern

. Density of questing Ixodes ricinus nymphs and adults infected by Borrelia burgdorferi sensu lato in Switzerland: spatio-temporal pattern at a regional scale. Vector Borne Zoonot Dis, 2004; 4:23–32.

27.

Killilea

, Swei

, Lane

, Briggs

. et al. Spatial dynamics of Lyme disease: a review. Ecohealth, 2008; 5:167–195.

28.

Kurtenbach

, Peacey

, Rijpkema

, Hoodless

. et al. Differential transmission of the genospecies of Borrelia burgdorferi sensu lato by game birds and small rodents in England. Appl Environ Microbiol, 1998; 64:1169–1174.

29.

Lindgren

, Gustafson

. Tick-borne encephalitis in Sweden and climate change. Lancet, 2001; 358:16–18.

30.

Maetzel

, Maier

, Kampen

. Borrelia burgdorferi infection prevalences in questing Ixodes ricinus ticks (Acari: Ixodidae) in urban and suburban Bonn, western Germany. Parasitol Res, 2005; 95:5–12.

31.

Matuschka

, Fischer

, Musgrave

, Richter

. et al. Hosts on which nymphal Ixodes ricinus most abundantly feed. Am J Trop Med Hyg, 1991; 44:100–107.

32.

Mejlon

, Jaenson

TGT

. Seasonal prevalence of Borrelia burgdorferi in Ixodes ricinus in different vegetation types in Sweden. Scand J Infect Dis, 1993; 25:449–456.

33.

Pichon

, Kahl

, Hammer

, Gray

. Pathogens and host DNA in Ixodes ricinus nymphal ticks from a German forest. Vector Borne Zoonot Dis, 2006; 6:382–387.

34.

Piesman

, Gern

. Lyme borreliosis in Europe and North America. Parasitology, 2004; 129:S191–S220.

35.

Pietzsch

, Medlock

, Jones

, Avenell

. et al. Distribution of Ixodes ricinus in the British Isles: investigation of historical records. Med Vet Entomol, 2005; 19:306–314.

36.

Randolph

, Storey

. Impact of microclimate on immature tick-rodent host interactions (Acari: Ixodidae): implications for parasite transmission. J Med Entomol, 1999; 36:741–748.

37.

Rauter

, Hartung

. Prevalence of Borrelia burgdorferi sensu lato genospecies in Ixodes ricinus ticks in Europe: a metaanalysis. Appl Environ Microbiol, 2005; 71:7203–7216.

38.

Richter

, Schlee

, Allgower

, Matuschka

. Relationships of a novel Lyme disease spirochete, Borrelia spielmani sp. nov., with its hosts in Central Europe. Appl Environ Microbiol, 2004; 70:6414–6419.

39.

Scharlemann

, Johnson

, Smith

, Macdonald

. et al. Trends in ixodid tick abundance and distribution in Great Britain. Med Vet Entomol, 2008; 22:238–247.

40.

Schouls

, van de Pol

, Rijpkema

, Schot

. Detection and identification of Ehrlichia, Borrelia burgdorferi sensu lato and Bartonella species in Dutch Ixodes ricinus ticks. J Clin Microbiol, 1999; 37:2215–2222.

41.

Skarphedinsson

, Lyholm

, Ljungberg

, Sogaard

. et al. Detection and identification of Anaplasma phagocytophilum, Borrelia burgdorferi, and Rickettsia helvetica in Danish Ixodes ricinus ticks. APMIS, 2007; 115:225–230.

42.

Smit

, Dijk

, Kruidbos

, Schouls

. et al. Populatiedynamiek en fenologie van teken in Nederland. Infectieziekten Bull, 2003; 14:167–170.

43.

Tälleklint

, Jaenson

. Infestation of mammals by Ixodes ricinus ticks (Acari: Ixodidae) in south-central Sweden. Exp Appl Acarol, 1997; 21:755–771.

44.

Tälleklint

, Jaenson

. Seasonal variations in density of questing Ixodes ricinus (Acari: Ixodidae) nymphs and prevalence of infection with B. burgdorferi s.l. in south central Sweden. J Med Entomol, 1996; 33:592–597.

45.

Taragelová

, Koci

, Hanincová

, Kurtenbach

. et al. Blackbirds and song thrushes constitute a key reservoir of Borrelia garinii, the causative agent of borreliosis in Central Europe. Appl Environ Microbiol, 2008; 74:1289–1293.

46.

van Overbeek

, Gassner

, Lombaers van der Plas

, Kastelein

. et al. Diversity of Ixodes ricinus tick-associated bacterial communities from different forests. FEMS Microbiol Ecol, 2008; 66:72–84.

47.

Vennestrom

, Egholm

, Jensen

. Occurrence of multiple infections with different Borrelia burgdorferi genospecies in Danish Ixodes ricinus nymphs. Parasitol Int, 2008; 57:32–37.

48.

Vor

, Kiffner

, Hagedorn

, Niedrig

, Rühe

. Tick burden on European roe deer (Capreolus capreolus) Exp Appl Acarol, 2010; 51:405–417.

49.

Weeda

, Schaminée

, van Duuren

. Atlas van Plantengemeenschappen in Nederland. Utrecht: KNNV Uitgeverij, 2005.

50.

Wielinga

, Gaasenbeek

, Fonville

, de Boer

. et al. Longitudinal analysis of tick densities and Borrelia, Anaplasma, and Ehrlichia infections of Ixodes ricinus ticks in different habitat areas in The Netherlands. Appl Environ Microbiol, 2006; 72:7594–7601.