Tb

JSa

b

c

d

e

ARRA

KIBVMPS

1

th

0h

Veterinary Parasitology 197 (2013) 318– 325

Contents lists available at SciVerse ScienceDirect

Veterinary Parasitology

jou rn al h om epa ge: www.elsev ier .com/ locate /vetpar

he propensity of voles and mice to transmit Borreliaurgdorferi sensu lato infection to feeding ticks

ana Radzijevskajaa, Algimantas Paulauskasa,∗, Olav Rosefa,b,aulius Petkeviciusc, Vytautas Mazeikaa, Tomas Rekasiusd,e

Vytautas Magnus University, Department of Biology, Vileikos str. 8, Kaunas LT-44404, LithuaniaTelemark University College, 3800 Bø i Telemark, NorwayVeterinary Academy of Lithuanian University of Health Sciences, Department of Infectious Diseases, Tilzes g. 18, Kaunas, LithuaniaVilnius Gediminas Technical University, Department of Mathematical Statistics, Sauletekio al. 11, Vilnius, LithuaniaVytautas Magnus University, Department of Mathematics and Statistics, Vileikos str. 8, Kaunas LT-44404, Lithuania

a r t i c l e i n f o

rticle history:eceived 10 May 2012eceived in revised form 31 May 2013ccepted 3 June 2013

eywords:xodes ricinusorrelia burgdorferi s.l.olesice

revalencepecific infectivity

a b s t r a c t

Lyme borreliosis (LB) caused by the spirochete Borrelia burgdorferi sensu lato is the mostcommon tick-borne zoonosis in the Northern Hemisphere. B. burgdorferi s.l. can infecthumans and wild and domestic animals. Ixodes ricinus is the main vector, and small rodentsare the most important mammalian reservoirs hosts of B. burgdorferi s.l. in Europe. Theprevalence of B. burgdorferi s.l. in I. ricinus ticks from captured rodents, calculated specificinfectivities, and transmission coefficients were estimated in order to investigate the roleof voles and mice in transmission of the LB causative agent. A total of 12.3% (53 out of 431)of immature I. ricinus ticks from rodents in Lithuania and 3.25% (21 out of 646) in Norwaywere infected with B. burgdorferi s.l. In Lithuania a total of 40% infested Microtus arvalis,29% of Myodes glareolus and 4.8% of Apodemus flavicollis carried infected larvae and 67% ofM. glareolus, 36% of M. arvalis but none of A. flavicollis carried infected nymphs. In Norway,2.4% of larvae and 12.1% of nymphs feeding on A. flavicollis were infected. A total of 9% ofinfested A. flavicollis carried infected larvae and 13% – infected nymphs. Borrelia afzelii wasthe single genospecies identified in ticks feeding on rodents in Lithuania, and was predom-inant in ticks collected from rodents in Norway. According to calculated indices of specificinfectivity and tick-to host transmission coefficient, M. arvalis and M. glareolus voles werefound to be more efficient in transmitting B. burgdorferi s.l. to ticks than A. flavicollis mice.

GLMM analysis showed that rodent species significantly influenced the probability of alarva being infected with B. burgdorferi s.l. The larvae feeding on M. arvalis and M. glareoluswere more likely to be infected with B. burgdorferi s.l. than those feeding on A. flavicollis.This is the first study to report the quantitative roles of voles and mice in the transmission

. to larv

of B. burgdorferi s.l

. Introduction

Vector-borne zoonotic pathogens, which are transmit-ed by arthropod vectors, can infect several vertebrateost species, including man. The hosts could be the long

∗ Corresponding author. Tel.: +370 37327905; fax: +370 37327916.E-mail address: a.paulauskas@gmf.vdu.lt (A. Paulauskas).

304-4017/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.vetpar.2013.06.008

al ticks in Lithuania and Norway.© 2013 Elsevier B.V. All rights reserved.

term reservoirs of vector-borne pathogens. Different hostspecies, however, vary in their susceptibility to infection,propensity to infect feeding vectors and consequently varyin their importance to pathogen transmission and thedynamics of infection (Brunner et al., 2008).

Lyme borreliosis (LB) caused by the spirochete Borre-lia burgdorferi sensu lato is the most common tick-bornezoonoses in the Northern Hemisphere. The B. burgdorferis.l. complex currently includes 18 named species that are

ary Para

J. Radzijevskaja et al. / Veterin

capable of infecting a wide range of hosts (Margos et al.,2009; Vollmer et al., 2011). Three species have been provento be pathogenic for humans: Borrelia afzelii, Borrelia gariniiand B. burgdorferi sensu stricto. These species appear to beresponsible for causing localized, disseminated and chronicmanifestations of Lyme borreliosis (Johnson et al., 1984;Baranton et al., 1992; Canica et al., 1993; Rauter et al., 2002).The recognizable signs of disease caused by B. burgdor-feri s.l. have been observed in domestic animals like dogs,horses and cattle (Blowey et al., 1994; Bushmick, 1994;Pejchalová et al., 2006).

According to the concept of host-specific Borrelia asso-ciation, which is explained by distinct types of innateimmunity of different groups of vertebrates, various reser-voir hosts harbor different Borrelia species (Kurtenbachet al., 2002). In Europe, the most important mam-malian reservoirs hosts of B. burgdorgferi s.l. are smallrodents, which are mainly associated with B. afzelii(Kurtenbach et al., 1998; Humair et al., 1999; Gray et al.,1999; Hanincova et al., 2003). Other important reser-voirs of B. burgdorferi s.l. are insectivores, such as shrewsand hedgehogs; lagomorphs (hares), and several birdspecies (Kurtenbach et al., 1998; Humair and Gern, 2000;Majlathova et al., 2008). Large mammals, such as cervids,cattle and horses, considered as reservoir-incompetent butimportant for pathogen transmission as they are frequentlyinfested with ticks, therefore contributing to higher tickabundance (Humair and Gern, 2000; Kurtenbach et al.,2002).

Ixodes ricinus tick, the main arthropod vector of B.burgdorferi s.l. in Europe, acquire B. burgdorferi s.l. infectionduring a blood meal on a reservoir host. Among reser-voir competent rodents the pathogen multiplies and mostticks become infected during their first meal at the larvalstage. Then larvae molt to nymphs and pass the infectionon to new vertebrate hosts. The transstadial transmissionin ticks is essential for the maintenance of the B. burgdor-feri s.l. (Humair and Gern, 2000). Because unfed larval ticksare rarely infected, it seems that transovarial transmissionof B. burgdorferi s.l. is uncommon (Piesman et al., 1986;Matuschka et al., 1992; Hubalek and Halouzka, 1998).

Detection of antibodies to Borrelia sp. or the presence ofBorrelia sp. DNA confirms that animals have been exposedto infection, but these techniques are insufficient for deter-mining the reservoir status of a host, i.e. the host’s abilityto transmit pathogen to feeding vectors. Rather, the con-cept of ‘reservoir potential’ was introduced by Mather et al.(1989) as a means of identifying the contribution of dif-ferent rodent hosts to transmission of B. burgdorferi s.l. byestimating the number of infected ticks found on poten-tial hosts. Reservoir potential depends on host infestationby ticks and “realized reservoir competence” (Dizij andKurtenbach, 1995). Reservoir competence depends on theability of host to maintain the pathogen internally andthe probability that a vector feeding on that host speciesbecomes infected. Reservoir competence for Lyme diseasecould be estimated by xenodiagnosis (the infestation of

infected hosts by uninfected ticks in the laboratory andfollow-up demonstration of the pathogen in the ticks), andby determination of the infection prevalence in ticks, espe-cially larvae, collected from free living animals (Kurtenbach

sitology 197 (2013) 318– 325 319

et al., 1998; Humair et al., 1999). The probability of ahost to transmit the infection to a feeding vector (specificinfectivity) can be calculated from individual infectivities(Mather et al., 1989; Schauber and Ostfeld, 2002; Kahlet al., 2002; Brunner et al., 2008). However, there are thelimitations of estimating reservoir competence using thismethod, as possibility of transmission of infection by co-feeding and a low rate of transovarial transmission mightcontribute to infection prevalence of larvae collected fromhosts.

The aim of this study was to define the propensity ofsmall rodents from different geographical regions (Lithua-nia and Norway) to transmit B. burgdorferi s.l. pathogens tofeeding ticks. For this purpose we investigated the naturalprevalence of B. burgdorferi s.l. in I. ricinus ticks collectedfrom individual rodent hosts, compared the results withthe infection prevalence in sampled rodents, and calcu-lated the specific infectivity and host-to-tick transmissioncoefficient.

2. Materials and methods

2.1. Sample collections and taxonomical identification

The study was conducted in Lithuania and southernNorway during summer 2005–2006. Small rodents werelive-trapped with wooden traps (16 cm × 7.5 cm × 8 cm)baited with sunflower oiled bread in different habitattypes: in deciduous and mixed forest, brushed, and eco-tonal areas. In Lithuania seven sampling sites were locatedon the continental area in central, eastern, and northern-eastern parts (Kaunas 54◦87′ N, 23◦97′ E; Zarasai 55◦45′ N,25◦51′ E; and Birzai 56◦15′ N, 25◦55′ E) and on the coastalarea of Curonian Lagoon (Silute 55◦25′ N, 21◦15′ E) inthe western part of country. In Norway, four samplingsites were located on continental area (Hinnebu 58◦35′ N,08◦28′ E), and on coastal area (Tjore 58◦19′ N, 08◦31′ E; Jom-fruland 58◦52′ N, 09◦36′ E; and Lista 58◦07′ N, 06◦40′ E) insouthern and southern-eastern parts of country, respec-tively. A total of 285 rodents belonging to 7 species weretrapped in Lithuania and 134 rodents of 4 species in Norway(Table 1).

Attached engorged and moving ticks were carefullyremoved from the rodents by using sterile tweezers. Quest-ing I. ricinus ticks were collected from the vegetation (at thesame time and place where the rodents were captured) byusing standard ‘flagging’ and ‘dragging’ methods (Hillyard,1996). All ticks were placed in 70% ethanol and stored at4 ◦C until molecular analysis.

Ticks were identified using both morphological tax-onomic keys (Filippova, 1977; Hillyard, 1996) andmolecular-based approach. A molecular approach wasused for identification of all larvae and nymphs whosemouthparts had been damaged after removal from thehost. Species-specific primers Ixri-F and Ixri-R, which

amplify a 150 bp segment of the 5.8S rRNA gene, wereused in PCR for identification of immature stages ofI. ricinus (Fukunaga et al., 2000; Radzijevskaja et al.,2005).

320 J. Radzijevskaja et al. / Veterinary Parasitology 197 (2013) 318– 325

Table 1Species and number of small rodents captured in Lithuania and Norway, and their infestation with immature stages of Ixodes ricinus ticks.

Location Rodent species Prevalence of infestation(%)a

No of rodents infested byinfected ticks (%)

Infected ticks(n)/examined (N)b

NN/n LN/n

LithuaniaKaunas A. agrarius 100 (4/4) 0 0/0 8/0

A. flavicollis 73.5 (25/34) 1 12/0 57/2M. arvalis 33.3 (4/12) 2 3/0 10/3M. glareolus 5.9 (1/17) 1 1/1 0/0M. musculus 0 (0/1)

Zarasai A. flavicollis 0 (0/1)M. arvalis 37.8 (14/37) 5 17/7 20/3M. glareolus 42.9 (3/7) 0 0/0 4/0

Silute A. agrarius 26.9 (7/26) 0 1/0 39/0A. flavicollis 40.9 (18/44) 1 9/0 121/2M. agrestis 11.1 (1/9) 0 1/0 0/0M. arvalis 0 (0/4)M. glareolus 47.8 (11/23) 7 49/15 73/20

Birzai A. agrarius 100 (1/1) 0 0/0 2/0A. flavicollis 2.4 (1/42) 0 0/0 2/0M. agrestis 0 (0/10) 0/0 1/0M. arvalis 0 (0/2)M. glareolus 10 (1/10) 0 0/0 1/0A. sylvaticus 0 (0/1)

Total 31.9 (91/285) 17 93/23 338/30

NorwayHinnebu A. sylvaticus 42.9 (6/14) 0 0/0 10/0Tjore A. sylvaticus 61.1 (11/18) 0 9/0 43/0

M. agrestis 100 (1/1) 1 1/0 4/2M. glareolus 100 (1/1) 1 4/2 10/0

Jomfruland A. flavicollis 98 (48/49) 9 24/4 467/13Lista A. flavicollis 62.7 (32/51) 0 9/0 65/0

r of infe

2

Ec(tp2atwtwFbwPup

2

b2

Total 73.9 (99/134)

a Number of infested rodents/number of captured rodents.b Number of nymphs (NN) and larvae (LN) collected on rodents/numbe

.2. DNA extraction

For DNA preparation single, whole ticks were used.xtraction of DNA from unfed or less engorged ticks wasarried out by lyses in ammonium hydroxide (NH4OH)Rijpkema et al., 1996; Stanczak et al., 1999). Ticks wereaken from the 70% ethanol solution, briefly dried, andut in 0.5 ml microcentrifuge tube. We added 40 �l of.5% ammonia solution for larvae and 80 �l for nymphsnd heated at 99 ◦C for 25 min in thermostat to releasehe DNA. After being cooled, the microcentrifuge tubesith the lysate were left open and incubated for 20 min

o evaporate the ammonia. DNA from engorged ticksas extracted using Genomic DNA purification Kit (MBI

ermentas, Lithuania) according to protocol suggestedy the manufacturer. The DNA concentration and purityere measured by spectrophotometer (Eppendorrf Bio-

hotometer, Germany). The lysates were stored at 4 ◦Cntil use as templates for PCR or at −20 ◦C for longereriods.

.3. Detection of B. burgdorferi s.l. DNA

Detection of B. burgdorferi s.l. infection was performedy using direct PCR amplification of B. burgdorferi s.l. DNA. A76 bp segment of the conserved chromosomal fla gene of

11 47/6 599/15

cted (n).

B. burgdorferi (Picken, 1992) was amplified using oligonu-cleotide primers FL6 (5′ TTC AGG GTC TCA AGC GTC TTGGAC T 3′) and FL7 (5′ GCA TTT TCA ATT TTA GCA AGT GATG-3′) (Stanczak et al., 1999; Paulauskas et al., 2008). The25 �l of the PCR mixture contained 12.5 �l of 2× PCR Mas-ter Mix (MBI Fermentas, Lithuania), 10 �M of each primer(FL6 and FL7), 6.5 �l double-distilled water and 4 �l of theprocessed tick sample. The PCR was performed in ther-mal cycler (Mastercycler personal, Eppendorf PCR system,Germany). The samples were initially denatured for 1 minat 94 ◦C. Subsequent cycles were: denaturation at 94 ◦C for20 s, annealing at 55 ◦C for 30 s, and extension at 72 ◦C for30 s. Thirty-nine cycles were performed. Final extendingwas at 72 ◦C for 3 min. A control plasmid containing thetarget sequence was used as a positive control, while ster-ilized double distilled water was used as negative controlin each PCR run. PCR amplification products were sepa-rated on 1.5% agarose gel stained with ethidium bromide,and visualized under UV light. Positive results were scoredwhen a specific product of the correct size was amplified.

2.4. Detection of B. burgdorferi s.l. genospecies

DNA samples of Borrelia positive ticks were used in fur-ther analysis in order to determine the genospecies. Primersets of GI-L/GI-R, GII-L/GII-R and GIII-L/GIII-R designed by

ary Para

(1 | SpeciIC l04 -8 16 -7

J. Radzijevskaja et al. / Veterin

Demaerschalck et al. (1995) based on differences existing inthe ospA sequences were used in multiplex PCR to amplifythe DNA of B. burgdorferi s.s., B. garinii, and B. afzelii. Reac-tion mixture of 30 �l contained 15 �l Multiplex Master PCRMix (Qiagen, USA), 20 �l of each primers (GI, GII and GIII)(Roth, Germany), and 3 �l of the processed tick sample. Thecycles consisted of initial denaturation at 95 ◦C – 15 minfor HotStart Taq Polymerase activation, and 39 cycles ofdenaturation at 95 ◦C for 30 s, annealing at 65 ◦C for 90 s,and extension at 72 ◦C for 60 s. PCR amplification productswere resolved on 2.0% agarose gel and visualized under UVlight. The specific products of 544 bp (B. burgdorferi s.s.),345 bp (B. garinii) and 189 bp (B. afzelii) were considered torepresent positive results.

2.5. Statistical analysis

For evaluation of the propensity of voles and miceto transmit B. burgdorferi s.l. pathogens to feeding tickswe calculated prevalence of infection in ticks, meanhost individual infectivity and specific infectivity (Matheret al., 1989), and host-to-larval transmission coefficient(Randolph and Craine, 1995; Randolph, 1998; Kahl et al.,2002; Hanincova et al., 2003). Individual infectivity (Ii)

was measured as proportion of infected larvae (li) to totalnumber of larvae (ln) removed from an individual rodent:Ii = li/ln. The specific infectivity (Is) of a reservoir hostspecies was calculated as the sum of individual infec-tivities divided by the number of sampled individuals ofparticular rodent species (infested with infected and non-infected larvae) (Is =

∑Ii/N). To measure the degree of

infectiousness of infected rodents, modified host-to-larvaltransmission coefficient (ˇH–T) was calculated according toformula ˇH–T =

∑Ii/ni, where ni is the number of individual

hosts that gave rise to at least 1 infected tick (according toHanincova et al., 2003).

2.6. Statistical model

Our data have hierarchical structure and describedwith 3 levels of variables: larvae collected from particu-lar rodent, which belong to particular species. To describesuch data we used Generalized Linear Mixed Models(GLMM) with two random factors: species and individuals.Individual-level factor was included as a random effect toaccount for non-independence of larvae samples from thesame rodent. Probability of larvae to be infected on differ-

Models:model.null: Borrelia ~ 1 +model: Borrelia ~ 1 + (1 |

Df AIC Bmodel.null 2 167.81 175.model 3 163.31 174.

ent rodent individual and different species of rodents wereanalyzed using Generalized Linear Mixed Models (GLMM)R 2.15.1 version using the package lme4 (Bates et al., 2012;R Core Team, 2012).

sitology 197 (2013) 318– 325 321

We tested probability of larvae to be infected with B.burgdorferi s.l. by feeding on different individual of rodentsand different species of rodents using the logit model:

logit(�kij) = ln

(�kij

1 − �kij

)= � + ˛k + ˇki. (1)

Let Ykij is binary variable, were k-indicates each ofrodent species (1, 2, 3), i − ith individual of rodents, andj − jth measurement of B. burgdorferi s.l. infected larvae oneach host i. Ykij = 0 (if larva was not infected), or Ykij = 1 (iflarva was infected).

�kij – the probability of B. burgdorferi s.l. infection inlarva from ith individual of rodent belonging to kth species.Probability is mean of Ykij (�kij = E(Ykij)). Then:

� – is the intercept for this model (1), and indicatesoverall infection rate.

Parameter ˛k indicates the effect of species-level factorfor odds of infection.

Parameter ˇki indicates the effect of ith individual(individual-level factor) from kth species for odds of infec-tion.

On propose to check the species-level effect, we com-pared our model with null model, in which species factorwas ignored (˛1 = ˛2 = ˛3 = 0 in model (1)). In proof of it,likelihood-ratio test was used.

Individual)es) + (1 | Individual)ogLik Chisq Chi Df Pr(>Chisq) 1.905 8.654 6.501 1 0.01078

R code

As p value (p = 0.01078) is <0.05, we could suggest, thatinvolvement of species factor to our model is significant.

3. Results

A total of 431 immature I. ricinus ticks (338 larvae and 93nymphs) were collected from rodents in Lithuania and 646I. ricinus (599 larvae and 47 nymphs) in Norway (Table 1).There were rodents infested with both larval and nymphalstages. A total of 313 questing I. ricinus ticks were collectedfrom the vegetation at places where the rodents were cap-tured.

3.1. Prevalence of B. burgdorferi s.l. infection in hostattached ticks

In Lithuania, 12.3% (53 of 431; 95% confidence interval(CI): 11.3–13.3%) of I. ricinus ticks collected on rodents wereinfected with B. burgdorferi s.l. and 3.25% (21 of 646; 95%CI: 1.9–4.6%) were infected in Norway (Table 1).

In Lithuania, rodents with infected ticks were trappedin three locations (Kaunas, Zarasai and Silute) representedsylvatic and ecotonal (forest-field) habitats. Infected tickswere found on M. glareolus, M. arvalis, and A. flavicollis

(Table 1). Out of 285 rodents trapped in Lithuania, 91 wereinfested with ticks (31.9%; 95% CI: 26.5–37.3%) and 17 ofthese 91 rodents were infested with infected ticks (18.6%;95% CI: 10.6–26.6%). Infected larvae were predominantly

322 J. Radzijevskaja et al. / Veterinary Parasitology 197 (2013) 318– 325

Table 2Infestation of voles and mice with Borrelia burgdorferi s.l.-infected ticks and prevalence of infection in larvae and nymphs collected from these rodents.

Rodent species Infestation with infectedticksa (%)

No of infectedticks/examined (%)

Mean number of infectedticksb

Larvae Nymphs Larvae Nymphs Larvae Nymphs

Lithuania A. flavicollis 2/41 (4.8) 0/9 (0.0) 3/180 (1.7) 0/21 (0.0) 1.5M. glareolus 5/14 (29) 4/6 (67) 20/78 (25.6) 16/50 (32) 4 4M. arvalis 4/10 (40) 4/11 (36) 6/30 (20) 7/20 (35) 1.5 1.75

Norway A. flavicollis 8/78 (9) 3/23 (13) 13/532 (2.4) 4/33 (12.1) 1.62 1.33

a Number of rodents infested by infected ticks/number of infested host.b Number of infected ticks/number of hosts infested by infected ticks.

Table 3Specific infectivities (Is) and host-to-tick transmission coefficients (ˇH–T).

Lithuania Norway

A. flavicollis M. glareolus M. arvalis A. flavicollis

(ˇH-T)

0.514

fptw(va9c(oag

af9(ciofcn1sisoNw

3

stiwB

vae. Single individual parameters are varied about zero,and only some from them markedly enhance risk ofinfection (Fig. 1).

Table 4Parameter estimates for the model (1).

Parameter Estimate Std. error z Value Pr(>|z|Chi-square)

Is (ˇH–T) Is

B. afzelii 0.008 0.085 0.184

ound on infested M. arvalis and M. glareolus, with a smallercentage found on infested A. flavicollis (Table 2). Thisrend was also observed for infected nymphs, all of whichere found solely on infested M. glareolus and M. arvalis

Table 2). The overall prevalence of B. burgdorferi s.l. in lar-ae feeding on rodents differed between voles and mice,nd was found higher in larvae from M. glareolus (25.6%;5% CI: 11.1–28.9) and M. arvalis (20%; 95% CI: 5.7–34.3)ompared with A. flavicollis mice (1.7%; 95% CI:-0.2–3.6)Table 2). The mean number of infected larvae was highestn M. glareolus (Table 2). The infection prevalence in hostttached nymphs was highest in M. arvalis, followed by M.lareolus (Table 2).

The species of rodents captured in Norway were not asbundant as in Lithuania, with A. flavicollis being the mostrequently trapped. Of the 134 rodents collected, 99 (73.9%;5% CI: 66.5–81.3%) were infested but only 11 of these11%; 95% CI: 4.8– 17.2%) carried infected ticks. Eighty per-ent (95% CI: 72.2–87.8%) of all captured A. flavicollis werenfested with ticks, with 11.3% (9 of 80) (95% CI: 4.4–18.2%)f these carrying infected ticks, all of which were sampledrom Jomfruland (Table 1). In average, infested A. flavi-ollis yielded almost equal number of infected larvae andymphs (Table 2). Out of 565 ticks feeding on A. flavicollis,7 (3%; 95% CI: 1.6–4.4%) were infected with B. burgdorferi.l. The infection prevalence in nymphs feeding on A. flav-collis was five times higher than in larvae (Table 2). Onepecimen of M. glareolus harbored 2 infected nymphs, andne of Microtus agrestis harbored 2 infected larvae (Table 1).one of the ticks feeding on Apodemus sylvaticus in Norwayas infected with B. burgdorferi s.l. (Table 1).

.2. B. burgdorferi s.l. genospecies in host attached ticks

A total of 74 ticks were found positive for B. burgdorferi.l. In Lithuania, B. afzelii was the single genospecies in 53

icks. In Norway, 15 (13 larvae and 2 nymphs) out of 17nfected ticks from A. flavicollis were infected with B. afzelii,

hile 2 nymphs were co-infected with two genospecies:. afzelii and B. garinii, and B. afzelii and B. burgdorferi s.s.,

Is (ˇH–T) Is (ˇH-T)

0.235 0.587 0.012 0.108

respectively. Two infected larvae from M. agrestis harboredB. garinii and B. afzelii, respectively, and two nymphs fromM. glareolus, harbored B. afzelii.

3.3. Specific infectivity and host-to-tick transmissioncoefficient

For rodents captured in Lithuania, the calculated val-ues of specific infectivity (Is) and host-to-tick transmissioncoefficient (ˇH–T) were found higher in voles M. glareolusand M. arvalis, than in A. flavicollis mice (Table 3).

3.4. GLMM analysis

We tested how species of rodent could influence theprobability for larvae acquire B. burgdorferi infection.GLMM analysis showed that compared with overall prob-ability of B. burgdorferi s.l. infection in larval ticks, theprobability for larvae to be infected with B. burgdor-feri s.l. when they feed on M. glareolus and M. arvalishas markedly increased (model intercept ˛1 = 1.0339493and ˛2 = 0.9378652 opposite to overall infection rate,� = −2.4175). However, feeding on A. flavicolis decreasethe probability for larva to be infected (model inter-cept ˛3 = −1.2909153) (Table 4 and Fig. 1). This isshow high species-level effect on infection rate in lar-

Intercept � −2.4175 0.7911 −3.056 0.00224M. glareolus ˛1 1.0339493M. arvalis ˛2 0.9378652A. flavicolis ˛3 −1.2909153

J. Radzijevskaja et al. / Veterinary Parasitology 197 (2013) 318– 325 323

Fig. 1. Parameters of individuals (ˇki) – A, and species of rodent (˛k) – B.

Table 5Borrelia burgdorferi s.l. infection in questing Ixodes ricinus ticks collected in rodents sampling places in Lithuania and Norway.

Location No positive ticks (%)/no ticks examined B. burgdorferi s.l. genospeciesa

Female Male Nymphs Total (%)

LithuaniaKaunas 4 (18.2)/22 13 (36.1)/36 0 (0)/12 17 (24.3)/70 B. a. (87.5); B. g. (12.5)Zarasai 3 (20)/15 0 (0)/9 3 (12.5)/24 B. a. (100)Silute 1 (2.8)/36 6 (26.1)/23 7 (11.9)/59 B. a. (100)Total 8 (11.0)/73 19 (27.9)/68 0 (0)/12 27 (17.6)/153

NorwayJomfruland 0 (0)/8 0 (0)/8 7 (9.2)/76 5 (5.4)/92 B. a. (40); B. b. s.s. (60)

4 (12.5)11 (11.2

Hinnebu 4 (9.5)/42 4 (12.5)/32

Total 4 (8)/50 4 (10)/40

a B. a. – B. afzelii, B. g. – B. garinii, B. b. s.s. – B. burgdorferi s.s.

3.5. B. burgdorferi s.l. infection in questing I. ricinus tickscollected in rodents sampling places

A total 27 out of 153 (17.6%; 95% CI: 11.6–23.6%) quest-ing I. ricinus ticks in Lithuania and 17 out of 198 (8.6%; 95%CI: 4.7% to 12.5%) ticks from Norway was infected with B.burgdorferi s.l. Both nymph and adult stages were infected.The prevalence of infection and distribution of genospeciesare presented in Table 4. The infection prevalence of B.afzelii in questing ticks in Kaunas (21.4%) was found higherthan in attached ticks (6.5%), conversely in Zarasai andSilute the infection prevalence of B. afzelii was found higherin engorged ticks (24.4% and 16.0% respectively), comparedwith questing ticks (Table 5).

4. Discussion

The results of our studies confirmed that immaturestages of I. ricinus ticks feed on a variety of wild rodentspecies, but not all of these species are competent forthe transmission of the B. burgdorferi s.l., and the relative

/32 12 (11.3)/106 B. a. (100))/98 17 (8.6)/198

contribution of each species to the overall reservoir compe-tence could be different (Östfeld and Keesing, 2000; Sinskiet al., 2006).

In Lithuania, the most frequently captured rodentspecies in woodland habitats were M. glareolus and A. flavi-collis, while M. arvalis were the most frequent in grasslands.In Norway, the trapped rodents were comprised almostentirely of A. flavicollis, which known to be the most abun-dant rodents in woodland habitats in southern Norway(Mehl, 1983). According to the previous our investiga-tion (Paulauskas et al., 2008) the higher prevalence of B.burgdorferi s.l. infection was detected in voles M. arvalis, M.agrestis and M. glareolus (53%, 22%, and 21%, respectively),compared with Apodemus mice (10.5% in Lithuania and 5.9%in Norway). As showed this study, voles accordingly carrieda higher percentage of infected ticks than mice. Infectionrates of larvae and nymphs from bank voles in Lithuania

are similar to those reported in Slovakia (Haninkova et al.,2003), but infection rates of larvae and nymphs from yellownecked mice in Lithuania and Norway are much less thanthat in Slovakia. Prevalence of infections reported in Poland

3 ary Para

(fi

gw

stitlfihfvhi(trife

cnBiMisvlrfebo

babrtob

mstg(iirMeemo

24 J. Radzijevskaja et al. / Veterin

Sinski et al., 2006) are markedly less than those reportedor voles and mice and for attached ticks from these rodentsn Lithuania.

The majority of infections in Lithuania and Norway wereenotyped as B. afzelii. In Lithuania all ticks fed on rodentsere genotyped as this species.

Combined the results on prevalence of B. burgdorferi.l. in rodents (Paulauskas et al., 2008) and in feedingicks we observed, that the ratio of infected voles carry-ng non-infected: infected larvae in Lithuania was 2, whilehe ratio of infected mice carrying non-infected: infectedarvae in Lithuania and Norway combined was 12. Thesendings suggest, that bank voles and common voles haveigher propensity to transmit B. burgdorferi s.l. infection to

eeding larvae, than yellow-necked mice. Although amongoles, only infected individuals (Paulauskas et al., 2008)arbored infected ticks, a low rate (0.6% and 1.2%) of

nfected larvae was detected on non-infected A. flavicollisPaulauskas et al., 2008) in Lithuania and Norway, respec-ively. It could be explained by possibility that non-infectedodents were occasionally infested with transovariallynfected larvae (Piesman et al., 1986), or through co-eeding transmission from infected nymphs (Randolpht al., 1996).

Different rates of B. burgdorferi s.l. infection in rodentsould depend on differences in exposure to infectedymphs or species-specific differences in susceptibility to. burgdorferi s.l. infection (Brunner et al., 2008). Observed

n our studies high infection prevalence in M. arvalis and. glareolus (Paulauskas et al., 2008) and relatively high

nfestation of these species with infected nymphs (Table 2)uggest that M. arvalis and M. glareolus have high reser-oir potential to produce infected ticks. Feeding uninfectedarvae and nymphs could acquire infection from infectedodents and additionally could be infected through co-eeding transmission from infected nymphs (Randolpht al., 1996). All these factors have increased the contri-ution of M. arvalis and M. glareolus voles in transmissionf Lyme diseases in Lithuania.

The common vole has the highest reservoir competenceased on a combination of calculated specific infectivitynd host-to-tick transmission coefficients (Table 3). Theank vole follows closely in second. This suggests that theseodents have a high probability of transmitting the parasiteo uninfected ticks feeding on these hosts, supporting a rolef the common vole and bank vole in transmission of Lymeorreliosis in Lithuania.

We have compared specific infectivities and the trans-ission coefficients of populations of voles and mice

ampled in different localities. The values of specific infec-ivity and the transmission coefficient calculated for M.lareolus (0.1845 and 0.5981, respectively) in SlovakiaHanincova et al., 2003) were almost identical to calculatedn our study (Table 3). However, the same indices for A. flav-collis were much higher in Slovakia (0.0851 and 0.3954,espectively) than in Lithuania and Norway (Table 3). Inajlathova et al. (2008) study, where the role of lizard Lac-

rta agilis in the transmission cycle of Borrelia lusitaniae wasxamined, the values of specific infectivities and the trans-ission coefficient much differed for distinct populations

f L. agilis.

sitology 197 (2013) 318– 325

Although the present study indicated that voles havea higher reservoir competence than mice, it demonstratedthat A. flavicollis is implicated in transmission of B. burgdor-feri s.l. in Lithuania and in Norway. However, it is knownthat the transmission and the persistence of tick-borneinfection, such as Lyme borreliosis is influenced by thedensities and the structure of vector (ticks) and host popu-lations, and specific community composition is also animportant for tick abundance and the prevalence of infec-tion (Östfeld and Keesing, 2000; Rosà and Pugliese, 2006;Randolph and Dobson, 2012).

5. Conclusion

This is the first study to report the B. burgdorferi s.l.prevalence in ticks removed from rodents and quantitativeroles of different species of rodents in the transmission ofB. burgdorferi s.l. to larvae ticks in Lithuania and Norway.Our findings confirmed that voles and mice vary in theirimportance to transmit B. afzelii. M. arvalis and M. glare-olus in Lithuania are more efficient reservoir hosts for B.burgdorferi s.l. than A. flavicollis. Detected differences ininfectivity of A. flavicollis in Lithuania and Norway, andin other countries showed than the reservoir competenceof this species of rodent vary between investigated geo-graphic regions and habitats.

Acknowledgement

This work was supported by the Research Council ofLithuania (Grant no. MIP-071/2011).

References

Baranton, G., Postic, D., Saint Girons, I., Boerlin, P., Piffaretti, J.C., Assous,M., Grimon, P.A.D., 1992. Delineation of Borrelia burgdorferi, Borreliagarinii sp. nov., and group VS461 associated with Lyme borreliosis. Int.J. Syst. Bacteriol. 42, 378–383.

Bates, D., Maechler, M., Bolker, B., 2012. lme4: Linear mixed-effects models using S4 classes. R package version 0.999999-0.http://CRAN.R-project.org/package=lme4

Blowey, R.W., Carter, S.D., White, A.G., Barnes, A., 1994. Borrelia burgdorferiinfections in UK cattle: a possible association with digital dermatitis.Vet. Rec. 135, 577–578.

Brunner, J.L., Lo Giudice, K., Ostfeld, R.S., 2008. Estimating reservoircompetence of Borrelia burgdorferi hosts: prevalence and infectivity,sensitivity, and specificity. J. Med. Entomol. 45, 139–147.

Bushmick, S.L., 1994. Lyme borreliosis in domestic animals. J. Spirochetal& Tick-Borne Dis. 1, 24–30.

Canica, M.M., Nato, F., Du Merle, L., Mazie, J.C., Baranton, G., Postic, D.,1993. Monoclonal antibodies for identification of Borrelia afzelii sp.nov. associated with late cutaneous manifestations of Lyme borrelio-sis. Scand. J. Infect. Dis. 25, 441–448.

Demaerschalck, I., Ben Messaoud, A., De Kesel, M., Hoyois, B., Lobet, Y.,Hoet, P., Bigaignon, G., Bollen, A., Godfroid, E., 1995. Simultaneouspresence of different Borrelia burgdorferi genospecies in biological flu-ids of Lyme disease patients. J. Clin. Microbiol. 33 (3), 602–608.

Dizij, A., Kurtenbach, K., 1995. Clethrionomys glareolus, but not Apodemusflavicollis, acquires resistance to Ixodes ricinus L., the main Europeanvector of Borrelia burgdorferi. Parasite Immunol. 17, 177–183.

Filippova, N., 1977. Ixodovye kleshi podsem. Ixodinae (Ixodid ticks of thesubfamily Ixodinae). Fauna SSSR, Nauka, Leningrad.

Fukunaga, M., Yabuki, M., Hamase, A., Oliver, J., Nakao, M., 2000. Molecu-lar phylogenetic analysis of Ixodic ticks based on the ribosomal DNA

spacer, internal transcribed spacer 2, sequences. J. Parasitol. 86 (1),38–43.

Gray, J., Kirstein, F., Robertson, J., Stein, J., Kahl, O., 1999. Borrelia burgdorferis.l. in Ixodes ricinus ticks and rodents in a recreational park in south-western Ireland. Exp. Appl. Acarol. 23, 717–729.

ary Para

J. Radzijevskaja et al. / Veterin

Hanincova, K., Schafer, S.M., Etti, S., Sewell, H.S., Taragelova, V., Ziak, D.,Labuda, M., Kurtenbach, K., 2003. Association of Borrelia afzelii withrodents in Europe. Parasitology 126, 11–20.

Hillyard, R., 1996. Ticks of North-West Europe. London.Hubalek, Z., Halouzka, J., 1998. Prevalence rates of Borrelia burgdorferi

sensu lato in host-seeking Ixodes ricinus ticks in Europe. Parasitol. Res.84, 167–172.

Humair, P.F., Rais, O., Gern, L., 1999. Transmission of Borrelia afzeliifrom Apodemus mice and Clethrionomys voles to Ixodes ricinus ticks:differential transmission pattern and overwintering maintenance.Parasitology 118, 33–42.

Humair, P.F., Gern, L., 2000. The wild hidden face of Lyme borreliosis inEurope. Microbes Infect. 2, 915–922.

Johnson, R.C., Hyde, F.W., Schmid, G.P., Brenner, D.J., 1984. Borreliaburgdorferi sp. nov.: etiologic agent of Lyme disease. Int. J. Syst. Bacte-riol. 34, 496–497.

Kahl, O., Gern, L., Eisen, L., Lane, R.S., 2002. Ecological research on Borreliaburgdorferi sensu lato: terminology and some methodological pitfalls.In: Gray, J., Kahl, O., Lane, R.S., Stanek, G. (Eds.), Lyme Borreliosis:Biology, Epidemiology and Control. CABI, New York, pp. 29–46.

Kurtenbach, K., Peacey, M., Rijpkema, S., Hoodless, A., Nuttall, P., Ran-dolph, S., 1998. Differential transmission of the genospecies of Borreliaburgdorferi sensu lato by game birds and small rodents in England.Appl. Environ. Microbiol. 64, 1169–1174.

Kurtenbach, K., De Michelis, S., Etti, S., Schäfer, S.M., Sewell, H.S., Brade, V.,Kraiczy, P., 2002. Host association of Borrelia burgdorferi sensu lato –the key role of host complement. Trends Microbiol. 10, 74–79.

Majlathova, V., Majlath, I., Hromada, M., Tryjanowski, P., Bona, M., Antczak,M., Víchova, B., Dzimko S, Mihalca, A., Pet’ko, B., 2008. The role ofthe sand lizard (Lacerta agilis) in the transmission cycle of Borreliaburgdorferi sensu lato. Int. J. Med. Microbiol. 298, 161–167.

Margos, G., Vollmer, S.A., Cornet, M., Garnier, M., Fingerle, V., Wilske, B.,Bormane, A., Vitorino, L., Pereira, M.C., Drancourt, M., Kurtenbach, K.,2009. A new Borrelia species defined by multilocus sequence analysisof housekeeping genes. Appl. Environ. Microbiol. 75 (16), 5410–5416.

Mather, T.N., Wilson, M.L., Moore, S.I., Ribeiro, J.M., Spelman, A., 1989.Comparing the relative potential of rodents as reservoirs of theLyme disease spirochete (Borrelia burgdorferi). Am. J. Epidemiol. 130,143–150.

Matuschka, F.R., Fische,r, P., Heiler, M., Blumcke, S., Spielman, A., 1992.Stage-associated risk of transmission of the Lyme disease spirocheteby European Ixodes ticks. Parasitol. Res. 78, 695–698.

Mehl, R., 1983. The distribution and host relations of Norwegian ticks(Acari, Ixodides). Fauna Norvegica Serie B. Norw. J. Entomol. 30, 46–51.

Östfeld, R.S., Keesing, F., 2000. The function of biodiversity in the ecologyof vector-borne zoonotic diseases. Can. J. Zool. 78, 2061–2078.

Paulauskas, A., Ambrasiene, D., Radzijevskaja, J., Rosef, O., Turcinaviciene,J., 2008. Diversity in prevalence and genospecies of Borrelia burgdorferi

sensu lato in Ixodes ricinus ticks and rodents in Lithuania and Norway.Int. J. Med. Microbiol. 298 (S1), 180–187.

Pejchalová, K., Zákovská, A., Fucík, K., Schánilec, P., 2006. Serologicalconfirmation of Borrelia burgdorferi infection in dogs in the CzechRepublic. Vet. Res. Commun. 30, 231–238.

sitology 197 (2013) 318– 325 325

Picken, R., 1992. Polymerase chain reaction primers and probes derivedfrom flagellin gene sequences for specific detection of the agents ofLyme diseases and North American relapsing fever. J. Clin. Microbiol.30, 99–114.

Piesman, J., Donahue, J., Mather, T., Spielman, A., 1986. Transovariallyacquired Lyme disease spirochetes (Borrelia burgdorferi) in field-collected larval Ixodes dammini (Acari: Ixodidae). J. Med. Entomol.23, 21.

R Core Team, 2012. R: A Language and Environment for Statistical Com-puting. R Foundation for Statistical Computing, Vienna, Austria, ISBN3-900051-07-0 http://www.R-project.org/

Radzijevskaja, J., Indriulyte, R., Paulauskas, A., Ambrasiene, D.,Turcinaviciene, J., 2005. Genetic polymorphism study in Ixodesricinus L. ticks populations of Lithuania using RAPD markers. ActaZool. Lith. 15, 341–348.

Randolph, S., Craine, G., 1995. General framework for comparative quan-titative studies on transmission of tick-borne diseases using Lymeborreliosis in Europe as an example. J. Med. Entomol. 32, 765–777.

Randolph, S.E., Gern, L., Nuttall, P.A., 1996. Co-feeding ticks: epidemiologi-cal significance for tick-borne pathogen transmission. Parasitol. Today12, 472–479.

Randolph, S., 1998. Ticks are not insects: consequences of contrastingvector biology for transmission potential. Parasitol. Today 14,186–192.

Randolph, S.E., Dobson, A.D.M., 2012. Pangloss revisited: a critique of thedilution effect and the biodiversity-buffers-disease paradigm. Para-sitology 139, 847–863.

Rauter, C., Oehme, R., Diterich, I., Engele, M., Hartung, T., 2002. Distributionof clinically relevant Borrelia genospecies in ticks assessed by a novel,single-run, real-time PCR. J. Clin. Microbiol. 40, 36–43.

Rijpkema, S., Golubic, D., Molkenboer, M., Verbeek-De Kruif, N.,Schellekens, J., 1996. Identification of four genomic groups of B.burgdorferi sensu lato in I. ricinus ticks collected in a Lyme borreliosisendemic region of Northern Croatia. Exp. Appl. Acarol. 20, 23–30.

Rosà, R., Pugliese, A., 2006. Effects of tick population dynamics and hostdensities on the persistence of tick-borne infections. Math. Biosci.,http://dx.doi.org/10.1016/j.mbs.2006.10.002.

Schauber, E.M., Ostfeld, R.S., 2002. Modeling the effects of reservoir com-petence decay and demographic turnover in Lyme disease ecology.Ecol. Appl. 12, 1142–1162.

Sinski, E., Pawełczyk, A., Bajer, A., Behnke, J., 2006. Abundance of wildrodents, ticks and environmental risk of Lyme borreliosis: a longitu-dinal study in an area of mazury lakes district of Poland. Ann. Agric.Environ. Med. 13, 295–300.

Stanczak, J., Racewicz, B., Kubica-Biernat, W., Kruminis-Lozowska, J.,Dabrowski, A., Markowska, M., 1999. Prevalence of Borrelia burgdor-feri sensu lato in Ixodes ricinus ticks (Acari, Ixodidae) in different Polishwoodlands. Ann. Agric. Environ. Med. 6, 127–132.

Vollmer, S., Bormane, A., Dinnis, R., Seelig, F., Dobson, D.M.A., Aanensen,D., James, M., Donaghy, M., Randolph, S., Feil, E., Kurtenbach, K., Mar-gos, G., 2011. Host migration impacts on the phylogeography of LymeBorreliosis spirochaete species in Europe. Environ. Microbiol. 13 (1),184–192.