Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 577

_____________________________ _____________________________

Host-seeking Activity of Ixodes ricinus in Relation to the Epidemiology of Lyme

Borreliosis in Sweden

BY

HANS MEJLON

ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2000

Dissertation for the Degree of Doctor of Philosophy in Entomology presented at UppsalaUniversity in 2000

AbstractMejlon, H. 2000. Host-seeking activity of Ixodes ricinus in relation to the epidemiology ofLyme borreliosis in Sweden. Acta Universitatis Upsaliensis. Comprehensive Summaries ofUppsala Dissertations from the Faculty of Science and Technology 577. 42 pp.Uppsala.ISBN 91-554-4842-9

This thesis examines seasonal, diel and vertical distribution patterns of activity of host-seeking Ixodes ricinus (L.) ticks at three localities in south-central Sweden. In addition, byexamining the prevalence of infection in ticks with Lyme borreliosis (LB) spirochetes,Borrelia burgdorferi s.l, information for estimating relative LB risk in humans and the effectof control measures directed against this tick vector is provided.

The seasonal activity pattern of I. ricinus was, in general, bimodal with peaks ofactivity in May-June and August-September. Tick densities were generally high at Torö andlow at Kungshamn-Morga. The greatest variation in tick density occurred at the sample sitelevel, which indicated a patchy distribution of ticks. The diel activity of adult I. ricinussampled at Bogesund showed a distinct nocturnal activity peak while nymphal ticksexhibited no particular diel variation. At the meadow site, there was a strong negativeassociation between activity of each tick stage and ambient air temperature, and larval ticksalso showed a nocturnal activity peak. I. ricinus of all stages were present in the vegetationup to at least 140 cm above ground level. At Torö, host-seeking larvae were found atsignificantly lower levels (below 20 cm) in the vegetation compared to nymphs and adults(50-59 and 60-79 cm, respectively). Vegetation structure is likely to be the main factorgoverning tick vertical distribution at this locality. The northern limit of the geographicaldistribution of I. ricinus in Sweden corresponds with the southern boundary of the taigazone, as well as with several other climatic or vegetational isoclines primarily associated withthe vegetation period.

The prevalence rates of Borrelia spirochetes, recorded by phase-contrastmicroscopy in host-seeking I. ricinus, were 0% in larvae, 5.8-13.1% in nymphs and 14.5-28.6% in adult ticks. The human LB risk, estimated by the number of Borrelia-infectednymphs per hectare, was greater at Torö than at Kungshamn-Morga and greater inwoodland than in open areas. The risk also possessed a bimodal seasonal pattern similar tothat of subadult host-seeking activity. Controlling the number of infected nymphs throughde-ticking of reservoir hosts seems not to be an effective control measure in Sweden due tothe ubiquitous availability of alternative reservoir hosts.

Key words: Borrelia burgdorferi, control, Ixodes ricinus, Lyme borreliosis, seasonality.

Hans Mejlon, Department of Systematic Zoology, Evolutionary Biology Centre,Norbyvägen 18D, SE-752 36 Uppsala, Sweden

© Hans Mejlon 2000

ISSN 1104-232XISBN 91-554-4842-9

Printed in Sweden by University Printers, Uppsala 2000

Preface

This thesis is based on the following papers, which will be referred to in the text by their Roman

numerals:

I. Mejlon, H.A. 1997. Diel activity of Ixodes ricinus (Acari: Ixodidae) at two locations

near Stockholm, Sweden.

- Experimental & Applied Acarology 21: 247-255.

II. Mejlon, H.A. & Jaenson, T.G.T. 1997. Questing behaviour of Ixodes ricinus (Acari:

Ixodidae).

- Experimental & Applied Acarology 21: 747-754

III. Mejlon, H.A. & Jaenson, T.G.T. 1993. Seasonal prevalence of Borrelia burgdorferi in

Ixodes ricinus (Acari: Ixodidae) in different vegetation types in Sweden.

- Scandinavian Journal of Infectious Diseases 25: 449-456.

IV. Jaenson, T.G.T, Mejlon, H.A., Tälleklint-Eisen, L, Olsén, B. and Bergström, S. Factors

affecting the abundance of Borrelia-infected Ixodes ricinus in different biotopes - an

overview.

- (Manuscript to be submitted for publication in Journal of Medical Entomology)

V. Mejlon, H.A, Jaenson, T.G.T. & Mather, T.N. 1995. Evaluation of host-targeted

applications of permethrin for control of Borrelia-infected Ixodes ricinus (Acari:

Ixodidae).

- Medical and Veterinary Entomology 9: 207-210.

Reprints of papers I-III and V were made with permission from the publishers.

Contents

Introduction................................................................................................................................ 7

Systematics and general biology of ticks ................................................................................... 7

The Ixodes ricinus complex ............................................................................................... 8

Life of I. ricinus in Sweden................................................................................................ 9

Tick-host relations and LB..................................................................................................... 10

Other pathogenic agents and diseases transmitted by I. ricinus..................................... 13

Control of tick-borne diseases ........................................................................................ 13

Aims of this thesis .................................................................................................................. 14

Materials and Methods ........................................................................................................... 15

Study areas............................................................................................................................ 15

Sampling methods.................................................................................................................. 16

Detection and identification of the study organisms .................................................................. 17

Statistical treatment ................................................................................................................ 18

Results and discussion

Temporal and spatial activity patterns of I. ricinus .................................................................. 20

Diel activity..................................................................................................................... 20

Vertical distribution ........................................................................................................ 22

Seasonal activity ............................................................................................................. 25

Prevalence of spirochetes in host-seeking ticks

Local and seasonal prevalence of spirochetes in I. ricinus ............................................. 27

Controlling I. ricinus in high-risk areas .......................................................................... 29

Conclusions ........................................................................................................................... 34

Acknowledgements .................................................................................................................. 35

References ............................................................................................................................... 36

Man förlorar bara en massa tid på att vara punktlig,

Storm P

7

INTRODUCTION

Ticks continue to earn great attention world-wide despite their relatively low number of

species. The main reason for this is related to their role as obligate blood-feeders, parasitizing

man and animals. Apart from the negative effects of blood loss, allergies and toxicoses, a vast

number of different infectious agents can be transmitted between ticks and their hosts. The

variety of these agents is unmatched even by mosquitoes. Some recently discovered diseases

of humans seem to have increased to nearly epidemic proportions, e.g, Lyme disease/ Lyme

borreliosis (LB1; Steere et al. 1977) and ticks also severely limit livestock production in many

areas of the world. Therefore, it is of great importance to identify the biological factors

responsible for the transmission and maintenance of these diseases if meaningful control

measures are to be developed.

Systematics and general biology of ticks

The vast majority of all chelicerate arthropods belong to the class Arachnida. The subdivisions

of this class include such wellknown groups as scorpions, spiders, harvestmen, mites and

pseudoscorpions. The major subclass Acari (mites), is commonly referred to as “mites and

ticks”. This is however inconsistent, because of the different levels of classification: all ticks are

mites but very few mites are ticks!

The world fauna of ticks (Ixodida) comprises ~850 described species in three families

(Sonenshine 1991). The hard ticks (Ixodidae) are the most important family, both with respect

to number of species and to veterinary and medical importance. Soft ticks (Argasidae) – which

lack the sclerotized scutal shield present in hard ticks, hence their name – have a predominantly

tropical or subtropical distribution and are also of some veterinary and medical importance.

The third family (Nutalliellidae) comprises only one single species and shares morphological

traits with both Argasidae and Ixodidae (Sonenshine 1991).

All ticks have at least four developmental stages: egg, 6-legged larva, 8-legged nymph

and adult. In hard ticks there is only a single nymphal instar whereas soft ticks and other

members of the Acari possess several (often 3-4). No sexually differentiating characters are

present until the adult stage. Soft ticks are usually associated with xeric environments, a

concealed, nidicolous behaviour and short feeding periods on multiple hosts (Sonenshine

1 throughout this thesis, LB will be used as a collective term for both expressions and makes no distinctionbetween the two

8

1991). Hard ticks, on the other hand, usually occur in more humid environments, quest freely

on the vegetation, and feed only one time in each active developmental stage on either one,

two or three different hosts. During this prolonged period of blood-feeding in hard ticks

(several days), new cuticle is synthesised as to accommodate for the great expansion of body

surface (Balashov 1972); in feeding females, the body weight can increase more than 100-fold

(Sonenshine 1991). In adult males however, the sclerotized scutal plate that covers the entire

dorsal surface restricts any major increase of surface area. The life cycle of an individual tick

is completed during a rather extended period of time, ~3 years in the case of Ixodes ricinus

(L.) in Northern Europe (Lees & Milne 1951; Gray 1991).

Ticks can be found parasitizing virtually any terrestrial vertebrate, and have been

found on a number of reptiles and amphibians. Ticks do not primarily detect their host by

vision (at least in the genus Ixodes which lacks eyes), but rather through sensillae or receptors

responding to mechanical, chemical, heat radiative or auditory stimuli (Waladde & Rice

1982). Generally, mating takes place on the host (in Argasidae and most Ixodidae) or off the

host (in most Ixodes species) prior to the female’s bloodmeal (Sonenshine 1991); I. ricinus

has been confirmed to mate on- as well as off-host (Gray 1987). Sperm transfer is

accomplished via a spermatophore (Oliver 1982). As argasid females mate and feed multiple

times, they also go through multiple gonotrophic cycles and thus lay multiple batches of fewer

eggs compared to ixodid females. The latter have a single gonotrophic cycle ending by

oviposition in ground litter in a single batch of several thousand eggs (Sonenshine 1991).

The Ixodes ricinus species complex

In temperate regions of the Northern hemisphere, the four members of the I. ricinus complex

are among the most wellknown tick species and those directly involved in the transmission of

LB (Sonenshine 1993). The complex consists of two Eurasian species with the following

geographic distribution: I. persulcatus Schulze from Eastern Europe to Japan and I. ricinus

from Western Europe to Northern Africa. From Poland, the Baltic countries and Eastern

Finland and eastwards to Kazakstan both species are present (Sonenshine 1993), but seem to

be associated with different biotopes; I. persulcatus is less sensitive to desiccation compared

to I. ricinus and is mainly found in the drier taiga forest environment while the latter species

mainly occurs in areas with denser ground vegetation. In the Nearctic, the complex is

9

represented by I. scapularis Say in Eastern America and I. pacificus Cooley & Kohls in

Western America. There is no overlap in the geographical distribution of these species.

Life of I. ricinus in Sweden

Due to the extended longevity of this tick, a considerable amount of time is spent off the host,

where the environment imposes a varying degree of threat to the survival of individual ticks.

As for several other animals and plants in Sweden, restraints caused by climate become

evident as the vegetation period gets shorter to the north. At some point this combined effect

of increasing latitude/altitude will prohibit completion of the tick’s life-cycle. This is one of

the reasons why I. ricinus is absent from the mountain regions in northern Sweden (Jaenson et

al. 1994; IV). The effect of altitude on climate and thereby on tick occurrence was confirmed

in mountain regions of central Europe. In Bohemia, I. ricinus was not able to complete its life-

cycle in areas exceeding 700 m a.s.l. (Daniel 1993).

The likewise universal problem of terrestrial arthropods in preventing dehydration also

plays an important role for the smaller scale distribution and survival of I. ricinus. Although

this species has the ability to extract water from a subsaturated atmosphere (i.e, 80-96%

relative humidity [R.H.]; Gray 1991; Knülle & Rudolph 1982), the physiological stress of

such an activity will reduce the life-span and decrease the chances of finding a host (Daniel

1993). The desiccation tolerance is generally smaller in the larval stage and greater in the

adult (Knülle & Rudolph 1982). Throughout its distributional area, I. ricinus seems to be

associated mainly with woodland areas on richer soils. Such areas will generally have a

buffering effect on microclimate with less fluctuations in temperature and R.H, thus enabling

ticks to restore their water balance within the well-developed ground vegetation where the

content of atmospheric water is usually high (Lees 1948). Additionally, the number of tick

hosts is also generally higher in such areas and, obviously, so is the number of potential tick-

host contacts.

Only part of the time spent off the host is actually used for active host-seeking

behaviour (i.e, questing; Lees & Milne 1951). The main part is allocated for other reasons,

such as restoring the water balance, behavioural and morphogenetic diapause or egg-laying (in

the case of an adult female). Actively host-seeking ticks climb the vegetation and usually

come to rest at the tips of vegetational parts (Lees 1948). After contact with a suitable host,

the tick can spend considerable time in search for an acceptable attachment site which is

usually located on the host’s extremities or head region (Nilsson 1981). After completion of

10

the feeding process (from 3-5 days in larval to 7-11 days in adult female ticks; Balashov

1972) the tick detaches and drops to the ground. There is only one blood-meal in each active

developmental stage, i.e, as a larva, nymph and adult female tick, respectively. The male does

not require blood as an adult (Balashov 1972), but spends the time in search of females.

Generally, if the blood-meal is completed in spring or summer, subadult ticks will moult and

eggs from fed females will hatch within the next one to three months whereas ticks fed in late

summer/autumn will moult the following year due to diapause behaviour (Daniel et al.

1976,1977; Gray 1982, 1991) induced by the changing photoperiod (Belozerov 1982). The

threshold temperature above which activity commences is ~5°C for I. ricinus nymphs and

adults (Gray 1984). This temperature is - by meteorological definition - coincident with the

onset of spring and the ending of autumn, respectively. Thus, in south-central Sweden, ticks

are generally active from March-April to October-November.

Tick-host relations and LB

Borrelia burgdorferi sensu lato (s.l.) is a complex of (so far) 10 different species/genomic

groups of spirochetal bacteria.Three of these, i.e, B. burgdorferi sensu stricto (s.s.), B. garinii

and B. afzelii, are implicated as causative agents of LB and each is associated with both

different clinical manifestations of the disease and with different tick-host systems (Baranton

et al. 1992; Marconi & Garon 1992; Gern & Falco 2000). The significance of the remaining

species is unknown, i.e, B. andersonii and B. bissettii in USA (Marconi et al. 1995; Postic et

al.1998),B. lusitaniaeand B.valaisiana in Europe (Le Fleche et al. 1997; Wang et al.1997),

and B. japonica, B. tanukii and B. turdae in Japan (Fukunaga et al. 1996; Kawabata et al.

1993). The distribution of B. burgdorferi s.l. is probably world-wide, but so far verified

clinical cases have only been reported from the northern hemisphere, where it is in fact the

most common vector-borne disease. In Sweden, 5,000-10,000 cases are estimated to occur

each year (Berglund et al. 1995; Tälleklint 1996b). LB is multi-systemic, and affects both

skin, heart, joints and the nervous system. The cardinal sign of the disease is erythema

migrans (EM), a gradually expanding rash (caused by migrating spirochetes) with – often –

subsequent fading in the centre. EM appears on the site of a tick bite some days or weeks after

infection (Gern & Falco 2000). In those cases where EM is absent, which is more common in

Europe compared to North America, the infection might go on unnoticed. However, for

diagnosed cases the prognosis is very good, especially if treatment with antibiotics is

commenced during the acute phase. Certain domestic animals such as dogs and horses have

11

also been shown to suffer from manifestations of LB (May et al. 1991). In these cases,

problems have mainly been joint-associated. In Sweden, infection is typically caused by

nymphal ticks which is the most abundant of the infective stages (since adult female ticks are

less abundant and also probably more easily detected and removed).

Although pronounced host-specificity is common within the genus Ixodes, all

members of the I. ricinus complex can be considered “opportunistic” and can potentially use a

broad spectrum of host species. The number of host species recorded for I. ricinus and I.

persulcatus total 317 and 241, respectively (Anderson 1991) and for I. scapularis nearly 100

(Anderson 1988) rendering these species potentially powerful vectors of disease. Regarding

LB, Borrelia spirochetes are maintained within several systems of ticks and a certain segment

of their hosts capable of acquiring, maintaining and transmitting them, i.e, reservoir hosts. The

interchange between these systems is important for understanding the epidemiology of the

disease, and apart from the human-biting members of the I. ricinus complex, several

additional tick species may be of some importance: i) One system, present in both

hemispheres, involves I. uriae White and several colony-nesting seabirds which have been

shown capable of maintaining as well as transmitting spirochetes (i.e, B. garinii; Olsén et al.

1993, 1995a). Humans do not generally come into contact with this tick, but I. uriae will

occasionally bite humans if given the opportunity. ii) A species that readily bite humans and

occurs together with I. ricinus on Islands in the Baltic Sea is Haemaphysalis punctata

Canestrini & Fanzago. Spirochetes have been detected in a small (2%) proportion of nymphs

of this species (Tälleklint 1996a), but their ability to transmit spirochetes further remains

unknown. iii) Ticks that do not bite humans, such as I. hexagonus Leach, I. canisuga

Johnston, I. trianguliceps Birula and some additional species, have been shown to harbour B.

burgdorferi s.l in Europe (Akimov & Nebogatkin 1995; Estrada-Peña et al.1995;Gorelova et

al. 1996). However, as for H. punctata, it is not clear whether these ticks also manage to

transmit spirochetes to hosts.

In Western Europe, the systems involving I. ricinus and its hosts are well studied and

presently major contributors to the overall enzootic circulation of LB spirochetes. The

reservoir hosts are found among small and medium-sized mammals and occasionally birds. In

the UK, grey squirrel (Sciurus carolinensis Gmelin) and pheasant (Phasianus colchicus L.)

seem to be the major reservoirs in some areas (Craine et al. 1997), while red squirrel (S.

vulgaris L.) is suspected to play a significant role in some areas in Switzerland (Humair &

Gern 1998). However in many studies, shrews (Sorex spp, Neomys fodiens Pennant), voles

(Clethrionomys glareolus [Schreber], Microtus agrestis [L.]), mice (Apodemus spp.), hares

12

(Lepus spp.) and possibly passerine birds (Turdus spp, Erithacus sp.) dominate the mainland

area (Gern et al. 1998; Hovmark et al. 1988; Humair et al. 1993; Jaenson & Tälleklint 1996,

1997; Kurtenbach et al. 1995; Matuschka et al. 1990; Olsén et al. 1995b; Tälleklint 1996b).

The competency of a certain reservoir host is a function of its density, tick load and infectivity

to feeding tick larvae. With reference to a study area in Sweden, density of the common shrew

(Sorex araneus L.) and bank vole (C. glareolus) seemed to be the most important factor

(Tälleklint et al. 1993). However, immune responses directed against feeding ticks or

invading spirochetes are additional factors regulating host competency. For instance, acquired

resistance against feeding ticks has been shown to develop in bank voles (Dizij & Kurtenbach

1995), thereby reducing the number of successfully engorged larvae. Many of the primary

reservoir hosts, i.e, shrews and rodents, do not pass spirochetes to its progeny and also have a

shorter life-span compared with I. ricinus. In order to maintain the infection over time, this

means that each generation of hosts have to be reinfected by nymphal ticks. Thus, spirochete-

containing nymphs themselves actually become the primary overwintering reservoirs for B.

burgdorferi s.l. (Tälleklint & Jaenson 1995).

Although small mammals are usually the main hosts for I. ricinus larvae and a small

proportion of nymphs, they do not support feeding by any adult female ticks or by large

numbers of nymphs. Instead, these ticks feed on larger host animals such as hares and roe deer

(Capreolus capreolus L.). The segment of tick hosts that can support feeding by female ticks

is defined as maintenance hosts, and will indirectly determine the production of new larvae

through blood-fed females. These larger hosts are also fed upon by larval ticks, and can by

themselves support continuous populations of I. ricinus. This was shown to occur on islands

in the Baltic Sea where hares were the only tick hosts present, but did nevertheless support

both the tick population as well as LB spirochetes (Jaenson & Tälleklint 1996). Areas where

the roe deer is the only mammal present are unlikely, but could, at least in theory support tick

populations over time. However, roe deer, and cervids in general [e.g, moose (Alces alces L.),

red deer (Cervus elaphus L.), white-tailed deer (Odocoileus virginianus [Zimmerman]), sika

(Cervus nippon Temminck) and fallow deer (Cervus dama [L.]), have all been shown

incapable of transferring LB spirochetes to feeding ticks (Gray et al. 1992, 1996; Jaenson &

Tälleklint 1992; Telford III et al. 1988). Therefore, these hosts are not regarded as reservoirs

in any tick-host system. On the contrary, in addition to the “diluting” effect on nymphal

spirochete prevalence in areas with an overrepresentation of cervids (Gray et al. 1992), even

pre-existing spirochetal infection in I. ricinus nymphs has occasionally been shown to vanish

after feeding on cervid hosts (Matuschka et al. 1993).

13

Other pathogenic agents and diseases transmitted by I. ricinus

In Sweden, the viral disease Tick-Borne Encephalitis (TBE) has a yearly incidence of about

50 to 80 human cases (Gustafson 1994). The main reservoirs for the TBE virus are probably

small mammals and the distribution seems restricted to certain areas around Lake Mälaren and

to some islands in the Baltic Sea. However, the recent changes in the general climate of

Sweden may affect this distribution (Lindgren 2000; Lindgren et al. 2000).

Besides LB, severalother diseases caused bybacterial agents are known from Sweden.

Among pathogens belonging to the spotted-fever group, Rickettsia helvetica (previously

described from Switzerland, see Rehacek 1993) have been isolated from Swedish I. ricinus.

This pathogen is present in ~20 % of nymphal and adult ticks, and could possibly be involved

in sudden cardiac failure under certain conditions (Nilsson et al. 1999a, 1999b). Another

rickettsia of the genus Ehrlichia cause granulocytic ehrlichiosis in cattle, sheep, dogs and

horses. Monocytic ehrlichiosis and human granulocytic ehrlichiosis (HGE) are human

diseases transmitted by ticks in Northern America. The HGE agent is probably also

transmitted by I. ricinus in Sweden. Another rickettsia (Coxiella burnetii) causes Q-fever in

both livestock and humans. In Sweden, this disease occurs on the Island of Gotland and

competent reservoirs are found among birds, small mammals and reptiles. In addition to tick

bites, the disease agent can also be transmitted via air-borne dust particles associated with tick

faeces or infected livestock. Tularaemia is a zoonotic disease among hares and rodents caused

by Francisella tularensis. Several routes of infection exist, usually via the bite of a blood-

sucking arthropod but also by direct contact, aerial and oral ingestion of infectious material.

The protozoan haemoparasite Babesia divergens usually infects cattle in Sweden, but

severe disease in splenectomized humans have also occurred.

Control of tick-borne diseases

Since the discovery in 1982 of I. scapularis ticks acting as vectors of human LB in Northern

America (Burgdorfer et al. 1982), research directed at controlling the transmission of

spirochetes at the tick-host interface has been intensified. Traditionally, methods aimed at

controlling the spread of disease in livestock have primarily included treating the animals with

acaricide (Amoo et al. 1993; Hardeng et al. 1992). For wild animals, control by host-targeted

applications of acaricide, e.g. permethrin-impregnated cottonballs used as rodent nesting

material (Deblinger & Rimmer 1991; Mather et al. 1988) or acaricide automatically applied to

14

deer visiting food stations (Sonenshine et al. 1996), reduced tick numbers in some cases but

were less effective in others (Daniels et al. 1991; Leprince & Lane 1996; Stafford III 1991b,

1992; V). Acaricidal, insecticidal or desiccating compounds have also been used for large

scale application in the natural habitats of vector tick populations, where several studies show

at least a short-term decrease in tick numbers (Allan & Patrician 1995; Monsen et al. 1999;

Schulze et al. 1991, 1992; Solberg et al. 1992; Stafford III 1991a). On a local scale however,

indirect control of ticks such as habitat modification (e.g. burning, mowing or grazing by

livestock; Bloemer et al. 1990; Fourie et al. 1996; Mather et al. 1993; Schulze et al. 1988)

and restriction of host access by fencing (Bloemer et al. 1990; Daniels et al. 1993; Gray et al.

1992; Stafford III 1993) reduced tick numbers considerably but did not always result in

decreased LB risk. In fact, the efficiency of such control measures are probably highly

dependent on the local tick-host system present. Potential candidates for biological control of

ticks are scarce. The most studied of ixodid parasitoids is the encyrtid wasp Ixodiphagus

hookeri (Howard), associated with several Ixodes species including I. ricinus (Mwangi et al.

1991). As a parasitoid of I. scapularis however, this wasp seems only to be associated with

extremely dense tick populations (Hu et al.1993;Mather et al. 1987; Stafford III et al. 1996).

Aims of this thesis

In order to better understand how pathogens of humans transmitted by I. ricinus in Sweden

are maintained and how such microparasites circulate (in this case LB) it is crucial to

investigate in natural environments several ecological parameters of this important tick

species, particularly:

a) What is the density range of I. ricinus ticks generally encountered in a south-central

Swedish locality, and how does the density of the tick population vary temporally and

geographically?

b) What parameters can be used to describe these variations in tick density and what factors

may be most important in limiting the northern geographical distribution of I. ricinus on a

macro-geographical scale ?

c) How prevalent is Borrelia burgdorferi s.l. in I. ricinus in habitats where this tick has

become established ? How can the risk for human LB be assessed based on such data ?

d) What are the main characteristics of high-risk areas for LB in Sweden and can host-

targeted de-ticking by acaricide reduce this risk ?

15

MATERIALS AND METHODS

Study areas

The majority of the field-work consisted of tick sampling, i.e, quantifying the number of

actively host-seeking I. ricinus present in a

natural environment. This was performed in

1988 and 1989 at the Kungshamn-Morga (KM)

nature reserve just south of Uppsala (III; Fig. 1),

generally once monthly from April/May to

September/October, but in 1989 however,

sampling was made twice monthly from March

to October. Additionally, tick samplings were

conducted at Röskär, Bogesund – a peninsula in

Lake Mälaren 10 km south of Stockholm –

during 1991 to 1993 (I; Fig. 1), and during 1995

(II), respectively. In parallel with the tick

samplings at KM in 1988 and 1989, samplings

were also made at Torö, an Island of 17 km²

situated 57 km south of Stockholm (III; Fig. 1).

In 1990 and 1991, trapping of small mammals

was also performed at this location (V). Data from tick samplings at three additional localities

situated in south-western Sweden were used in paper II (sampled in 1995) and in paper IV

(sampled in 1992-95): Änggårdsbergen (City of Gothenburg), Dagsås, situated in the central

part of the province of Halland approximately 10 km from the coast and Hallands Väderö, an

island three km outside the north-western corner of the province of Skåne.

The vegetation at the sample sites at Torö was mainly mixed coniferous/deciduous

woodland. In 1988 and 1989 (III), four rectangular sites of approximately 1500 m² each were

sampled.Two of these sites were located in woodland dominated by spruce (Piceaabies [L.]),

and with scattered birch (Betula pubescens Ehrh.), aspen (Populus tremula L.) and alder

(Alnus glutinosa [L.]) trees. The ground layer at these sites was mostly dense, and

occasionally reached considerable heights where stinging nettle (Urtica dioica L.) or

loosestrife (Lysimachia vulgaris L.) occurred. The remaining two sites were located in an area

adjacent to the sea shore and were dominated by alder, reed (Phragmites australis [Cav.]),

Fig. 1. Map of Southern Swedenshowing three main study localities(Kungshamn-Morga, Bogesund andTorö) and three additional referencelocalities (Änggårdsbergen, Dagsås andHallands Väderö) where I. ricinus tickswere sampled.

16

raspberry (Rubus idaeus L.) and stinging nettle. In 1990 and 1991 (IV,V), four circular

sampling sites of 1 ha each were sampled. All of these sites were located in the same type of

woodland area as described above and were separated with at most 500 m. In 1992 and 1993

sampling was performed in a clearing made for electrical power lines, a narrow transect

flanked by mixed spruce/deciduous woodland. The vegetation here consisted of young alder

and birch trees, and a high and dense layer of grasses and ferns (II, 'high' vegetation). At KM

(III), three biotopes constituting two sites each were sampled: 1) a marshy pasture with

scattered aspen trees and low ground vegetation due to grazing by cattle and horses 2)

blueberry-spruce forest with less developed ground vegetation and 3) mixed

deciduous/coniferous woodland along the north-eastern shore of Lake Ekoln. At the Bogesund

locality two rectangular sites of 480 m² each were sampled in 1991-1993 (I) and 1995 (II,

'low' vegetation). One approximately 40-year old pine (Pinus sylvestris L.) plantation with a

medium-dense ground layer, and an open meadow, respectively. The sampling sessions at

Dagsås, Änggårdsbergen and Hallands Väderö in 1992 to 1995 (II,IV) were all performed in

broad-leaf (mainly Fagus sylvatica L.) forest areas with sparse ground cover.

Most of the primary tick hosts in Sweden, i.e, shrews (Sorex spp.), rodents (C.

glareolus, Apodemus spp.), hares and roe deer are naturally occurring within all areas

sampled. Generally, however, quantitative data on host densities for the particular areas have

not been available, especially concerning the small and medium-sized mammals. One

exception is the Bogesund locality where the population dynamics of roe deer has been

extensively studied. This area harboured an extremely dense population of deer, in the autumn

of 1992 it reached 0.36 animals/ha (courtesy of Mr. P. Kjellander, Grimsö wildlife research

station 1999). According to hunting statistics, a similarly high density (0.3-0.4 animals/ha) of

roe deer is suspected to have occurred at Torö during the period 1988 to 1992. The roe deer

density at KM at this time was, however unsubstantiated, presumably lower.

Sampling methods

The sampling procedure generally adopted (papers I, III, IV, V) was to use blanket-dragging

to obtain the number of actively host-seeking I. ricinus per unit area, that is, those ticks that

attached to a 1.5 m² white, woollen flannel cloth dragged over the ground vegetation and

checked every 10 to 20 m. In paper II the same material was used in a dress at ordinary

walking pace to obtain a measure of the vertical distribution of host-seeking ticks. In general,

the effective area sampled at each sampling occasion ranged from 50 to 250 m², depending on

17

which study was performed and on the density – and the subsequent time-consuming counting

– of host-seeking I. ricinus larvae. Sampling was made during daylight hours, mainly between

10 a.m. and 6 p.m. (DST), but was avoided shortly after or during rainfall. In paper I however,

I. ricinus was sampled at intervals of 1 month, each time during 24 hours. Trapping of small

mammals (V) was accomplished using the rodent live trap “Ugglan special”. The traps were

baited with apple, carrot and oat cereals and set for one night on each occasion.

Fig. 2. Sampling by blanket-dragging at Bogesund, 1993.Photo: L. Tälleklint

Detection and identification of the study organisms

The species status of all ticks brought back to the laboratory was noted before dissection.

Those specimens counted but left in the field were still regarded as I. ricinus since similar tick

species are host-specific and tend only to occur in the nests or burrows of their particular host.

Ticks were put in tubes with moistened tissue paper and kept at +4°C. Dissection was

performed within six weeks of collection.

18

The initial method for detecting Borrelia spirochetes in collected ticks was by phase-

contrast microscopy at 400-500X magnification. The body content of each tick was put into a

drop of phosphate-buffered saline on a glass slide and screened for spirochetes for 10 min. By

this technique, spirochetes are readily visible and can also be quantified for each individual

tick specimen. Quantitative data of this kind was obtained for ticks collected in 1988 and 1989

(unpublished data), but not for those collected in 1990-1992. The slides, with cover slip

removed, were then air-dried, fixed in ice-cold

acetone and stored at -20°C. Subsequent examination

of a subset of the stored slides with indirect

immunofluorescence assay (using monoclonal

antibody H6831 directed at the outer surface protein

B, ospB, of B. burgdorferi) revealed a small fraction

of these slides as positive (III). B. afzelii was later

confirmed by PCR from ticks collected at Torö (IV).

The poor result of the immunofluorescence test can be

explained by the more recent discovery that the

target protein ospB is generally lacking in the

Borrelia species prevailing in Sweden (i.e, B. garinii

and B. afzelii), and is usually present only in the

nominate species B. burgdorferi s.s, which was

detected in ticks on migrating birds in Sweden (Fig. 3; Olsén et al. 1995b). However, changes

in the osp profile can occur after passage of spirochetes through tick vectors and their hosts,

so it is possible that the positive slides observed represent spirochetes which have acquired

ospB but do not belong to B. burgdorferi s.s. The actual status of the B. burgdorferi s.l.

species occurring in especially Kungshamn-Morga and Bogesund remains unknown, but is

likely to consist of B. garinii and/or B. afzelii.

Statistical treatment

When sampling host-seeking ticks, the resulting variable is a measure of ticks per unit area –

provided that sampling was performed in a standardised manner. This kind of data is the

statistical basis for all material published in this thesis. However, some problems with this

kind of measurements that will potentially violate one or more of the critical assumptions of

Fig. 3. Scanning electronmicrograph of B. burgdorferi s.s.spirochetes (strain B31; ~ 6,500X).Photo: H. Mejlon

19

general parametric methods are evident: i) Counts of individuals can only adopt even numbers

and are therefore not truly continuous variables. This will affect the (statistical) distribution,

which is generally found non-normal. ii) Due to the host dependence and restricted lateral

movement off-host, ticks cannot be considered randomly distributed in the vegetation. In fact,

clumped/over-dispersed distribution of ticks is generally the case (Petney et al. 1990). For the

above reasons, the statistical procedures generally performed here are those adapted for

distribution-free or "non-parametric" data. Even though there are ways to transform data to

meet the assumptions of parametric tests, it was found that the success of such activities was

often limited to certain subsets of the data, e.g, only to nymphal ticks or to samples collected

during a particular year. Similar problems to those discussed above also arise when analysis is

performed on frequencies, such as the Borrelia infection prevalence in nymphs. In some cases

however, the situation was remedied by transformation of original data (IV). In this context, it

might have been more appropriate to use the median as a measure of location instead of the

arithmetic mean and standard deviation (I-V) which is valid only for continuous, normally

distributed data. However, this choice would not affect the outcome of the statistical

procedure used.

The following tests were used when evaluating tick density (i.e, ticks per unit area;

Sokal & Rohlf 1981): the Kruskal-Wallis test with χ²-approximation for multiple comparisons

between groups (I-III,V) and the Mann-Whitney U-test (V) and Wilcoxon matched pairs test

(I) for pairwise comparisons. For comparisons of B. burgdorferi s.l. frequencies found in I.

ricinus nymphs the G-test was used (I,III,V), and when not applicable, the Fisher's exact test

(III). A 1-way ANOVA was also performed on arcsine-transformed mean frequencies (IV).

Correlation was performed on ranked variables, i.e, the Spearman ranked correlation

coefficient (rs) was used (I,II,IV).

The computer software generally used was the statistical package CSS (Statsoft™) and

MS Excel 5 add-ins for data analysis.

20

RESULTS AND DISCUSSION

Temporal and spatial activity patterns of I. ricinus

There are several expressions more or less related to quantity of ticks used in the separate

papers of this thesis (i.e, density, activity, abundance, availability, number of) that essentially

denote the same thing: the fraction of unfed, host-seeking ticks present in a particular area and

moment, that will also attach to the sampling device. By no means do these expressions imply

the total number of ticks present in an area, a quantity that is virtually unattainable.

Diel activity (I)

The results of the 24h sampling sessions at Bogesund in 1991-1993 primarily indicated a

nocturnal peak activity (i.e, from 23.00-03.00) for all active stages of I. ricinus at the meadow

site. However, differences between the six time periods were not significant for nymphs at

this site. At the forest site, no significant differences between time periods were detected for

any tick stage (Table 1). Generally, subadult ticks (i.e, larvae and nymphs) were more

abundant in the forest (82-124 and 3.5-5.3 per 40m², respectively) compared to the meadow

(0.6-6.1 and 2.0-4.7, respectively), while adult numbers were similar (ranging from 0.4-0.8;

pooled data). When the proportions of ticks sampled in each time interval were compared

between sites, those of larval ticks differed significantly (G-test, G=147, d.f.=5 and p<0.0001;

Table 1). This indicated a more irregular diel activity pattern in the meadow site, although the

peak activity at both sites (6.1 and 124 for meadow and forest, respectively) were recorded

during the same time interval (23.00-03.00).

For all developmental stages of ticks sampled in the meadow site, a negative

association with ambient air temperature was recorded (Spearman rank correlation, p≤0.005),

while a corresponding positive association with R.H. was established only for adult ticks

(p<0.05). In the forest site, there was no association between nymphs or adults and these

variables, whereas for larvae, the situation was reversed (i.e, a positive correlation with

temperature [p<0.05] and a negative correlation with R.H. [p=0.005], respectively). In many

cases, macroclimatic variables (e.g, temperature, R.H.) measured at the time of sampling or

obtained post-sampling from nearby meteorological stations, will often produce weak or non-

existent associations with tick activity data. However, this must not necessarily mean that

21

there is no association, but merely that estimates obtained in this way either are too crude, or

their interrelationship separated by a time lag.

Table 1. Mean no. per 40 m² of I. ricinus larvae, nymphs and adults (y) and fraction of totalsample (%) sampled in meadow and forest sites during six time intervals (DST): I=2300-0259,II=0300-0659, III=0700-1059, IV=1100-1459, V=1500-1859, VI=1900-2259; n= no. of40 m² drag-samples during 1991-93 at Bogesund, Stockholm.

Time Larvae Nymphs Adults

interval Biotope n y % y % y %

I Meadow 14 6.1 31 4.7 22 0.6 31

II Meadow 13 3.4 16 3.7 17 0.4 17

III Meadow 17 1.7 11 2.0 11 0.1 7

IV Meadow 17 0.6 4 2.3 13 0.1 7

V Meadow 17 1.1 7 2.6 14 0.1 3

VI Meadow 20 4.4 32 3.2 23 0.4 34

Sum¹ 98 * 101 ns 100 ** 99

I Forest 13 124 15 5.3 18 0.6 30

II Forest 13 118 16 3.8 13 0.3 13

III Forest 16 113 17 3.6 14 0.2 13

IV Forest 20 119 22 3.5 17 0.1 3

V Forest 16 89 13 3.8 15 0.2 13

VI Forest 21 82 16 4.4 23 0.4 27

Sum 99 ns 99 ns 100 ns 99

I Both 27 - - 0.6 31

II Both 26 - - 0.4 15

III Both 33 - - 0.2 10

IV Both 37 - - 0.1 5

V Both 33 - - 0.2 8

VI Both 41 - - 0.4 31

Sum 197 *** 100

¹ Due to rounding error, sums of fractions do not always equal 100%. Significance levels bythe Kruskal-Wallis test: ns, p>0.05; *p<0.05; **p<0.01; ***p<0.001

22

An example of the latter was shown by Gigon (1985), where a model for I. ricinus activity

that involved measures of daily R.H. minima during the week before the actual sampling

showed a strong predictive value.

The pattern of diel activity recorded for ticks collected at the meadow site is most

likely linked to the daily fluctuations of temperature, indicated by the pronounced association

with this variable. This is consistent with Belozerov (1982), who reviewed previous studies on

the daily rhythms of activity for several tick species. Here, temperature was the main leading

factors for tick diel activity, followed by R.H. and solar radiation. At Bogesund, in response to

the more adverse microclimatic conditions encountered in open areas (e.g, increased solar

radiation during the day and heat escape during the night), the probability of survival of

individual ticks may be decreased. An indication of increased mortality of larval ticks was the

fact that the catch ratio between larvae and nymphs differed markedly between the open

meadow site (close to 1:1) and the forest site (~25:1). Additionally, some previous studies

have actually shown that the mortality of I. ricinus larvae was higher in an open area

compared with a woodland site (Daniel et al. 1976, 1977). Alternative explanations include

those regarding sampling efficiency and, although less likely, host-associated differences: i)

Due to differences in the vegetation structure between meadow and forest (i.e, denser, grass-

dominated vegetation in the meadow) it is possible that a smaller proportion of the host-

seeking larvae was available for sampling. ii) For some reason the production of fed female

ticks from hosts (and hence host-seeking larvae) was lower in the meadow.

Vertical distribution (II)

In contrast to the host-seeking behaviour of I. scapularis in North America (which is usually

questing at ground level; Ginsberg & Ewing 1989a, 1989b), I. ricinus ticks are mainly found

questing on the vegetation, i.e, at some distance above ground level (Gigon 1985). Potential

questing places are ultimately restricted by vegetation structure, but the gradual decline of

atmospheric humidity with increasing height above ground level is likely to affect the

longevity of questing ticks depending on their location (Lees 1948). Because of the

differential desiccation tolerance of I. ricinus ticks, i.e, larvae being the least tolerant and

adult ticks the most tolerant (Knülle & Rudolph 1982), a similar vertical distribution of these

stages was expected. Additionally, the different "preferences" of larvae, nymphs and adults,

respectively, for hosts of different sizes, was also expected to impact tick vertical distribution.

23

Such a host-size dependent vertical distribution has previously been proposed both for I.

ricinus (Gigon 1985) and I. pacificus adults (Loye & Lane 1988).

In high vegetation, ticks of all stages were recorded from 10 to 140 cm above ground

level, while the lowest interval (0-9 cm) could not be sampled. The intervals with the greatest

mean number of ticks were 10-19 cm for larvae, 50-59 cm for nymphs and 60-79 cm for

adults, respectively (Fig. 4). Here, larvae occurred at a significantly lower level (mean

height=50 cm) compared to nymphs (59 cm) or adults (66 cm).

In low vegetation (i.e, from 0 to 80 cm), adult ticks were absent while larvae were

sampled from the interval 0-59 (n=96) and nymphs from 0-491 cm (n=5). For larvae, the

interval with the greatest mean number was 0-9 cm and mean height=18 cm.

The fact that larval I. ricinus was generally found to quest at lower heights in the vegetation

than did nymphs or adults is similar to the results of Gigon (1985). In that study on Swiss

populations of I. ricinus, both subadult stages were found closer to the ground (7-11 cm)

compared to adult ticks (10-50 cm). However in my study, the mean questing height for larvae

1 absent between 20 and 29 cm

0

5

10

15

20

25

30

35

40

45

50

0-10

10-2

0

20-3

0

30-4

0

40-5

0

50-6

0

60-7

0

70-8

0

80-9

0

90-1

00

100-

110

110-

120

120-

130

130-

140

Height above ground (cm)

Est

imat

ed p

ropo

rtio

n of

mai

n ho

sts'

su

rfac

e ar

eas

or v

eget

atio

n he

ight

s (%

)

1,0

10,0

100,0

Mea

n no

. tic

ks p

er 1

00 m

² +1

Fig. 4. Estimated proportion of main hosts’ surface area (shaded bars) and vegetation heights(open bars) in high vegetation at Torö. Lines represent mean numbers per 100 m² of questingI. ricinus larvae (dotted), nymphs (broken) and adults (solid) at the vertical range 10-140 cm.Ticks were sampled on eight occasions in 1992.

24

in high vegetation (50 cm) may seem unreasonably high, but is most likely biased due to the

exclusion of the interval 0-9 cm. Probably for the same reason the larval to nymphal ratio

(1.4:1) in high vegetation appears skewed, both compared with blanket-dragging data from

the same area (7.5:1-25:1; V, unpubl. data) as well as with vegetation-free areas (8:1-32:1). In

a study by Tälleklint & Jaenson (1994) it was estimated that the majority of host-seeking I.

ricinus larvae were fed on small, ground-dwelling mammals which is also a strong indication

of high larval densities close to the ground, i.e, within the interval 0-9 cm.

Table 2. Spearman rank correlations between tick availability, i.e. the numbers of questing I.ricinus collected, and the estimated vertical proportion in 10-cm increments of host surfacearea or vegetation heights.

Host surface area Vegetation height

Tickstage Vegetation

Numberof ticks rs p value n rs p value n

Larva High 796 0.37 ns 10 0.96 0.0001 13

Nymph High 550 0.60 ns 10 0.80 0.001 13

Adult High 35 0.47 ns 10 0.34 ns 13

Larva Low 96 -0.14 ns 8 0.92 0.01 8

Nymph Low 5 -0.17 ns 8 0.45 ns 8

Adult Low 0 No ticks observed

Ticks were sampled in the interval 0-140 cm above ground level in high and low vegetation types at twolocalities (Torö and Bogesund) near Stockholm. rs, Spearman rank correlation coefficient. n, no. of verticalincrements. ns, not significant at the 0.05 level.

The vertical distribution pattern of I. ricinus was also compared to the distribution of

vegetational apices, i.e, potential questing places, as well as to the distribution of the potential

hosts' surface areas (Fig. 4, Table 2). Estimates of the distribution of vegetational apices were

visually quantified for each 10-cm interval, henceforth referred to as vegetation height. The

surface areas of "preferred" I. ricinus hosts, i.e, the bank vole, yellow-necked field mouse (A.

flavicollis [Melchior]), wood mouse (A. sylvaticus [L.], common shrew, varying hare (L.

timidus L.), European hare (L. europaeus Pallas) and roe deer; Tälleklint & Jaenson 1994)

were calculated by combining available data on mean tick host densities per hectare with the

cross-sectional area in cm² (frontal projection) of each host species. Subsequently, the

fractions of total host area (7740 cm²/ha) for each height interval were calculated. The results

25

from ranked correlation between the vertical distribution of ticks and vegetation height

showed significant association for larvae and nymphs in high vegetation and for larvae in low

vegetation, respectively (Table 2). No association was detected between subadult ticks and

host area, or between adult ticks and any parameter. The microclimatic variables measured at

the time of sampling, i.e, ambient air temperature and R.H, showed almost no variation

between the different measurements on 0, 10, 50 and 100 cm above ground level,

respectively. Thus, the variation in these variables was not sufficient to explain any

differences in vertical distribution of ticks.

Seasonal activity (unpubl. data relating to III and V)

Studies on the seasonal occurrence of host-seeking I. ricinus ticks were conducted at three

different biotopes (i.e, open pasture, blueberry-spruce and mixed woodland, respectively) at

KM in 1988-1989, and at two different biotopes (semi-open brush/meadow and mixed

woodland, respectively) at Torö in 1988-1990. Generally, ticks of all stages were more

abundant at Torö compared to KM, as well as more abundant in woodland compared to more

exposed areas.

Generally, two types of seasonal activity patterns were distinguished in the case of

host-seeking subadults: i) the unimodal pattern characterised by a single peak of activity

usually lasting from spring/early summer to late summer/autumn, or one peak occurring either

in spring or autumn. ii) the bimodal pattern characterised by decreased activity during the

end of June to early August period thus resulting in one spring peak (April-Midsummer), and

one autumn peak (late August-September), respectively (Fig. 5). The bimodal pattern

prevailed for the subadult ticks sampled in 1988 and 1989 at both KM and Torö. In 1990

however, the numbers of larval ticks were extremely high in both sample sites at Torö. Here,

the unimodal pattern dominated activity (i.e, spring peak for area C and autumn peak for area

B, respectively), while that of nymphs was mainly bimodal. The activity of adult ticks showed

no distinct seasonality, however the numbers sampled were generally low. Some early studies

however, suggested the seasonal activity of adult I. ricinus to be of a multimodal type (Nass

1975).

26

1

10

100

1000

M A M J J A S O

No

. tic

ks p

er 1

00 m

² +

1

Pasture Spruce MixedKM

Larvae

1

10

100

1000

M A M J J A S O

No

. tic

ks p

er 1

00 m

² +

1

Nymphs

1

10

100

1000

M A M J J A S O

No

. tic

ks p

er 1

00 m

² +

1

Adults

1

10

100

1000

M A M J J A S O

Shrub MixedTorö

1

10

100

1000

M A M J J A S O

1

10

100

1000

M A M J J A S O

Fig. 5. Mean numbers per 100 m² of I. ricinus larvae, nymphs and adults in three biotopesat KM (pasture, spruce forest and mixed coniferous/deciduous woodland; left column) and intwo biotopes at Torö (shrub/marsh and mixed coniferous/deciduous woodland; right column).The ticks were sampled during March-October 1989.

27

The modality of seasonal patterns are frequently different between different years

and/or areas and in an article by Gray (1991), the modulation of these pattern of I. ricinus was

reviewed: Firstly, the longevity of ticks in open areas will be short due to higher desiccation

rates and a more rapid activation caused by high insolation. Thus, the resulting seasonal

activity peaks will tend to be more distinct as well as separated, i.e, a bimodal pattern. In

woodland areas however, the longevity of ticks will be more extended which will tend to

produce a unimodal pattern. Secondly, climatic parameters will determine the onset and

termination, respectively, of tick activity as well as tick development rates. Thirdly,

photoperiodic cues regulates the diapause behaviour of ticks and fourthly, host fauna density

and composition will affect the density of ticks. The pattern of seasonal appearance of

subadult ticks at Torö and KM is most likely related to some of the above parameters, and

especially to duration of the vegetation period (IV) and to the density of the host faunas (i.e.,

roe deer).

Prevalence of spirochetes in host-seeking ticks

Local and seasonal prevalence of spirochetes in I. ricinus (III)

Several methods for detecting B. burgdorferi s.l. spirochetes in ticks can be used. However,

all methods are not necessarily comparable to each other. For instance, PCR-based detection

is more sensitive, i.e, yields a higher proportion of infected ticks, compared with direct or

indirect microscopic methods (moderately sensitive) and cultivation in BSK medium (least

sensitive; Hubalek & Halouzka 1998).

All larval ticks examined by phase-contrast microscopy (n=421) were negative for LB

spirochetes (Table 3). Similar findings were presented by Tälleklint & Jaenson (1993), where

none out of 191 larvae dissected contained spirochetes. In comparison with a European-wide

study however, an average of 1.9 % (n=5699, range 0-11 %) of larval I. ricinus were positive

for Borrelia (Hubalek & Halouzka 1998). This would imply that at least in some cases,

transmission of spirochetes by the transovarial route does occur. However, the number of

pathogens transferred via single eggs is likely to be small. For nymphal ticks, the most

abundant of the infective stages, the mean prevalence of spirochetes was similar (~7 %) at

KM in both 1988 and 1989, while significantly different (13 and 8 %, respectively) between

these years at Torö (Table 3). For adult ticks, the mean Borrelia prevalence in males was

similar at both KM (15 %) and Torö (14 %), as well as for females collected at these localities

28

(26 and 29 %, respectively). A significant difference between sexes was only established for

adults sampled at Torö in 1988 (12 and 36 % for males and females, respectively). In fact,

there is no reason to expect a general difference between sexes in this respect, especially since

the number of previous blood-meals ingested are the same. However, in a study by Alekseev

& Dubinina (1996) adult I. persulcatus kept separately, i.e, without access to the opposite sex,

showed lower spirochete prevalences compared with male-female pairs which showed an

almost 2-fold increase. The routes of infection in this case are probably venereal from male to

female via the spermatophore, and by cannibalistic feeding from female to male. Venereal

transmission is also the most likely explanation for the elevated spirochete prevalence in I.

ricinus females in this study, especially since adult ticks were not kept separately before

dissection. The mean Borrelia frequencies reported here, both for I. ricinus nymphs (9 %) and

adults (20 %), are close to the European mean values of 10.8 and 17.4 %, respectively

(Hubalek & Halouzka 1998) although considerable variation is present due to different sample

sizes, detection methods and composition of host faunas.

Table 3. Prevalence (%) of Borrelia infection in I. ricinus from two localities in south-centralSweden. Total number of ticks examined in parentheses. The ticks were collected duringMay-September 1988 and March-October 1989 at both Torö (T) and Kungshamn-Morga(KM).

Stage of tick 1988 1989 1988+1989T KM T KM T KM

Larva 0 (147) 0 (53) 0 (189) 0 (32) 0 (336) 0 (85)

Nymph 12.9 (333) 7.1 (126) 8.5 (622) 6.8 (324) 10.1 (955) 6.9 (450)

Male 12.2 (41) 27.3 (11) 21.4 (14) 6.7 (15) 14.5 (55) 15.4 (26)

Female 36.4 (22) 16.7 (6) 15.4 (13) 29.4 (17) 28.6 (35) 26.1 (23)

Generally, there were no detectable differences in Borrelia prevalence between

biotopes sampled at either the KM or Torö locality. However, differences in this parameter

were detected within two of these biotopes, i.e, between the two sample sites in mixed forest

at Torö (for nymphs and adult males) and in the pasture (for nymphs), respectively. These

results may reflect both variation in the monthly incidence of infected ticks as well as

clumped distribution of infected ticks. However the seasonal variation, i.e, the difference in

proportions of infected ticks between months, was not significant for either KM or Torö.

Similar results were obtained from the Stockholm area and despite a tendency towards higher

29

nymphal prevalences in late spring-early summer (Tälleklint & Jaenson 1996), these

differences were not significant by G-test for either of the three years investigated.

Additionally, data from a Czech study showed no significant seasonal variation in Borrelia

prevalence for either nymphal or adult I. ricinus (Hubálek et al. 1994). A clumped distribution

of infected ticks seems a likely explanation for the pattern found at KM and Torö, and is also

indicated by the large within-biotope variations in numbers of Borrelia-infected nymphs

(Table 4). Aggregated distribution of both nymphs as well as nymphs infected with Borrelia

spirochetes was previously described for I. scapularis in North-Western America (Telford III

et al. 1992). Additionally, the importance of spatial scale in assessing LB risk and the small-

scale distribution of infected ticks was also demonstrated for I. pacificus nymphs (Tälleklint-

Eisen & Lane 1999).

Table 4. Estimated infection percentage (no. examined) and mean number of host-seeking,Borrelia-infected I. ricinus nymphs per hectare of 2 different biotopes at Torö (T) and 3different biotopes at Kungshamn-Morga (KM) during 1988-1989.

% (n) mean perhectare

% (n) mean perhectare

Locality Biotope 1988 1989

T Mixed 1 12.7 (55) 135 6.6 (211) 73.5

T Mixed 2 12.0 (92) 139 13.6 (191) 155

T Shrub 1 14.9 (47) 68 1.6 (62) 6.3

T Shrub 2 12.9 (139) 222 7.6 (158) 80.7

KM Pasture 1 16.7 (12) 12.9 37.5 (16) 29.4

KM Pasture 2 0 (17) 0 0 (14) 0

KM Spruce 1 0 (18) 0 7.5 (40) 12.2

KM Spruce 2 16.7 (12) 21.2 3.5 (57) 10.5

KM Mixed 1 6.5 (46) 32 5.9 (118) 52.8

KM Mixed 2 9.5 (21) 26 5.1 (79) 22.4

Controlling I. ricinus in high-risk areas (IV,V)

LB spirochetes have been shown to occur wherever I. ricinus ticks occur (Jaenson 1991).

Therefore, a factor obviously limiting the distribution of LB in Sweden is the geographical

distribution of this tick species. In Sweden, the northern limit of I. ricinus is close to the

Limes Norrlandicus, an important zoogeographical zone which marks the northern

30

distribution of several plant and animal populations. To the north of this zone begins the true

boreal or taiga zone. In paper IV, the northern limit of the distributional area of I. ricinus

(Jaenson et al. 1994) was compared with several climatic and vegetational isoclines on a

national scale. Data show that I. ricinus is present in areas with an annual snow cover of 150

days or less and in areas where the duration of the vegetation period exceeds 170 days,

respectively. Three additional isoclines showed rather close correspondence with the northern

distributional limit of I. ricinus: 1) the northern limit of the southern boreal zone (Fig. 6)

2) the average date of first frost = 15 September and 3) the temperature sum = 1100 degree-

days. Thus, the importance of increasing latitude/altitude is again demonstrated to be of key

importance, reflected in all of the above-mentioned parameters. However, along the east coast

of central and northern Sweden I. ricinus seem to generally extend its distribution beyond that

of nearly all isoclines mentioned above. A likely explanation is that the Baltic Sea has a

buffering effect on the climate so that the vegetation period in the area is increased.

Additionally, the higher humidity in the atmosphere around the Baltic Sea could positively

affect the survival of ticks. Tick survival has been shown to increase in the vicinity of large

water bodies.

The basis for evaluating the risk for human LB is usually the number of B. burgdorferi

s.l. infected I. ricinus nymphs present in a particular area, an index which can be extracted

from tick field-sampling data. To estimate the actual risk to contract the infection and to

develop clinical LB symptoms, however, it is necessary to acquire data on several “post-

attachment” parameters such as human immunological status, number of spirochetes

transmitted, their virulence etc., factors which cannot easily be quantified. Therefore, the

measure of LB risk used in the present papers is defined as the number of infected I. ricinus

nymphs per unit area, and high-risk areas are thus characterised by large numbers of infected

nymphs. In addition, the frequency of visits by people in all study areas were estimated and

categorised as low (+), intermediate (++) or high (+++) human utilisation (Table 5). High-risk

areas were mainly associated with deciduous or mixed deciduous/coniferous woodland along

the south-eastern coast of Sweden (Bedarö/++, Bogesund/+++, Grönö/+ and Torö/++; Table

5), while the lowest numbers of Borrelia-infected nymphs per 100 m² (i.e, less than 1.0) were

found in two of three localities investigated on the Swedish west coast (Dagsås/+ and

Änggårdsbergen/+++), in an inland area close to Uppsala (KM/+++) and in one locality along

31

Fig. 6. Comparative matching of the estimated distribution of I. ricinus in Sweden (shaded)with the northern limit of the south-boreal zone (solid line) and northern limit of the boreo-nemoral zone (hatched line), respectively. Data for vegetational zone distribution from 1961-90 (Raab & Vedin 1995).

Tab

le 5

.D

ensi

tyan

d in

fect

ion

perc

enta

ge o

fI. r

icin

us n

ymph

s an

d B

. bur

gdor

feri

s.l.

gen

omos

peci

es d

iagn

osed

at 1

2 st

udy

loca

litie

s in

Sw

eden

incl

udin

g ke

y m

amm

al s

peci

es a

nd ty

pe o

f veg

etat

ion.

Stu

dy lo

cali

ty[P

rovi

nce]

Lon

g,L

atit

ude

Veg

etat

ion/

woo

dlan

d ty

peM

amm

als

Hum

anut

ili-

sati

on

Nym

phs

(NN

)

per 1

00 m

²(%

inf.)

inf.

NN

per 1

00m

²B

.b.s

.l.ge

nosp

.***

Ref

eren

ces*

Nor

rbys

kär[

Ång

erm

anl.]

63°3

3’N

, 19°

52’E

alde

rL.

timid

usvo

les,

mic

e,sh

rew

s++

ND

(2.4

)N

DB

gB

ergs

tröm

et a

l. 19

92,J

aens

onet

al.

1989

Går

dskä

r[U

ppla

nd]

60°3

6’N

, 17°

35’E

mix

ed**

roe

deer

, moo

se, h

ares

,vo

les,

mic

e, s

hrew

s+

7.9

(6.3

)0.

5B

gT

älle

klin

t & J

aens

on 1

996a

Grö

nö [

Upp

land

]60

°30’

N, 1

7°45

’Em

ixed

**-“

-+

11.7

(28.

6)3.

4B

gT

älle

klin

t & J

aens

on 1

996a

;T

älle

klin

t et a

l. 19

96

Kun

gsha

mn-

Mor

ga[U

ppl.]

59°4

7’N

,17°

40’E

spru

cem

ixed

**-“

--“

-++ ++

+

2.2

8.3

(5.8

)(6

.5)

0.1

0.5

ND

ND

III;

Täl

lekl

int e

t al.

1998

Bog

esun

d[U

ppla

nd]

59°2

3’N

,18°

20’E

deci

duou

s/oa

km

ixed

**de

cidu

ous

-“-

-“-

-“-

+++

++ ++

19.7

20.3

13.9

(25.

0)(1

8.2)

(18.

0)

4.9

3.7

2.5

ND

ND

ND

Täl

lekl

int &

Jae

nson

199

6a,

1996

b; T

älle

klin

t et a

l. 19

98

Tor

ö[S

öder

man

land

]58

°50’

N,1

7°51

’E

deci

duou

sm

ixed

**88

-89

mix

ed**

90-9

2

roe

deer

, (m

oose

), h

ares

,vo

les,

mic

e, s

hrew

s++ ++ ++

9.3

12.6

57.3

(9.9

)(1

0.0)

(12.

2)

0.9

1.3

7.0

Ba

Ba

Ba

III

Bed

arö

n[S

öder

man

land

]58

°54’

N,1

7°58

’Ede

cidu

ous

mix

ed**

fall

ow d

eer,

(ha

res)

, vol

es,

mic

e, s

hrew

s++ ++

39.4

43.4

(8.6

)(1

3.1)

3.4

5.7

Ba

Täl

lekl

inte

t al.

1998

;Täl

lekl

int

& J

aens

on 1

996a

, 199

6b;

Got

ska

San

dön

[Got

land

]58

°23’

N,1

9°12

’Ede

cidu

ous

L. t

imid

us++

+22

.1(1

1.2)

2.5

1 B

g (i

n 1

Ir)

1 B

g (i

n 2

Lt)

Jaen

son

& T

älle

klin

t 199

6

Stor

a K

arls

ö[G

otla

nd]

57°1

7’N

,17°

57’E

Juni

peru

sbr

ush

L. t

imid

us++

+9.

4(1

4.9)

1.4

3 B

g (i

n 3

Ir)

1 B

g (i

n 1

Lt)

Jaen

son

& T

älle

klin

t 199

6

Hal

land

s V

äder

ö[S

kåne

]56

°26’

N,1

2°34

’Ede

cidu

ous

L. t

imid

us, h

edge

hog,

vole

s, m

ice,

shr

ews

+++

16.8

(13.

1)2.

21

Bg

& 1

Ba

(in

1 Ir

)B

g

Dag

sås

[Hal

land

]57

o 4’N

,12o 30

’Ebe

ech

roe

deer

, vol

es, m

ice,

shre

ws

+4.

7(5

.2)

0.2

ND

Äng

gård

sber

gen

[Väs

terg

ötl.]

57o 41

’N,1

1o 57’E

beec

hro

e de

er, v

oles

, mic

e,sh

rew

s++

+4.

2(1

6.1)

0.7

Bg

Bg

Ref

eren

ce(s

) for

a s

tudy

loca

lity

indi

cate

s th

at a

t lea

st s

ome

of th

e in

form

atio

n in

this

pap

er h

as b

een

publ

ishe

d pr

evio

usly

in th

e re

fere

nce(

s).*

*mix

ed =

mix

edco

nife

rous

and

dec

iduo

us v

eget

atio

n;**

*A f

igur

e de

note

s nu

mbe

r of

pos

itiv

e ti

ck s

peci

men

s (I

r,I.

ric

inus

) or h

ares

(Lt,

L. t

imid

us);

Bb,

B. b

urgd

orfe

ri; B

a, B

. afz

elii

;B

g,B

. gar

inii

;ND

, no

data

ava

ilabl

e

33

the east coast (Gårdskär/+). The northernmost area investigated was the Island of Norrbyskär

(in the Baltic Sea east of Umeå). This is most likely also a low-risk area as indicated by the

low spirochete prevalence in nymphs (2.4%; Table 5).

The two components of the risk index, i.e, nymphal density and spirochete prevalence

in nymphs, respectively, seemed not to be associated when analysed by ranked correlation,

neither for Swedish data alone nor when including data from several European and American

studies (IV). However as discussed earlier, the lack of association may be an effect of the non-

random distribution of (infected) ticks. The inherent variability in these parameters, i.e,

average values of spatially and temporally widely scattered data, might obscure any potential

relations. Also, by comparing several different tick-host systems in Europe and North

America will certainly add to the above variability.

At Torö, considered a high-risk area for LB, a field experiment was performed in 1990

to 1992 aimed at reducing the numbers of Borrelia-infected I. ricinus nymphs in natural

habitats (V). The approach was to distribute commercially available Damminix™ tubes filled

with permethrin-treated cottonballs in the natural habitat of small mammal reservoir hosts. If

this cotton was then used by rodents as nesting material (i.e. mice and bank voles), their

ectoparasite load and the production of Borrelia-infected nymphs from these hosts would

likely be reduced. The results showed that the tick load of bank voles was significantly

reduced in treated areas compared to control areas but no such difference was detected for

yellow-necked field mice (Table 6). No clear differences between treated and untreated areas

could be detected regarding the mean numbers of host-seeking nymphs infected with

spirochetes. A possible reason for this uncertainty is the significant increase in mean numbers

of infected I. ricinus in the untreated control areas (2.0, 6.2, 10.6 infected nymphs per 100 m²

for 1990, 1991 and 1992, respectively), while the density of infected ticks did not change

significantly in the treated areas (6.4, 3.0 and 3.9, respectively).

Treatment with host-targeted acaricide has in several studies been shown to greatly

reduce the number of host-seeking Borrelia-infected nymphs of I. scapularis in North-Eastern

America. However the main reservoir host for spirochetes in these areas is the white-footed

mouse, a single important host which will also readily use cotton as nesting material.

Additionally, the area of treatment in question is often limited to peoples' backyards and is

thus more easily saturated with acaricidal material. In Sweden, the ecological situation is more

complex. First, there are several alternative reservoir hosts for LB spirochetes, all of which are

not likely to respond in the same way to a particular treatment. Second, the vast areas

34

inhabited by ticks makes it virtually impossible to saturate the environment with, for instance,

acaricidal nesting material.

Table 6. Mean number of I. ricinus larvae per rodent and number of rodents examined (n) ofthe bank vole (C. glareolus) and the yellow-necked field mouse (A. flavicollis).

No. larvae per No. larvae per

Year Area vole n mouse n

1990 Treated 0.9 12 21.6 12

Control 16.3 6 16.5 6

1991 Treated 3.5 24 1.0 3

Control 13.0 13 3.2 4

Conclusions

The spatial and temporal activity patterns of host-seeking I. ricinus seemed to be primarily

associated with the physical environment, i.e, with vegetation structure and climatic

parameters, and the strength of these associations generally seemed greater for the larval stage

and smaller or non-existent for adult ticks. Lower numbers of larvae were generally found in

open areas compared to woodland, indicating a lower survival rate of larvae in these areas.

The northern limit of the distributional area of I. ricinus in Sweden corresponded closely with

several climatic isoclines, in particular to those relating to the vegetation period.

The LB-like spirochetes (B. afzelii and B. garinii) were detected in both nymphal and

adult I. ricinus while not in larval ticks. The frequency of infection in nymphs did not vary

seasonally but occasionally between localities and was not associated with nymphal density.

The risk for human LB was estimated as the number of host-seeking, Borrelia-infected

nymphs per unit area.

The greatest risk for human LB was estimated to occur in southern Sweden, in

deciduous or mixed deciduous/coniferous woodland preferably close to the south-east coast.

In a high-risk area at Torö where the rodent reservoir was provided with acaricide-

impregnated nesting material, no significant reduction in number of infected nymphs could be

detected. Thus, control measures aimed controlling the tick infestation rate on only a segment

35

of the numerous species of tick hosts in areas with many potential hosts are not likely to

reduce the overall risk for human LB.

ACKNOWLEDGEMENTS

My greatest thanks go to my supervisor Thomas Jaenson for his never failing support and

interest in my work, despite my frequent habit of (sometimes considerably) underestimating

the time required to complete manuscripts (or thesis). Equally important is the patience and

love from my family Rose-Marie and Anna-Maria who have endured 10 and 5 years,

respectively, of intermittent periods of irritation and mumbling conversations. My mother and

father Gun and Göte have always supported me in my projects (as parents tend to do), but

nevertheless encouraged me in my biological interest.

During this rather extended period of time required to finish this thesis I have met with

a huge number of good colleagues and friends, both at the former Department of Zoology and

later also at the Department of Population biology and – at present – the Department of

Systematic Zoology and the Museum of Evolution. If I would try to name them all I would

surely forget some important persons, so thank You all. I will however, give special thanks to

a few people that at some point have been directly involved in my work: My present and

former professors Fredrik Ronquist, Jan Ekman and Christine Dahl; Bo W Svensson for

interesting discussions and advice; my colleague Lars Tälleklint for help with all sorts of tick-

related issues and good company during conference trips; Sven Bergström and Björn Olsén at

Umeå University for ruining my mouse ears (i.e., ‘Mejlons Musöron’); Göran Sahlén for help

with manuscript revisions and all sorts of practical (jokes) stuff; Tom Mather for his great

hospitality during my short stay in America; Lars-Erik Lindell, Gunilla Olsson, Gillis

Aronsson for field and laboratory help; Marianne, Lars-Erik, Gary, Nisse, Mats, Henrik, Siv,

Gun for administrative and practical skills. My thanks also to the people at the section of

Entomology, Museum of Natural History in Stockholm for encouraging my tick studies.

This work was financed by grants from the Royal Swedish Academy of Sciences,

Stiftelsen för Zoologisk forskning at Uppsala University and by grants to Dr. Thomas G.T.

Jaenson from the Swedish Natural Science Research Council.

In the last minute of October 3, 2000 / Hans M.

36

References

Akimov I.A & Nebogatkin I.V (1995) Ixodid ticks (Acarina, Ixodidae) and Lyme disease inUkraine. Vestnik Zoologii 1, 73-75.

Alekseev A.N & Dubinina H.V (1996) Exchange of Borrelia burgdorferi between Ixodespersulcatus (Ixodidae: Acarina) sexual partners. Journal of Medical Entomology 33,351-354.

Allan S.A & Patrican L.A (1995) Reduction of Immature Ixodes scapularis (Acari: Ixodidae)in Woodlots by Application of Desiccant and Insecticidal Soap Formations. Journal ofMedical Entomology32, 16-20.

Amoo A.O.J, Dipeolu O.O, Capstick P.B, Munyinyi D.M, Gichuru L.N & Odhiambo T.R(1993) Ixodid ticks (Acari: Ixodidae) and livestock production: Effect of varyingacaricide treatments on ticks and productivity in East Coast fever-immunized weanerand dairy cattle. Journal of Medical Entomology30, 503-512.

Anderson J.F (1988) Mammalian and avian reservoirs for Borrelia burgdorferi. In: Benach,J.L. & Bosler, E.M. (eds.), Lyme disease and related disorders. Annals ofthe New York Academy of Science 539, 180-191.

Anderson J.F (1991) Epizootiology of Lyme borreliosis. Scandinavian Journal of InfectiousDiseases Suppl. 77, 23-34.

Balashov Y.S (1972) (A translation of) Bloodsucking ticks (Ixodoidea) - vectors of diseasesof Man and animals. Miscellaneous Publications of the Entomological Society ofAmerica 8, 159-376.

Baranton G, Postic D, Saint Girons I, Boerlin P, Piffaretti J.-C, Assous M & Grimont P.A.D(1992) Delineation of Borrelia burgdorferi sensu stricto, Borrelia garinii sp. nov, andgroup VS461 associated with Lyme borreliosis. International Journal of SystematicBacteriology 42, 378-383.

Belozerov V.N (1982) Diapause and biological rhythms in ticks. In Physiology of ticks. -(Current themes in tropical science), ed. Obenchain F.D & Galun R, pp. 469-500.Oxford: Pergamon Press.

Berglund J, Eitrem R, Ornstein K, Lindberg A, Ringner A, Elmrud H, Carlsson M, RunehagenA, Svanborg C & Norrby R (1995) An epidemiologic study of Lyme disease in southernSweden. New England Journal of Medicine 333, 1319-1324.

Bergström S, Olsén B, Burman N, Gothefors L, Jaenson T.G.T, Jonsson M & Mejlon H.A.(1992) Molecular characterization of Borrelia burgdorferi isolated from Ixodes ricinusin Northern Sweden. Scandinavian Journal of Infectious Diseases 24, 181-188.

Bloemer S.R, Mount G.A, Morris T.A, Zimmerman R.H, Barnard D.R & Snoddy E.L (1990)Management of lone star ticks (Acari: Ixodidae) in recreational areas with acaricideapplications, vegetative management, and exclusion of white-tailed deer. Journal ofMedical Entomology27, 543-550.

Burgdorfer W, Barbour A.G, Hayes S.F, Benach J.L, Grunwaldt, E and Davies, J.P (1982)Lyme disease – a tick-borne spirochetosis? Science 220: 321-322.

Craine N.G, Nuttall P.A, Marriott A.C & Randolph S.E (1997) Role of grey squirrels andpheasants in the transmission of Borrelia burgdorferi sensu lato, the Lyme diseasespirochaete, in the UK. Folia Parasitologica 44, 155-160.

Daniel M (1993) Influence of the microclimate on the vertical distribution of the tick Ixodes

37

ricinus (L.) in Central Europe. Acarologia 34, 105-113.

Daniel M, Cerny V, Dusbabek F, Honzakova E & Olejnicek J (1976) Influence ofmicroclimate on the life cycle of the common tick Ixodes ricinus (L.) in thermophilic oakforest. Folia Parasitologica 23, 327-342.

Daniel M, Cerny V, Dusbabek F, Honzakova E & Olejnicek J (1977) Influence ofmicroclimate on the life cycle of the common tick Ixodes ricinus (L.) in an open area incomparison with forest habitats. Folia Parasitologica 24, 149-160.

Daniels T.J, Fish D & Falco R.C (1991) Evaluation of host-targeted acaricide for reducing riskof Lyme disease in southern New York State (USA). Journal of MedicalEntomology28, 537-543.

Daniels T.J, Fish D & Schwartz I (1993) Reduced abundance of Ixodes scapularis (Acari:Ixodidae) and Lyme disease risk by deer exclusion. Journal of MedicalEntomology30, 1043-1049.

Deblinger R.D & Rimmer D.W (1991) Efficacy of a permethrin-based acaricide to reduce theabundance of Ixodes dammini (Acari: Ixodidae). Journal of Medical Entomology28,708-711.

Dizij A & Kurtenbach K (1995) Clethrionomys glareolus, but not Apodemus flavicollis,acquires resistance to Ixodes ricinus L, the main European vector of Borreliaburgdorferi. Parasite Immunology 17, 177-183.

Estrada-Peña A, Oteo J.A, Estrada- Peña E, Gortazar C, Osacar J.J, Moreno J.A & CastellaJ (1995) Borrelia burgdorferi sensu lato in ticks (Acari: Ixodidae) from two differentfoci in Spain. Experimental and Applied Acarology 19, 173-180.

Fourie L.J, Kok D.J, Krugel L, Snyman A & Van der Lingen F (1996) Control of Karooparalysis ticks through vegetation management. Medical and Veterinary Entomology10, 39-43.

Fukunaga M, Hamase A, Okada K and Nakao M. (1996) Borrelia tanukii sp. nov. andBorrelia turdae sp. nov. found from Ixodid ticks in Japan: rapid species identification by16S rRNA. Microbiology and Immunology 40: 877-881.

Gern L, Estrada-Peña A, Frandsen F, Gray J.S, Jaenson T.G.T, Jongejan F, Kahl O,Korenberg E, Mehl R. and Nuttall P.A. (1998) European reservoir hosts of Borreliaburgdorferi sensu lato. Zentralblatt für Bakteriologie 287, 196-204.

Gern L. and Falco R.C (2000) Lyme disease. Rev. sci. tech. Off. int. Epiz. 19, 121-135

Gigon F. (1985) Biologie d’Ixodes ricinus L. sur le Plateau Suisse – Une contribution àl’écologie de ce vecteur. PhD dissertation, Université de Neuchâtel, Neuchâtel 1985.

Ginsberg H.S & Ewing C.P (1989a) Comparison of flagging, walking, trapping, and collectingfrom hosts as sampling methods for northern deer ticks, Ixodes dammini, and lone-starticks, Amblyomma americanum (Acari, Ixodidae). Experimental and AppliedAcarology 7, 313-322.

Ginsberg H.S & Ewing C.P (1989b) Habitat distribution of Ixodes dammini (Acari: Ixodidae)and Lyme disease spirochetes on Fire Island, New York. Journal of MedicalEntomology26, 183-189.

Gorelova N.B, Korenberg E.I, Kovalevskii Y.V, Postic D & Baranton G (1996) Isolation ofBorrelia from the tick Ixodes trianguliceps (Ixodidae) and the significance of thisspecies in epizootiology of ixodid tick-borne borrelioses. Parazitologiya 30, 13-18.

38

Gray J.S (1982) The development and questing activity of Ixodes ricinus (L.) (Acari:Ixodidae) under field conditions in Ireland. Bulletin of Entomological Research 72,263-270.

Gray J.S (1984) Studies on the dynamics of active populations of the sheep tick, Ixodesricinus L. in Co. Wicklow, Ireland. Acarologia 25, 167-178.

Gray J.S (1987) Mating and behavioural diapause in Ixodes ricinus L. Experimental andApplied Acarology 3, 61-71.

Gray J.S (1991) The development and seasonal activity of the tick Ixodes ricinus: a vector ofLyme borreliosis. Review of Medical and Veterinary Entomology 79, 323-333.

Gray J.S, Hayden T.J, Casey S, Kirstein F, Rijpkema S & Curtin S (1996) A Lyme diseaseserosurvey of deer in Irish National Parks. Biology and Environment 96B, 27-32.

Gray J.S, Kahl O, Janetzki C & Stein J (1992) Studies on the ecology of Lyme disease in adeer forest in County Galway, Ireland. Journal of Medical Entomology29, 915-920.

Gustafson R (1994) Epidemiological studies of Lyme borreliosis and tick-borne encephalitis.Scandinavian Journal of Infectious Diseases Suppl 92, 1-63.

Hardeng F, Bølsrud K.J & Overnes G (1992) Controlling tick infestations and diseases insheep by pour-on formulations of synthetic pyrethroids: A field study. VeterinaryResearch Communications 16, 429-436.

Hovmark A, Jaenson T.G.T, Åsbrink E, Forsman A & Jansson E (1988) First isolations ofBorrelia burgdorferi from rodents collected in northern Europe. APMIS 96, 917-920.

Hu R, Hyland K.E & Mather T.N (1993) Occurrence and distribution in Rhode Island ofHunterellus hookeri (Hymenoptera: Encyrtidae), a wasp parasitoid of Ixodesdammini. Journal of Medical Entomology30, 277-280.

Hubalek Z & Halouzka J (1998) Prevalence rates of Borrelia burgdorferi sensu lato in host-seeking Ixodes ricinus ticks in Europe. Parasitology Research 84, 167-172.

Hubalek Z, Halouzka J, Juricova Z & Svobodova S (1994) Seasonal distribution of borreliaein Ixodes ricinus ticks. Zentralblatt für Bakteriologie 280, 423-431.

Humair P.F & Gern L (1998) Relationship between Borrelia burgdorferi sensu lato species,red squirrels (Sciurus vulgaris) and Ixodes ricinus in enzootic areas in Switzerland.Acta Tropica 69, 213-227.

Humair P.F, Turrian N, Aeschlimann A & Gern L (1993) Borrelia burgdorferi in a focus ofLyme borreliosis: Epizootiologic contribution of small mammals. Folia Parasitologica40, 65-70.

Jaenson T. G. T, Bergström S, Burman N, Chirico J and Jonsson M. (1989)Spiroketinfekterade fästingar (Ixodes ricinus) - risk för angrepp även i Norrland[Spirochete-infected ticks (Ixodes ricinus) - risk of infection even in Northern Sweden].Läkartidningen 86, 2584.

Jaenson T.G.T & Tälleklint L (1992) Incompetence of roe deer as reservoirs of the Lymeborreliosis spirochete. Journal of Medical Entomology29, 813-817.

Jaenson T.G.T & Tälleklint L (1996) Lyme borreliosis spirochetes in Ixodes ricinus (Acari:Ixodidae) and the varying hare on isolated Islands in the Baltic sea. Journal of MedicalEntomology33, 339-343.

Jaenson T.G.T & Tälleklint L. (1997 The reservoir hosts of Borrelia burgdorferi sensu lato in

39

Europe. In: Acarology IX (Needham, G.R, Horn, D.J, Welbourn, W.C, eds.).Columbus, Ohio.

Jaenson T.G.T (1991) The epidemiology of Lyme borreliosis. Parasitology Today 7, 39-45.

Jaenson T.G.T, Tälleklint L, Lundqvist L, Olsén B, Chirico J & Mejlon H (1994) Geographicaldistribution, host associations, and vector roles of ticks (Acari: Ixodidae, Argasidae) inSweden. Journal of Medical Entomology31, 240-256.

Kawabata H, Masuzawa T & Yanagihara Y (1993) Genomic analysis of Borrelia japonicaSp. Nov. Isolated from Ixodes ovatus in Japan. Microbiology and Immunology 37,843-848.

Knülle W & Rudolph D (1982) Humidity relationships and water balance of ticks. InPhysiology of ticks. - (Current themes in tropical science), ed. Obenchain F.D &Galun R, pp. 43-70. Oxford, Pergamon Press Ltd.

Kurtenbach K, Kampen H, Dizij A, Arndt S, Seitz H.M, Schaible U.E & Simon M.M (1995)Infestation of rodents with larval Ixodes ricinus (Acari: Ixodidae) is an important factorin the transmission cycle of Borrelia burgdorferi s.l. in German woodlands. Journal ofMedical Entomology32, 807-817.

Le Fleche A, Postic D, Girardet K, Péter O. and Baranton G. (1997) Characterization ofBorrelia lusitaniae sp. nov. by 16S ribosomal DNA sequence analysis. InternationalJournal of Systematic Bacteriology 47, 921-925.

Lees A.D & Milne A (1951) The seasonal and diurnal activities of individual sheep ticks(Ixodes ricinus L.). Parasitology 41, 189-208.

Lees A.D (1948) The sensory physiology of the sheep tick, Ixodes ricinus L. Journal ofExperimental Biology 25, 145-207.

Leprince D.J & Lane R.S (1996) Evaluation of permethrin-impregnated cotton balls aspotential nesting material to control ectoparasites of woodrats in California. Journal ofMedical Entomology33, 355-360.

Lindgren E, Tälleklint L. and Polfedt T. (2000) Impact of climatic change on the northernlatitude limit and population density of the disease-transmitting European tick Ixodesricinus. Environ. Health Perspectives 108, 119-123.

Lindgren E. (2000) The new environmental context for disease transmission, with case studieson climate change and tick-borne encephalitis. Ph.D. thesis, Stockholm University.

Loye J.E & Lane R.S (1988) Questing behavior of Ixodes pacificus (Acari: Ixodidae) inrelation to meteorological and seasonal factors. Journal of Medical Entomology25,391-398.

Marconi R.T and Garon C.F. (1982) Phylogenetic analysis of the genus Borrelia – acomparison of North American and European isolates of Borrelia burgdorferi. Journalof Bacteriology 174, 241-244.

Marconi R.T, Liveris D. and Schwartz I. (1995) Identification of novel insertion elements,restriction fragment length polymorphism patterns, and discontinuous 23S rRNA in Lymedisease spirochetes: phylogenetic analyses of rRNA genes and their intergenic spaces inBorrelia japonica sp. nov. and genomic group 21038 (Borrelia andersonii sp. nov.)isolates. Journal of Clinical Microbiology 33, 2427-2434.

Mather T.N, Duffy D.C & Campbell S.R (1993) An unexpected result from burning vegetation

40

to reduce Lyme disease transmission risks. Journal of Medical Entomology30, 642-645.

Mather T.N, Piesman J & Spielman A (1987) Absence of spirochaetes (Borreliaburgdorferi) and piroplasms (Babesia microti) in deer ticks (Ixodes dammini)parasitized by chalcid wasps (Hunterellus hookeri). Medical and VeterinaryEntomology 1, 3-8.

Mather T.N, Ribeiro J.M.C, Moore S.I & Spielman A (1988) Reducing transmission of Lymedisease spirochetes in a suburban setting. In: Benach, J.L. & Bosler, E.M. (eds.), Lymedisease and related disorders. Annals of the New York Academy of Science 539, 402-403.

Matuschka F.-R, Heiler M, Eiffert H, Fischer P, Lotter H & Spielman A (1993) Diversionaryrole of hoofed game in the transmission of Lyme disease spirochetes. American Journalof Tropical Medicine and Hygiene 48, 693-699.

Matuschka F.-R, Lange R, Spielman A, Richter D & Fischer P (1990) Subadult Ixodesricinus (Acari: Ixodidae) on rodents in Berlin, West Germany. Journal of MedicalEntomology27, 385-390.

May C, Carter S.D, Barnes A, Bell S & Bennett D (1991) Serodiagnosis of Lyme disease inUK dogs. J Small Anim Pract 32, 170-174.

Monsen S.E, Bronson L.R, Tucker J.R & Smith C.R (1999) Experimental and fieldevaluations of two acaricides for control of I. pacificus (Acari: Ixodidae) in northernCalifornia. Journal of Medical Entomology36, 660-665.

Mwangi E.N, Dipeolu O.O, Newson R.M, Kaaya G.P & Hassan S.M (1991) Predators,parasitoids and pathogens of ticks: a review. Biocontrol Science and Technology 1,147-156.

Nass W (1975) Untersuchungen zur aktivität der zecke Ixodes ricinus L. in zweinaherholungsgebeiten der stadt Halle. Hercynia 12, 325-340.

Nilsson A (1981) Spatial differentiation of ectoparasites on small mammals. HolarcticEcology 4, 184-190.

Nilsson K, Lindquist O, Liu A.J, Jaenson T.G.T, Friman G. and Påhlson C. (1999a)Rickettsia helvetica in Ixodes ricinus ticks in Sweden. Journal of ClinicalMicrobiology 37, 400-403.

Nilsson K, Lindquist O. and Påhlson C. (1999b) Rickettsia helvetica associated with chronicperimyocarditis in sudden cardiac death. Lancet 354, 1169-1173.

Oliver Jr J.H (1982) Tick reproduction: Sperm development and cytogenetics. In Physiologyof ticks. - (Current themes in tropical science), ed. Obenchain F.D & Galun R, pp.245-276. Oxford: Pergamon Press Ltd.

Olsen B, Duffy D.C, Jaenson T.G.T, Gylfe A, Bonnedahl J & Bergström S (1995a)Transhemispheric exchange of Lyme disease spirochetes by seabirds. Journal ofClinical Microbiology 33, 3270-3274.

Olsén B, Jaenson T.G.T & Bergström S (1995b) Prevalence of Borrelia burgdorferi sensulato-infected ticks on migrating birds. Applied and Environmental Microbiology 61,3082-3087.

Olsén B, Jaenson T.G.T, Noppa L, Bunikis J & Bergström S (1993) A Lyme borreliosis cyclein seabirds and Ixodes uriae ticks. Nature 362, 340-342.

41

Petney T.N, Van Ark H & Spickett A.M (1990) On sampling tick populations: the problem ofoverdispersion. Onderstepoort Journal of Veterinary Research 57, 123-127.

Postic D, Marti Ras N, Lane R.S, Hendson M. and Baranton G. (1998) Expanded diversityamong Californian Borrelia isolates and description of Borrelia bissettii sp. nov.(formerly Borrelia group 127). Journal of Clinical Microbiology 36, 3497-3504.

Rehacek J (1993) Rickettsiae and their ecology in the alpine region. Acta Virol 37, 290-301.

Schulze T.L, Parkin W.E & Bosler E.M (1988) Vector tick populations and Lyme disease: Asummary of control strategies. In: Benach, J.L. & Bosler, E.M. (eds.), Lyme disease andrelated disorders. Annals of the New York Academy of Science 539, 204-211.

Schulze T.L, Taylor G.C, Jordan R.A, Bosler E.M & Shisler J.K (1991) Effectiveness ofselected granular acaricide formulations in suppressing populations of Ixodes dammini(Acari: Ixodidae): Short-term control of nymphs and larvae. Journal of MedicalEntomology28, 624-629.

Schulze T.L, Taylor G.C, Vasvary L.M, Simmons W & Jordan R.A (1992) Effectiveness ofan aerial application of carbaryl in controlling Ixodes dammini (Acari: Ixodidae) adultsin a high-use recreational area in New Jersey. Journal of Medical Entomology29,544-547.

Sokal R.R. and Rohlf F.J. (1981) Biometry, 2 edn. W.H. Freeman and Co, New York.

Solberg V.B, Neidhardt K, Sardelis M.R, Hoffmann F.J, Stevenson R, Boobar L.R & HarlanH.J (1992) Field evaluation of two formulations of cyfluthrin for control of Ixodesdammini and Amblyomma americanum (Acari: Ixodidae). Journal of MedicalEntomology29, 634-638.

Sonenshine D.E (1991) Biology of ticks. Oxford, Oxford University Press.

Sonenshine D.E (1993) Biology of ticks. Oxford, Oxford University Press.

Sonenshine D.E, Allan S.A, Norval R.A.I & Burridge M.J (1996) A self-medicating applicatorfor control of ticks on deer. Medical and Veterinary Entomology 10, 149-154.

Stafford III K.C (1991a) Effectiveness of carbaryl applications for the control of Ixodesdammini (Acari: Ixodidae) nymphs in an endemic residential area. Journal of MedicalEntomology28, 32-36.

Stafford III K.C (1991b) Effectiveness of host-targeted permethrin in the control of Ixodesdammini (Acari: Ixodidae). Journal of Medical Entomology28, 611-617.

Stafford III K.C (1992) Third-year evaluation of host-targeted permethrin for the control ofIxodes dammini (Acari: Ixodidae) in south-eastern Connecticut. Journal of MedicalEntomology29, 717-720.

Stafford III K.C (1993) Reduced abundance of Ixodes scapularis (Acari: Ixodidae) withexclusion of deer by electric fencing. Journal of Medical Entomology30, 986-996.

Stafford K.C.I, Denicola A.J & Magnarelli L.A (1996) Presence of Ixodiphagus hookeri(Hymenoptera: Encyrtidae) in two Connecticut populations of Ixodes scapularis (Acari:Ixodidae). Journal of Medical Entomology33, 183-188.

Steere A, Malawista S.E, Snydman D.R, Sjope R.E, Andiman W.A, Ross M.R. and SteeleF.M. (1977) Lyme arthritis; an epidemic of oligoarticular arthritis in children and adults inthree Connecticut communities. - Arthritis Rheum. 20, 7-17.

42

Telford III S.R, Mather T.N, Moore S.I, Wilson M.L & Spielman A (1988) Incompetence ofdeer as reservoirs of the Lyme disease spirochete. American Journal of TropicalMedicine and Hygiene 39, 105-109.

Telford III S.R, Urioste S.S & Spielman A (1992) Clustering of host-seeking nymphal deerticks (Ixodes dammini) infected by Lyme disease spirochetes (Borrelia burgdorferi).American Journal of Tropical Medicine and Hygiene 47, 55-60.

Tälleklint L (1996a) Lyme borreliosis spirochetes in Ixodes ricinus and Haemaphysalispunctata ticks (Acari: Ixodidae) on three islands in the Baltic Sea. Experimental andApplied Acarology 20, 467-476.

Tälleklint L, Jaenson T.G.T & Mather T.N (1993) Seasonal variation in the capacity of thebank vole to infect larval ticks (Acari: Ixodidae) with the Lyme disease spirochete,Borrelia burgdorferi. Journal of Medical Entomology30, 812-815.

Tälleklint L, Jaenson T.G.T and Mejlon H.A. (1998) Transmission of the Lyme borreliosisspirochete at the mammal reservoir - tick vector interface in Sweden, pp. 91-103. In L.Coons and M. Rotschild [eds.], The Second International Conference on Tick BornePathogens at the Host-Vector Interface: A Global Perspective, August 28 - September1 1995, Kruger National Park, South Africa: proceedings and abstracts.

Tälleklint L. (1996b) Transmission of Lyme borreliosis spirochetes at the tick vector – mammalreservoir interface. PhD thesis, Uppsala University, Uppsala 1996.

Tälleklint L. and Jaenson T.G.T. (1994) Transmission of Borrelia burgdorferi s.l. frommammal reservoirs to the primary vector of Lyme borreliosis, Ixodes ricinus (Acari:Ixodidae), in Sweden. Journal of Medical Entomology 31, 880-886.

Tälleklint L. and Jaenson T.G.T. (1995) Is the small mammal (Clethrionomys glareolus) orthe tick vector (Ixodes ricinus) the primary overwintering reservoir for the Lymeborreliosis spirochete in Sweden? Journal of Wildlife Diseases 31, 537-540.

Tälleklint L. and Jaenson T.G.T. (1996a) Seasonal variations in density of questing Ixodesricinus (Acari: Ixodidae) nymphs and prevalence of infection with B. burgdorferi s.l. insouth central Sweden. Journal of Medical Entomology 33, 592-597.

Tälleklint L. and Jaenson T.G.T. (1996b) Relationship between Ixodes ricinus (Acari:Ixodidae) density and the prevalence of infection with Borrelia-like spirochetes and thedensity of infected ticks. Journal of Medical Entomology 33: 805-811.

Tälleklint-Eisen L. & Lane R.S. (1999) Variation in the density of questing Ixodes pacificus(Acari: Ixodidae) nymphs infected with Borrelia burgdorferi at different spatial scales inCalifornia. Journal of Parasitology 85, 824-831.

Waladde S.M & Rice M.J (1982) The sensory basis of tick feeding behaviour. In Physiologyof ticks. - (Current themes in tropical science), ed. Obenchain F.D & Galun R, pp.71-118. Oxford: Pergamon Press Ltd.

Wang G, van Dam A.P, Le Fleche A, Postic D, Péter O, Baranton G, de Boer R, SpanjaardL. and Dankert J. (1997) Genetic and phenotypic analysis of Borrelia valaisiana sp.nov. (Borrelia genomic groups VS116 and M19). International Journal ofSystematic Bacteriology 47, 926-932.