Prevalence of Borrelia burgdorferi Sensu Lato in Ticks Collected from Migratory Birds in Switzerland

Abstract

The prevalence of ticks infected by Borrelia burgdorferi sensu lato on birds during their migrations was studied in Switzerland. A total of 1,270 birds captured at two sites were examined for tick infestation. Ixodes ricinus was the dominant tick species. Prevalences of tick infestation were 6% and 18.2% for birds migrating northward and southward, respectively. Borrelia valaisiana was the species detected most frequently in ticks, followed by Borrelia garinii and Borrelia lusitaniae. Among birds infested by infected ticks, 23% (6/26) were infested by B. lusitaniae-infected larvae. Migratory birds appear to be reservoir hosts for B. lusitaniae.

It is now clearly established that birds play a role as reservoir hosts in the ecology of Lyme borreliosis (16). In Europe, seven Borrelia species belonging to the complex B. burgdorferi sensu lato (Borrelia burgdorferi sensu stricto, Borrelia garinii, Borrelia afzelii, Borrelia valaisiana, Borrelia lusitaniae, Borrelia bissetti, and Borrelia spielmani) have been detected in the tick vector, Ixodes ricinus. The association of at least two of them, B. garinii and B. valaisiana, with birds is now well documented in Europe (12, 17, 21, 22). Birds play an important role not only in maintaining B. burgdorferi sensu lato in areas of endemicity; in fact, some of them, through their migration, also play a role by spreading ticks within and between continents (13, 19, 23-25, 30, 34, 35). In Europe, the role of migratory birds in the dispersal of B. burgdorferi sensu lato has been investigated in Scandinavia (26, 27) and on the Baltic Sea (3).

In Switzerland, studies on the infection of birds by B. burgdorferi sensu lato are scarce (17, 18). The reservoir competence of the European blackbird (Turdus merula) for B. garinii and B. valaisiana was demonstrated by Humair et al. (17). Recently, Papadopoulos et al. (28) reported the presence of I. ricinus on 51 bird species captured in this country. These authors listed 11 tick species infesting birds, with 2 (Rhipicephalus simus and Hyalomma marginatum marginatum) having been imported by migratory birds (28). Since nothing is known about the role of migratory birds in Switzerland, we investigated the tick infestation of birds during their pre- and postnuptial migrations and the prevalence of B. burgdorferi sensu lato in ticks on migratory birds.

The first survey took place at Bolle di Magadino in April and May 2002. This site is a marsh area at 194 m above sea level south of the Alps (Canton Ticino), where local birds and birds migrating further north were caught. The second survey was carried out at the Col de Bretolet (Canton Valais, at the Swiss-French border) from 6 September to 4 November 2002. At this site (1,923 m above sea level), birds were caught during their southwestward migration while crossing this Alpine pass. Birds were caught in mist nets during routine bird-catching operations of the Swiss Ornithological Institute (Sempach, Switzerland). All birds belonging to the Turdidae were examined for ticks as well as other species, such as Anthus trivalis, Anthus pratensis, Anthus spinoletta, Prunella modularis, Sylvia borin, Sylvia atricapilla, Sylvia communis, and Sturnus vulgaris. Ticks were removed from birds with forceps upon visual inspection and placed into plastic vials (one vial/bird). The species and stage of each tick were identified.

Live fully engorged ticks were screened for B. burgdorferi sensu lato after their molt by cultivation in BSKII medium (10, 33) either as individuals or in pools of ticks. Pools contained ticks collected from a bird at one time. Culture medium was examined by dark-field microscopy for 45 days. All tubes were investigated for Borrelia DNA detection by PCR-restriction fragment length polymorphism according to the method of Postic et al. (29).

Partially engorged ticks stored in 70% ethanol were examined for Borrelia DNA by PCR-reverse line blot (RLB) according to the method of Schouls et al. (32). Borrelia DNA was extracted using 0.7 M ammonium hydroxide. Primers used to amplify the variable spacer region between two repeated genes encoding ribosomal 23S and 5S were 5S-Bor and 23S-Bor (4). DNA amplification was performed using a Whatman (Göttingen, Germany) Biometra TGradient Thermocycler 96. Isolates of B. burgdorferi sensu stricto (B31), B. garinii (NE11), B. lusitaniae (PotiB3), B. afzelii (NE632), and B. valaisiana (VS116) were used as positive controls. Negative controls were included during extraction and PCR. For RLB, the PCR products were hybridized to seven oligonucleotide probes (75 pmol) blotted in lines on an activated Biodyne C membrane (Pall Europe Ltd., Portsmouth, United Kingdom) using a Miniblotter 45 (Immunetic, Cambridge, MA). B. burgdorferi sensu lato, B. burgdorferi sensu stricto, B. afzelii, and B. garinii probes were described by Schouls et al. (32) and Alekseev et al. (4). For this study, three new B. burgdorferi genospecies-specific oligonucleotide probes were designed: B. valaisiana (VSNE) (5′ amino-TATATCTTTTGTTCAATCCATGT), B. garinii (GANE) (5′ amino-CAAAAACATAAATATCTAAAAACATAA), and B. lusitaniae (LusiNE) (5′ amino-TCAAGATTTGAAGTATAAAATAAAA). Hybridization was visualized after exposing the membrane to X-ray film (Hyperfilm ECL; Amersham Biosciences, Otelfingen, Switzerland).

At Bolle di Magadino, in spring, 283 birds were examined for ticks (Table 1). Seventeen birds of four species were infested by ticks, giving a prevalence of 6%. We observed a mean density (total number of ticks observed divided by total number of examined hosts [20]) of 0.12 tick per bird (33 ticks/283 birds) and an intensity of infestation (total number of ticks observed divided by the number of infested hosts [20]) of 1.94 ticks per bird (33 ticks/17 birds). Intensity of infestation varied from 2.23 to 1 according to bird species (Table 1). All 18 larvae (not identified) were found on Erithacus rubecula. Among 15 infesting nymphs, 3 were H. marginatum (2 on Phoenicurus phoenicurus and 1 on E. rubecula), 10 were Ixodes frontalis (9 on E. rubecula and 1 on Luscinia megarhynchos), and 2 were Ixodes ricinus (1 on E. rubecula and 1 on P. modularis). None of the 33 ticks was infected by Borrelia.

TABLE 1.

Prevalence and intensity of tick infestation of migratory birds captured at Bolle di Magadino

Bird species (name)	No. of birds infested/ examined (%)	No. of ticks	Infestation intensity
Anthus spinoletta (water pipit)	0/1
Prunella modularis (dunnock)	1/1 (100)	1	1.00
Erithacus rubecula (European robin)	13/219 (5.9)	29	2.23
Luscinia svecica (bluethroat)	0/1
Luscinia megarhynchos (nightingale)	1/10 (10.0)	1	1.00
Phoenicurus phoenicurus (black redstart)	2/32 (6.3)	2	1.00
Saxicola rubetra (whinchat)	0/13
Turdus merula (blackbird)	0/5
Sylvia communis (common whitethroat)	0/1
Total	17/283 (6.0)	33	1.94

In autumn, at the Col de Bretolet, 987 birds (21 species) were examined for ticks. One hundred eighty birds of 11 species were infested by 417 ticks (369 larvae and 48 nymphs), giving a prevalence of infestation of 18.2% (Table 2), a mean density of 0.42 tick per bird (417 ticks/987 birds), and an intensity of infestation of 2.32 ticks per infested bird (417 ticks/180 birds). Prevalences and intensity of infestation varied according to bird species (Table 2). The highest intensity of infestation was 7.0 ticks. The maximum number of ticks collected was 30 larvae on one European robin captured in October. The most abundant tick species was I. ricinus (n = 411). I. frontalis (n = 3) was collected from one European robin. Three additional ticks could not be identified.

TABLE 2.

Prevalence and intensity of tick infestation of migratory birds captured at the Col de Bretolet

Bird species (name)	No. of birds infested/examined (%)	No. of ticks	Infestation intensity
Alauda arvensis (skylark)	0/2
Anthus pratensis (meadow pipit)	0/5
Anthus spinoletta (water pipit)	0/1
Anthus trivialis (tree pipit)	0/18
Erithacus rubecula (European robin)	142/636 (22.3)	327a	2.30
Luscinia megarhynchos (nightingale)	1/1	1	1.00
Luscinia svecica (bluethroat)	0/1
Oenanthe oenanthe (northern wheatear)	1/6 (16.7)	1	1.00
Phoenicurus ochruros (black redstart)	1/20 (5.0)	7	7.00
Phoenicurus phoenicurus (common redstart)	3/24 (12.5)	9	3.00
Prunella modularis (dunnock)	2/17 (11.8)	1	0.50
Saxicola rubetra (whinchat)	0/1
Sturnus vulgaris (starling)	2/55 (3.6)	1	0.50
Sylvia atricapilla (blackcap)	1/31 (3.2)	1	1.00
Sylvia borin (garden warbler)	0/1
Turdus iliacus (redwing)	1/4 (25.0)	2	2.00
Turdus merula (blackbird)	8/29 (27.6)	23	2.88
Turdus philomelos (song thrush)	18/107 (16.8)	44	2.44
Turdus pilaris (fieldfare)	0/1
Turdus torquatus torquatus (ring ouzel)	0/19
Turdus viscivorus (mistle thrush)	0/8
Total	180/987 (18.2)	417	2.32

^aIncluding 3 I. frontalis ticks.

I. ricinus larvae and nymphs collected from 95 birds (76 E. rubecula, 2 P. phoenicurus, 1 Oenanthe oenanthe, 12 T. philomelos, 3 T. merula, and 1 P. ochruros bird) were examined for Borrelia infection. Among these birds, 26/95 (27.4%) harbored infected ticks: 14/76 (18.4%) of E. rubecula, 7/12 (58.3%) of T. philomelos, 2/3 (66.7%) of T. merula, 2/2 (100%) of P. phoenicurus, and 1/1 (100%) of P. ochruros. Most birds (15/26, 57.7%) were infested by larvae only (11 of E. rubecula, 2 of T. merula, 1 of P. phoenicurus, and 1 of P. ochruros), 7 (26.9%) were infested by larvae and nymphs (3 of E. rubecula, 1 of P. phoenicurus, and 3 of T. philomelos), and 4 (15.4%) were infested by nymphs only (T. philomelos).

Borrelia was observed in 21/127 (16.5%) larvae and 9/26 (34.6%) nymphs examined individually (chi-square = 4.475; P < 0.05). Five Borrelia species were detected in ticks. Mixed infections were observed in six larvae (B. lusitaniae and B. garinii; B. lusitaniae and B. burgdorferi sensu stricto; B. garinii and B. valaisiana [n = 2]; B. lusitaniae, B. valaisiana, and B. garinii; and B. lusitaniae, B. valaisiana, and B. burgdorferi) and one nymph (B. garinii and B. valaisiana). In addition to these individually examined ticks, 68 larvae and 2 nymphs collected on 9 E. rubecula, 2 T. merula, and 1 T. philomelos bird were examined as 12 pools (1 for each bird). B. valaisiana and B. afzelii were detected in one pool of two larvae collected from one E. rubecula bird, B. valaisiana and B. lusitaniae in one pool of two larvae collected from one T. merula bird, and B. valaisiana in 1 pool of 3 larvae collected from one T. merula bird.

B. valaisiana was the most frequent species in ticks (n = 14), followed by B. garinii (n = 8), B. lusitaniae (n = 7), B. afzelii (n = 5), B. burgdorferi sensu stricto (n = 5), and unidentified species (n = 5) (Table 3). Five Borrelia species were detected in ticks feeding on E. rubecula and T. philomelos, whereas T. merula and P. phoenicurus were infested by ticks infected by B. garinii, B. valaisiana, and B. lusitaniae. Ticks collected from P. ochruros were infected by B. valaisiana and B. burgdorferi sensu stricto only. E. rubecula birds harboring infected ticks were infested in 11 cases by larvae only and by larvae and nymphs in 3 cases. Five isolates were obtained from ticks: NE3928 (B. afzelii), NE3929 (B. garinii), NE3930 (B. valaisiana and B. lusitaniae), and NE3931 and NE3032 (B. valaisiana).

TABLE 3.

B. burgdorferi sensu lato infection in I. ricinus feeding on the different bird species (including mixed infections and infections of pooled ticks)

Bird species	Frequency of Borrelia species in larvae and nymphs (no. of birds with infected ticks)a
	Ba	Bg	Bv	Bb	Bl	Bsp
Erithacus rubecula	4*/0 (4)	1/0 (1)	7*/0 (6)	3/0 (3)	2/1 (3)	2/1 (3)
Turdus philomelos	0/1 (1)	1/3 (4)	1/1 (2)	1/0 (1)	2/0 (2)	1/1 (2)
Turdus merula	0/0 (0)	0/1 (1)	2**/0 (2)	0/0 (0)	1**/0 (1)	0/0 (0)
Phoenicurus phoenicurus	0/0 (0)	2/0 (2)	1/0 (1)	0/0 (0)	1/0 (1)	0/0 (0)
Phoenicurus ochruros	0/0 (0)	0/0 (0)	2/0 (1)	1/0 (1)	0/0 (0)	0/0 (0)
Total	4/1 (5)	4/4 (8)	13/1 (12)	5/0 (5)	6/1 (7)	3/2 (5)

^a*, included a pool of two larvae infected by B. afzelii and B. valaisiana and considered as one larva with a mixed infection; **, included one pool of three larvae infected by B. valaisiana (considered as one larva) and one pool of two larvae infected by B. valaisiana and B. lusitaniae and considered as one larva with a mixed infection. Ba, B. afzelii; Bg, B. garinii; Bv, B. valaisiana; Bb, B. burgdorferi sensu stricto; Bl, B. lusitaniae; Bsp, unidentified Borrelia species.

I. ricinus, I. frontalis, and H. marginatum were infesting migratory birds during the northward and southward migrations. I. ricinus clearly dominated on birds captured during the southward autumn migration, whereas I. frontalis dominated among ticks identified from local or northward spring-migrating birds. I. ricinus is very frequently found on birds in Switzerland (2, 17, 18, 28). In this country, I. ricinus is known as the vector of B. burgdorferi sensu lato, the tick-borne encephalitis virus, Rickettsia helvetica, Babesia divergens, Babesia microti, Dipetalonema rugosicauda, and a trypanosome related to Trypanosoma theileri (1, 5, 8, 28).

I. frontalis is widely distributed in Europe and is rather rare in Switzerland compared to regions with a milder climate, such as Spain (28). Robins appear to be its most common hosts (28), as confirmed here. Papadopoulos et al. (28) also described its presence on a song thrush, and we observed I. frontalis for the first time in Switzerland on L. megarhynchos.

All H. marginatum ticks were collected from northward spring-migrating birds, as reported by Papadopoulos et al. (28). All individuals of this Mediterranean Hyalomma species were fully engorged, showing that they had been attached for a few days and were ready to drop off. This suggests that exotic ticks are imported into Switzerland by birds coming from the south. This tick species is considered the main vector of Crimean-Congo hemorrhagic fever virus in Eastern Europe, the Balkan countries, and Transcaucasia (14) and has been found infected by the West Nile virus (6).

During the southward migration in autumn, the prevalence of tick infestation was higher in birds (18.2%) than during the northward migration in spring (6.0%). In autumn, E. rubecula and T. merula had a tick infestation prevalence of 22% and 28%, respectively. For E. rubecula, this prevalence is much higher than that reported by Olsen et al. (26) in Scandinavia (2.2%). The mean density of ticks on birds was also higher in autumn (0.45 tick per bird) than in spring (0.12 tick per bird). A previous study in Egypt (15) also showed a higher infestation of birds in autumn than in spring. In contrast, other studies have shown a higher prevalence of tick infestation and mean density in spring than in autumn (3, 26). In our study, considering bird species, the rate of infestation of E. rubecula by I. ricinus was almost five times higher in autumn than in spring, while the sample size was too small with the other species for comparisons between seasons. The higher proportion of birds carrying ticks and the higher mean density in autumn than in spring or vice versa reflect the activity of ticks at these times of the year at stopover sites used by birds. It is clear that the lower the latitude (as with Switzerland compared to Sweden, Denmark, and Baltic Sea), the higher the infestation in autumn, since birds captured during the southward autumn migration over the Alps traveled through areas where I. ricinus biotopes are widespread and where I. ricinus may show a second peak of questing activity at this period of the year. At the opposite, in spring, birds traveled through the southern part of Europe before arriving in Switzerland and therefore had few contacts with I. ricinus. This explains why birds migrating to the south were three times more infested by ticks than those migrating northwards.

Most of the migrants caught in autumn originate from areas of the northeast of Switzerland, up to Fennoscandia and Russia. Among the species with a sample size >20 and which carried infected ticks, E. rubecula, T. philomelos, T. merula, and P. ochruros are wintering in southern France, Italy, the Iberian peninsula, Morocco, and Algeria, whereas P. phoenicurus is wintering in sub-Saharan Africa.

Among birds migrating southward, five species (E. rubecula, T. philomelos, T. merula, P. ochruros, and P. phoenicurus) carried B. burgdorferi sensu lato-infected ticks. Infection rates of examined I. ricinus ticks were 16.5% for larvae and 34.6% for nymphs. These results are similar to those observed in Scandinavia (26).

In Europe, B. garinii, B. burgdorferi sensu stricto, and B. afzelii were reported in ticks from migrating birds (3, 26). In our study, B. valaisiana, B. garinii, and B. lusitaniae were the most frequently found species, followed by B. afzelii and B. burgdorferi sensu stricto. Interestingly, robins and song thrushes were harboring ticks infected by five Borrelia species. Most robins (78.6%) were infested by larvae alone, suggesting that birds transmitted these Borrelia species to feeding larvae. The origin of the Borrelia species found in ticks feeding on song thrushes is less clear; larvae may have been infected through cofeeding transmission (9). The role played in our results by the reactivation of Borrelia infection in migrating birds (11) is unknown.

The most surprising part of our results is the abundance of B. lusitaniae in ticks feeding on birds. Not only was this Borrelia species identified in larvae in most cases, but it was above all present in larvae feeding on birds (6/7) infested by larvae only. This strongly suggests that birds are reservoir hosts for B. lusitaniae. This Borrelia species appears to be infrequent in I. ricinus in most areas of Europe (10), but B. lusitaniae has been described as the dominant species in southwest Europe (7) and in North Africa (31, 36, 37). Since some bird species, such as E. rubecula and T. philomelos, are migrating between southwestern Europe/North Africa and northern Europe, their role in the dispersal of B. lusitaniae, according to our results, is highly suspected.