Borrelia burgdorferi sensu lato infection in passerine birds from the Mazurian Lake region (Northeastern Poland)

Abstract

The aim of the present study was to evaluate a potential role of different passerine birds species in Mazurian Lake region (northeast Poland) in the spread of Borrelia burgdorferi sensu lato, the spirochaete that causes Lyme disease. A total number of 1254 birds (representing 42 species) were captured during the 3-year study period. Blood samples were collected from birds and analyzed with a nested polymerase chain reaction technique in order to detect fragments of the pathogen DNA. Positive results were obtained in 4.2% of all blood samples. Specifically, B. burgdorferi s.l. were detected in tree pipit (Anthus Erivialis; 21.1% of 19 birds), dunnock (Prunella modularis; 15.8% of 19 birds), chaffinch (Fringilla coelebs; 12.7% of 166 birds), song thrush (Turdus philometos; 9.3% of 54 birds), nuthatch (Sitta euopea; 7.7% of 26 birds), hawfinch (Coccothraustes coccothroustes; 6.7% of 15 birds), robin (Erithacus rebecula; 5.1% of 256 birds), blackbird (Turdus merula; 4.2% of 71 birds) and wren (Troglodytes troglodytes; 3.7% of 27 birds). Additionally, the incidence of the infection was analyzed in relation to the habitat in which the birds resided (mixed coniferous forest or alder swamp forest), months of the study (from April to October), age and sex, but the differences were not statistically significant.

1 Introduction

Lyme disease, caused by the spirochaete Borrelia burgdorferi sensu lato, is the most common tick-borne zoonosis in the world, manifesting with multisystemic disorders in humans (Steere, 1989). B. burgdorferi s.l. is transmitted mainly by ticks of the Ixodes ricinus group. These tick species parasitize various mammals, birds, and reptiles. Infected nymphs or adult females may transmit B. burgdorferi s.l. to vertebrates, which have a reservoir role (i.e. the spirochaetes persist for a considerable time), and they can infect engorging larval ticks, non-infected nymphs as well as adult ticks. Small and medium-sized mammals are the main vertebrate reservoirs of B. burgdorferi s.l. (Mather et al., 1989; Humair et al., 1993a; Tälleklint & Jaenson, 1994). The importance of birds as additional reservoir hosts is a subject of growing interest.

It is well documented that numerous species of passerine birds are frequently infested with ticks (Pruett-Jones & Pruett-Jones, 1991; Battaly & Fish, 1993; Humair et al., 1993b; Hubálek et al., 1996; Kinsey et al., 2000). Moreover, the role of these avian hosts in spreading ticks to other sites has been shown (Olsén et al., 1995; Smith et al., 1996; Durden et al., 2001; Scott et al., 2001; Slowik & Lane, 2001). On the contrary, B. burgdorferi s.l. has been isolated from the blood (Burgess et al., 1993; Humair et al., 1993b; McLean et al., 1993; Olsén et al., 1996; Durden et al., 1997), from skin biopsies (Miyamoto et al., 1997; Humair et al., 1998; Durden et al., 2001) and internal organs (e.g. liver; Anderson et al., 1986; Miyamoto et al., 1997) of passerine birds. The pathogen has been also detected in blood and foot web tissues of seabirds (Olsén et al., 1993; Gylfe et al., 1999). However, there is insufficient data to assess whether birds have the ability not only to maintain and multiply, but also to transmit the pathogen to ticks.

The aim of the present study was to establish whether the spirochaete is found in the passerine bird community resident in the Mazurian Lake region, and if there are any differences between avian species in the presence of B. burgdorferi s.l.

The most valuable investigative approach is larval xenodiagnosis (Humair et al., 1998; Kurtenbach et al., 1998; Richter et al., 2000), but not all vertebrate species are amenable to it because they are difficult to capture and maintain under controlled conditions. For the latter reasons, in the present study detection of B. burgdorferi s.l. DNA in the birds blood was employed.

The Mazurian Lake region, located in northeastern Poland, was selected as the study site as it is considered as an endemic area of B. burgdorferi s.l. infection in ticks (Sinski & Karbowiak, 1994; Wegner et al., 1995). Second, it is also a region of abundance of the most common tick (I. ricinus L.) in Poland (Lachmajer, 1967; Siuda, 1993). Furthermore, many studies revealed B. burgdorferi or antibodies to B. burgdorferi in humans living in this region and reporting frequent contacts with ticks (Prokopowicz, 1995; Prokopowicz et al., 1995; Pancewicz et al., 1996).

2 Materials and Methods

2.1 Bird sampling

Birds were captured in two types of habitats: mixed coniferous forest (Pino—Quercetum) and alder swamp forest (Carici elongatae—Alnetum), in places where the abundance of undergrowth and shrub plants was similar. The catches were performed in 1996, 1997 and 1998 during the seasons of the highest tick activity; that is, from April to October (permissions OP 4072/230/97, OPog 4201/103/98). Twenty standard ‘mist nets’ were placed in both habitats and examined at 1-h intervals. Investigations of both forests took place during a subsequent full 6 days and 6 nights for 7 consecutive months.

The species and, if possible, the age and sex of the captured birds were determined according to Svensson (1992). A small amount of blood (approximately 50 μl) was collected from the brachial vein by the method of Kruszewicz (1995) from individuals presenting in good physical condition. The blood samples were frozen at –20°C. After blood sampling the birds were ringed. Then the birds were released into the same habitats in which they had just been caught. Individuals recaptured within the same month were released immediately and were not recorded in the data analysis, whereas birds recaptured in other months or years of the study were analyzed separately and the distinct samples were tested each time.

2.2 Polymerase chain reaction

All the blood samples were analyzed with a nested plymerase chain reaction (PCR) technique in order to detect any fragments of B. burgdorferi s.l. DNA. The DNA was isolated by the procedure of Stanczak et al. (1995). Briefly, the DNA was extracted once with an equal volume of water-saturated phenol and once with an equal volume of chloroform. Centrifugation was at 15 000×g for 15 min at 4°C. The supernatant was placed into a fresh tube and an equal volume of isopropyl alcohol was added to precipitate the DNA. Then the reaction mixture was incubated overnight at –20°C and the DNA was spun down at 15 000×g for 15 min at 4°C. The supernatant was then removed, DNA was washed with 200 μl cold 70% ethanol, centrifuged (at 15 000 x g for 15 min at 4°C), dried (for 3 min at 72°C) and dissolved in 60 μl sterile water. Amplification of DNA was performed by the nested PCR technique according to Valsangiacomo et al. (1996) and based on two-stage amplification of the fragment of hbb gene coding a highly preserved histon protein of B. burgdorferi s.l. The first stage product is 433 base pairs in size and contains the whole hbb gene, and the second stage product is a 184 base pairs fragment of the hbb gene. Genomic DNA of B. burgdorferi (DNA; Gdansk II) was used as a positive control, and reagents as a negative control. The cycling conditions were 94°C for 1 min, 52°C for 1 min and then 72°C for 30 sec for a total of 35 cycles in the first stage and 25 cycles in the second stage in an automated thermal cycler (GeneAmp PCR System 9600; Perkin Elmer).

The second-stage products were separated in 2% agarose gel and visualized by 3 μl ethidium bromide staining. Electrophoresis was performed during 30 min with the current intensity of 120 mA. The M1 marker (pUC19/MspI DNA; Gdansk II) was used as a molecular mass standard of DNA (Fig. 1).

Fig. 1 Agarose gel demonstrating PCR detection of Borrelia burgdorferi s.l. DNA in passerine bird blood.

3 Results

3.1 Birds investigated

During the 3-year study period a total number of 1254 blood samples were collected (Table 1) (277 blood samples in 1996, 448 in 1997 and 529 in 1998). Of these, 100 individuals were recaptures. In the study site, the most numerous species were robin, great tit and chaffinch; 256, 183 and 166 specimens were caught and investigated, respectively.

Passerine birds examined for B. Burgdorferi s.l. infection in northeastern Poland

Family	Species	Number of birds examined	Number (%) of birds infected
Sylviidae	Blackcap (Sylvia atricapilla [L., 1758])	120	0^a
	Garden warbler (Sylvia borin [Boddaert, 1783])	55	0
	Whitethroat (Sylvia communis [Latham, 1787])	1	0^a
	Barred warbler (Sylvia nisoria [Bechstein, 1783])	2	0
	Lesser whitethroat (Sylvia curruca [L., 1758])	5	0
	Goldcrest (Regulus regulus [L., 1758])	4	0
	Icterine warbler (Hippolais icterina [Vieillot, 1817])	5	0
	Sedge warbler (Acrocephalus schoenobaenus [L., 1758])	1	0
	Great reed-warbler (Acrocephalus arundinaceus [L., 1758])	3	0
	Reed warbler (Acrocephalus scirpaceus [Hermann, 1804])	2	0
	Willow warbler (Phylloscopus trochilus [L., 1758])	9	0
	Chiffchaff (Phylloscopus collybita [Vieillot, 1817])	16	0^a
	Wood warbler (Phylloscopus sibilatrix [Bechstein, 1793])	3	0
Turdidae	Robin (Erithacus rubecula [L., 1758])	256	13 (5.1)^a
	Song thrush (Turdus philomelos Brehm, 1831)	54	5 (9.3)^a
	Blackbird (Turdus merula L., 1758)	71	3 (4.2)^a
	Mistle thrush (Turdus viscivorus L., 1758)	2	0
	Redwing (Turdus iliacus L., 1766)	1	0
	Redstart (Phoenicurus phoenicurus [L., 1758])	5	0^a
	Thrush nightingale (Luscinia luscinia [L., 1758])	1	0^a
Paridae	Great tit (Parus major L., 1758)	183	0^a
	Willow tit (Parus montanus Conrad, 1827)	32	0
	Blue tit (Parus caeruleus L., 1758)	25	0
	Crested tit (Parus cristatus L., 1758)	6	0
	Coal tit (Parus ater L., 1758)	1	0
	Marsh tit (Parus palustris L., 1758)	27	0
Fringillidae	Chaffinch (Fringilla coelebs L., 1758)	166	21 (12.7)^a
	Scarlet grosbeak (Carpodacus erythrinus [Pallas, 1770])	1	0
	Hawfinch (Coccothraustes coccothraustes [L., 1758])	15	1 (6.7)^a
	Bullfinch (Pyrrhula pyrrhula [L., 1758])	8	0
	Siskin (Carduelis spinus [L., 1758])	2	0
Muscicapidae	Spotted flycatcher (Muscicapa striata [Pallas, 1764])	26	0
	Pied flycatcher (Ficedula hypoleuca [Pallas, 1764])	8	0
Emberizidae	Yellowhammer (Emberiza citrinella L., 1758)	14	0
Motacillidae	Tree pipit (Anthus trivialis [L., 1758])	19	4 (21.1)^a
	White wagtail (Motacilla alba L., 1758)	1	0
Sittidae	Nuthatch (Sitta europea L., 1758)	26	2 (7.7)
Prunellidae	Dunnock (Prunella modularis [L., 1758])	19	3 (15.8)
Troglodytidae	Wren (Troglodytes troglodytes [L., 1758])	27	1 (3.7)^a
Certhiidae	Tree creeper (Certhia familiaris L., 1758)	22	0
Corvidae	Jay (Garrulus glandarius [L., 1758])	3	0
Oriolidae	Golden oriole (Oriolus oriolus [L., 1758])	1	0
Total		1254	53 (4.2)
^a Passerine bird species listed by EUCALB as effective reservoirs of the bacteria B. burgdorferi s.l.

3.2 Detection of B. burgdorferi s.l. by PCR

B. burgdorferi s.l. was found in blood of 4.2% of all analyzed birds (n=1254). Positive samples were observed in nine of the 42 avian species examined (Table 1). Of the recaptured birds (n=100) only one bird (chaffinch) that was caught in July and August 1998 was positive for B. burgdorferi s.l. each time. The general prevalence of bird infections with B. burgdorferi s.l. was similar within the three consecutive seasons: in 1996 it was 3.6% of the 277 individuals, in 1997 it was 5.1% of 448 individuals, and in 1998 it was 3.8% of 528 individuals.

The highest prevalence (more then 10%) was measured for the tree pipit, dunnock and chaffinch; 21.1% of 19 birds, 15.8% of 19 birds and 12.7% of 166 birds, respectively, were infected (Table 1). For groups of frequently captured birds (more than 15 representatives), statistical analysis revealed significant differences of infection rates between the bird species (χ²=75, degrees of freedom (df)=19, P<0.0001).

Of all infected birds, 57% (n=53) were captured in the mixed coniferous forest. These infected birds represented 5.2% of the total number of the birds captured and examined in this type of forest (n=577), compared with 3.4% of total number of the birds captured and examined in alder swamp forest (n=677); however, the difference was not statistically significant (χ²=2.5, df=1, not significant (NS)). In the next step, analysis was made only for the bird species that significantly preferred one of the forest habitat. This analysis revealed that birds preferring mixed coniferous forest were more frequently B. burgdorferi s.l. carriers than birds preferring alder swamp forest (χ²=13.85, df=1, P<0.0002).

The infection rates changed by month; however, the alterations were not statistically significant (χ²=4.4, df=6, NS) (Fig. 2).

Fig. 2 Changes in Borrelia burgdorferi s.l. infection by season.

Analyses with respect to age and sex were also performed. Young birds (up to 1 year old; n=520) were infected in 5% of cases, and the older population (n=662) was infected in 4.2%; however, the difference was not statistically significant (χ²=0.4, df=1, NS). Similarly, the sex analysis (only species with sex dimorphism were considered) revealed no significant differences. Bacterial DNA was found in 4.3% of 185 females and 5.2% of 347 males (χ²=0.2, df=1, NS).

4 Discussion

The etiology and transmission of tick-borne diseases are subjects of rapidly growing interest during recent years. Research studies have attempted to determine not only which vertebrate species are the most frequent tick hosts, but have also focused on their potential role as reservoir hosts; that is, their ability to participate in the circulation of the pathogen in nature (Mather et al., 1989; Tälleklint & Jaenson, 1994). Another important factor is also quantitative analysis to determine the infection rate with the pathogen. The overall findings of the present study suggest that the infection rate with B. burgdorferi s.l. is markedly lower in birds when compared with rodents (4.2% versus 34.8 to 75%) (Humair et al., 1993a; Sinski & Karbowiak, 1994). This may suggest the lower importance of birds as hosts and reservoirs of B. burgdorferi s.l. in comparison with small forest rodents. A key consideration, apart from the low average prevalence rates for all the birds, is whether there are differences in infection rate between different bird species. The present results suggest that within a resident passerine bird community of the study area there are some bird species potentially spreading the bacteria B. burgdorferi s.l. more than others (Table 1), making it possible that some avian species could be considered important in maintaining the B. burgdorferi s.l. population in nature. The European Union Concerted Action on Lyme Borreliosis (EUCALB) (Smith et al., 1998) has published a list of vertebrate species considered effective reservoirs of the bacteria B. burgdorferi s.l. (Gern et al., 1998). In the present study, seven of the species of passerine birds listed by EUCLAB as reservoirs were found to be infected. Additionally, infection was detected in two other bird species frequently found in the study area (i.e. nuthatch and dunnock). Some bird species listed by EUCALB as participating in the survival of B. burgdorferi s.l. population were not found to have the pathogen in the present study. However, three of these species (whitethroat, redstart and thrush nightingale) are only incidentally found in the study area (up to five caught individuals), which may explain negative results. However, both the great tit and the blackcap were very frequently captured birds in this study (183 and 120 specimens, respectively), but none was shown to be pathogen-positive. This may reflect very low rates of infestation of both bird species by ticks in the study area (Gryczynska et al., 2002). Great tits are of a bird group preferring upper zones of forest that limit their contact with ticks. Blackcaps do not avoid bottom zones of the forest; their low level of infestation may be due to their migration behavior. These birds are found in the study area relatively late (i.e. during May) and start their autumn return rather early (i.e. August). As a consequence they may evade contact with ticks, which are not found to show marked activity before September.

The differences in infection rate between bird species may result from many reasons. However, the most frequently infected bird species are those with the highest prevalence of infestation of the vector (i.e. I. ricinus tick) (Gryczynska et al., 2002).

In mixed coniferous forest, there was a higher percentage of B. burgdorferi s.l.-infected birds than in the alder swamp forest. Probably, this results from a higher prevalence of tick infestation in birds of this type of forest (Gryczynska et al., 2002). The hypothesis that passerine birds play a greater role in spreading of B. burgdorferi s.l. in mixed coniferous forest than in alder swamp forest is supported by the differences in infection prevalence between two bird species preferring one habitat or the other. Such comparison was performed for tree pipits and blackbirds, preferring mixed coniferous forest and alder swamp forest, respectively. The presence of B. burgdorfresi s.l. in blood was found more frequently in tree pipits than in blackbirds (χ²=5.92, df=1, P<0.015), even though tick infestation was found at similar frequency in these two species (Gryczynska et al., 2002).

The changes in the incidence of infected birds during consecutive months suggest that passerine birds may be a carrier of the pathogen only for short period of time. The incidence of infected birds increased during the first 3 months of the study in the both type of a forest, and this corresponds with the increase of the total population of ticks in their environment (Gryczynska et al., 2002). Then, the incidence of infected birds went down during the period of lower tick activity (July, August). Accordingly, the incidence of infected individuals of the birds living in the same area does not increase as months go by, but it strictly depends on the presence (and quantity) of infected ticks in this area. This hypothesis is also supported by the fact that older birds did not appear to be more frequently infected than 1-year-old individuals. On the contrary, however, it may be that the reducing pathogen incidence in birds with time may be related to an increased mortality of birds carrying the pathogen and earlier elimination of infected individuals in the environment. This hypothesis is supported by the fact that of 100 birds captured more then once only the one individual was positive both during the first and second catching (1-month interval), whereas other infected birds were not caught again.

In conclusion, it is suggested that passerine birds may play a role as a reservoir in the process of B. burgdorferi s.l. spread in the forest environment; however, their host competence is dependent on the bird species. There is a need to determine whether B. burgdorferi s.l.-positive avian species are able to transmit the spirochaete to ticks and thus to have reservoir competence. There is also a need to analyze B. burgdorferi s.l. genomospecies, as a few reports suggest that different groups of vertebrates are reservoirs for specific bacteria genomospecies (Hanincova et al., 2003a,b, Humair et al., 1995; Olsén et al., 1995). It is also known that ticks may be carriers both of one and many bacteria genomospecies (mixed infections) (Strle et al., 1995; Rijpkema et al., 1995; Saint Girons et al., 1998), and the possibility exists that there are two or more independent cycles maintaining stable specific populations of the bacteria genomospecies in their environment providing distinct zoonotic reservoirs (Kurtenbach et al., 2002). Various genomospecies of bacteria B. burgdorferi s.l. may be responsible for different forms of borrelliosis (Canica et al., 1993; van Dam et al., 1993).

Acknowledgments

The study was partially supported by Warsaw University grants BW-1345/15, BW-1385/9 and BW-98/1. M.S. is supported by a Young Scientist Award from the Foundation for Polish Science, Warszawa, Poland.