Introduction

Borrelia burgdorferi sensu lato (Bbsl) is a complex of spirochaetal species, which includes at least 23 genospecies, mostly associated to hard ticks of the genus Ixodes (Casjens et al. 2011; Ivanova et al. 2014; Margos et al. 2011; Scott et al. 2017; Stanek and Reiter 2011). Within this complex, Borrelia burgdorferi sensu stricto, Borrelia afzelii, and Borrelia garinii are the major etiological agents of Lyme borreliosis (Baranton et al. 1992; Burgdorfer et al. 1982), a tick-borne infectious disease of humans. Additionally, other species as B. spielmanii, B. mayonii B. bavariensis, B. bissettiae, B. kurtenbachii, B. lusitaniae and B. valaisiana have also been associated to Lyme borreliosis in humans (Le Fleche et al. 1997; Margos et al. 2010, 2013, 2016; Pritt et al. 2016; Richter et al. 2006; Wang et al. 1999). Four tick species belonging to the Ixodes ricinus complex are major vectors for the transmission of Bbsl to humans: I. scapularis and I. pacificus in North America, and I. ricinus and I. persulcatus in Europe and Asia (Steere et al. 2016). In Uruguay, there are four Ixodes species currently identified: I. auritulus, I. longiscutatus, I. loricatus and I. aragaoi (Nava et al. 2017; Onofrio et al. 2014). The only member of the I. ricinus complex in Uruguay is I. aragaoi (Nava et al. 2017), and to the present there are no reports of it parasitizing humans (Guglielmone et al. 2014). In the southern cone of America, different haplotypes of Bbsl have been detected in Ixodes spp. from Argentina, southern Brazil, Chile, and Uruguay (Barbieri et al. 2013; Cicuttin et al. 2019; Dall’Agnol et al. 2017; Ivanova et al. 2014; Nava et al. 2014; Saracho-Bottero et al. 2017; Sebastian et al. 2016). Uruguayan Borrelia genotypes isolated from I. aragaoi belong to five different haplotypes of the flagellin (fla) gene, called A, B, C, D and E. From those five haplotypes, haplotypes A to C are similar to Borrelia bissettiae, whereas haplotypes D and E are associated with Borrelia americana (Barbieri et al. 2013). In Argentina, two genospecies belonging to the Bbsl complex were detected and isolated from I. pararicinus: one highly similar to haplotypes A to C from Uruguay, and another, similar to haplotypes D and E (Nava et al. 2014; Saracho-Bottero et al. 2017). The similarities between the Argentinian and Uruguayan Borrelia haplotypes detected in the ticks may be due to the fact that I. aragaoi and I. pararicinus are closely related to one another (Nava et al. 2017; Onofrio et al. 2014; Saracho-Bottero et al. 2017; Venzal et al. 2005a). Furthermore, in Chile, Argentina and Brazil, Bbsl have been detected infecting Ixodes species that are not members of the I. ricinus complex. In Chile, Borrelia chilensis was described in association with I. stilesi and Borrelia sp. Navarino in I. auritulus (Ivanova et al. 2014; Muñoz-Leal et al. 2019). In the Argentinean Patagonia, I. neuquenensis and I. sigelos, both conforming a phylogenetic group with I. stilesi, were found to be infected with a new genospecies of Bbsl, named Borrelia sp. haplotype Patagonia, phylogenetically related to the corresponding sequence of B. chilensis (Sebastian et al. 2016). In Buenos Aires city, Argentina, I. auritulus was found infected with a Borrelia sp. related to a Borrelia sp. detected in I. auritulus from Canada (Cicuttin et al. 2019). In Rio Grande do Sul, the southernmost state of Brazil, Borrelia sp. haplotype Pampa has been detected in I. longiscutatus (Dall’Agnol et al. 2017). It is worth noting that these reports do not necessarily involve a risk for human health, since the pathogenicity of these new genospecies is unknown. With the exception of a single record of a I. pararicinus nymph parasitizing human in the Yungas forests of Argentina (Saracho-Bottero et al. 2018), none of the Ixodes species, from which the Borrelia genospecies were detected, has been found parasitizing humans.

Ixodes auritulus is a tick species with a worldwide distribution that parasitizes several orders of birds, with passerine birds probably the main hosts (González-Acuña et al. 2005). Rodents are considered exceptional hosts for this tick, there are no records of I. auritulus on other mammals (including humans) (Guglielmone et al. 2014; Nava et al. 2017). It has been reported transstadial transmission of B. burgdorferi in the larva-nymph and nymph-adult molts of I. auritulus suggesting vector competence for Bbsl (Scott et al. 2015, 2018a). Furthermore, Scott et al. (2015, 2018a) showed that I. auritulus is involved in the enzootic maintenance cycle of Bbsl in British Columbia, Canada.

The aim of this study was to determine the presence of Bbsl in I. auritulus collected from birds and vegetation in southeastern Uruguay.

Materials and methods

Tick collection and identification

Ticks were retrieved from birds and collected on vegetation from April 2013 to December 2014. Eight tick samplings were made, two per season, in two localities of southeastern Uruguay: Reserva Natural Salus (Lavalleja Department; 34° 25′ S, 55° 18′ W) and Laguna Negra (Rocha Department; 34° 03′ S, 53° 40′ W). Both localities belong to the ecoregion Sierras del Este sensu Brazeiro et al. (2012). Free-living ticks were collected by flagging vegetation along animal trails and footpaths. Birds were captured using mist nets, which remained active from dawn to dusk. Birds were caught with permission of Uruguayan authorities from the Departamento de Fauna, Ministerio de Ganadería, Agricultura y Pesca (Resolution 368/14). Bird species were determined in the field following Narosky and Yzurieta (2003) and Olmos (2011) taxonomic keys. Nomenclature follows the convention of Clements et al. (2016). Each bird was examined for ticks using entomological forceps, and then released. The ticks obtained were immediately stored in 95% ethanol. At the laboratory, ticks were morphologically identified using a stereoscopic microscope and keys for larval, nymph and adult stages (Durden and Keirans 1996; Keirans and Clifford 1978; Kleinjan and Lane 2008; Nava et al. 2017; Onofrio et al. 2006).

DNA extraction and PCR amplification

For molecular analysis, adult ticks and nymphs were pooled by stage (1–20 adult or nymphs per pool) according to source (bird/vegetation) and date of collection. Larvae were not included in this study. Ticks were bisected longitudinally using sterile scalpel blades and forceps, rinsed with distilled water to remove ethanol, and crushed with a homogenization pestle. DNA was extracted using Pure LinkTM Genomic DNA Kit (InvitrogenTM USA) following the manufacturer’s instructions. Molecular screening of Borrelia spp. was done as previously described by Barbieri et al. (2013). Briefly, nested-PCR targeting the flagellin gene (fla) of Borrelia spp. was performed using primers FlaRL (5′-GCA ATC ATA GCC ATT GCA GAT TGT-3′) and FlaLL (5′-ACA TAT TCA GAT GCA GAC AGA GGT-3′) that amplify a fragment of 665 bp, and for nested amplification primers FlaRS (5′-CTT TGA TCA CTT ATC ATT CTA ATA GC-3′) and FlaLS (5′-AAC AGC TGA AGA GCT TGG AAT G-3′) that targets a fragment of 354 bp of fla gene (Barbour et al. 1996). Some positive samples to fla gene were further analyzed by PCR for the presence of a 225 to 255 bp fragment of the rrfA-rrlB intergenic spacer region (IGS) using primers IGSb (5′-AGC TCT TAT TCG CTG ATG GTA-3′) and IGSa (5′-CGA CCT TCT TCG CCT TAA AGC-3′) (Derdáková et al. 2003). All PCR reactions were performed including water and B. anserina DNA as negative and positive control, respectively. PCR products were analyzed in a 1.5% agarose gel by electrophoresis. Amplicons were purified and sent to the Institut Pasteur de Montevideo (Uruguay) for sequencing.

Sequence comparison and phylogenetic analysis

The sequences were assembled and compared using Lasergene software (DNAStar, Madison, WI). The alignments and phylogenetic analysis were performed using MEGA 6.06 (Tamura et al. 2013). The fla and IGS partial sequences (354 bp and 252 bp, respectively) of Borrelia spp. obtained in this study were aligned with the respective sequences of Bbsl genotypes retrieved from the GenBank. The best fitted nucleotide substitution model (GTR + gamma) for ours datasets was selected using jModelTest (Posada 2008). Maximum likelihood trees were conducted with 1000 bootstrap replicates. Sequences of Borrelia hermsii and B. anserina were included in fla phylogenetic inference as out-groups.

Results

Ticks

During the study, 392 birds corresponding to 43 species belonging to five orders and 18 families were captured (Table 1). Of those, 108 (27.5%) were captured in autumn, 90 (23.0%) in winter, 133 (33.9%) in spring, and 61 (15.6%) in summer. Of the 392 birds sampled, 355 (90.6%) corresponded to the order Passeriformes, split in 14 families. During this study, we found bird-feeding ticks parasitizing only specimens of the order Passeriformes. Seventy-eight of the total birds examined (19.9%), were found parasitized with I. auritulus (Table 1). The prevalence of I. auritulus infestation was 21.3% in autumn, 23.3% in winter, 12.8% in spring and 27.9% in summer (Table 1). Three hundred and six I. auritulus ticks (167 larvae, 115 nymphs and 24 females) were retrieved from bird specimens belonging to eight families of passerines (Table 2a), and a total of 1110 I. auritulus free-living ticks were collected from vegetation: 847 larvae, 186 nymphs and 77 females (Table 2b).

Table 1 Birds collected and infested with Ixodes auritulus in southeastern Uruguay (2013–2014)
Full size table
Table 2 Larvae (L), nymphs (N) and adults (A) of Ixodes auritulus collected from (a) passerine birds and (b) vegetation
Full size table

Among the bird families parasitized, the Turdidae family, represented by three species, accumulated 79.7% (244/306) of the total I. auritulus collected on birds in the study. The prevalence of I. auritulus infestation in Turdus spp. was 41.6% in autumn, 71.4% in winter, 46.7% in spring and 94% in summer.

We described only the results about I. auritulus ticks, although during the study I. aragaoi, Haemaphysalis juxtakochi and Amblyomma spp. were also retrieved from birds and collected on vegetation, which information will be used in future analyzes.

Molecular detection of Borrelia spp.

Seventy-three pools (55 pools of nymphs and 18 pools of females), corresponding to 139 I. auritulus (115 nymphs and 24 females) retrieved from 60 birds, were tested by fla PCR for detection of Borrelia spp. Fifteen pools of nymphs (minimum infection rate: 13%) and two pools of females (minimum infection rate: 8.3%) resulted in Bbsl positives. They corresponded to ticks obtained from 17 birds (Table 3a). Furthermore, fla PCR analysis of 263 I. auritulus (186 nymphs and 77 females) collected from vegetation revealed the presence of Borrelia spp. in 12 pools of nymphs (minimum infection rate: 6.4%) and two pools of females (minimum infection rate: 2.6%) of 35 pools tested (19 and 16 of nymphs and females, respectively) (Table 3b). There were Borrelia spp. positive samples through all seasons. The Borrelia spp. fla gene fragments obtained from all positive tick pools were sequenced. They revealed the presence of four different fla haplotypes; named as Borrelia sp. I. auritulus UY1, UY2, UY3, and UY4 (registered in GenBank accession numbers MK160129, MK160130, MK160131, and MK160132, respectively). We did not observe a correlation between haplotype and bird species or collection season. UY4 was the most common haplotype, found in 11 pools of ticks collected from birds and in all pools from vegetation (Table 3). As the screening was done using tick pools we don’t discard the possible presence of more haplotypes. In the phylogenetic reconstruction of the fla gene, those four Uruguayan haplotypes formed a well-supported monophyletic clade within the Bbsl complex. The four haplotypes are closely related to sequences of Borrelia spp. detected in I. auritulus from Argentina (haplotypes G1138 and G1000; GenBank accession number MK984829 and MK984824) and Canada (isolate Cn186-BbPCR2; GenBank accession number KT827332) (Fig. 1). The fla haplotypes UY3 and UY4 were subjected to an additional PCR targeting an IGS fragment. Both of them resulted positive and were successfully sequenced (GenBank accession numbers MK160133, MK160134). The IGS phylogenetic tree showed the Uruguayan haplotypes clustering together with Bbsl British Columbia genotype 1 sequences obtained by Scott et al. (2010) from I. auritulus of Canada (GenBank accession numbers EU019112.1, EU019121.1 and EU019120.1) (Fig. 2).

Table 3 Detection of Borrelia spp. by nested PCR targeting the fla gene in Ixodes auritulus pools of nymphs (N) and adults (A) collected from (a) birds and (b) vegetation
Full size table
Fig. 1
figure 1

Maximum likelihood tree constructed of Borrelia spp. fla partial sequences. Numbers represent bootstrap support generated from 1000 replications. The sequences obtained in this study were highlighted bold and GenBank accession numbers are in brackets. B. anserina and B. hermsii were included as outgroup

Full size image
Fig. 2
figure 2

Maximum likelihood tree constructed of Borrelia spp. IGS partial sequences. Numbers represent bootstrap support generated from 1000 replications. The sequences obtained in this study were highlighted bold and GenBank accession numbers are in brackets

Full size image

Discussion

Ixodes auritulus, the avian coastal tick, is found in the Australian, Ethiopian, Nearctic and Neotropical zoogeographical regions (Guglielmone et al. 2014). During this survey, we captured 78 passerine birds infested with I. auritulus. Among those, members of the family Turdidae represented 77% (n = 60) of the parasitized birds. Moreover, 79.7% of the total ticks collected were found in specimens from this family (Table 2a). In a previous study about ticks on wild birds of Uruguay, birds from the genus Turdus were the most infested with larvae and nymphs of ixodids (Venzal et al. 2005b). Wild birds are frequently infested by ticks, and some bird species act as reservoirs of zoonotic pathogens (Richter et al. 2000). There are several reports discussing the role of members of the family Turdidae as hosts of I. auritulus (da Cunha Amaral et al. 2013; Arzua and Barros-Battesti 1999; González-Acuña et al. 2005; Scott et al. 2012, 2015). These authors and other researchers highlighted the role of birds belonging to the genus Turdus, as reservoirs of Borrelia spp. and their participation in the dissemination of infected ticks (Rudenko et al. 2014; Saracho-Bottero et al. 2017; Scott et al. 2012; Scott and Foley 2016). In our study, another bird species found highly infested with I. auritulus was Poospiza cabanisi. Previous studies about bird-tick relationships have shown that tick infestation is more common among birds that forage primarily on the ground and in the shrub layer (Morshed et al. 2005). In this context, the genera Turdus and Poospiza include birds characterized by living in low forest stratum and are frequently found on the ground (Narosky and Yzurieta 2003), which may explain the high infestation levels reported in this study.

The previous findings of Bbsl genotypes in I. auritulus were made in ticks collected on Canadian birds (Morshed et al. 2005; Scott et al. 2010, 2012, 2015, 2018a; Scott and Foley 2016). Borrelia haplotypes found in these birds corresponded to B. burgdorferi s.s., British Columbia genotype 1, British Columbia genotype 2, British Columbia genotype 3 (Morshed et al. 2005; Scott et al. 2010) and, similarly, B. americana (Scott and Foley 2016). Recently, Muñoz-Leal et al. (2019) found a Borrelia genospecies belonging to the Bbsl complex (Borrelia sp. Navarino) retrieved from I. auritulus collected on a bird, Troglodytes musculus, in Chile.

In the present study, the results of the PCR using fla gene revealed the presence of Borrelia in 17 pools, (minimum infection rate: 12.2%), of nymphs and females of I. auritulus, collected from birds (Table 3a). Nine positive pools of I. auritulus were obtained from passerine birds of Laguna Negra and eight from Reserva Natural Salus. At the same time that our study was carried out, Cicuttin et al. (2019) studied the presence of Borrelia spp. in a protected urban area of Buenos Aires city, Argentina. They found a prevalence of Bbsl in I. auritulus collected on birds of 27.3%. The phylogenetic tree generated with the fla sequences in our study suggests that the Borrelia haplotypes UY1-4 detected in I. auritulus belongs to the Bbsl complex and are closely related to the haplotypes G1000 and G1138 detected by Cicuttin et al. (2019) in I. auritulus, and are related to an isolate of Borrelia sp. obtained from I. auritulus in Canada (Isolate Cn186-BbPCR2) (Fig. 1). Cicuttin et al. (2019) showed that the same haplotypes from Argentina, Uruguay and Canada conform a monophyletic group. The IGS phylogenetic reconstruction showed that Bbsl sequences of I. auritulus from Uruguay form a well-supported clade with British Columbia genotype 1 of I. auritulus from Canada (Fig. 2). The difference between the IGS fragment sequences obtained from I. auritulus of Canada and Uruguay is a single nucleotide. However, the fragment of IGS used is very short, so further analyses using other genes (or longer fragments of the genes) are required to properly determine the genetic variation between haplotypes from the two geographic locations. Migratory birds could be involved in the probable relation between the Canadian haplotypes and the haplotypes found in Argentina and Uruguay. Migratory birds are increasingly considered important in the global dispersal of zoonotic pathogens; they can transport the tick vectors as well as the pathogens (Ogden et al. 2008). Some species of migratory birds have been described as efficient reservoirs of some genotypes of Bbsl (Ogden et al. 2008; Scott et al. 2018b). Neotropical passerines migrate across national and intercontinental borders, and become long-range vectors for any zoonotic pathogen that they harbor. Overall, dispersal of Bbsl-infected ticks along migration routes is an important mechanism in the establishment of new endemic foci of tick-borne diseases (Scott et al. 2014, 2018b). Ixodes auritulus parasitizes several orders of birds, with passerine birds probably the main hosts, sustaining tick populations throughout the Neotropic and the Nearctic (González-Acuña et al. 2005; Scott et al. 2015).

The spirochete detected in I. auritulus during this study represents the third genospecies of Bbsl reported for Uruguay. The two previous Uruguayan genospecies were associated with I. aragaoi ticks obtained in the same locations of this study (Barbieri et al. 2013). Until now, Lyme borreliosis does not represent a problem for public health in the southern cone of South America. In Brazil, it has been described a disease named as Lyme disease-like syndrome, or Baggio-Yoshinari syndrome, with some clinical manifestations similar to those observed for Lyme disease (Mantovani et al. 2007; Miziara et al. 2018; Yoshinari et al. 2010). However, a more recent study performed a critical evaluation of the diagnostic methods that were used for this Baggio-Yoshinari syndrome, and concluded that they might not represent a Borrelia-caused disease (de Oliveira et al. 2018). In Uruguay, although one case described in 1996 and many suspected clinical cases with Lyme disease-like symptoms, tick-borne borreliosis has not been diagnosed (Conti-Díaz 2001; Nava et al. 2014 Protasio et al. 1996). Furthermore, even though in the Holarctic region ticks bites in humans by Ixodes spp. are very common, the situation in South America is different, with only a few reports of human bites by ticks of the genus Ixodes (Guglielmone et al. 2006, 2014; Nava et al. 2014; Saracho-Bottero et al. 2017). However, Scott et al. (2018a) described that if there are two or more tick species feeding concurrently on a host, they can transmit Bbsl, via the reservoir host, from one cofeeding tick species to another cofeeding tick species. Alternatively, one tick species can infect a reservoir-competent host and, after the blood meal, another tick species (that could be one that bite humans) can subsequently acquire Bbsl from this spirochetemic host. Also, although I. auritulus only parasitizes avifauna, both birds and mammals eat these ixodid ectoparasites, and may become systematically infected by oral inoculation (Scott et al. 2018a). These findings mean that medical implications due to the presence of Bbsl in I. auritulus cannot be dismissed. The isolation and culture of these Borrelia genospecies will be essential to study their pathogenicity.

The growing number of publications describing the presence of new genospecies of Bbsl in the southern cone region of America (Barbieri et al. 2013; Ivanova et al. 2014; Nava et al. 2014; Saracho-Bottero et al. 2017; Sebastian et al. 2016), and the scarce information about its pathogenicity, reservoirs and vectors, highlights the importance of further studies about spirochetes presence in Uruguay and the region.