Introduction

Formerly, state borders and legal travel restrictions confined the movement of dogs within Europe, effectively limiting the introduction of vector-borne infections to non-endemic regions. The founding of the European Union facilitated pet travel and importation of dogs by private persons and animal welfare organisations from all over Europe to Central and Western European countries. Such dogs often originating from endemic areas may act as carriers for different vector-borne infections, e.g. Babesia spp., Hepatozoon canis, Leishmania spp., Dirofilaria spp. or Ehrlichia canis, into hitherto non-endemic countries.

Such infections necessitate significantly extended differential diagnostic investigations in small animal practice in non-endemic regions in Central and Western Europe (Deplazes et al. 2006). In Germany, Babesia spp., Leishmania spp., Dirofilaria immitis and E. canis have repeatedly been recorded in travelling and imported dogs (Dongus and Gothe 1996; Zahler et al. 1997; Glaser and Gothe 1998; Gothe 1998; Gothe and Wegerdt 1991; Menn et al. 2010; Hamel et al. 2011; Röhrig et al. 2011). Only few cases of leishmaniosis, hepatozoonosis or ehrlichiosis, however, have been considered as autochthonous infections based on anamnesis, clinical features and consecutive diagnostics (Gothe 1998; Jensen et al. 2007a; Kellermaier et al. 2007; Gärtner et al. 2008). The transmission of these pathogens in the absence of competent vectors in Germany still remains an open question.

Canine vector-borne infections are recognised as a major health problem in Mediterranean countries. However, despite intensive clinical and epidemiological research, information on the actual distribution and seasonal occurrence of such infections is still lacking for most South-Eastern European countries (Trotz-Williams and Trees 2003). Nonetheless, Babesia canis spp. is widely recognised as an important clinical canine disease in Hungary and in Romania (Földvari et al. 2005; Hornok et al. 2006; Ilie et al. 2010; Ionescu et al. 2010). The objective of this study was to obtain more detailed information on the epidemiology and extent of such infections. The added scientific value will contribute to appropriately adjusting routine diagnostic screening of dogs imported from these countries to the current epidemiological situation. Therefore, a total of 216 dogs from Hungary and Romania (29 from a veterinary clinic in Bucharest, Romania and 187 from Romania and Hungary imported to Germany) were tested for the presence of vector-borne infections by direct and indirect diagnostic techniques.

Material and methods

Animal data and clinical assessment

Canine EDTA blood samples obtained during March 2009 to April 2010 were screened for the presence of canine vector-borne infections by direct and indirect methods. One hundred eighty-seven EDTA blood and serum samples of stray dogs from Hungary (n = 78) and Romania (n = 109) imported into Germany were tested. One hundred seventy-nine samples of clinically apparently healthy animals were supplied by animal welfare organisations and eight by private owners for routine diagnostic to the diagnostic centre at the Chair of Comparative Tropical Medicine and Parasitology. These imported dogs consisted of 117 mixed breed dogs, 22 pure bred dogs and 48 without information on breed. The dogs’ age ranged from 4 months to 10 years [mean 2.9 years, standard deviation (SD) 2.2 years]; 64 were male dogs, 68 bitches and for 55 animals no gender was specified.

Additional 29 EDTA blood and serum samples were taken in April 2010 from pet dogs submitted to the Salvavet Veterinary Clinic in Bucharest, Romania. The material was obtained from surplus blood samples taken for routine diagnostic. The age of local Romanian dogs was specified for 28 animals, ranging from 3 months to 13 years (mean 4.8 years, SD 4.0 years). Eight bitches and 21 male dogs were tested. Twenty-eight dogs were from Bucharest and vicinity and one from the Carpathian Mountains region. They were presented at the Salvavet Veterinary Clinic for various reasons, e.g. gastric symptoms (seven), surgery/accidents (six), babesiosis (five), renal problems (three), parasites (two) or other (six). Eight owners reported the use of prophylactic measures against ectoparasites and 12 owners recognised “ectoparasites” on their dogs prior to clinical consultation.

Laboratory testing

Samples were tested with standard techniques established in our laboratory such as blood/buffy coat smear, Knott’s test, DiroChek®-ELISA, Indirect Immunofluorescent Antibody Test (IFAT) for anti-B. canis spp., anti-Leishmania spp. and anti-E. canis antibodies and by the conventional and real-time PCRs. Giemsa-stained blood/buffy coat smears were performed for all animals for the detection of pathogens in the blood. For technical reasons, only the blood samples of imported dogs were tested by Knott’s test for microfilaria. DiroCheck®-ELISA Canine/Feline Antigen Test Kit (Synbiotics Corp., USA) was applied for screening for the antigen of female adult D. immitis.

Molecular methods

DNA was extracted from 200 μl of the EDTA blood samples with the Qiagen DNA MiniKit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. Conventional PCRs were carried out for the detection of B. canis spp. and Babesia gibsoni, H. canis and Mycoplasma spp. and for discriminating different species of microfilaria in Knott’s test-positive samples (Birkenheuer et al. 2003; Inokuma et al. 2002; Watanabe et al. 2003; Rishniw et al. 2006). The reactions were carried out on an Eppendorf Mastercycler gradient (Eppendorf, Hamburg, Germany) or an Applied Biosystems GeneAmp 9700 (Applied Biosystems, Darmstadt, Germany) using the Qiagen HotStarTaq Polymerase Kit (Qiagen, Hilden, Germany) for Babesia spp. and the HotMaster Taq Polymerase (5-Prime, Darmstadt, Germany) for H. canis, Mycoplasma spp. and differentiation of microfilaria. Successful amplification of the gene targets in the conventional PCRs was verified with 2.0% agarose gel electrophoresis after staining with GelRed™ (Biotium, Hayward, USA) and visualized under UV light. For Babesia spp., the primers 455-479F and 793-722R were used for the first amplification, and for the heminested reaction, the reverse primer 793-722 R was used with BCC-F for Babesia canis canis-specific detection, with BCV-F for Babesia canis vogeli-specific detection and with B.gib.ASIA-F for B. gibsoni-specific detection. The bands of 185–200 bp in the respective heminested reaction were considered positive. For the first about 30 samples, also the first reaction was run on a gel. For H. canis, a band of 666 bp was considered positive, for Mycoplasma spp., a band of 202 bp indicates Candidatus Mycoplasma haematoparvum and of 276 bp, Mycoplasma haemocanis, and for the differentiation of microfilaria, 484 bp were considered as Dirofilaria repens, 542 bp as D. immitis and 578 bp as Acanthocheilonema reconditum. The products of the filarial PCR were sequenced with one primer for confirmation.

Amplicons of two Babesia spp. and of the filarial PCR were purified with the QIAquick PCR Purification Kit according to manufacturer’s instruction (Qiagen) and sequenced at Eurofins MWG Operon (Martinsried, Germany). Sequence homology searches were made by BLASTn analysis of GenBank (www.ncbi.nlm.nih.gov.library.vu.edu.au/BLAST/).

Real-time PCRs were used for the detection of Leishmania infantum, Anaplasma phagocytophilum and E. canis DNA (Mary et al. 2004; Courtney et al. 2004; Silaghi, unpublished data). A CT value smaller than 39 was considered positive for A. phagocytophilum and E.canis, and smaller than 38 for L. infantum.

Serological methods

Antibodies against B. canis spp., E. canis and A. phagocytophilum were tested using commercial IFAT tests (A. phagocytophilum, Focus Technologies, Cypress, USA; E. canis and B. canis ssp., MegaCor, Hörbranz, Austria). An in-house anti-Leishmania spp. IFAT was used for the detection of Leishmania spp. antibodies (Mancianti et al. 1995). Titres of 1:64 were considered as seroreactive to B. canis spp., Leishmania spp. and A. phagocytophilum and of 1:32 as borderline titres. Titres of 1:40 were seropositve in E. canis-IFAT, respectively.

Statistical analysis

To obtain information on the incidence of infections in pet and stray dogs, the dogs were grouped into those presented at the veterinary clinic and imported animals. Contingency table analysis (p < 0.05) and calculation of confidence intervals was performed with Graph Pad Prism (Graph Pad Software, USA).

Results

The screening for vector-borne infections of 216 canine EDTA blood samples resulted in the detection of 11 different pathogens by molecular methods—B. canis canis, B. canis vogeli, B. gibsoni, Babesia felis-like, Leishmania spp., H. canis, D. immitis, D. repens, A. reconditum, M. haemocanis and A. phagocytophilum. A total of 56.0% of dogs under investigation were positive for infections by direct and/or 21.3% by indirect methods. B. canis canis was by far the most prevalent pathogen (43.1%) while the others were found at lower prevalence rates of <0.1% up to 8.3%. Except for M. haemocanis, there was no statistical difference in the presence of pathogens in imported dogs and those from the veterinary clinic (p < 0.05). B. felis-like sequences have been submitted to GenBank: JN543169 and JN543170.

Imported dogs

Of all samples, 55.1% were positive by direct methods and 16.5% were positive for antibodies (Table 1). Microscopic examination of blood/buffy coat smears gave no evidence for vector-borne pathogens. Overall, 97 dogs were positive in the Babesia spp.-nested PCR. Two products gave bands in the first reaction, but not in the nested reaction. Sequencing of these products of the first reaction with subsequent BLASTn analysis resulted in two B. felis-like (99% similarity to GenBank accession nos. AF244912 and AY452702). The diagnostic procedure was repeated two times in the case of B. felis-like for confirmation including bidirectional sequencing and revealed each time the same result. Microfilariae were detected in six samples by Knott’s technique. Additional PCR analysis followed by sequencing for species identification revealed one dog from Hungary infected with D. immitis, one from Romania with A. reconditum and two with D. repens (Romania and Hungary). Sequencing further two microfilaraemic samples failed due to low DNA yield. DiroChek®-ELISA was positive in two microfilaraemic samples. Multiple infections were detectable by direct methods in 13 samples: B. canis canis and B. gibsoni (four times), B. canis canis and B. canis vogeli (three times), B. canis canis and M. haemocanis (three times), H. canis and M. haemocanis (once) and B. canis canis and A. phagocytophilum (once), one animal was triple positive for B. canis canis, B. canis vogeli and M. haemocanis. Only 8.0% of the imported dogs had antibodies against B. canis spp., Leishmania spp. (<0.1%) and/or A. phagocytophilum (9.4%). Twenty-nine dogs had borderline titres against Leishmania spp. (16) or B. canis (13). One dog had antibodies against B. canis and Leishmania spp. and one against A. phagocytophilum and B. canis.

Table 1 Results of 216 dogs from Romania and Hungary tested for arthropod-borne pathogens by direct and indirect methods
Full size table

Dogs submitted to Salvavet Veterinary Clinic in Bucharest

Overall detection rate for pathogens in dogs sampled at the Salvavet Veterinary Clinic was 49.5% by direct methods and 48.3% by indirect methods (Table 1). Six Giemsa-stained blood/buffy coat smears were positive for gamonts of large Babesia spp. Eight animals were positive in the Babesia-nested PCR for B. canis canis and all eight had detectable anti-Babesia spp. antibodies. Multiple infections were not detected by direct methods. Seven (24.0%) animals had antibody titres against two pathogens: three (10.3%) against Leishmania spp./B. canis spp., three (10.3%) against A. phagocytophilum/B. canis spp. and one (3.4%) against E. canis/B. canis spp. Three samples presented borderline titres against Leishmania spp. There was no statistical difference between animals treated and not treated against ectoparasites considering presence of infection by direct and/or indirect methods. However, the presence of ectoparasites as reported by pet owners and detection of pathogens by PCR correlated significantly (p < 0.05).

Discussion

A total of 11 different protozoan, helminthic and bacterial pathogens with predominance of Babesia spp. were detected in 187 imported stray dogs from Romania and Hungary and 29 pet dogs from the Bucharest area in Romania.

B. canis canis

More than 40% of all dogs were tested positive for B. canis canis by PCR. B. canis canis has already been characterized by PCR in dogs from Hungary (Földvari et al. 2005; Furlanello et al. 2005). Large Babesia was only detected in Giemsa-stained blood/buffy coat smears from dogs submitted to the Salvavet Veterinary Clinic in Bucharest, Romania, as several of those dogs were presented due to acute signs of clinical disease. Gamonts of large Babesia spp. have also been previously detected in blood/buffy coat smears in Romanian dogs (Ilie et al. 2010). The vector tick Dermacentor reticulatus is a common tick species in these two countries with considerable vector potential (Teodorescu and Popa 2002; Földvari and Farkas 2005; Hornok 2009; Tudor et al. 2010). D. reticulatus has a focal distribution in Germany and some areas in South-Western parts are considered endemic for B. canis canis (Gothe and Wegerdt 1991; Dautel et al. 2006; Heile et al. 2006; Barutzki et al. 2007). Imported dogs infected with B. canis canis could potentially act as reservoir hosts when introduced into non-endemic areas of Germany with a presence of the vector tick. Therefore, PCR screening of imported dogs from countries highly endemic for B. canis canis and treatment of PCR-positive animals should be recommended to avoid an introduction of the pathogen into non-endemic regions of Germany.

B. canis vogeli

Five dogs from Hungary and Romania were positive for the B. canis vogeli-DNA. In Europe, B. canis vogeli primarily occurs in the Mediterranean area, coinciding with distribution of its vector tick, Rhipicephalus sanguineus (Dantas-Torres 2008). This tick is present in Romania and in Hungary, although the latter finding on a Hungarian farm is supposed to be linked to the introduction of R. sanguineus ticks by animal transport from Croatia (Hornok and Farkas 2005; Tudor et al. 2010). While the presence of R. sanguineus in Romania supports the occurrence of B. canis vogeli, positive dogs from Hungary in this study have most likely attained an infection outside of Hungary prior to transfer to Germany.

Despite comparable detection rates of B. canis spp. (∼40%) by PCR in stray dogs and dogs presented at the Salvavet Veterinary Clinic in Bucharest, seroprevalence in B. canis-IFAT did clearly differ. While only 12% of the imported dogs were seroreactive, more than 41% of the dogs presented at the veterinary clinic had detectable antibodies against B. canis spp. Based on these PCR results, we expected a higher seroprevalence in the examined imported dogs. Similar results have been published in dogs from Kosovo and Albania (Lazri et al. 2008; Hamel et al. 2009). The authors assumed that B. canis canis infections must have occurred very recently as seroconversion was not detectable (Lazri et al. 2008). A publication from Hungary, although applying a different cut-off for anti-Babesia antibodies, also reported a relatively low seroprevalence of anti-B. canis spp. antibodies in dogs; PCR was not applied (Hornok et al. 2006). The IFAT result therefore might evolve either from low avidity of antibodies to the applied B. canis antigen, or recent infection, although the animals were considered as healthy at the time of testing.

Our results in dogs from the Salvavet Veterinary Clinic clearly indicate a high incidence of infected vector ticks in the Bucharest area. In a recent study, a high infestation rate of dogs with D. reticulatus was reported, supporting the serological and molecular biological results on B. canis canis (Tudor et al. 2010)

B. gibsoni

Based on morphological criteria, infections attributed to “small” Babesia and B. gibsoni in dogs have previously been published in various areas of Europe, e.g. Italy, Spain, Hungary and Romania (Casapulla et al. 1998; Suarez et al. 2001; Farkas et al. 2004; Ilie et al. 2010). B. gibsoni has only recently been characterized by molecular techniques and may present an underdiagnosed or possibly emerging disease in dogs and in addition also requires a different treatment regime than B. canis spp. (Boozer and Macintire 2005; Garcia 2006). Furthermore, other “small” Babesia spp., Theileria annae or Babesia microti-like Babesiae, are increasingly diagnosed in blood smears and consecutive PCR examination in dogs from Southern Europe (Camacho et al. 2001; Beck et al. 2009; Dezdek et al. 2010; Simoes et al. 2011). Autochthonous B. gibsoni infections in dogs, mainly in American Pitbull Terriers and dogs used in dog fighting, have been confirmed by PCR in several European countries within the last 10 years, e.g. in Croatia, Germany, Italy and Spain (Suarez et al. 2001; Criado-Fornelio et al. 2003a; Garcia 2006; Hartelt et al. 2007; Beck et al. 2009; Trotta et al. 2009). In this study, we present the first molecular identification of B. gibsoni in dogs from Hungary and Romania, confirming previous reports on “small” Babesia based on morphological criteria (Farkas et al. 2004; Ilie et al. 2010). These findings stress the need for further research on B. gibsoni and other “small” Babesia: despite their increasing detection rate both in the New and in the Old World, in particular, the transmission and identification of vector tick species still need to be further elucidated (Solano-Gallego and Baneth 2011).

B. felis-like

BLASTn analysis of the Babesia spp. sequences derived from two imported Romanian dogs revealed 99% similarity to sequences deposited in GenBank as belonging to the species B. felis detected from domestic cats and lions in South Africa (GenBank accession nos. AF244912 and AY452702). The 18S rRNA gene of piroplasma is highly conserved and small amplicons of this gene might hinder accurate species determination by sequencing due to high similarities amongst many species (Baneth et al. 2004). Therefore, it is difficult to interpret this result accurately and further investigations on the potential occurrence of further Babesia species in dogs in South-Eastern European countries will be necessary. B. felis has a major health impact in felids, especially in southern Africa, while only a few cases in cats have been reported elsewhere (Penzhorn et al. 2004). There is no information available on the occurrence of B. felis in canids. In contrast, there is evidence for the susceptibility of cats to B. canis spp. infections (Criado-Fornelio et al. 2003b; Baneth et al. 2004).

Leishmania spp.

A total of six dogs from Romania and Hungary were positive for anti-Leishmania spp. antibodies (four) or for the DNA of Leishmania spp. (two). Despite the recent detection of vector-competent phlebotomine sandfly species in Hungary, screening of indigenous dogs and wild canids gave up to date no evidence for Leishmania infections (Farkas et al. 2011). Vector-competent Plebotomus perfiliewi have been identified in southern Romania but there are no current reports on authochthonous canine or human leishmaniosis available (Danesco 2008; Neghina et al. 2011). Although treatment does not eliminate Leishmania spp. in dogs, treatment regimes with allopurinol alone or in combination with meglumine antimony have been proven to successfully inhibit infection of sandflies due to absence or low presence of amastigotes in the canine host (Miró et al. 2011).

H. canis

Four blood samples from Romanian dogs were positive for H. canis by PCR and consecutive sequencing of the PCR product. However, gamonts of Hepatozoon spp. were not observed in the corresponding blood/buffy coat smears. As the vector tick R. sanguineus has been reported in Romania, the endemic occurrence of H. canis is highly likely (Tudor et al. 2010). In another study, 144 dogs from Timiş County in western Romania screened for vector-borne pathogens also did not reveal the presence of H. canis in blood smears (Ilie et al. 2010). This result might be attributed to the low diagnostic sensitivity of this technique (Karagenc et al. 2006). Imported dogs from Hungary were all tested negative for H. canis. Despite the single finding on a farm, the vector tick can up to date be considered as absent in Hungary (Hornok and Farkas 2005). H. canis has also been identified in dogs and foxes in other countries, e.g. Albania and Croatia and was supposed to be introduced by wild canids, e.g. foxes or golden jackals (Canis aureus), into Slovakia (Majlathova et al. 2007; Lazri et al. 2008; Hamel et al. 2009; Vojta et al. 2009). Golden jackals occur widely in Romania, Serbia and Bulgaria and have expanded their range into Central Europe recently (Nikolova et al. 2001; Blaga et al. 2008; Zachos et al. 2009). Further spreading of vector-borne canine pathogens associated with golden jackals cannot be ruled out in the future.

Filarial infections

A total of seven dogs were positive for filarial infections by Knott’s test, DiroChek®-ELISA and PCR. Three different filarial nematode species, D. immitis, D. repens and A. reconditum, were identified from six microfilaraemic dogs in this study. Our results indicate low infection rates when compared to results from Italy with more than 20% of tested dogs positive for filarial infections (Magi et al. 2011). Canine filarial infections occur worldwide in temperate and tropical regions, whereas D. immitis and D. repens are the predominant species in Europe (Simon et al. 2009). Infections with D. immitis, D. repens and A. reconditum in dogs imported into Germany have been published previously (Glaser and Gothe 1998; Zahler et al. 1997). Single cases of heartworm disease and several cases of canine cutaneous filariosis have been described previously in dogs and a ferret in Hungary and in indigenous Romanian dogs (Olteanu 1996; Coman et al. 2007; Ilie et al. 2009; Jacsó et al. 2009; Tudor et al. 2009; Ciocan et al. 2010a, b; Fernoagă et al. 2010; Fok et al. 2010; Molnar et al. 2010). A. reconditum has until now not been reported from dogs from Hungary and Romania, but has been detected in 4 out of 193 dogs in the neighbouring Vojvodina Province in Serbia (Tasić et al. 2008). Due to the long incubation period and the fact that animals in this study have only been tested once, data presented here are preliminary and require further confirmation.

E. canis

Only one dog from Romania was positive for antibodies against E. canis. As the vector tick is present in Romania, further research might reveal a wider occurrence of E. canis in this country (Tudor et al. 2010). This infection is more common in close-by Balkan countries and prevalence rates of 50% have been detected by IFAT and 17% by PCR in Albania (Hamel et al. 2009).

A. phagocytophilum

A. phagocytophilum is a widespread tick-borne infection with Ixodes ricinus as main vector in Europe (Woldehiwet 2010). Several dogs (10.6%) were seroreactive to A. phagocytophilum antigen in IFAT while only four dogs were positive by PCR (1.9%). Despite the ubiquitous occurrence of this tick in Europe, prevalence rates in dogs vary considerably between different countries, e.g. studies from Germany report around 6% of the dogs positive by PCR (Jensen et al. 2007b; Kohn et al. 2010) and only up to 1% in Poland (Skotarzack et al. 2004; Welc-Faleciak et al. 2009). Recently, A. phagocytophilum-DNA has also been detected by PCR with a minimum prevalence of 1.3% in ticks removed from dogs and foxes in Hungary (Sreter et al. 2004). This prevalence is similar to that (2.9%) reported in I. ricinus collected in southern Germany, although peak prevalence in city parks may reach around 10% (Silaghi et al. 2008)

M. haemocanis

Seventeen dogs from Romania and one from Hungary were tested positive for M. haemocanis-DNA by PCR. M. haemocanis is transmitted by the brown dog tick R. sanguineus and is often found in immunocompromised and splenectomized dogs (Seneviratna et al. 1973; Kemming et al. 2004; Willi et al. 2010). To our knowledge, this is the first report on this agent in dogs from Romania and Hungary. As the brown dog tick is de facto not present in Hungary, the transmission or possible import of this dog into Hungary remains an open question (Hornok and Farkas 2005).

Conclusion

The results of our investigations suggest the need for a routine laboratory screening for vector-borne canine blood infections, particularly in dogs imported from South-Eastern European countries into Germany or other Central and Western European countries. Apart from standard serological methods, more advanced molecular biological techniques should be applied to detect new, hitherto unknown pathogens in these canine populations. Results of such studies should be communicated to veterinarians, dog owners and animal welfare organisations in order to create awareness of impending health risks. Furthermore, prophylactic treatment with recommended parasiticides should be employed on dogs when brought to countries with high incidence of vectors. More research on existing and potential arthropod vectors and their distribution as well as improved identification of arthropod-borne pathogens is deemed necessary in view of increased tourism along with Europe—wide pet trade and imports.