DETECTION OF HERPESVIRUSES IN PASSERINE BIRDS CAPTURED DURING AUTUMN MIGRATION IN SLOVENIA

Some avian herpesviruses (HVs) are well researched and known to cause diseases such as Marek's disease and infectious laryngotracheitis in gallinaceous birds (Galliformes; Kaleta and Redmann 1997; Voelckel et al. 1999; Pennycott et al. 2003), duck virus enteritis in Anseriformes (Jansen and Wemmenhove 1965; Keymer and Gough 1986; Spieker et al. 1996; Campagnolo et al. 2001), Pacheco's disease in parrots (Psittaciformes; Randall et al. 1979), and inclusion body disease or herpesvirus hepatitis in pigeons (Columbiformes), owls (Strigiformes), and birds of prey (Falconiformes; Saik et al. 1986; Gailbreath and Oaks 2008). In passerines, HVs have been detected in sick or dead captive-bred Canaries (Serinus canaria f. domestica; Widen et al. 2012), Gouldian Finches (Erythrura gouldiae; Wellehan et al. 2003; Paulman et al. 2006; Widen et al. 2012), and caged Superb Starlings (Lamprotornis superbus; Tomaszewski et al. 2004). All avian HVs are members of the genera Iltovirus and Mardivirus of the subfamily Alphaherpesvirinae; however, many viruses detected in wild birds have not been completely characterized and therefore remain unassigned in the family Herpesviridae (Kaleta and Docherty 2007). Clinical manifestations from HV infection in passerines are poorly understood; however, in captive-bred finches, clinical signs such as reduced activity; weight loss; labored breathing; respiratory snicks; unilateral or bilateral conjunctivitis with red, swollen, crusty, adherent eyelids; and high morbidity and mortality were reported (Paulman et al. 2006). In addition to dead or sick birds, HVs have also been detected in healthy captive-bred White-rumped Munias (Lonchura striata), Bronze Mannikins (Lonchura cucullata), Northern Cardinals (Cardinalis cardinalis), and Zebra Finches (Taeniopygia guttata; Widen et al. 2012). Recently, two partial DNA polymerase gene and UL16 gene sequences of HV were published: passerine HV1 (PaHV1), detected in captive-bred caged finches in Canada and the US, and psittacid HV1 (PsHV1), detected in caged Superb Starlings (Wellehan et al. 2003; Tomaszewski et al. 2004; Paulman et al. 2006).

The isolation and molecular characterization of HVs from free-living passerine birds is rarely described (Widen et al. 2012), and only one report appears to be published about the detection and partial characterization of HV in free-living passerine birds; namely, the Hooded Crow (Corvus cornix) and the Song Thrush (Turdus philomelos; Woźniakowski et al. 2013). The viruses detected were classified as columbid HV1 through analysis of partial DNA polymerase gene sequences (Woźniakowski et al. 2013). Because the current understanding of HV in free-living passerine birds is quite limited, we sought to expand upon the current knowledge of HV infections in free-living passerine birds. Our aims were to investigate the occurrence of HV in free-living passerine birds and to characterize and compare the virus sequences with previously known HV of birds. We used PCR to detect HV DNA in free-living passerine birds captured and sampled during the autumn migration in Slovenia. The HV DNA that we detected was characterized by direct sequence analysis of the PCR product to provide greater insight into the phylogenetic relationships and epidemiology of the HVs detected.

Free-living passerine birds were captured with mist nets during the autumn migration, from the end of September to early October 2014 and 2017, at the bird ringing station near Vrhnika in central Slovenia as part of a bird migration study (Vrezec and Fekonja 2018). Each bird was clinically examined by a veterinarian, and 525 cloacal swabs from 32 different passerine species were collected (Table 1). A total of 292 cloacal swabs from 30 species and 233 cloacal swabs from 16 species was collected in 2014 and 2017, respectively. All birds were sampled with special sterile microswabs (Deltalab, Barcelona, Spain) to avoid cloacal damage. All sampled birds were clinically healthy, and no birds were harmed during the sampling procedure. All birds were released after ringing at the site of capture soon thereafter. Dry swabs were placed in the refrigerator at 4 C stationed at the ringing station facility. At the end of each day, samples were transported in an isolated cooling bag and finally stored at –20 C for up to 10 d until analyzed.

Cloacal swabs were individually vortexed in 2 mL of phosphate-buffered saline for 2 min, and 100-µL aliquots of each swab in phosphate-buffered saline were pooled to produce 500-µL samples for genomic nucleic acid extraction. Samples were pooled by the date of sampling.

Total DNA and RNA were extracted from 140 µL of pooled samples by the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Viral DNA was detected by nested PCR with a set of primers that target the HV DNA polymerase gene region as described by VanDevanter et al. (1996). The PCR volume was 20 µL, and it contained 10 µL of 2X DreamTaq Green PCR Master Mix (Thermo Scientific, Dreieich, Germany), 1 µM of each PCR primer, 2 µL of isolated DNA, and deionized water up to 20 µL. The parameters for prime and nested PCR were denaturation at 95 C for 5 min, followed by 45 cycles of denaturation at 94 C for 30 s, annealing at 46 C for 1 min, extension at 72 C for 1 min, and final extension at 70 C for 7 min. Individual samples from positive pools of samples were tested individually, as previously described.

Amplified products from individual samples were separated by electrophoresis on a 1.8% agarose gel (Sigma-Aldrich, St. Louis, Missouri, USA) containing ethidium bromide. The PCR products of 215–315 base pairs were excised and purified with a FastGene Gel/PCR Extraction Kit (Nippon Genetics, Duren, Germany) and sent for sequencing to the Macrogen Laboratory (Macrogen Inc., Amsterdam, the Netherlands). The nucleotide sequences obtained were first analyzed by BLAST (Altschul et al. 1990) to identify sequences relevant for further analyses (National Center for Biotechnology Information 2019). Nucleotide alignments were constructed in Geneious Prime 2019 software suite version 1.3 (Biomatters Ltd., Auckland, New Zealand) with MAFFT translation alignment (Katoh and Standley 2013). Phylogenetic analysis was performed using the maximum likelihood method with the Tamura 3-parameter model and 1,000 bootstrap replicates by MEGA 7.0 (Kumar et al. 2016). The percentage of similarity among sequences was calculated by the p-distance model (pairwise distance) in MEGA 7.0. Mustelid HV1 (accession AF376034) was selected as the outgroup sequence. The accession numbers of HV sequences obtained in this study are MN274972–MN274975.

The HVs were detected by nested PCR with a set of primers that target a region of the HV DNA polymerase gene in four out of 525 (0.76%) free-living passerine birds. Herpesviruses were detected in three passerine species: a Eurasian Blue Tit (Cyanistes caeruleus; 1/8; 12.60%) and two Eurasian Blackcaps (Sylvia atricapilla; 2/131; 1.53%) in 2017 and a Common Blackbird (Turdus merula; 1/37; 2.70%) in 2014. Herpesvirus was not detected in 29 other passerine species (Table 1).

The partial nucleotide HV sequences (174 nt) detected in four Slovenian free-living passerine birds in 2014 and 2017 were compared with the sequences of DNA polymerase gene of other avian and mammal HVs to determine their phylogenetic relationship. The accession numbers of other HV sequences used for phylogenetic analysis are included in Figure 1. The phylogenetic analyses showed that these four HV sequences are most closely related to the alphaherpesvirus sequences detected in other bird species (Fig. 1). More precisely, the sequences of detected HVs clustered together with novel HV sequences detected in the Sulphur-crested Cockatoo (Cacatua galerita) and Neotropic Cormorant (Phalacrocorax brasilianus), as well as with known HVs such as gallid HV1 (GaHV1), PaHV1, PsHV1, and PsHV2 (Fig. 1). Known HV sequences detected in pigeons and birds of prey, such as columbid HV1, falconoid HV1, and GaHV2 and meleagrid HV1, as well as novel HV sequences detected in owls, were grouped together in the other cluster (Fig. 1). The sequences detected in 2017 in Eurasian Blackcaps were identical and were most similar (73.1%) to the sequence detected in the Common Blackbird in Slovenia in 2014. When compared with known HV sequences, the highest identity (72.5%) was observed with the currently unclassified PsHV2 sequence detected in the African Grey Parrot (Psittacus erithacus). Lower similarity (70.8% and 67.8%) was observed between the HV sequence detected in the Eurasian Blue Tit in 2017 and the other three HV sequences detected in Eurasian Blackcaps and the Common Blackbird in Slovenia, respectively. When compared with known HV sequences, the HV sequence detected in the Eurasian Blue Tit shared the highest identity (66.1%) with the HV sequence detected in the Gouldian Finch, whereas the HV sequence detected in the Common Blackbird shared the highest identity (72.5%) with the sequence of PsHV1.

This study detected HVs in free-living passerine birds caught and sampled during the autumn migration in Slovenia. The occurrence of HVs in free-living passerines in our study was low (0.76%), with only four birds out of 525 testing positive. Generally, only a few studies have been published on the prevalence of HVs in free-living birds, which is apparently low but present, as this study also shows. However, it appears that the prevalence of HV-associated infections differs among free-living bird species. In wild raptor species, detections of HV infections range from 14.5% (8/55; Žlabravec et al. 2018) to 17.8% (8/45; Woźniakowski et al. 2013) for owls in Slovenia and for selected birds of prey in Poland, respectively. A slightly lower HV prevalence ranging from 3.85% (4/104; Verdugo et al. 2019) to 5.6% (14/250; Niemeyer et al. 2017) was detected in seabirds. The reasons for these differences in HV prevalence detected among different bird species could be complex, involving factors such as the biology and ecology of the bird species tested, intermittent shedding of HVs by the cloaca, trachea, or both, and the immune status of infected birds (Kaleta and Lierz 2016). The differences may also arise from varying study approaches; in previous studies, HVs were mostly detected in dead birds (Woźniakowski et al. 2013; Žlabravec et al. 2018), in which the prevalence might be higher than in living and clinically healthy birds, and our findings were based on detection of HVs only in cloacal swabs. Hence, the results should be treated with considerable caution because a higher occurrence could be expected if oropharyngeal swabs had also been investigated (Phalen et al. 2017). Moreover, the low detection rate could be linked to some technical reasons. Collection, storage, and transport conditions of samples are important factors for an accurate and reproducible diagnosis (Borsanyiova et al. 2018). Although swabs with viral transport media are better for virus recovery and detection than dry swabs (Spackman et al. 2013), some studies have reported successful usage of dry swabs in the diagnosis of viral infection such as in the families Orthomyxoviridae (Druce et al. 2012) and Paramyxoviridae (Moore et al. 2008), and in the family Herpesviridae (Boppana et al. 2011). Because the samples from passerine birds were taken and transported similarly to procedures described in these studies, we believe that only storage of the samples at –20 C for up to 10 d could have some influence, but not significant, on the low detection rate of HV.

Although HVs were mostly detected in sick or dead passerines in past studies, some reports describe the detection of HV in apparently healthy captive-bred caged passerine birds (Widen et al. 2012). As in these cases, all HV-positive passerines in our study lacked clinical signs of disease. It is assumed that most natural HV infections are acquired during early life, and many healthy birds could remain viremic or persistently infected for prolonged periods of time. Some internal or environmental stress factors could probably alter the host–viral balance and lead to later development of the disease (Ritchie 1995; Ramis et al. 1996; Phalen 1997).

On the other hand, a potential benefit of HV as part of a symbiotic relationship with the host needs to be considered. Numerous studies have shown that asymptomatic HV infection has a protective effect on hosts against viral and bacterial infection (Barton et al. 2007). Mice infected with murine gammaherpesviruses were more resistance to infection with Listeria monocytogenes and Yersinia pestis and had a higher survival rate for influenza A virus infection (Barton et al. 2007; Saito et al. 2013; Furman et al. 2015). Under certain circumstances, chronic HV infection was related to increased resistance to tumors in mice (White et al. 2010; Raffegerst et al. 2015). Therefore, detection of HVs in cloacal swabs of healthy animals suggests that the relationship between host and virus likely includes both symbiosis and infection that could result in disease.

The use of a relatively short fragment of the DNA polymerase gene of mammalian and avian HVs for phylogenetic analyses is well supported by numerous studies (Ehlers et al. 1999; Li et al. 2000; Pagamjav et al. 2005; Gailbreath and Oaks 2008; Woźniakowski et al. 2013). The phylogenetic analyses showed that the HV sequences detected in four free-living passerine birds in our study are the most closely related to the alphaherpesvirus sequences detected in different bird species (Fig. 1). More precisely, the HV sequences detected clustered together with HV sequences detected in Sulphur-crested Cockatoo and Neotropic Cormorants, as well as with better-known HVs such as GaHV1, PaHV1, PsHV1, and PsHV2. Despite the limited sequence data on HVs detected in passerine birds, it could be that different sequences of HVs with relatively low identity were detected (Fig. 1). This finding points to the probability that different HVs were circulating in the passerine population in our study. Our detection of identical HV sequences in two Eurasian Blackcaps is consistent with the detection of identical HV sequences in two separate outbreaks of the disease in captive-bred Gouldian Finches (Wellehan et al. 2003; Paulman et al. 2006; Fig. 1). These findings suggest that HVs detected in passerine birds might be species specific; however, their ability to cross the tissue or host barriers described for HVs remains unknown (Gerlach 1994).

All infected species in our study are short-distance migrants or sedentary species, with the highest occurrence being detected in the Eurasian Blue Tit (12.6%), but this could be a consequence of the small sample size of this species (n=8). Surprisingly, no infected long-distant migrants were found, although they were included in the study in reasonably large numbers, including Reed Warblers (Acrocephalus spp.), Garden Warblers (Sylvia borin), and Common Chiffchaffs (Phylloscopus collybita). Migration has extremely selective power on migrating birds (Newton 2003), selecting against less fit birds (i.e., infected specimens). Selective pressure is high, even in short-distance migrants, as indicated by the very low HV infection occurrence we found. Susceptibility to HV infection should be studied in the future by combining sampling on living birds as well as dead birds, in which the occurrence is expected to be higher if the species is susceptible. Our phylogenetic analysis also showed that HV sequences detected in free-living passerine birds in Slovenia are grouped together with HV sequences detected in exotic parrots, passerines, and poultry. Future studies should reveal whether the HVs found in passerines are of indigenous European origin or if they were transferred from imported exotic birds to wild avifauna, which is already a known transmission pathway for various diseases (Lockwood et al. 2013).

Our findings indicate that HVs with different partial DNA polymerase gene sequences have been circulating in the population of free-living songbirds caught during the autumn migration in Slovenia. Further work needs to be done to establish whether the partial viral sequences detected are novel HVs circulating in the passerine population and to study the epidemiology of HVs and their potential effects on the health of free-living songbirds.

The authors thank Tjaša Sernel and Darja Krelj from the Institute of Poultry, Birds, Small Mammals, and Reptiles, Faculty of Veterinary Medicine (University of Ljubljana, Slovenia), and Tea Knapič from the Slovenian Museum of Natural History (Ljubljana) for technical assistance. We are also grateful to Urška Kuhar from the Institute of Microbiology and Parasitology, Faculty of Veterinary Medicine (University of Ljubljana, Slovenia), for help with phylogenetic analysis. The study was supported by the Slovenian Research Agency, Junior Researchers grant 50525, Environment and Food Safety project group P4-0092, and it was part of the program “Communities, relations, and communications in ecosystems” (P1-0255) financed by the Ministry of Education, Science, and Sports of the Republic of Slovenia. The authors declare that they have no competing interests. This study was carried out with animal ethics approval by the Ministry of the Environment and Spatial Planning, Slovenian Environment Agency (documents 35601-125/ 2009-8 and 35601-10/2010-6).