Abstract
Climate change, with warmer temperatures and altered precipitation patterns, has affected the distribution of vectors and vector-borne diseases. In the northern hemisphere, vectors are spreading north, and with them, pathogens of zoonotic and animal health impact. Eurasian moose (Alces alces alces) are physiologically and anatomically adapted for cold climate, and are rarely considered ideal hosts of vectors, apart from deer keds (Lipoptena cervi). To investigate the presence of vector-borne pathogens, spleen samples from 615 moose were collected in southern Sweden from 2008 to 2015. The samples were analyzed with a high-throughput PCR method for 24 bacterial, and 12 parasitic pathogens. Anaplasma (82%), Borrelia (3%), Babesia (3%), and Bartonella (1%) DNA was found, showing that moose are exposed to, and can act as hosts of some of these pathogens, which can have an impact of both animal and human health. These results show that Swedish moose are exposed to pathogens that in some instances are more commonly found in regions with warmer climate, and highlights the importance of also considering moose as sentinels of vector-borne pathogens. Further research is needed to understand the effect of these pathogens on the health of individual moose and to elucidate whether climate change and moose population density interact to create the pattern observed.
Introduction
Globally, different types of vector-borne pathogens, such as virus, bacteria, and parasites, with potential impact on human and animal health, are more frequently detected (Kilpatrick and Randolph 2012, Suk and Semenza 2014). The change in climate during the last decades has been suggested as a plausible cause for the increase in distribution and presence of pathogens and their vectors, such as ticks (Gray et al. 2009, Medlock and Leach 2015). Higher temperatures and shorter winters positively affect survival, abundance, and spread of vectors (Ogden and Lindsay 2016). Concurrently, vector hosts are also increasing in different parts of the world (Brites-Neto et al. 2015), and in Europe in particular (Gray et al. 2009, Apollonio et al. 2017).
Ungulates, especially cervids (Cervidae), can act as key players in some host–vector–pathogen systems (Rosef et al. 2009), and it is shown that Europe today has larger cervid populations than ever before (Deinet et al. 2013). Thus, there are several factors to consider, when elucidating the suggested increase and distribution of vector-borne diseases.
Eurasian moose (Alces alces alces), is a large ungulate browser found in boreal forests of the northern parts of Europe and Asia. In Fennoscandia (Norway, Sweden, Finland), ∼30% of the population is harvested annually (∼160,000 animals, Olaussen and Skonhoft 2011). Moose prefer cool summers and cold winters due to the species' physiological and anatomical characteristics (McCann et al. 2013). In Fennoscandia, deer keds (Lipoptena cervi), hematophagous arthropods, are found in large numbers on moose (Samuel et al. 2012). Deer keds are not considered to be as important as ticks in the spread of pathogens, mostly because they spend their entire adult life on their host animals (Samuel et al. 2012). However, they have been shown to harbor Bartonella spp. (Duodu et al. 2013, Korhonen et al. 2015, Perez Vera et al. 2016), and Anaplasma spp. (Hornok et al. 2011, Víchová et al. 2011).
As climate changes, ticks (Ixodes and Haemaphysalis spp.) are expanding their range northward in Europe, and Ixodes ricinus can now be observed in most parts of Scandinavia (Medlock et al. 2013), including northern Sweden (Jaenson et al. 2012). The deer ked has also increased its distribution range in northern Europe over the last decades (Välimäki et al. 2012).
A variety of vector-borne pathogens have been discovered in ticks collected by flagging in Scandinavia, for example, Borrelia spp., Anaplasma phagocytophilum (Skarphédinsson et al. 2007, Rosef et al. 2009, Mysterud et al. 2013), and Candidatus Neoehrlichia mikurensis, Bartonella spp., and Babesia sp. (Stensvold et al. 2015) collected from dogs. Previous studies from Fennoscandia have shown that moose likely can act as natural hosts, and possibly also reservoirs of A. phagocytophilum (Milner and van Beest 2013, Malmsten et al. 2014, Pūraite et al. 2015). This may have implications from a zoonotic and domestic animal health perspective. In addition, it suggests that moose may be a previously overlooked species in the spread and maintenance of emerging vector-borne zoonotic pathogens in northern Europe.
The aim of this study was to investigate the occurrence of selected known and hitherto unknown bacterial and parasitic vector-borne pathogens in moose from southern Sweden.
Materials and Methods
Spleen samples from 615 hunter-harvested moose in southern Sweden were collected from 2009 to 2015 during ordinary hunting from October to January. Participation of hunters in the sample collection was voluntary, and samples were taken from culled animals. The data collection was thus not biased toward diseased individuals. Hunters were instructed by researchers in how to collect samples in the field, and in some cases, field staff were on site to collect samples. Of the sampled animals, 42% (n = 258) were juveniles, 19% (n = 117) were subadults 1–2 years of age, and 39% (n = 240) were adults. The sex distribution among the sampled animals was 48% female and 52% male. Small (5 × 5 cm) pieces of the spleen were put in plastic zip-lock bags, before being frozen at −20°C. After thawing, DNA was extracted as previously described (Michelet et al. 2014). DNA was then preamplified using the PerfeCTa PreAmp SuperMix Kit (Quanta Biosciences, Beverly) according to the manufacturer's instructions. Thermal cycling conditions were as follows: 1 cycle at 95°C for 10 min, 14 cycles at 95°C for 15 s, and 4 min at 60°C (Michelet et al. 2014).
High-throughput real-time PCRs, screening for 24 bacterial species, 12 parasitic species, and 3 bacterial genera (Table 1), were then performed using TaqMan probes with TaqMan Gene Expression Master Mix, in accordance to the manufacturer's instructions (Applied Biosystem, Thermo Fisher Scientific, Courtaboeuf Cedex, France). Twenty-seven different target genes were used in accordance with Michelet et al. (2014). PCR cycling comprised 5 min at 95°C, 45 cycles at 95°C for 10 s, 15 s at 60°C, and 10 s at 40°C (Michelet et al. 2014). Data were acquired on the BioMark Real-Time PCR system and analyzed using the Fluidigm Real-Time PCR Analysis software to obtain crossing point values. The assays were performed in duplicate using two negative water controls per chip and Escherichia coli strain EDL933 was added in each run to control for internal inhibition (Michelet et al. 2014).
Species of Bacteria and Parasite Analyses with a High-Throughput Real-Time PCR Screening of Spleens from Hunter-Harvested Swedish Moose from 2009 to 2015
To identify the bacterial and parasitic species present in the samples, nested PCRs or real-time PCRs using different primers than those of the BioMark® system were used (Table 2). Amplicons were sequenced by Eurofins MWG Operon (Ebersberg, Germany) and assembled using the BioEdit software (Ibis Biosciences, Carlsbad). An online BLAST (National Center for Biotechnology Information) was used to identify the sequenced organism.
Primers Used to Confirm the Presence of Bacterial and Parasitic DNA in Moose Samples
Results
In the 615 samples, DNA from 4 bacterial and 2 parasitic pathogens was detected (Table 3). A. phagocytophilum DNA was found in the majority (82.1%, n = 505) of the samples, whereas Borrelia, Rickettsia belonging to the Spotted fever group, and Bartonella, were found in 0.2–3.3% of the samples (n range: 1–20, Table 3). Parasite DNA (Babesia venatorum and Babesia spp.) was found in 3.3% and 1.1%, respectively (Table 3). In 83.4% (n = 513) DNA from at least one pathogen was found. In 8.0% (n = 49) DNA from at least two, and in 0.3% (n = 2) DNA from three pathogens were found.
Results of a High-Throughput Real-Time PCR Screening of Spleens from Hunter-Harvested Swedish Moose from 2009 to 2015
Bartonella schoenbuchensis.
Species identifications were performed by nested PCRs or real-time PCRs and sequencing. A. phagocytophilum and B. venatorum analyses were identical to the A. phagocytophilum str. Norway variant2 (GenBank CP015376; acc. no. MG646677) and the B. venatorum isolate Bv4 (GenBank KX008041; acc. no. MG646678), respectively. Babesia divergens was not confirmed by nested PCR and sequencing. Bartonella schoenbuchensis DNA was identified in the four positive Bartonella spp. samples, and had a nine 7% similarity with the B. schoenbuchensis R1 strain (GenBank CP019789; acc. no. MG646676). Nested PCRs targeting Borrelia spp. and Rickettsia spp. did not reveal the Borrelia and Rickettsia species, possibly due to the inability to provide a band of expected size, or a lower-than-expected intensity of the band.
Discussion
In this large-scale screening for vector-borne bacteria and parasites, we provide the first reported finding of DNA presence of Borrelia spp., and B. venatorum in moose. Babesia spp., is with a high certainty B. divergens, which makes this result novel in moose. The presence of these pathogens, along with previous findings of DNA from A. phagocytophilum and Bartonella DNA in Swedish (Guy et al. 2013, Malmsten et al. 2014) and Finnish moose (Perez Vera et al. 2016), show that subarctic ungulates may have the capability to act as hosts of vector-borne pathogens. From a European perspective, and acknowledging the fact that moose also are expanding their distribution range on the continent (Deinet et al. 2013), our findings indicate that moose can act as indirect vectors of pathogens known to infect mammals—domestic and wild. The finding of Borrelia DNA in spleen samples can reflect a high tick burden, which corresponds with increasing tick abundance (Jaenson et al. 2012) and increasing incidence of human borreliosis cases in southern Sweden (Bennet et al. 2006). Contamination of samples during evisceration of moose, or when handling samples in the laboratory is unlikely, but cannot be completely excluded. False positive results are unlikely, based on the high specificity and sensitivity of the method used. Babesia (B. capreoli, B. odocoilei-like Babesia), and Bartonella have been reported in moose in Norway (Duodu et al. 2013, Pūraite et al. 2016), as well as in Finland (Korhonen et al. 2015), and the present study confirms that these pathogens also are present in southern Sweden, although, and interestingly, different species of Babesia (B. spp./divergens and B. venatorum) were detected. B. divergens was, as stated previously, not confirmed by nested PCR and sequencing, which may have occurred due to a poor DNA signal in the sample. In addition, preamplification of DNA followed by real-time PCR is more sensitive than a nested PCR, and the previously confirmed high specificity of the method (Michelet et al. 2014) used in this study, provide satisfactory evidence that the Babesia spp. finding indeed is B. divergens. B. divergens is present in Swedish cattle (Andersson et al. 2017), and is reported to infect roe deer (Duh et al. 2005). Nevertheless, the low prevalence observed in the present study, and given the fact that moose and grazing cattle sometime share habitat, it is not likely that moose have or will play an important role in the epidemiology of B. divergens in Sweden. Swedish roe deer are reported to carry B. venatorum (Andersson et al. 2016), but to a small degree, but have higher tick loads, compared with moose, which can explain the low observed prevalence. As with B. divergens, there are zoonotic implications to consider, but as a reservoir or natural host, moose likely play a minor or no role.
A. phagocytophilum is seemingly common in the southern Swedish moose population, and compared with a previous study conducted in this region, the prevalence of DNA-positive animals has seemingly increased substantially from 26% (Malmsten et al. 2014) to 82% (this study). However, the difference in prevalence more likely reflects the differences in sensitivity of the two methods used to detect bacterial DNA: preamplification of DNA and the use of real-time PCR (this study) versus real-time PCR without preamplification (Malmsten et al. 2014). Subpopulation differences in prevalence could account for the increase observed, although this is not as likely, given the fact that both studies partly include the same time span, adjacent areas, and sometimes overlapping sampling areas. Anaplasma can cause clinical disease in domestic animals (Stuen 2007, Granquist 2016), and may also have a negative effect on moose health (Jenkins et al. 2001, Malmsten et al. 2014). As moose are present all over Fennoscandia, habitat overlap with grazing domestic animals occur, which may facilitate transmission of Anaplasma from wildlife to domestic animals, and vice versa. Furthermore, and considering the high prevalence in moose in this study, its zoonotic potential and risk of indirect transmission to humans should not be neglected, as Anaplasma-related disease is presumed to be underdiagnosed (Bakken and Dumler 2015).
Besides ticks, other potential vectors, such as deer keds, may play a role in the spread of pathogens, as is the case for Bartonella (Duodu et al. 2013, Korhonen et al. 2015, Perez Vera et al. 2016). Thus, the findings of some of the pathogens, like B. schoenbuchensis in the present study, can also be a result of the significant increase in the presence of, and distribution range of deer keds (Välimäki et al. 2012). To our knowledge, only Bartonella, (Duodu et al. 2013, Korhonen et al. 2015), including B. schoenbuchensis, and Anaplasma-DNA (Víchová et al. 2011) have been found in deer keds. This implies that further investigations of the role of deer keds as vectors of pathogens are warranted. The high prevalence (73%) of Bartonella in Finnish moose (Perez Vera et al. 2016) compared with the present study, indicates a substantial geographic difference not only dependent if it is a deer ked zone or not. However, deer keds are presumed to have been present in Finland longer than in Sweden (Välimäki et al. 2012), and are discussed to be a primary vector of different Bartonella species (Perez Vera et al. 2016), including B. schoenbuchensis. Further geographical spread of vector-borne pathogens through ticks and deer keds can be assisted by moose. Moose often migrate distances exceeding 200 km in Sweden (Singh et al. 2012), and may therefore account for a more extensive spread of pathogens than lagomorphs or rodents.
Conclusion
The present study shows that Swedish moose are exposed to pathogens that are commonly found in other species, and regions in Europe (Hartemink and Takken 2016). Further research is needed to elucidate to what extent moose can enhance the spread and distribution of vectors and relevant pathogens, and act as sentinels of the same. In addition, studies on the clinical effect of these pathogens on moose health is warranted, as well as how climate change and the general increase of cervids and vectors interact.
Footnotes
Acknowledgments
The authors thank the many hunters who volunteered to collect samples for this study. They also thank the Swedish Veterinary Institute and the ANSES-Animal Health Laboratory. J.M. and A-.M.F. were funded by the Swedish Environmental Protection Agency and the Södra Research Fund.
Author Disclosure Statement
No conflicting financial interests exist.
