Genospecies of Borrelia burgdorferi sensu lato detected in 16 mammal species and questing ticks from northern Europe

Lyme borreliosis is the most common vector-borne zoonosis in the northern hemisphere, and the pathogens causing Lyme borreliosis have distinct, incompletely described transmission cycles involving multiple host groups. The mammal community in Fennoscandia differs from continental Europe, and we have limited data on potential competent and incompetent hosts of the different genospecies of Borrelia burgdorferi sensu lato (sl) at the northern distribution ranges where Lyme borreliosis is emerging. We used qPCR to determine presence of B. burgdorferi sl in tissue samples (ear) from 16 mammalian species and questing ticks from Norway, and we sequenced the 5S–23 S rDNA intergenic spacer region to determine genospecies from 1449 qPCR-positive isolates obtaining 423 sequences. All infections coming from small rodents and shrews were linked to the genospecies B. afzelii, while B. burgdorferi sensu stricto (ss) was only found in red squirrels (Sciurus vulgaris). Red squirrels were also infected with B. afzelii and B. garinii. There was no evidence of B. burgdorferi sl infection in moose (Alces alces), red deer (Cervus elaphus) or roe deer (Capreolus capreolus), confirming the role of cervids as incompetent hosts. In infected questing ticks in the two western counties, B. afzelii (67% and 75%) dominated over B. garinii (27% and 21%) and with only a few recorded B. burgdorferi ss and B. valaisiana. B. burgdorferi ss were more common in adult ticks than in nymphs, consistent with a reservoir in squirrels. Our study identifies potential competent hosts for the different genospecies, which is key to understand transmission cycles at high latitudes of Europe.


Results
Infection of B. burgdorferi sl was found in all rodent and shrew species except in the single house mouse (Mus musculus) found (Table 1). With the exception of red squirrels, all sequences of the IGS in rodents and shrews were consistent with infections of B. afzelii. In 17 red squirrels, we found high infection rates of B. burgdorferi sl (88%), of which 9 sequences yielded four B. afzelii, two B. garinii and three B. burgdorferi ss. Infection of B. burgdorferi sl was also found in red fox (Vulpes vulpes) and badger (Meles meles), but not in the single hare (Lepus timidus) included. Attempts to determine the genospecies in red fox and badger were unsuccessful. No infection of B. burgdorferi sl was found in 111 moose, 28 roe deer or 141 red deer (Table 1). In the eastern region, the infection levels were higher in squirrels (Z = 4.060, P < 0.001) and lower in the Eurasian pygmy shrew (Sorex minutus) (Z = −2.922, P = 0.003) compared to the common shrew (S. araneus), the latter had similar infection levels as the bank vole (Myodes glareolus) (Z = 0.206, P = 0.837) and the wood mice (Apodemus sylvaticus) (Z = 0.184, P = 0.854).
In questing nymphal ticks with infection of B. burgdorferi sl (Table 2), the genospecies B. afzelii dominated both in the Sogn & Fjordane (66.7%) and Møre & Romsdal (75.2%) county in the western region, with B. garinii being common in both Sogn & Fjordane (26.5%) and Møre & Romsdal (21.0%), while B. valaisiana was less commonly found (Sogn & Fjordane: 6.1%, Møre & Romsdal: 3.8%). Sample sizes for the eastern region was low, and yielded only B. afzelii in nymphs (n = 8). The same main geographic pattern in terms of genospecies distribution was found in adult male and female ticks. The exception was B. burgdorferi ss that were found in 6 adult ticks and www.nature.com/scientificreports www.nature.com/scientificreports/ one nymphal tick overall, and a logistic regression confirmed a higher frequency of B. burgdorferi ss in adults than nymphs compared to B. afzelii (Z = −2.071, P = 0.044).
To ensure that the B. garinii classification based on sequencing of 5S-23S rDNA intergenic spacer region (IGS) from positive samples was correct and not included B. bavariensis (see Methods), we also ran another genetic marker, namely pepX, on the 83 positive samples for B. garinii, of which 51 yielded successful sequences. We confirmed B. garinii in 49 of the 51 samples, while two cases (one red squirrel in Akershus, one tick nymph in Sogn & Fjordane) came out as B. afzelii, likely due to infection with both B. garinii and B. afzelii (see Discussion). www.nature.com/scientificreports www.nature.com/scientificreports/

Discussion
Determining the competence of vertebrate hosts to pathogens is one of several keys necessary to estimation of disease hazard. A competent host is defined by the ability to be a source of infection to ticks 14,28 . A competent host must be fed on by infected ticks and take up a critical number of the pathogen, and it must allow the pathogen to multiply and to pass on infection to new ticks 28 . Hence, establishing infection is necessary, but on its own does not prove amplification of a pathogen in a host or transmission to a vector. In some cases, DNA from B. burgdorferi sl can be found without evidence of amplification or transmission to vector ticks 29 , and PCR cannot separate live and dead pathogen DNA. Several other technical issues, such as the sensitivity and specificity of the PCR, tissue tropisms of the pathogen and timing of host collection can also impede the identification of competent hosts. Although the absence of infection can confirm a vertebrate as an incompetent host, lack of pathogen detection by PCR due to technical limitations, cannot firmly establish this. Ideally, laboratory-based transmission experiments 30,31 , as well as field studies focusing on vertebrate communities 32 , are both necessary and complimentary approaches to acquire full understanding of the transmission dynamics of tick-borne pathogens. Our field study hence provides one step towards gaining the required information to identify important potential competent and incompetent hosts at northern latitudes of Europe.
A new potential competent host of B. afzelii. Due to its medical importance and relevance of determining disease hazard, there are a huge number of studies reporting frequency of genospecies of B. burgdorferi sl in Europe. The dominating genospecies in questing ticks is the genospecies B. afzelii almost universally across continental Europe 33 , while B. garinii dominates in the UK 8 . Our estimates of 75% and 67% B. afzelii in the two western counties are close to previous estimates from Norway both in south with 62% 34 , 86% in east and south 35 , and 68% further north of the western study site reported here 36 . This main picture was also found in Sweden 37 . The B. garinii is typically the second most common genospecies mainly linked to birds 38,39 . Our estimated prevalence in B. burgdorferi sl infected ticks of 21% and 26% B. garinii in the two western counties are also close to previous estimates of 23.4% B. garinii in south 34 , 12% in east and south 35 and 20.8% in west 36 of Norway. Further, as reported in these studies, we also found a low prevalence of the bird-borne, but less pathogenic, B. valaisiana.
The commonness of B. afzelii in ticks is linked to the high abundance and spread of small mammals, which is its main reservoir. As commonly reported, the bank vole and the wood mice are dominant hosts of larvae and have often high infection levels with B. afzelii. Wood mice and yellow-necked mice are also considered to be important hosts of B. burgdorferi sl in urban environments 40 . As we reported earlier 41 , the common shrew is abundant and important in feeding tick larvae in Norway, as also found in Scotland 42 . Based on this study, we can add the taiga shrew (S. isodon) to the list of potential competent hosts for B. afzelii in Europe. The taiga shrew has larval tick loads similar to the common shrew (own unpublished data), but prevalence of B. burgdorferi sl was lower (Table 1). This is a rare species in Norway and Red Listed as 'data deficient' . However, the species is common in Finland 43 and further east in Russia. Therefore, the host species involved in transmission may differ in the north relative to continental Europe. There is also a high level of endemism in the Mediterranean small mammals 44 , suggesting that species involved in circulation of B. afzelii differs with latitude.
Medium-sized hosts infections. Squirrels are sufficiently large in size to be commonly bitten by nymphal ticks [14][15][16] , which is required to obtain infections of B. burgdorferi sl 8 . Their movements on the ground make them regularly exposed to ticks. We indeed found a high infection prevalence in red squirrels. Interestingly, we found infection of three genospecies in red squirrels: B. afzelii, B. burgdorferi ss, and B. garinii. Squirrels seem to regularly be infected by several genospecies of B. burgdorferi sl. In Switzerland, a mixture of B. afzelii and B. burgdorferi ss was found in red squirrels 14 . In France, prevalence of 18.9% B. burgdorferi s.s., 11.9% B. afzelii, and 3.5% B. garinii was found in red squirrels 15 . All four genospecies (B. afzelii, B. burgdorferi ss, B. garinii, B. valaisiana) that occur in the United Kingdom were detected in gray squirrels, and the commonly bird-associated B. garinii was most common in the gray squirrels 16 . Our few reports of B. burgdorferi ss in questing ticks, and mostly in adult ticks, are consistent with a reservoir like squirrels typically having nymphs attached and which are less abundant than small mammals. Another study at the west coast of Norway also found B. burgdorferi ss more commonly in adult than nymphal ticks 36 .  www.nature.com/scientificreports www.nature.com/scientificreports/ Infection with several genospecies seem common in medium-sized mammals. More B. burgdorferi sl genospecies were found in Siberian chipmunks (Tamias sibericus) than in the native bank vole 45 . We only retrieved one hare and it was without infection, but infection of both B. afzelii and B. burgdorferi ss in hare was documented in south Norway 46 . Hares were also infected with B. garinii in Sweden 37 . We also documented infection in badger and red fox, but prevalence was low ( Table 1). Infection of B. burgdorferi sl has previously been found in red fox 47 , and badger was reported infected with both B. afzelii and B. valaisiana 48 . Medium-sized mammals may hence potentially play roles for transmission of multiple genospecies. Hedgehogs are also important for circulation of B. bavariensis 49 , and that this genospecies have not been found in Norway may be due to the low population numbers of hedgehogs.
Amplification and the challenge of co-infections. A quite high proportion of our positive B. burgdorferi sl samples did not amplify (Tables 1 and 2), which is a requirement in our approach for determining genospecies. The sequencing success decreases with increasing Ct values (i.e., the number of PCR cycles before getting a positive signal). However, only 0.35% (6 of our 865 samples) had Ct values above 40. We cannot identify the cause of lack of amplification with certainty. The most straightforward explanation is that our qPCR-based assay is more sensitive than the conventional PCR 50 . It is also possible that infections with multiple B. burgdorferi sl genospecies can play a role. Indeed, typing using the pepX marker of our B. garinii-positive samples based on IGS (see Methods) came up with two B. afzelii (one in a red squirrel and one in a tick nymph). As B. garinii and B. afzelii are readily separated by both methods, this is indicative of co-infection. The aim of our study was not to quantify extent of co-infections, but these kind of co-infections may cause detection bias if amplification success by PCR differs between genospecies, or when B. burgdorferi sl genospecies have differences in tissue tropism. That we in the small sample of squirrels identified all 3 genospecies using IGS alone would suggest this was not a strong bias. It is therefore likely a robust result that only B. afzelii was detected in smaller mammals with a much larger sample size, and that this was not due to failure of B. burgdorferi ss to amplify. Multiple infections in medium-sized hosts may provide a platform for genetic re-arrangements between genospecies, becoming a potential 'melting-pot' 51 . It can also cause co-infection of genotypes of B. burgdorferi sl in ticks already at the nymphal stage.

Cervids as incompetent hosts.
A controversial issue is the extent to which biodiversity dilutes disease in general, and how important incompetent hosts are for Lyme borreliosis hazard 52,53 . Several lines of evidence have been used to infer that cervids most likely are incompetent hosts in the enzootic cycle of B. burgdorferi sl. Commonly, ticks engorged on cervids are used as evidence 54 . However, ticks removed from cervids often contain a low level of infection with B. burgdorferi sl 55 , which may be due to a number of different mechanisms. In our sizeable sample, the complete lack of detection suggest using ear skin infections are likely better in substantiating that cervids are incompetent hosts than using ticks feeding on cervids. The absence of infection are evidence that cervids are incompetent hosts, but not sufficient evidence that they are dilution hosts. The mechanism of the dilution hypothesis is that when competent hosts are outnumbered by incompetent hosts, the probability of vectors feeding on competent hosts is reduced and so the abundance of infected vectors is lowered 52 . The degree of dilution by cervids depends not only on the relative proportion of the immature tick population fed relative to competent host groups, but also on the absolute number of ticks fed with a change in host population density 52 . High deer density was linked to increased incidence of Lyme disease in Norway 24 , which document that deer mainly act as an amplification (or reproduction) host of tick populations rather than as a dilution host that interferes with the Borrelia transmission cycle.
Our study identifies potential competent (or reservoir) and incompetent hosts towards the northern distribution range of ticks in Europe with a different mammal community compared to continental Europe. Future studies are necessary to determine that they are indeed reservoir competent, vector competent and whether the presence of competent and incompetent hosts has a substantial effect on the hazard (density of questing infected nymphs), which will depend on a number of factors including the population densities of the different species involved 52 .

Material and Methods
Study areas. The study areas are spread across the southern part of Norway (Fig. 1). The western coast has a warmer and rainier climate compared to eastern areas of Norway. The location of samples and the main habitat types are given in Fig. 1. A more detailed description of the main habitat types can be found in ref. 24 .
Small mammal trapping. Small mammals were captured in spring and fall during 2014-16 in Vestby municipality, Akershus county and 2013-16 in Førde and Askvoll municipalities, Sogn & Fjordane county, Norway. Small mammals captured were sacrificed (cervical dislocation) and transferred to an individual zip-lock plastic bag, marked with station number, trap number and date. All bags were stored in a freezer for later observation. The temporal and regional variation in tick load and infection prevalence of B. burgdorferi sl will be presented elsewhere (own unpublished results).
Medium and large sized mammals. We used different methods to sample as many different species of mammals as possible. In Vestby municipality in Akershus county and surrounding areas, we sampled road-kills of roe deer, badger, fox, red squirrels and hare from spring 2016. Ears from red deer and moose were sampled in connection with ongoing surveillance for Chronic Wasting Disease fall 2016. For these data, hunters provided full heads to the Norwegian Food Safety Authority, which removed an ear that was then frozen and later transported to CEES, University of Oslo. We obtained further red squirrels from hunting (August 2016).
Ethics statement. Permissions to capture rodents and shrews and hunt red squirrels outside of hunting season were given by the Norwegian Environment Agency. Small mammals captured were sacrificed on site (see above). The licence by Norwegian Environment Agency is the only requirement to conform to the Norwegian laws and regulations.
www.nature.com/scientificreports www.nature.com/scientificreports/ Questing ticks. Questing ticks were sampled regularly along 34 transects in the Sogn & Fjordane county and along 42 transects and 10 other plots in Møre & Romsdal county, western Norway 56,57 . The temporal and regional variation in prevalence of B. burgdorferi sl in questing nymphs has been presented elsewhere 24,58 . The questing ticks were all identified to I. rinicus based on morphology.
DNA extraction. Tissue samples from mammals were derived from the ears, sometimes with surrounding skin for the smallest shrew species. The DNA was extracted with Qiagen blood and tissue kit according to the manufactures recommendations. To easier prepare a high number of samples that later could be extracted, we froze the samples after the overnight incubation with ATL buffer and proteinase K. A total of 94 samples were extracted at the time, leaving two spaces empty for controls. The DNA was stored at −80 °C till later use. qPCR protocol. The extracted DNA were screened for B. burgdorferi sl by realtime PCR (qPCR) in a duplex 59  To ensure that a product from the PCR was obtained, gel electrophoreses was utilized with 8 µl of DNA on a 1.5% agarose gel. The gel was colored with SYBR ™ Gold Nucleic Acid Gel Stain (Invitrogen ™ ). If the PCR was successful showing a clear band on the gel, the DNA was cleaned with ExoSAP-IT ™ PCR Product Cleanup Reagent (Applied Biosystems ™ ) and sent to sequencing by the firm BaseClear. The chromatographs of the sequences were visually inspected and the primers sites were trimmed in Bionumerics. Our sequences were used to identify the B. burgdorferi sl genospecies by comparison to sequences of known genospecies from GenBank (See Supplementary Table 1). The cluster analyses were performed in Bionumerics 7.4 (Applied Math, Belgium) as described previously 62 . sequencing of pepX of B. garinii positive samples. As it is not always simple to discriminate between B. bavariensis and B. garinii based on IGS data, the nested PCR pepX was done for confirmation following Margos et al. 63 . We used the forward primer pepXF362 (5′-ACAGAGACTTAAGCTTAGCAG-3′) and reverse primer pepXR1172 (5′-GTTCCAATGTCAATAGTTTC-3′) for the initial PCR, and added 1 µl of this product to the nested PCR. The nested primers are the forward primer pepXF449 (5′-TTATTCCAAACCTTGCAATCC-3′) and reverse primer pepXR1115 (5′-TGTGCCTGAAGGAACATTTG-3′).
A total of 10 µl of the nested product was analyzed using the QIAxcel DNA High Resolution Kit on the QIAxcel Advanced System (Qiagen, Hilden, Germany 2018), and samples with a clear band were sent to sequencing by the firm BaseClear. The primer sites of our sequences were manually trimmed and processed. The sequences from this study were used to identify the B. burgdorferi sl genospecies by genetic comparison to sequences of established isolates, as described previously 62 . PepX sequences of known B. burgdorferi sl genospecies can be retrieved from GenBank or from a Borrelia MLST Database (https://pubmlst.org/borrelia/).

Statistical analysis.
We used logistic regression in R vs. 3.4.4 to analyse variation in prevalence of B. burgdorferi sl with mammal species as a factor. However, the presence of species was not balanced across counties, so that we cannot run a full model. For the most common small mammals, we were mainly interested in comparing infection levels relative to medium-sized mammals. Due to the issue of perfect separation for some species (no positives in cervids), we initially applied a one-step-estimator (maxit = 1), but failed to get convergence. We therefore ended up with an analysis restricted to the eastern region and including rodents and shrews, except water shrew due to low sample size. For questing ticks, we ran a logistic regression to test if genospecies B. afzelii and B. burgdorferi ss in questing ticks was found equally often in nymphs or adults (males and females combined) and with county included as a 3-level factor variable.

Data Availability
All data are reported in Tables 1 and 2 within the paper. Sequences obtained in this study are in Genbank with accession numbers MK108437 to MK108914.