Field samplings of Ixodus ricinus ticks from a tick-borne encephalitis virus micro-focus in Northern Zealand, Denmark

In 2008-2009 a tick-borne encephalitis virus (TBEV) micro-focus was detected in Northern Zealand, Denmark. No new cases of TBE with an epidemiological link to Northern Zealand has been reported since. Here we undertook to investigate Ixodus ricinus ticks from this endemic micro-focus in 2016 and 2017. In addition to TBEV, I. ricinus ticks may host other pathogens that include Borrelia spp., Babesia spp., Rickettsia spp. and Neoehrlichia mikurensis , together with various endosymbiont microorganisms. To detect multiple organisms we used a metagenomics PanVirus microarray and next-generation sequencing to examine the persistence and evolution of other emerging viruses, bacteria and parasites. Here we report the rise and fall of the Danish TBEV micro-focus in Northern Zealand. However, we identify for the first time in Danish I. ricinus ticks the presence of Uukuniemi virus in addition to a tick-borne phlebovirus and a range of bacteria.


Introduction
Tick-borne encephalitis (TBE) is a viral infectious disease that causes severe neurological sequelae in humans and has mortality rates of up to 3.0% (Lenhard et al., 2016). In contrast to the evenly distributed and widespread Borrelia bacteria, the tick-borne encephalitis virus (TBEV) tends to occur in micro-foci A C C E P T E D M A N U S C R I P T within endemic areas. TBE is a growing public health concern in Europe and the overall notification rate in EU/EEA rose from 0.4 cases per 100,000 population in 2015 to 0.6 cases per 100,000 population in 2016 (ECDC 2018). The incidence and geographical range of TBE is similarly increasing in Scandinavia (Jaenson et al., 2012;Sidorenko et al., 2018). In Denmark, TBE has been reported since the 1950s only from the Bornholm Island east of mainland Denmark in the Baltic Sea with an incidence of 3.8 cases per 100,000 persons per year (Ocias et al., 2017;Fomsgaard 2018). In 2008 and 2009, two cases of TBE were reported from Tokkekøb Hegn in Northern Zealand, where after TBEV was identified in a pool of Ixodus ricinus nymphs collected from the edge of the forest and water surrounding a patient's property and a public recreational area (Fomsgaard et al., 2009). TBEV was detected at the same micro-focus in 2010 and 2011, which confirms the establishment of an additional micro-focus of TBEV in Denmark (Fomsgaard et al., 2013). Since 2009 there has been no new cases of TBE in Denmark with an epidemiological link to Northern Zealand. We therefore undertook to investigate this endemic TBEV micro-focus again in 2016 and 2017.
Since ticks may also host other pathogenic microorganisms including Borrelia spp., Babesia spp., Neoehrlichia mikurensis and Rickettsia spp. (Fertner et al., 2012), the aim of the study was not only to reexamine the area for persistence of TBEV but also to look for other viruses, bacteria and parasites.

Study design and Nucleic Acids extraction
In total 4,100 ticks were collected by flagging from Tokkekøb Hegn, Northern Zealand (Fomsgaard et al., 2009;Fomsgaard et al., 2013) at four sampling times in 2016 and 2017 (Table 1). All ticks were identified as I. ricinus on the basis of morphology and sorted according to stage (nymphs vs adults). The ticks were then divided into 71 pools of nymphs, 6 pools of adult males and 5 pools of adult females containing 50 ticks in each pool (Table 1) as previously described (Fomsgaard et al., 2009;Fomsgaard et al., 2013). Each tick pool were homogenized in 1 ml of MagNA Pure Lysis/Binding Buffer (Roche Diagnostics A/S, Risch-Rotkreuz, Switzerland), incubated overnight at 5 °C before nucleic acid (NA) was extracted with the MagNA Pure

Virus-specific PCR
Five µl of extracted NA were used in a TBEV-specific real-time RT-PCR as previously described (Fomsgaard et al., 2009;Fomsgaard et al., 2013). A specific real-time RT-PCR assay for the M-segment (NC_005220) of

PanVirus Microarray
Twenty-five µl of extracted NA from 20 randomly selected tick pools were treated with 4 U TURBO™ DNase (ThermoFisher, Waltham, MA, USA) and enriched for RNA virus using an unpublished I. ricinus tick-modified version of the published protocol (Matranga et al., 2016). Five µl of enriched RNA were used for random whole transcriptome amplification (WTA) using the Repli-g Cell WGA & WTA kit (Qiagen, Hilden, Germany) according to the manufactures instructions. Amplified cDNA was purified using the QiaAmp DNA mini kit (Qiagen, Hilden, Germany) and 1.5 µg of purified cDNA was labelled with Cy5 using the SureTag DNA labelling kit (Agilent, Santa Clara, CA, USA). The labelled DNA was pooled according to the sampling date and tick life-cycle stage and hybridized to a PanVirus microarray (SurePrint G3 Custom Gene Expression Microarray (4x180K) (Agilent, Santa Clara, CA, USA)) containing 60-160 probes for each sequenced virus described in GenBank (2016). Microarray data was analysed using a simple Excel-based data analysis method as described previously (Erlandsson et al., 2011;Rosenstierne et al., 2014).

Microbiome analysis
The microbial content of pro-and eukaryotes was analysed using a modified version of the in-house custom build 16S/18S ribosomal ribonucleic acid (rRNA) amplicon-based approach, previously described by Ring et al., 2017 andKrogsgaard et al., 2018. In brief, 2 µl of extracted NA from the 20 randomly selected tick pools were PCR amplified using 4 different primer sets (the modified 341F/806R primers targeting the prokaryotic V3-V4 16SrDNA region, and the G3F1/G3R1, G4F3/G4R3 and G6F1/G6R1 primers targeting the hypervariable regions V3-V5 of the eukaryotic 18SrDNA gene (Krogsgaard et al., 2018). Library preparation was performed as preciously described (Krogsgaard et al., 2018). The resulting amplicons were quantified using the Quant-ITTM dsDNA High Sensitive Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA), pooled in equimolar amounts and sequenced on an Illumina MiSeq desktop sequencer (Illumina Inc., San Diego, CA, USA) using the 500-cycle MiSeq Reagent Kit V2 (Illumina Inc., San Diego, CA, USA). Sequences were taxonomically mapped using BION, a k-mer-based mapping software described previously (Krogsgaard et al., 2018). Query sequences originating from prokaryotes were compared with the 340-807bp rRNA positions derived from Escherichia coli in RDP 11.04 (https://rdp.cme.msu.edu), and eukaryotic query sequences compared against the same region in SILVA version 123 (www.arb-silva.de). A unique taxon was defined by denotation to a specific evolutionary taxonomic group.

Results and Discussion
None of tick pools were positive for TBEV using either TBEV specific PCR or the PanVirus microarray (Table   1). We have previously shown the presence of TBEV using a TBEV-specific PCR in nymph pools containing 50-100 nymphs (Fomsgaard 2009;Fomsgaard 2013) indicating that the lack of detection of TBEV in the nymph pools was not due to assay performance. The negative TBEV finding and the absence of clinical cases of TBE with epidemiological link to Northern Zealand since 2009 (Ocias et al., 2017) Randolph et al., 2008;Knap et al., 2009). Even though the micro-focus may have been temporary, continued surveillance of the site will be performed and TBE should still be considered in Denmark outside Bornholm if clinically relevant symptoms are present. There are indications that TBEVs geographical range is increasing in Sweden, Norway and Denmark (Jaenson et al., 2012;Sidorenko et al., 2018;Andersen 2019). It is thus important to be aware of the establishment of new micro-foci in Denmark outside Bornholm.
Two phleboviruses were detected with the PanVirus microarray in the collected I. ricinus ticks (Table 1) (Table 1). This is the first report of UUKV in Danish I. ricinus ticks. UUKV has been identified in I. ricinus ticks in Finland (Saikku and Brummer-Korvenkontio, 1973), Norway (Traavik and Mehl, 1977), the Czech Republic (Kolman, 1970) and Sweden and the Netherlands (Gondard et al., 2018). To date, UUKV has not been associated with disease in humans. However, other tick-borne phleboviruses very similar to UUKV, such as severe fever with thrombocytopenia syndrome virus and Heartland virus (Palacios et al., 2013) has been associated with severe and often fatal disease in humans (Yu et al., 2011;McMullan et al., 2012). A possible explanation for the lack of UUKV-associated disease in humans is that the non-structural protein is a weak interferon antagonist that is incapable of counteracting the interferon response (Rezelj al., 2015).
A positive signal for Blacklegged tick phlebovirus (BTPV) was also detected by the PanVirus microarray in the pool of nymphs from August 8 th and June 26 th 2017 (Table 1). In contrast to the microarray signal for UUKV, the microarray signal for BTPV was only positive for a partial region of the BTPV L-segment (data not shown) which indicated cross-hybridization of probes to an unidentified phlebovirus. The presence of BTPV was not verified in this study. BTPV was first identified in I. scapularis in the United States (Tokarz et al., 2014); however new tick-borne phleboviruses are continuously being identified all over Europe which cluster to BTPV (Tokarz et al., 2014;Papa et al., 2016;Prinz et al., 2017). The exact identification and molecular characterization of the phlebovirus in Danish I. ricinus ticks from Northen Zealand remains to be performed.
The 20 randomly selected tick pools were also analysed for the presence of other microorganisms using 16S/18S rRNA amplicon-based metagenomics. Only species detected with at least 500 sequence reads in the 16S/18S rRNA amplicon-based metagenomics are presented (Table 2). Rickettsia spp. was the most prevalent bacteria species detected in all 20 pools with the consistent highest number of reads (Table 2,   Supplementary Table). Rickettsia may be present in I. ricinus ticks either as an endosymbiont (mutualistic relationship) or as a pathogen, including R. helvetica and R. monacensis. The metagenomics analysis was not able to differentiate the Rickettsia species as they are almost identical in the amplified area of 16S.
Other prevalent bacteria were Borrelia spp., including, B. afzelii (8 pools Table 2). No parasites, including protozoan, were detected in any of the tested pools. Some fungi were detected, but not analysed further in this study.
We detected two human pathogens in our tested tick pools: Borrelia spp. and Anaplasma phagocytophilum. B. miyamotoi was also detected, although we are not aware that B. miyamotoi disease has been diagnosed in Denmark. Other well-known tick-borne pathogens such as Babesia, Francisella, Bartonella, Ehrlichia, and Candidatus Neoehrlichia mikurensis were not detected in this study. This is in agreement with the previous studies from Denmark where only A. phagocytophilum and Borrelia spp. were detected in I. ricinus ticks from Jutland, Funen and Bornholm (Andersen et al., 2019;Skarphédinsson et al., 2007). However, two Babesia species (B. divergens and B. venatorum) were demonstrated in 2012 in I.
We did not report any Midichloria mitochondrii, an endosymbiont frequently associated to I. ricinus (Cafiso et al., 2019). The RDP 11.04 database only contains sequences from Candidatus Midichloria, taxonomically defined at subtype level without any genus and species names and were therefore ignored by BION.
However, when mapping raw sequence data to the entire RDP database, including taxonomically undefined references, reads from 18 of 20 samples mapped to Candidatus Midichloria sp. Ixholo1. Endosymbiont microorganisms exhibit both negative and beneficial effects on tick fitness (de la Fuente et al., 2017;Bonnet et al., 2017), in addition to providing protection against other bacteria, viruses and eukaryotic parasites, which indicates that they can interact with other microorganisms sharing the same host environment (de la Fuente et al., 2017;Bonnet et al., 2017).
In conclusion, we find that a previously identified TBEV micro-focus in Northern Zealand, Denmark seems to have disappeared for undefined reasons. We propose continued surveillance of known TBEV micro-foci. We also demonstrated for the first time UUKV in Danish I. ricinus ticks and the possible presence of a new phlebovirus. We confirm that I. ricinus ticks are vectors of pathogens of public health concern.

Conflict of interests
The authors declare that there is no conflict of interest in this study.

Acknowledgment
We thank Susanne Lopez Rasmussen, Louise Borup, Randi Thøgersen for technical assistance and Maria Magdalena Lassaunière for editorial assistance. This study is part of the "Metagenomic Array Detection of emerging Virus in EU" (MAD-Vir) project, which is part of the European Joint Programme One Health EJP