Mosquito surveillance of West Nile and Usutu viruses in four territorial units of Slovakia and description of a confirmed autochthonous human case of West Nile fever, 2018 to 2019

Background Despite the known circulation of West Nile virus (WNV) and Usutu virus (USUV) in Slovakia, no formal entomological surveillance programme has been established there thus far. Aim To conduct contemporaneous surveillance of WNV and USUV in different areas of Slovakia and to assess the geographical spread of these viruses through mosquito vectors. The first autochthonous human WNV infection in the country is also described. Methods Mosquitoes were trapped in four Slovak territorial units in 2018 and 2019. Species were characterised morphologically and mosquito pools screened for WNV and USUV by real-time reverse-transcription PCRs. In pools with any of the two viruses detected, presence of pipiens complex group mosquitoes was verified using molecular approaches. Results Altogether, 421 pools containing in total 4,508 mosquitoes were screened. Three pools tested positive for WNV and 16 for USUV. USUV was more prevalent than WNV, with a broader spectrum of vectors and was detected over a longer period (June–October vs August for WNV). The main vectors of both viruses were Culex pipiens sensu lato. Importantly, WNV and USUV were identified in a highly urbanised area of Bratislava city, Slovakias’ capital city. Moreover, in early September 2019, a patient, who had been bitten by mosquitoes in south-western Slovakia and who had not travelled abroad, was laboratory-confirmed with WNV infection. Conclusion The entomological survey results and case report increase current understanding of the WNV and USUV situation in Slovakia. They underline the importance of vector surveillance to assess public health risks posed by these viruses.


Introduction
West Nile virus (WNV) and Usutu virus (USUV) are members of the Flaviviridae family of viruses and are phylogenetically and antigenically related to the Japanese encephalitis virus complex. WNV and USUV share an enzootic transmission cycle involving amplification in avian reservoir hosts and mosquitoes as vectors. In Europe, the major vectors belong to the Culex pipiens complex. Mosquitoes can transmit both WNV and USUV to horses, humans, or other mammals. These are, however, considered dead-end hosts as the viraemia in such hosts does not reach a level sufficient to further infect vectors during a blood meal [1][2][3][4]. For WNV, Cx. modestus is considered as the principal bridge vector between birds and humans. cases in the European Union rose substantially in 2018 compared with previous years [6].
In Slovakia, the presence of WNV and WNV antibodies in dead-end (horses, sheep and human) and reservoir hosts (birds) has been investigated several times since 1967, mainly in the southern regions [7][8][9][10]. Nevertheless, until 2018, flavivirus screening of mosquitoes had only been conducted once, in 1972, in the country's western regions. In the 1972 study, WNV had been detected in a pool of Aedes cantans [11]. The geographical spread of USUV in Slovakia is at present not well understood. Csank et al. [12,13] detected USUV antibodies in birds from the Levice district (South Central (SC) Slovakia) between 2012 and 2014 and in lizards from the Slovak Karst National Park (South East (SE) Slovakia) between 2017 and 2018 [14]. In 2018, an entomological surveillance study conducted in the southern Slovak district of Komárno, revealed a remarkably high prevalence of WNV in mosquitoes with a minimum infection rate (MIR) of 0.46, as well as the first identification of USUV-infected mosquitoes in the country (MIR: 0.25). Furthermore, this work demonstrated the co-existence of both viruses in the same environment [15]. To our knowledge, no human case of USUV has been diagnosed in Slovakia so far.
In the current study, a contemporaneous nationwide entomological survey for WNV and USUV is reported as well as its results. In addition, the first autochthonous human West Nile fever (WNF) case in Slovakia, who was diagnosed in the summer of 2019 is described.

Usutu and West Nile virus vector surveillance
Geographical locations and period of mosquito sampling A nationwide WNV vector surveillance was established in four separate territorial units of the country including South Western (SW), SC, SE and North Eastern (NE) Slovakia ( Figure 1).
Within SW Slovakia, entomological surveillance was conducted at four locations. The first location, was in an urban setting (Bratislava city) between May and November 2018. Mosquitoes were sampled using one BG-Mosquitaire CO 2 (Biogents, Regensburg, Germany)  The NE Slovakia mosquito collection took place in the mountainous environment of the Tatras National

Vector species morphological identification and preparation of mosquito pools
Trapped mosquitoes were stored at − 80 °C before further processing. Female mosquitoes were morphologically distinguished using the identification keys from Becker et al. [16] and pooled according to species, sampling site, date and locality with a maximum of 50 individuals per pool.

Testing mosquito pools for viruses
In preparation for nucleic acid extraction, pools were homogenised in sterile phosphate buffered saline (PBS; pH = 7.2; 200 to 800 µL depending on the number of mosquitoes in a pool) using a Qiagen Tissue Lyser II (Qiagen, Hilden, Germany) with 5 mm stainless beads at 30 Hz/min. RNA was extracted from 140 µL of the supernatant using the GeneJet Genomic RNA Purification Kit (ThermoFisher, Dreieich, Germany), according to the manufacturer's instructions.
The presence of WNV and USUV nucleic acids was tested by two separate (one for WNV and one for USUV) one-step real-time reverse-transcription (RT)-PCR protocols (respectively described below). The PCRs were carried out on the Bio-Rad CFX96 Real-time system (Bio-Rad, Hercules, California, US) with a qScript XLT One-Step RT-qPCR ToughMix, ROX (Quantabio, Beverly, Massachusetts, US).
For the USUV real-time RT-PCR, primers targeting the non-structural protein 5 (NS5) region, USU-9721F, USU-9795R and TaqMan probe USU-9746 (final concentration 200 nM of each) were used without modification of the thermal profile [18]. Two samples of nuclease free water and USUV Europe 2, strain USUV220/2018/ SK RNA (dilution 1:1,000) were included with each run as negative and positive controls. USUV and WNV realtime RT-PCRs with a cycle threshold (Ct) values over 40 were considered negative.

Minimum infection rate estimation
The minimum infection rate (MIR) in an area, assuming a single positive mosquito in a pooled sample, was calculated by extrapolation from the real-time PCR results (the total number of positive pools in the area/total number of mosquitoes sampled in this area × 1,000).

Selective viral gene amplification and sequencing
To obtain sequence information for subsequent determination of USUV or WNV lineages positive samples were further tested by specific RT-PCRs targeting the WNV envelope (ENV) gene (position: 1,531-1,836) [19] and USUV NS5 protein (position: 9,177-9,689) [18], using a one-step RT-PCR kit (Qiagen). The amplicons were visualised on 1.5% agarose gel followed by Sanger sequencing in a commercial laboratory (Eurofins Genomics, Ebersberg, Germany).

Analysis of viral genetic sequences
All sequences obtained were verified using Basic Local Alignment Search Tool (BLAST). Phylogenetic analyses were conducted using a maximum likelihood (ML) algorithm Kimura-2 model (Molecular Evolutionary Genetics Analysis; MEGA 6.0 [20]) with bootstrap resampling of 1,000 replicates.

Confirmation of mosquito species by molecular methods
Furthermore, for positive real-time RT-PCR WNV and USUV pools, a molecular identification of the acetylcholinesterase 2 (ACE-2) and CQ11 loci of the pipiens complex group was performed, extensively described elsewhere [21][22][23].

Ethical statement
The human samples were tested in the National Reference Centre for Arboviruses and Haemorrhagic Fevers of the Public Health Authority of the Slovak

Africa 2 (outgroup)
The tree was generated the using Kimura-2 model with bootstrap resampling of 1,000 replicates. The tree is drawn to scale, with branch lengths corresponding to the number of substitutions per site. Bootstrap values are shown at the nodes. GenBank accession numbers of the sequences in the tree are provided and the sequences retrieved in the current study are in bold font and further indicated by black circles.

Mosquitoes sampled
Altogether, 421 separate pools including in total 4,508 individual female mosquitoes were trapped in 2018 and 2019 (  Figure 3). Moreover, weak clustering of some Slovak sequences was observed within the main cluster of Europe 2. The sequences obtained in this investigation were submitted to the GenBank under accession numbers MN912545-MN91254.

Human case of West Nile fever in Slovakia
At the end of August 2019, a patient presented to hospital with fever (40 °C) accompanied by headache, weakness, myalgia, distractibility and arthralgia. A papular exanthema had developed on their thorax. The patient was admitted to hospital and a subsequent neurological examination (lumbar puncture, magnetic resonance imagining) showed normal findings. Laboratory testing showed full blood count parameters within the normal range. Serological analyses for rubella virus, influenza viruses, parainfluenza viruses, Borelia spp., tick-borne encephalitis virus, Chlamydophila pneumoniae, Mycoplasma pneumoniae, Leptospira spp., hepatitis viruses and human immunodeficiency virus were all negative.
Four days prior, the patient had been bitten several times by mosquitoes, during a visit to a village in the Bratislava region in SW Slovakia (Figure 1). The patient reported not travelling outside Slovakia in the 14 days before symptom onset.
A consultation with an infectious disease specialist raised the suspicion of WNF and laboratory testing for WNV was requested. A first blood serum collection taken on the fourth day of hospitalisation was tested by WNV ELISA and was positive for WNV IgM but not WNV IgG. The second serum sample taken 1 week later was positive for WNV IgM and IgG, and had a low avidity for WNV IgG. The whole blood and urine samples collected at the same time, as well as both serum samples, were tested by a one-step WNV real-time RT-PCR with a positive result only from whole blood. The RT-PCR amplicon was sequenced and the sequence was found to belong to WNV lineage 2 (data not shown).
The limited remaining patient sample material did not allow sequencing of the WNV ENV protein gene for inclusion in the phylogenetic tree. Five days after hospitalisation, the patient began to recover and was released from hospital.
The patient represents the first autochthonous laboratory-confirmed case with clinical criteria of WNV infection from Slovakia (according the European Union case definition [24]). The case was reported to The European Surveillance System (TESSy), Rapid Alert Blood (RAB) and other European public health surveillance databases. The National Transfusion Service was immediately alerted and the Public Health Authority of the Slovak Republic published a guideline to raise medical practitioners' awareness in mid-September 2019.

Discussion
Ecological factors, vector and host composition in a natural environment are responsible for the clustered distribution of arboviruses. Targeted vector surveillance can reveal foci, seasonality and ecological niches of such viruses [25,26]. The information from the surveillance can assist in identifying areas and periods of high viral circulation in vectors and allow to preempt outbreaks [25,26]. Taking this into consideration, nationwide mosquito surveillance is of upmost importance for arbovirus control programmes.
While an emerging risk of human WNV infection in Slovakia has already been suggested since some time by prior investigations, no formal entomological surveillance programme is so far established in the country. In previous studies, WNV was detected in reservoir species as well as WNV antibodies in dead-end hosts [7][8][9][10], and the virus was also shown to circulate in neighbouring countries, particularly Hungary [27][28][29] and Austria [30,31], where human cases of WNV infection were confirmed.
In Hungary, an outbreak of WNV in birds occurred in 2003, followed by the first autochthonous human cases. Subsequently, the circulation of two WNV lineages (lineage 1 and 2) was recognised there [27]. Between 2003 and 2017, an average of 15 to 20 human cases of WNF and West Nile neuroinvasive disease (WNND) was diagnosed annually [28,29], and an extraordinary increase in human WNF cases was reported in 2018. During this year, the number of autochthonous (n = 215) and imported (n = 10) cases of WNV infection was nine times higher than in 2017 (n = 23), moreover, their cumulative number (n = 225) exceeded that of the total number of cases in the previous 14 years (n = 213) [28]. Hungary borders Slovakia to the south and both our data and previous records show that the majority of WNV infections in birds, horses and mosquitoes in Slovakia occurs along this southern border [32]. In the current study, we only detected WNV in a rural village (Podrečany) of SC Slovakia.
Austria borders Slovakia to the west and south-west. In Austria, WNV lineage 2 was first detected in birds and mosquitoes in 2008 and the first human WNF cases were diagnosed in 2006 and 2010 [30]. A total of 23 autochthonous human cases were recorded between 2008 and 2017. The number of recorded human cases of WNV infection in 2018 was 21 [31], demonstrating a similar fold increase as the one observed in Hungary that year. Our study identified WNV in mosquitoes in SW Slovakia. Of particular concern, was the finding of infected vectors in the highly urbanised centre of Bratislava city. Moreover, the first WNF human case recorded in Slovakia, occurred in the south-west part of the country, ca 20 to 25 km from both Hungary and Austria. This Slovak patient's clinical symptoms were typical of WNF infection. Clinical and laboratory diagnostic testing was appropriate and rapid, with the initial serological detection followed up by molecular confirmation. As the human cases in Hungary and Austria, the virus strain detected from the first autochthonous Slovak case was of lineage 2, similar to the Slovak mosquitoes.
Initially, the risk of USUV infections was not expected to be high in Slovakia due to the low number of records in the literature [12][13][14]. However, our results showed that USUV was considerably more prevalent in mosquito vectors (MIR: 3.55) than WNV (MIR: 0.67) and the spectrum of mosquito species infected with USUV was broader than for WNV. USUV was present in all three territorial units of southern Slovakia. USUV human cases are known to have occurred in surrounding countries [33]. In Hungary, the first human symptomatic case of USUV (aseptic meningitis) was diagnosed in 2018. The strain was identified as USUV lineage Europe 2 [28]. In Austria, 18 human USUV cases were recorded in the year 2018 among blood donors, with one donor co-infected with both WNV and USUV. The majority of USUV cases were asymptomatic. The USUV strains from the Austrian cases were related to the Europe 2 and Africa 3 lineages [31]. All USUV strains in Slovak mosquitoes from our study were related to lineage Europe 2. Nonetheless, circulation of more USUV lineages has been confirmed in other countries bordering Slovakia, such as Czechia [33].
As expected given the high altitude, no occurrence of WNV and USUV was detected in the mountainous site of High Tatras National Park. On the other hand, at some of the other study locations, co-circulation of WNV and USUV viruses was observed, such as in Bratislava city (SW Slovakia) and Podrečany village (SC Slovakia). Molecular identification confirmed that these positive pools contained of two ornithophilic species; Cx. pipiens pipiens and Cx. torrentium. However according recent studies, host seeking activity in the Cx. pipiens complex is not as strict as was previously assumed and all member of the complex may serve as bridge vectors for arboviruses [34]. Given the high prevalence in vectors and widespread circulation of USUV, its risk to public health in Slovakia should be reconsidered.
The entomological surveillance conducted in this study provides some preliminary indications on the seasonal prevalence of WNV and USUV in Slovakia. All USUV infected mosquitoes were trapped between the months of June and October. The first human WNF case occurred in August 2019, the same month as all WNV infected mosquitoes identified in this study were trapped in 2018. In the previous study conducted in Slovakia, WNV and USUV infection of mosquitoes was observed between July and September [15]. According to present data, the seasonal prevalence of USUV appears to be longer than WNV, but the peak incidence for both viruses in vector mosquitoes seems to be in August, as suggested by the abundance of positive mosquito pools in that month. This finding may help healthcare practitioners and treating physicians to better detect and diagnose WNF and WNND cases, who may present in a very similar manner to other diseases, e.g. tick-borne encephalitis. The data will also further inform public health and veterinary authorities, allowing better seasonal targeting of prophylaxes and prevention activities in affected areas to protect human and animal health.
Some limitations of our study should be noted. No comprehensive WNV and/or vector control measures are established in Slovakia, so the selection of sampling areas was challenging. Sampling sites also changed between the 2 years of the surveillance, which could have resulted in differences in the 2018 and 2019 trapping seasons. Also, trap types varied according to the mosquito habitat as well as in their way of use. EVS traps were employed mainly in marshes and fishponds and BG-traps in the more urbanised areas. Nevertheless, the principle of both traps is the same -they attract female mosquitoes via carbon dioxide -and the results are likely comparable. Despite these limitations, our study adds to the overview of USUV and WNV ecology in Europe.

Conclusion
We report the first human case of WNF from Slovakia and the results of a concurrent surveillance study undertaken to identify WNV and USUV foci, vectors, seasonal activities and the type of habitat where these viruses occur in Slovakia. The mosquito sampling sites were selected to cover three territorial units of southern Slovakia, where the circulation of these two arboviruses could occur, and one unit in the north east with a site at higher altitude. USUV was found at all investigated southern locations, while WNV was detected in the SC territorial unit as well as in the SW one, where the human WNF case occurred. The data obtained contribute to a preliminarily assessment of the public health risk of WNV and USUV in Slovakia. Continued mosquito surveillance will allow monitoring of these viruses' activity in the country and help to preempt future human cases and outbreaks.

License, supplementary material and copyright
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0) Licence. You may share and adapt the material, but must give appropriate credit to the source, provide a link to the licence and indicate if changes were made.