An update of the Culicoides (Diptera: Ceratopogonidae) checklist for the Balkans

Background The prime significance of species belonging to the genus Culicoides Latreille, 1809 (Diptera: Ceratopogonidae) is their ability to transmit viruses such as bluetongue virus (BTV) to wild and domestic ruminants. Prior to 1998, BTV was considered exotic in Europe, but according to recent history of its outbreaks, it has become endemic in southern and eastern European countries circulating beyond its expected historical limits, into the Balkan region. The wind-borne long-distance dispersal of Culicoides spp. over water bodies and local spreading between farms emphasize the necessity of filling in the information gaps regarding vector species distribution. In most Balkan countries, data on Culicoides fauna and species distribution are lacking, or information is old and scarce. Results During this study, 8586 specimens belonging to 41 species were collected. We present the first faunistic data on Culicoides species in the former Yugoslav Republic of Macedonia (FYROM), Kosovo, Montenegro and Serbia. For other countries (Bosnia and Herzegovina, Bulgaria and Croatia), all historical records were compiled for the first time and then expanded with our findings to various extents. In all countries, confirmed or suspected BTV vector species belonging to the subgenera Avaritia and Culicoides were collected. The total number of species sampled during our field collections was 20 in Bosnia and Herzegovina (15 new records), 10 in Bulgaria (2 new records), 10 in Croatia (5 new records), 13 in FYROM, 9 in Kosovo, 15 in Montenegro, and 28 in Serbia. Of these, 14 species were registered for the first time in this part of the Balkans. Conclusions This paper provides the first data about Culicoides fauna in FYROM, Kosovo, Montenegro and Serbia, as well as new records and an update on the checklists for Bosnia and Herzegovina, Bulgaria and Croatia. These findings provide preliminary insights into the routes of BTV introduction and spreading within the Balkans, and present a valuable contribution to further research related to Culicoides-borne diseases in Europe.

Although Culicoides spp. cause a significant biting nuisance (impacting tourism and forestry) [3][4][5], their greatest importance lies in the fact that they are biological vectors of human and animal pathogens (mainly viruses, but also protozoans and filarial nematodes) [1,5]. Indeed, the prime significance of Culicoides spp. is their ability to transmit viruses of domestic ruminants, such as bluetongue virus (BTV) or the novel Schmallenberg virus, which was first detected in 2011 in Germany [6].
Bluetongue (BT), which has to be reported to the Office International des Épizooties (OIE) [1,7], is caused by BTV, affecting domestic and wild ruminants, in particular certain breeds of sheep (causing severe clinical disease and in some cases significant mortality) [8], but also cattle [9] and goats [10]. Bluetongue causes considerable economic concern and poses a major risk in the international trade of animals and animal products [11].
Prior to 1998, the disease was considered exotic in Europe [11], with just a few sporadic incursions recorded in southern European countries (Spain, Portugal, Greece and Cyprus) [12][13][14], but the situation has since dramatically changed. In the seven-year period from 1998 to 2005, BTV outbreaks were recorded in 15 southern European countries [11] and followed by sudden and unexpected outbreaks in northern Europe during 2006-2008 [14][15][16][17]. According to the European Food Safety Authority (EFSA) [8], this was arguably the largest and the most continuous outbreak of a Culicoides-borne pathogen in the history of humankind and caused enormous economic losses. Interestingly, BTV serotypes affecting southern and northern Europe are different [14]. Thanks to the implementation of compulsory BT vaccination in Europe resulting in a massive reduction of BTV cases, outbreaks were limited and some countries were back to a free BTV status.
When observing the situation in the Balkans, BTV-9 first entered southeastern Europe (Greece) in 1998 [18]. During 1999, this serotype spread from Greece to southeastern Bulgaria [19], while further outbreaks caused by BTV-9 were reported during 2001 in the former Yugoslav Republic of Macedonia (FYROM) [20], Kosovo [21], Montenegro, Serbia [22,23] and Croatia [24,25]. During the following year (2002), the BTV-9 outbreak was confirmed in Bosnia and Herzegovina [26] and Albania [27]. Interestingly, in the following years other serotypes of the virus were isolated in the Balkans: BTV-16 in Croatia during 2004, and an important BTV-4 epizootic in Greece, Bulgaria, Croatia, Serbia, Montenegro and FYROM in the period 2014-2017 [18].
According to the recent BT outbreak records in Europe, it is reasonable to state that BT is endemic in the countries of southern and eastern Europe and that different serotypes of BTV are still circulating through Europe.
Bluetongue emergence in previously unaffected regions of the Mediterranean basin has been attributed to climate change, through an increasing abundance of Culicoides imicola Kieffer, 1913, a proven Afrotropical vector species [28,29]. However, this was not applicable to the Balkan region as C. imicola did not occur (except in Greece) where BT outbreaks were recorded [11,[30][31][32]. Meanwhile, the Palaearctic species belonging to the subgenera Avaritia and Culicoides were found implicated as BTV vector species in different European countries [11,33].
Laboratory and field studies reported the following species as probable BTV vectors: (i) subgenus  [43], and describing identification at the species "complex" level [44], both bring up taxonomic concerns.
In Serbia, Pavlović et al. [45] reported four species (C. obsoletus, C. pulicaris, Culicoides parroti Kieffer, 1922 and C. nubeculosus) in a conference abstract, but the lack of details on the identification method does not allow an assessment as to whether these authors were referring to species or groups of species. Comprehensive checklists are available only for Croatia with 16 species [42,46,47], for Albania with 20 species [48] and for Bulgaria with 37 species [49,50], for which Bobeva et al. [49] summed up previously published literature, mostly in Bulgarian language [51][52][53].
We assembled all Culicoides species records from the region, combining the data collected during field studies performed by several research groups and the historical records (for Bosnia and Herzegovina, Bulgaria and Croatia). In addition, we provide the first checklists for Culicoides species for FYROM, Kosovo, Montenegro and Serbia.

Culicoides sampling
Culicoides spp. collections in the study area ( Fig. 1) were conducted during sand fly (Diptera: Psychodidae: Phlebotominae) and/or mosquito (Diptera: Culicidae) surveys. Therefore, the selection of sites and trap types was not specifically adjusted to Culicoides trapping, except once in Serbia when an Onderstepoort Veterinary Institute blacklight suction trap (OVI trap) was used [54]. All traps were run for one night and set up from approximately 1 h before sunset to 1 h after sunrise.
In Bosnia and Herzegovina, 18 sites were surveyed over two periods. Five sites were sampled during a mosquito survey (17-22/06/2015) by NS2 dry ice-baited traps without light (constructed at the University of Novi Sad, Serbia), while another 13 sites were prospected during sand fly sampling (03-07/07/2015) using standard miniature Centre for Disease Control (CDC) light traps (John W. Hock, model 512, Gainesville, FL, USA) without dry ice.
In Bulgaria, 5 sites were sampled during sand fly collections (20-24/08/2015) using miniature CDC light traps with and without dry ice.
In Croatia, miniature CDC light traps without dry ice were used, during sand fly field studies (13-15/07/2015) at two sites.
In Serbia, collections were undertaken at 124 sites (urban, semi-urban and rural areas) (28/04/2015-13/ 09/2015) during a West Nile virus surveillance program (81 sites, using NS2 traps only) and sand fly collections (43 sites). Within these 43 sites, 14 sites were exclusively sampled for sand fly collections and only CDC light traps without dry ice were used; at the other 29 sites, NS2 and miniature CDC light traps with and without dry ice were used and an OVI trap was placed only once. At these 29 sites, all mentioned trap types were sometimes positioned to potentiate the collection of Culicoides as well (depending on the number of animals and their composition, as well as farmers' permission for trap placing). According to the study context, the traps were placed bi-weekly, monthly or just once per collecting season in different parts of Serbia.
Culicoides biting midges in other surveyed countries were collected in traps positioned to sample sand flies (all traps in Bulgaria, Croatia, FYROM, Kosovo, and 72% of the traps placed in Bosnia and Herzegovina) and/or mosquitoes (all traps in Montenegro and 28% of the traps in Bosnia and Herzegovina). In all countries, the sand fly surveys were carried out in farms (usually containing different mammal species and much rarely avian species). During the mosquito surveys, traps were positioned in Fig. 1 Culicoides collection sites in the study area. A background elevation map was created with data produced by Jonathan de Ferranti and downloaded from viewfinderpanoramas.org. Abbreviations: ME, Montenegro; FYROM, the former Yugoslav Republic of Macedonia different kinds of sites: highly urban (without any animal breeding); semi-urban (with different animal breeding); and rural (far away from human settlements and farms).
All sand fly-specific sites gained Culicoides spp. positive samples. Consequently, in the countries where only sand fly monitoring was performed, frequency of occurrence of different Culicoides spp. is expressed as a ratio of positive to all collection sites. As the sites for mosquito sampling were chosen according to different criteria, usually not comprising the courtyards, Culicoides spp. were not collected at all of them. Therefore, for Montenegro and Serbia, only the Culicoides spp. positive mosquito sampling sites were considered for the frequency of occurrence evaluation.

Culicoides identification
Species identification was carried out by observing morphological features under a stereomicroscope. Primarily, we focused on the wing pigmentation pattern and the distribution of wing macrotrichia. Then, we observed the antennal XI/X ratio (length of segment XI divided by length of segment X), and the shape and size of the 3rd palpal segment. Finally, we compared all observed traits with IIKC (interactive identification key for Culicoides) database pictures [55]. Slide mounting was performed in the cases of morphologically similar species, atypical variations of the wing pattern, identification confirmation, or if the specimen was damaged. Slide-mounted specimens were observed under a light microscope connected to a camera and attention was paid to the broad range of other morphological features including shape, size and the number of female spermathecae, the shape and form of male genitalia [56,57], as well as eye separation distance [55].
Culicoides obsoletus females cannot be morphologically separated with confidence from C. scoticus females or can be distinguished only by using many morphological measurements [58][59][60]. Thus, in our paper, we refer to C. obsoletus/C. scoticus females. However, males were recorded as C. obsoletus or C. scoticus, because they can be distinguished based on the appearance of the genitalia [55]. The Culicoides obsoletus taxon refers to C. obsoletus (sensu lato), i.e. potentially including the cryptic species, which have been recently discovered [61,62].
In some cases, morphologically closely related species, or specimens with an atypical wing pattern (i.e. variations and overlapping between two species) could not be morphologically separated and identified with confidence. We refer to these as pairs of species, such as  [63] since a formal redescription is missing [64]. We decided to associate sensu lato to Culicoides grisescens Edwards, 1939 to take into account the cryptic diversity highlighted recently within this taxon [61]. It is also worth noting that females of this species could be confused with Culicoides deltus Edwards, 1939. Although according to Borkent's catalog, Culicoides trivittatus Vimmer, 1932 belongs to the subgenus Oecacta [65], we decided to place it in "Miscellaneous unplaced species" (Table 1) since the male genitalia are very particular and very different from the type-species Oecacta furens [66]. However, in numerous trapping sites, the morphological identification to the species level of females and of males (confirmed by the observation of their species-specific genitalia appearance) was possible, providing us with information on the records of the members of the species complexes in most countries.
Furthermore, only a few specimens, too damaged to be assigned to a given species, were classified as Culicoides sp.

Results
A total of 8586 Culicoides specimens belonging to at least 41 species (at least 40 species classified in eight subgenera plus 1 miscellaneous unplaced species [55,[64][65][66]) were processed during our study ( Table 1). Species of an additional two subgenera were recorded by previous authors (Table 1).
Only C. punctatus, C. newsteadi and C. obsoletus/C. scoticus were recorded in all inspected countries during our study. Three patterns of species presence/absence and frequency of occurrence were observed: the first in Bosnia and Herzegovina, Bulgaria, Croatia, FYROM and Kosovo; the second in Montenegro and the third, an intermediate pattern, in Serbia.
The situation in Serbia seemed to be intermediate. Culicoides festivipennis was convincingly the most frequent species (recorded at 59% of sites), followed by Reported as Culicoides cubitalis Edwards, 1939 Key: ×, the first record of certain species or subgenera in the country; ×, species recorded in previous studies and the present study; *, species recorded by previous authors The subgenera and species are classified and presented in accordance with Borkent [65], Mathieu et al. [55], Ramilo et al. [64] and Szadziewski et al. [66]. C. indistinctus is presented in Borkent [65] as a synonym of C. odiatus. C. submaritimus is presented in Borkent [65] as a synonym of C. maritimus. Bobeva et al. [49] summed up previously published literature for Bulgaria, citing: Zilahi [51], Remm [52] and Nedelchev [53] Culicoides circumscriptus Kieffer, 1918 and C. griseidosum (51% and 49%, respectively). Culicoides obsoletus/C. scoticus was recorded at 30% of sites, while C. punctatus and C. newsteadi were each found at 25% of the sites. Culicoides minutissimus, C. alazanicus, and C. pulicaris were less frequent, observed at 21%, 16% and 9% of sites, respectively. All other species were recorded at less than 8% of sites.
In the north of Serbia, more species were recorded during the sand fly surveys (27 spp.) than during the mosquito surveys (16 spp.), even if sampling was carried out in more sites and more frequently during mosquito surveys than sand fly surveys. Only one species, Culicoides kibunensis Tokunaga, 1937, was collected only during the mosquito surveys and not during the sand fly surveys. Finally, C. fagineus was recorded in the south-west of the country where only sand fly targeted sampling was performed. Additionally, in the north of Serbia where both mosquito and sand fly surveys were carried out, Culicoides were found in all sites sampled for sand flies, and only in 83% of the sites (67 out of 81) sampled for mosquitoes.
The total number of species recorded during this study was as follows: at least 20 in Bosnia and Herzegovina, at least 10 in both Bulgaria and Croatia, at least 13 in FYROM, at least 9 in Kosovo, at least 15 in Montenegro and at least 28 in Serbia (Table 1).

Discussion
This paper provides the first Culicoides faunistic data for FYROM, Kosovo, Montenegro and Serbia, and an update of the Culicoides species checklists by adding several newly recorded species: at least 15 for Bosnia and Herzegovina, two for Bulgaria, and at least five for Croatia (Table 1), as a preliminary and mandatory step prior to any further research related to Culicoides-borne diseases in this region of Europe.
Until recently, there was very limited information about Culicoides species in the territory of former Yugoslavia. In a short note format article, Callot & Kremer [42] reported 12 Culicoides species based on adult emergence from mud samples collected in 1972, mainly along the coast of the Adriatic Sea (including present Bosnia and Herzegovina and Croatian territories). Additional studies in Bosnia and Herzegovina [43] and Croatia [46,47] ( Table 1) were carried out several decades later after this region was affected by a BTV outbreak [24][25][26], which acted as a trigger for Culicoides survey (re)starting. The same applies to Kosovo, where the "Obsoletus" and "Pulicaris" species complexes were reported, but without identification down to the species level [44], while for Serbia, due to the limited amount of published information (conference abstract) [45], it was not clear if the taxa mentioned were referring to species or to groups of species. Hence, we decided not to include these findings in the Serbian species checklist provided here. The research on Culicoides in Bulgaria has a much longer tradition and had been performed by several authors [49][50][51][52][53] (Table 1).
There were no previously published data on Culicoides in Montenegro. We recorded at least 15 species, but potential BTV vectors, such as C. obsoletus/C. scoticus or C. punctatus, which predominantly feed on large mammals [67] were very rare (only in a few localities), while C. newsteadi was present in more collection sites. The most frequently recorded species were C. griseidorsum, C. festivipennis, C. alazanicus, C. pictipennis and C. newsteadi. The species fauna was different compared to other countries surveyed in this study, which might be explained by the specific positioning of the traps aimed to sample mosquitoes. This sometimes may include rural locations (far away from human settlements and farms) or highly urban areas (without any cattle breeding). The two most common species found in Montenegro, namely C. griseidorsum and C. festivipennis, are known as predominant bird feeders [67], as well as C. alazanicus which was the third common species in our survey. Culicoides alazanicus shared the third position with C. pictipennis, which is an opportunistic feeder, and with C. newsteadi, which feeds exclusively on large mammals [50,67]. Culicoides specimens were found in 28% of the traps positioned in urban sites, 64% in semi-urban sites and 8% in rural sites. Although semi-urban sites (with animal breeding) were more frequently sampled, the number of mammals was usually low (≤ 2 in 69% of cases).
In the majority of countries, contrary to the study in Montenegro, the trap locations were selected mainly to meet requirements for sand fly sampling, i.e. all collection sites in Bulgaria, Croatia, FYROM and Kosovo, as well as 72% of the traps placed in Bosnia and Herzegovina. Finally, in Serbia, 11% of the traps were set up exclusively for sand fly sampling, 24% of the collection sites were chosen for both sand fly and Culicoides sampling, while 65% of the traps were intended to mosquito survey. The traps set up for mosquitoes and collecting Culicoides were positioned in urban (15%), semi-urban (66%) and rural (19%) sites.
Although Culicoides, sand flies and mosquitoes can be very often found in the same environment, the choice of the trap location (particularly presence of suitable hosts nearby) can influence the species composition of each group. Since all sand fly samplings were carried out on farms (usually containing different mammal species and, much more rarely, avian species), it is not surprising that Culicoides catches in surveyed countries other than Montenegro and Serbia were dominated by mammophilic species: C. punctatus, C. newsteadi, C. obsoletus/C. scoticus and C. pulicaris.
In contrast, in Serbia, where 65% of the traps were placed for mosquito sampling (including urban, semi-urban and rural areas), the species composition was dominated by predominately ornithophilic species, namely C. festivipennis, C. circumscriptus and C. griseidorsum, followed by predominately mammophilic feeders, namely C. obsoletus/C. scoticus, C. newsteadi and C. punctatus [67]. This kind of species composition and the occurrence frequency of Culicoides spp. indicates strong correlation with the host species present at each sampling site (the selection was tailor-made for the study purpose). Moreover, the sites sampled for sand flies may be more suitable for Culicoides inventory than those sampled for mosquitoes because we recorded more species in the first than in the second ones. Indeed, all sites sampled for sand flies were positive for Culicoides, whereas 17% of the sites sampled for mosquitoes were negative (when comparing catches in the north of Serbia). It should be noted, however, that these are preliminary estimations because the number of sampling sites and the sampling frequency were not the same for the different protocols.
Out of at least 41 species of Culicoides collected during our study, 14 species were recorded for the first time in this part of the Balkans (Table 1). Moreover, the finding of C. trivittatus in Croatia represents the first record of this species in Europe. This species is known to be distributed throughout Asia and middle Asia (Israel, Iran, Iraq, Armenia, Azerbaijan, Uzbekistan, Tajikistan, Turkmenistan and the south 50°N in Russia and Ukraine) [52].

Conclusions
We present the first data on Culicoides fauna in FYROM with at least 13 species, in Kosovo with at least 9 species, in Montenegro with at least 15 species and in Serbia with at least 28 species. For the other countries all previous Culicoides records (published independently by different author groups) were compiled for the first time and then expanded with our findings: at least 15 newly recorded species for Bosnia and Herzegovina, two for Bulgaria, and at least five for Croatia. Fourteen species were detected for the first time in the seven Balkan countries. Among the newly recorded species, C. trivittatus was recorded for the first time in Europe. Furthermore, due to sampling design, insights about species occurrence outside farm environments were provided. This can also serve as good starting point for research on ornithophilic Culicoides species, as well as surveillance of the bird pathogens they can transmit. In all countries, species of the subgenera Avaritia and Culicoides, which are proven or probable BTV vectors, were registered. This finding provides preliminary insights into the routes of BTV introduction and spreading within the Balkans. Authors' contributions DPu contributed to manuscript writing and carried out Culicoides identification, data analysis, sample collection and processing. DPe contributed to manuscript writing, sample collection and processing. XA and IR carried out Culicoides identification. BA, NA, AC, TG, FG, KH, AIĆ, MK, TL, OM, JS, SV and AZ carried out sample collection and processing. IP carried out sample collection and processing, and data analysis. CG contributed to manuscript writing. BM contributed to manuscript writing and carried out Culicoides identification. TB contributed to manuscript writing and data analysis. All authors read and approved the final manuscript.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.