Description of Culicoides (Culicoides) bysta n. sp., a new member of the Pulicaris group (Diptera: Ceratopogonidae) from Slovakia

Species of the genus Culicoides Latreille, 1809 (Diptera: Ceratopogonidae) are mainly known as vectors of arboviruses such as bluetongue (BTV) and Schmallenberg (SBV). Among the known vectors, few species within the subgenus Culicoides Latreille, 1809 have been implicated in the transmission of BTV and SBV. Nevertheless, phylogenetic studies had revealed the presence of cryptic and undescribed species in Europe, raising questions about their vectorial role. A previous integrative study, associating morphology and barcode data, raised the hypothesis of the presence of undescribed species in Slovakia. The present study, combining morphological and molecular approaches, is aimed to support the hypothesis and a description of Culicoides bysta n. sp. is provided. Series of male and female specimens were dissected and several of them were sequenced for the barcode region of the mitochondrial cytochrome c oxidase subunit 1 gene (cox1). Bayesian inference phylogenetic analyses based on 72 cox1 sequences of the species belonging to the Pulicaris group of the subgenus Culicoides, were carried out and the frequencies of intra/interspecific variations were analyzed. The morphology of abundant material of the new species (31 females and 12 males) was examined and compared with the paratypes of Culicoides boyi Nielsen, Kristensen & Pape, 2015 and with specimens of Culicoides pulicaris Linnaeus, 1758. For females, suture distances on the eyes were newly evaluated as a diagnostic character and for males we assessed a new measurement on the ninth tergite and on the apicolateral processes. Both phylogenetic analysis and barcode distances supported the distinct status of the new species, Culicoides bysta n. sp. described as a member of the Pulicaris group based on the morphology of males and females. The new species is closely related to C. boyi and C. pulicaris but can be distinguished on the basis of the wing pattern and the ratio between the two eye sutures. Both newly evaluated characters, i.e. eyes in females and male genitalia appeared to be diagnostic for distinguishing the new species described herein. The vector potential of the recently described species C. boyi and C. bysta n. sp. to transmit arboviruses, such as BTV and SBV, is unknown. When considering these two species as being close to C. pulicaris, the previous data, such as the vector implication for C. pulicaris in BTV transmission, should be revaluated in future.

The exact number of species belonging to the subgenus Culicoides in the Palaearctic region is unknown, as different authors include different species in the subgenus [12]. Other authors presented their disagreement regarding the subgeneric nomenclature and thus classified the species related to C. pulicaris as the species of the Pulicaris group [13] and eventually subgroups [14]. Regardless of the chosen classification, the species related to and "grouped" with C. pulicaris vary among the authors. In an attempt to clarify the classification and the identification of the species within the Pulicaris group, phylogenetic studies based on cox1 mitochondrial gene sequences [15][16][17][18][19], as well as on ITS2 rDNA region [12,20], revealed the presence of cryptic species. Following the discovery of this genetic diversity, in 2013, a new species, C. paradoxalis Ramilo & Delécolle, 2013, close to C. lupicaris was described from France and Portugal [21]. In Denmark, three species molecularly characterised in 2011 [18], have been recently described as new species, i.e. C. boyi, C. selandicus Nielsen, Kristensen & Pape, 2015 and C. kalix Nielsen, Kristensen & Pape, 2015 [18,22]. It is worth to mention that for the four above mentioned recently described species, i.e. the one from Portugal and the three from Denmark, morphological studies have highlighted the diagnostic characters for accurate identification [21][22][23]. In the light of the cryptic diversity detected in this group, further undescribed species may be expected [15,17,19]. Further studies, including morphological exploration, are still required to eventually lead to the formal description of new species. Currently, new species are frequently discovered by both genetic discrepancies detected by barcode analyses (cox1 region) and the presence of highly supported clade from phylogenetic studies [21,22]. Besides, morphological studies are highly recommended to be linked to these genetic studies, for a better understanding of the composition of subgenus Culicoides [19].
In the present study, we describe a member of the Pulicaris group, C. (Culicoides) bysta n. sp. The sample consisting of a unique female specimen found in our previous study [19] is now completed with a larger series of males and females. To support the distinct species status of the new species, we implemented a combination of morphological and phylogenetic (based on cox1 gene) approaches.

Sampling and identification of Culicoides
Specimens used in this study were collected at 3 permanent trapping sites in eastern Slovakia (game park in Rozhanovce and cattle farms in Michalany and Tulcik), where the CDC miniature light trap model 1212 (John Hock Company, USA) was operated weekly, from April to November 2011-2015. In addition, occasional collections were done across Slovakia: Bysta (game park), Pcoline (cattle farm), Velaty (farm with different animals), Antalka (family house with animals), Ziar (family weekend house with sheep). The collection and identification procedures were described in the previous paper [19]. Specimens from Bulgaria, Kosovo, and Denmark were collected and identified within the VectorNet project, whereas the ones from France were from the French surveillance network funded by the French Ministry.
Morphological terminology follows Mathieu et al. [24] and Sarvašová et al. [19]. The antennal trichodea ratio (AtR), described by Meiswinkel [25], was measured for females to evaluate its diagnostic potential within the Pulicaris group. The ratio of the third segment, calculated by dividing the length of the first flagellomere by its width, previously evaluated by Nielsen et al. [22], was measured for females. On the eyes we investigated the usefulness of the new characters illustrated in Fig. 1. The length of the inter-ocular suture of the joined eyes (Lios) and the distance between the latter and the transverse suture above the first inter-ocular seta (Dios-ts) were measured. Subsequently, we computed the ratio consisting in Lios divided by Dios-ts; the length of the transverse suture was also reported (Lts). In order to discriminate males from the Pulicaris group, new morphological characters of aedeagus were measured and analyzed (Fig. 2): the length (Lap) and width (Wap) of the apical processes of the ninth tergite, the distance separating the two tips of the apical processes (Dt), and the distance separating the base of the two apical processes (Db). All measurements are in micrometres and are provided as the mean followed by the range (minimum-maximum) and the number of measurements in parentheses. The difference between measurements was determined by Mann-Whitney test (P < 0.05) using R software [26]. Characters measured for more than two groups were tested by Kruskal-Wallis rank sum test prior to Mann-Whitney.
The holotype and 10 paratypes are deposited in the collection of the Institut de Parasitologie et de Pathologie Tropicale de Strasbourg (IPPTS), 10 paratypes in the Slovak National Museum and 11 paratypes are deposited in the collection of the University of Veterinary Medicine and Pharmacy in Košice.

DNA barcode and phylogenetics
The following 53 cox1 (mtDNA) sequences have been gathered from the Genbank: C. boyi (JF766293-96), C. bysta n. sp. (KJ624118) referred as speA in a previous study [19], C. kalix (JF766328-31), C. lupicaris (HQ 824431-33, KJ624097), C. newsteadi Austen GB (AM236 742-46), C. newsteadi N1 (GQ338915, KJ624101), C. newsteadi N2 (GQ338916-20), C. newsteadi N3  In addition, we extracted the DNA from thorax and first abdominal segments of 17 specimens of C. bysta n. sp. and two C. boyi, using the DNeasy blood and tissue kit (Qiagen, USA). At the start of this study, barcoding fragments of the first 11 specimens have been amplified following the protocol of Pagès et al. [15] using the primers C1J1718/C1N2191 [27]. The resulting cox1 sequence size of 472 bp is similar to the most cox1 sequences available for Culicoides species. To produce longer sized amplicons allowing more complete genetic studies, the cox1 of the last 8 specimens have been amplified using the primer pairs LCO1490/HCO2198 [28] allowing a sequence size of up 687 bp. PCR products have been purified and sequenced by the Eurofins MWG Operon (Ebersberg, Germany). Information associating the origin and the accession number of the 19 newly sequenced specimens are presented in Table 1.
The 72 above mentioned cox1 sequences were aligned using the ClustalW [29] and genetic distances were computed using the Jukes-Cantor model of MEGA version 6 [30]. The best-fit model of nucleotide substitutions was calculated as HKY + I+ Γ by the JModelTest v.2.1.4 [31].
The latter model was used to parameterise a phylogenetic analysis carried out under the Bayesian Inference (1,000,000 generations), using the MrBayes v3.1.2 [32]. Two-thousand and five-hundreds of the saved trees were discarded and the remaining 7500 trees were used to construct the phylogenetic tree. Clade posterior probabilities (CPP) estimates were used to assess the robustness of tree nodes.

Male. [Based on the holotype and 11 paratypes;
Figs. 6d-f and 8.] Head: Eyes (Fig. 8c) bare, contiguous. Antenna (Fig. 8e, Table 3): sensilla coeloconica present on flagellomeres 1, 11-13 ( Table 3) Table 3 Distribution of sensilla coeloconica along the flagellomeres. Data are provided for females and males of C. bysta n. sp. and C. pulicaris from Slovakia (SK) and France (FR); no males of C. bysta n. sp. were observed and measured in France   Species  Country  Flagellomere   1  2  3  4  5  6  7  8  9  10  11  12  13 Female C. bysta n. sp.   Significant differences (***P < 0.001 and *P < 0.05) between C. bysta n. sp. and C. pulicaris are indicated considering the cumulative data from Slovakia (SK) and France (FR). No differences within the species and between the countries, or between C. bysta n. sp. and C. boyi, were observed. Culicoides boyi and C. pulicaris were significantly different for the AtR ratio (P < 0.05) a Data from Nielsen et al. [23] (n = 14); ratio between lengths of first long flagellomere and last short R11/10 = 2.3 (n = 14). Palpus (Fig. 8c, d) slender, palpal segment I with 1 long chaetica, segment II with 4 short chaeticae, segment III slightly swollen, carrying 4.6 short chaeticae, with multiple irregular pits located in third apical part of segment, segment IV with 3 short chaeticae, segment V without chaetica but with 5 apical bristles (n = 9); lengths of segments I-V: 67 (I + II Table 5); gonocoxite swollen in its basal part, middle part of internal edges lined with thick spines; ventral apodeme small, hook-shaped, dorsal apodeme cylindrical, robust; gonostylus barely longer than gonocoxite and twice as wide at basal edge as at apex, width of basal part reduced abruptly from 1/3 length to a parallel shape till the last 2/3 of gonostylus. Aedeagus (Fig. 8f ) Y-shaped with round, short tip and long lateral arms, straight and curved at base; moderately sclerotized arch present in proximal part, where lateral arms join body of aedeagus. Parameres (Fig. 8f ) separated, slender, becoming gradually very thin from proximal to distal part, tip of parameres terminating in fine pubescence. Immatures. Unknown.

Differential diagnosis
The combination of three characters of the wing pattern in females is discriminant to separate C. bysta n. sp. from C. pulicaris and C. boyi: (i) absence of a dark spot on the cubital-anal fork, (ii) a dark rounded spot in the distal part of m2, separated from the dark area on the CuA1 and, (iii) an incomplete and narrow hour-glass shape dark spot in r3 with the posterior margin short and narrow (Figs. 5a and 7a-d). In addition, several quantitative characters revealed significant differences, the most important being the Dios-ts and the ratio Dios-ts/Lios on the eyes (Table 6); the AtR ratio for C. bysta n. sp. is significantly smaller than that in C. pulicaris [1.50 (n = 52) vs 1.77 (n = 17); see Table 4]; the P/H ratio for C. bysta n. sp. is significantly smaller than that in C. pulicaris [0.77 (n = 26) vs 0.85 (n = 10); see Table 4]; the ratios of the first flagellomere are varied among C. pulicaris, C. boyi and C. bysta n. sp.: 1.52 ± 0.07 [22], 1.78 ± 0.07 [22], and 1.62 ± 0.1 (n = 33), respectively. Considering the lack of description of the male in C. boyi, the following diagnosis comments on the male of C. bysta are focused on comparison with C. pulicaris. The wing pattern of C. bysta n. sp. is similar to that in the female and may be used with confidence for the   Table 5 Measurements of male genitalia in micrometres presented as the mean followed by the range in parentheses Abbreviations: Lap length of the apical processus, Wap width of the apical processus, Dt distance between the tips of the apical processes, Db distance between the bases of apical processes, N1and N2 numbers of observations for Lap/Wap and Dt/Db, respectively Significant differences (***P < 0.001 and **P < 0.01) between C. bysta n. sp. and C. pulicaris are indicated considering the cumulative data from Slovakia (SK) and France (FR); no significant differences were found between geographical populations of C. pulicaris a The data from C. pulicaris were considered as a whole (C. pulicaris) or separately, regarding the origin in Slovakia (C. pulicaris SK) and France (C. pulicaris FR) identification of males. In addition, the measurement of Db on the ninth tergite can discriminate C. bysta n. sp. from C. pulicaris with 77 (65-89, n = 12) μm and 59 (38-74, n = 8) μm, respectively, even though a small overlap exists there (  [22,23,34].

Remarks
We evaluated the variations of the wing pattern of C. boyi and hereafter we present the summary of our observations (Fig. 7e, f ): cubital-anal fork dark extended to at least a third of the vein (n = 14); a dark spot present in the distal part of the anal cell; a dark rounded spot present in the distal part of m2, usually connected from the dark area on the CuA1 (n = 14) with the dark spot and area rarely separated (4/14 wings); tips of the veins M1, M2, and CuA1 dark, with sometimes a slightly pale spot in M1 (10/14 wings). Based on these observations, we suggest to use first the wing pattern to discriminate between C. bysta n. sp. and C. boyi (see above).

Discussion
The phylogenetic and morphological differences presented here justify the distinct status of C. bysta n. sp. within the Pulicaris group. The maximum interspecific genetic distance between the new species and C. boyi is low (7.6%), although greater than between the C. selandicus and the C. kalix (5.9%) [23]. For the most part, cox1 distances between species are usually found to be higher than 10% [15,18,19,23,35]. Thus, within the subgenus Culicoides, the lowest pairwise genetic distance was 12% between C. fagineus F1 and C. subfagineus (s.s.) in [15]. Moreover, the comparison of five closely related species within the subgenus Avaritia showed even lower genetic distance of 9.5% between C. bolitinos Meiswinkel, 1989 and C. tutti-frutti Meiswinkel, Cornet & Dyce, 2003 [34]. Although the genetic distance recorded between C. bysta n. sp. and C. boyi is low, the frequency distribution of pairwise genetic distances evidence a barcode gap between the intra-and the interspecific distances (Fig. 4). A similar graph was plotted to confirm the hypothesis of C. scoticus being a race of C. obsoletus [36]. As for the intraspecific distances, Pagès et al. [15] presented very low values (smaller than 0.6%) for all clades analyzed, including C. newsteadi N1. Moreover, the added specimen of C. newsteadi N1 sequenced in the previous study [19] exhibits a higher intraspecific genetic distance of 3.4% within the latter species. Thus the four existing and highly supported clades within C. newsteadi (sensu lato) indicate a clear need of an in-depth revision. Genetically and morphologically, C. boyi is the closest species to C. bysta n. sp. and both exhibit morphological similarities to C. pulicaris. Culicoides bysta n. sp. can be distinguished from C. boyi and C. pulicaris by the combination of the following characters on the wing pattern: (i) the absence of a dark spot on the cubital-anal fork; (ii) the presence of a dark rounded spot in the distal part of m2 separated from the dark area on the CuA1 vein; and (iii) an incomplete and narrow hour-glass shape dark spot in r3 with short and narrow posterior margin. In addition to the ratio of the first flagellomere, the two newly evaluated characters are the most important for discrimination. Thus the females of C. bysta n. sp. possess a Dios-ts/Lios ratio significantly lower than that in C. boyi and higher than that in C. pulicaris. For males, Db measurements allow accurate discrimination between C. bysta n. sp. and C. pulicaris. Nevertheless, the male of C. boyi remains undescribed and the usefulness of Db as the discriminating character for males of C. boyi and C. bysta n. sp. should be investigated in future.
Detected in Slovakia, C. bysta n. sp. was recorded in areas from the eastern to the western parts of the country. This species appeared to be present in various environments such as farms with domestic ruminants or horses, in forests hosting game animals, as well as in Abbreviations: Lios length of the inter-ocular suture of the joined eyes, Dios-ts distance between the inter-ocular suture and the transverse suture above the first inter-ocular seta, Lts length of the transverse suture, Dios-ts/Lios ratio of the above mentioned characters measured zoological gardens and family houses with domestic animals and poultry. First recorded in Slovakia, C. bysta n. sp. was afterwards identified in France. This species appeared to not be as rare because the specimens were found in various localities in the north-east of France.
During the preparation of the present description, additional specimens were also recorded in Bulgaria and Kosovo, indicating that C. bysta n. sp. may be widespread in Europe. As C. pulicaris is known to exhibit morphological variation [12][13][14], special attention should be paid in future studies on the Pulicaris group. For instance, at the time of submission of the present manuscript, an article was published with a new record of C. boyi for the fauna of France [34]. In the latter study, the genetic data fit perfectly with the data for C. boyi from Denmark, but morphologically the specimens from France exhibited variation. However, the combination of the above mentioned three characters may still be used for the accurate identification. Several species morphologically similar to C. pulicaris and thus close to the newly described C. bysta n. sp. were considered to be competent vectors for BTV and SBV transmissions. As for BTV, C. pulicaris and C. lupicaris were involved in the transmission by virus isolation and RT-PCR, respectively [2,7]. While both studies processed the pools of specimens identified by morphology, eventual presence of cryptic species within those pools remains possible. Similarly, a recent study implicated C. punctatus in the transmission of SBV by RT-PCR from pools of morphologically identified specimens [11]. To avoid the doubt which can be raised afterwards, studies dealing with Culicoides spp. should include molecular controls for the identification, such as (i) use of the diagnostic PCR tool for cryptic species within the subgenus Culicoides, for example [15], or (ii) sequencing of the barcode cox1 region as in a recent study [37] where cox1 was used to confirm the identification of specimens orally exposed to SBV. In the light of the increasing number of studies describing Culicoides spp. diversity, such as those providing evidence for the presence of cryptic species complexes and descriptions of new species, all studies using these midges as biological material should associate the molecular ID to their morphological ID. While none of the recent cryptic or new species have been involved in the arbovirus transmission so far, future studies focused on the evaluation of the role of Culicoides spp. in the transmission, taking into account the entire recently described diversity, may lead to overhaul the current knowledge of Culicoides transmitting diseases.

Conclusions
We described here C. bysta n. sp. as a new species belonging to the Pulicaris group of the subgenus Culicoides. This species is closely related to the recently described C. boyi and to C. pulicaris. The phylogenetic analyses based on cox1 and the morphological differences justify C. bysta n. sp. as a distinct species. Female specimens of this new species described here can be distinguished by the wing pattern and by the ratio between two sutures on the joined eyes. This latter morphological character evaluated here for the first time, and the characters on the ninth tergite for males, are promising for species discrimination within the Pulicaris group. However, male of C. boyi remain unknown and comparison of males within the Pulicaris group requires further studies. The vector potential of the recently described species C. boyi and C.bysta n. sp. to transmit arboviruses, such as BTV and SBV, is unknown. The published data on vector implication of C. pulicaris in BTV transmission acquired prior the description of the two recently described species, C. boyi and C. bysta n. sp., should be re-evaluated in future.