A phylogenetic analysis of the biting midges belonging to Culicoides Latreille (Diptera: Ceratopogonidae) subgenus Avaritia using molecular data
Parasites & Vectors volume 13, Article number: 243 (2020)
Within the genus Culicoides (Diptera: Ceratopogonidae), the subgenus Avaritia is of particular interest as it contains a significant number of economically important vector species. Disagreements about the systematic classification of species within this subgenus have resulted in a taxonomic imbroglio.
A molecular phylogeny of the subgenus Avaritia was conducted to test the existing systematic classification, which is based on phenetic assessment of morphological characters. Three nuclear ribosomal markers, internal transcribed spacer 1 and 2 (ITS1, ITS2), 5.8S, and three mitochondrial markers, cytochrome c oxidase subunit 1 and 2, and cytochrome b (cox1, cox2 and cytb), were obtained for 37 species of the subgenus Avaritia from all six biogeographical regions. Phylogenetic reconstructions using these genes independently and in combination were implemented using Bayesian inference analysis and maximum likelihood methods.
Phylogenetic reconstructions gave strong support to several monophyletic groups within the subgenus Avaritia. Both C. actoni and C. pusillus formed a single clade with C. grahamii so their respective groups, the Actoni and Pusillus groups, have been merged with the Grahamii group. Some support was provided for the Boophagus and Jacobsoni groups. A group of species currently placed into the Orientalis group clustered in a clade with poor support. The Obsoletus group was defined as a sister clade to all other Avaritia groups. The clade including the Imicola group was well supported based on phylogenetic criteria.
This phylogenetic study combining five distinct molecular markers has provided meaningful insights into the systematic relationships of Culicoides (Avaritia) and highlighted future directions to continue the study of this subgenus. While the cox2 marker appeared to be useful to investigate closely related species, the 5.8S marker was highly conserved and uninformative. Further investigations including species absent from this work are needed to confirm the proposed systematic scheme. However, this systematic scheme can now serve as a foundation to investigate cryptic species affiliation within the subgenus. We advocate that future studies employ a combination of morphological and molecular analyses.
The genus Culicoides Latreille (Diptera: Ceratopogonidae) is one of the most diverse groups of ceratopogonids, with over 1400 described species worldwide . The subgeneric classification of the genus has traditionally relied on morphological characters such as variations in wing patterns and male genitalia . The current subgeneric classification consists of 31 subgenera containing 63% of extant species, 38 unplaced groups of species made of 24% of extant species and a further 13% of extant species that have not been placed into any of these groups [1, 2].
Attempts to subdivide this large genus into subgenera have been partially successful. Adult morphological characters have most often been used to define these groupings although Glukhova  also relied on larval characters. Root & Hoffmann  and Edwards et al.  first proposed division of the genus into two series based on the male genitalia and other external features. Khalaf  subsequently attempted a classification based on features of male genitalia and female spermathecae but these two classification schemes are divergent in their grouping delimitation and species composition.
Starting in the 1950’s, some authors felt uncomfortable assigning species to untested subgenera and preferred to gather species in groups, many of which were not assigned to a subgenus [7, 8]. These contributions to the subgeneric classification were often based on regional assessments with limited attempts to rationalize groupings leading to numerous different species groups some of which were likely synonymous [9,10,11]. This approach was first used by Root & Hoffman  who separated the North American species into two “series”. Khamala & Kettle  preferred to group the 61 East African species into 16 species groups whereas Wirth & Blanton  grouped 88 species from Panama into 21 species groups and Campbell & Pelham-Clinton  placed 41 British species in 11 species groups. Partly due to the difficulty to collaborate in those days in terms of accessibility to material and communication, the authors classified the fauna based on their own geographical studies rather than placing them in context with an overarching assessment. Therefore, without any accordance on species groupings and their composition, duplication of species groupings and names emerged .
Confounding this main issue is the proliferation in use of the terms species group and species complex [2, 11]. The difficulty is that the International Code of Zoological Nomenclature (ICZN) does not recognize these levels as having a formal status even though their use is prevalent within Culicoides literature . The knowledge of the phylogenetic relationships between species, complexes and groups within subgenera is therefore very incomplete and has not yet been evaluated worldwide using a multi-molecular markers approach. Recent molecular phylogenetic studies suggest that at least some of the current subgenera are polyphyletic [12,13,14,15], prompting Borkent to consider the current classification as in “chaos” .
Shortly after the classification attempt by Khalaf , Fox  erected the subgenus Avaritia to encompass Khalaf’s Chiopterus species group. This subgenus was subsequently validated [9, 17] with the definition by Wirth & Hubert  being the most widely accepted description of the subgenus. The definition of Wirth & Hubert was, however, not followed by Yu et al.  who adopted a very different diagnosis resulting in the inclusion of a large number of Chinese species that Wirth & Hubert had affiliated to different groups or subgenera. Indeed, Dyce et al.  removed the species represented in Australasia back into the groupings defined by Wirth & Hubert  and added the Jacobsoni complex to the growing number of natural groupings within subgenus Avaritia [3, 17, 20].
The subgenus Avaritia includes species known to be vectors of viruses of ruminants and equids distributed worldwide causing economically important diseases . The first attempt at a worldwide systematic scheme within subgenus Avaritia was by Meiswinkel and colleagues [11, 22] who listed all species into the six species “complexes” known at the time: the Andicola, Grahamii (=Actoni sensu Dyce et al. ), Imicola, Obsoletus, Pusillus and Orientalis complexes. A tentative Pseudopallidipennis “subgroup” of the Imicola group as well as two further complexes (Gulbenkiani and Suzukii) were proposed. Indeed, each of these groupings had been subject to revision from several authors throughout the years. We propose hereafter a clarification for each group.
Khamala & Kettle  used adult features to delimit East Afrotropical species groups and grouped seven species (C. imicola Kieffer (syn. C. pallidipennis), C. grahamii Austen, C. kibatiensis Goetghebuer, C. trifasciellus Goetghebuer, C. tororoensis Khamala & Kettle, C. kanagai Khamala & Kettle and C. glabripennis Goetghebuer (syn. C. spinifer)) into the Pallidipennis group, which they thought was related to the Obsoletus group of Root & Hoffman  and Campbell & Pelham-Clinton  and the subgenus Avaritia described by Wirth & Blanton . Glick  changed the name of the group to the Imicola group after the synonymisation of C. pallidipennis with C. imicola. Due to the economic importance of several species of the group, it has received great attention. Culicoides imicola and related species were placed into the Orientalis group by Wirth & Hubert , whereas other authors have clearly distinguished the Imicola and the Orientalis groups . Glick  did not refer to the subgenus Avaritia and placed all the Ethiopian species into one Imicola group, including species which are currently not considered closely related to C. imicola, e.g. C. kanagai. More recently Chang et al.  provided a checklist of the Chinese species belonging to the subgenus Avaritia but followed the systematic system of Yu et al.  which included species that do not comply with the more widely accepted definitions of the subgenus and without any mention of groups within the subgenus. Meiswinkel  redefined the Imicola and the Orientalis groups based on adult features assigning 11 species to the Orientalis group and 12 species (including 3 undescribed species) to the Imicola group. In addition, Meiswinkel’s Pseudopallidipennis subgroup has been considered as a full species complex  though needing possible amendment on its delimitation .
The Andicola group was originally and adequately defined by Wirth & Lee  for three new species collected from high altitude sites in the Colombian Andes. More recently, two new species, also collected at high altitudes, were assigned to the Andicola group [27, 28] which now includes five Neotropical species (C. andicola Wirth & Lee, C. hermani Spinelli & Borkent, C. puracensis Wirth & Lee, C. orjuelai Wirth & Lee, and C. suarez Rodriguez & Wirth).
Wirth & Hubert  first defined the Actoni species group to accomodate the Oriental, Afrotropical and Australasian regions encompassing C. actoni, C. dasyops Clastrier and C. minimus Wirth & Hubert, and transferred C. grahamii from the Imicola group. Their diagnosis, species with pubescent eyes and the second radial cell being completely dark, overlapped with that of the Pusillus Group creating some confusion about the boundaries of these groups. Meiswinkel et al.  subsequently renamed this group after C. grahamii following the convention of naming a group after the oldest known member.
The Pusillus species complex of Wirth et al.  includes three species (C. pusillus Lutz, C. pusilloides Wirth & Blanton, and C. impusilloides Spinelli & Wirth) but so far lacks a precise definition.
Meiswinkel et al.  defined two additional species groups, the Gulbenkiani group for C. brevipalpis Delfinado, C. gulbenkiani Caeiro and C. tororoensis, and the Suzukii group including Culidoides wadai Kitaoka. The latter was referred to as the Boophagus group by Dyce et al.  as C. boophagus is the oldest described member. More recently, Bellis et al.  supported the monophyly of the Imicola complex and excluded C. brevipalpis from that species group based on morphological criteria and molecular analysis. In 2001, Dyce  was the first to mention a Jacobsoni group but neither this article nor Dyce’s pictorial atlas published in 2007  provide a diagnosis of this group.
Campbell & Pelham-Clinton  revised the taxonomic scheme of the British Culicoides fauna and proposed the Obsoletus group comprising the Chiopterus group of Khalaf  but excluding C. austeni Carter, Ingram & Macfie and including C. sinanoensis Tokunaga, C. jacobsoni Macfie (syn. C. kitaokai Tokunaga) and C. brevitarsis Kieffer (syn. C. robertsi Lee & Reye). The Palaearctic species of this group have recently received increasing attention due to the association of several species with outbreaks of bluetongue and Schmallenberg diseases [32, 33]. Due to similarities in wing patterns with C. obsoletus (Meigen), C. dewulfi Goetghebuer was considered part of the Obsoletus group [10, 34].
The lack of agreement of morphological characters that define subgenus Avaritia and its component species groups is fundamental to the confusion around the status and composition of species groups within the subgenus. An independent means of testing the limits of these groups would lend confidence to their status and help clarify those morphological characters which have phylogenetic importance. Molecular phylogenetic studies have been used successfully to test systematic proposals of many types of organisms [35, 36] that are based on morphological evidence and could prove useful in clarifying the classification of subgenus Avaritia.
A range of markers have been used to test the phylogeny of Culicoides (Avaritia) with varying success. One study combining data from cox1 and cox2  did not support the monophyly of the subgenus, casting doubt over the usefulness of mitochondrial markers for higher-level phylogenetic studies. Other studies have, however, successfully used cox1 to investigate relationships at a lower level for example, within the Imicola group [13, 38]. Studies using either internal transcribed spacer 1 (ITS1)  or 2 (ITS2)  alone or in combination  have also been used successfully to investigate species relationships within subgenus Avaritia. Additionally, cox1 has been combined with other markers in a multi-marker approach to investigate the phylogeny of Culicoides (Avaritia). While a first study based on cox1 and CAD (carbamoyl phosphate synthetase 2) supported the monophyly of the Imicola complex , a second has reported support for several subgeneric groupings and highlighted the presence of cryptic species . Combining cox1, ITS1 and ITS2, provided evidence to exclude C. dewulfi from the Obsoletus group . In 2018, a comprehensive phylogenetic study of the Afrotropical fauna, including several species from subgenus Avaritia, used two mitochondrial (cox1 and 16S rDNA) and two nuclear (CAD and 28S rDNA) markers to support the monophyly of the Imicola and Dasyops groups .
This study aims to (i) test species group delimitation within subgenus Avaritia; (ii) elucidate the phylogenetic relationships among the different species groups; and (iii) propose a new systematic scheme of the subgenus Avaritia, at a worldwide level, using six genetic markers including nuclear and mitochondrial molecular targets (ITS1, ITS2, 5.8S, cox1, cox2, cytb).
Taxon sampling and species identification
A total of 191 specimens representing 35 species belonging to Culicoides (Avaritia) were collected from 21 countries from all six biogeographical regions of the world (Additional file 1: Table S1). The sampling effort was heavily represented by specimens from the Palaearctic and Afrotropical regions. Species identification was determined or confirmed by either the experts who kindly provided specimens or by the authors of this paper using the following morphological studies and keys [7, 8, 17, 19, 44,45,46,47]. Culicoides (Meijerehelea) arakawae (Arakawa), the only species of Culicoides with publicly available mitochondrial and rDNA sequences, was selected as the outgroup. Two additional species belonging to Culicoides (Trithecoides), i.e. C. (T.) humeralis Okada and C. (T.) matsuzawai Tokunaga, were also added as outgroups because of the availability of three of the five genetic markers used in this study.
DNA extraction, genome amplification by PCR and sequencing
A semi-destructive method was used to extract DNA from specimens. Each specimen was soaked in a saturated phenolic solution before dissection and mounted in Canada balsam following the procedure described in  except for the thorax and the first five abdominal segments that were retained for DNA extraction. Total genomic DNA was extracted from the tissues digested with Proteinase K at 10%, using the DNeasy Tissue kit (Qiagen, Valencia, CA, USA) following the manufacturer’s instructions.
In addition to the barcode region cytochrome c oxidase subunit 1 (cox1), cytochrome c oxidase subunit 2 (cox2) and cytochrome b (cytb) mitochondrial gene fragments (mt) were amplified. One ribosomal DNA (rDNA) gene fragment was studied: the internal transcribed spacer 1 and 2 (ITS1 and ITS2) including the coding 5.8S region. PCR reactions were performed in a total volume of 50 µl consisting of 1× PCR reaction buffer, 2.5 mM of MgCl2, 250 μM each of dATP, dCTP, dGTP and dTTP (Invitrogen, Carlsbad, CA, USA), 200 ng of each primer, and 2.5 U of HotStart Taq DNA polymerase (Qiagen) and 1 μl of template DNA. Primer sets used are listed in Table 1. PCR cycling conditions for mtDNA were as follows: an initial denaturation step at 95 °C for 15 min, then 5 cycles at 95 °C for 40 s, 45 °C for 40 s, 72 °C for 1 min, followed by 45 cycles at 95 °C for 40 s, 50 °C for 40 s, 72 °C for 1 min and a final extension step at 72 °C for 20 min. To amplify the ITS1-5.8S-ITS2 region, the PCR cycling conditions were as follows: an initial denaturation step at 95 °C for 15 min, then 40 cycles at 95 °C for 30 s, 54 °C for 1 min, 72 °C for 1 min and a final extension step at 72 °C for 20 min. PCR products were examined by electrophoresis in a 2% ethidium bromide agarose gel.
PCR products were sequenced in both directions by Eurofins MWG Operon (Edersberg, Germany). The rDNA ITS1-5.8S-ITS2 region was sequenced after a cloning step into the PCR-Blunt vector (Zero Blunt PCR Cloning Kit, Invitrogen), using chemically competent E. coli. Plasmid DNA was extracted using a Plasmid DNA Preparation Kit (Nucleospin® Plasmid, Macherey Nagel, USA). Bam HI (Biolabs, UK) restriction endonuclease digestion and quantification of DNA plasmid copies were carried out by measuring with a spectrophotometer DNA concentration (wavelength of 280 nm) before sequencing.
Among the 35 species investigated in this study, the number of specimens of each species successfully sequenced for each marker is presented in Table 2.
For each marker, the complete list of specimens, species names, specimen codes, accession numbers including the sequences retrieved from GenBank are provided in Additional file 1: Table S1. For only one species, C. dentiformis McDonald & Lu from Taiwan, were no sequences available for any of the 5 amplified genetic markers. cox1 barcodes for C. candolfii Delécolle, Paupy, Rahola & Mathieu, C. hui Wirth & Hubert and C. kwagga Meiswinkel were retrieved from GenBank and added to the dataset. As a result, the complete dataset includes 37 species. The cox1 dataset contained 174 specimens from 21 countries representing 34 species, while 101 cox2 sequences were generated for 22 species from 18 countries. For cox1 and cox2, respectively 15 and 8 sequences from GenBank were added. The cytb dataset included 138 sequences representing 27 species from 20 countries while the ITS1+ITS2 dataset contained 67 sequences representing 28 species from 17 countries, with an additional 9 sequences retrieved from GenBank. Sequences of C. arakawae for the five genetic markers of the study were added to the dataset.
Alignment and phylogenetic analyses
The mtDNA sequences (cox1, cox2 and cytb) were originally aligned using Vector Nti v11.5.0 (Invitrogen) and checked using ClustalW . Alignment of non-coding rDNA sequences (ITS1, 5.8S, ITS2) was performed with MAFFT  and gaps were removed using MEGA version 5 . Alignments of each of the mtDNA regions were translated into peptide sequences using MEGA version 5  in order to exclude putative NUMt copies.
Heterogeneity tests of nucleotide frequency among specimens were performed using DAMBE . Using the same software, potential saturation of each dataset was investigated graphically by plotting the absolute number of transitions and transversions against Tamura-Nei model (referred as ML Composite TN93) distance for all pairwise comparisons of sequences. The sequences revealing a saturated phylogenetic signal were discarded from the final dataset. The remaining sequences were concatenated using Seaview  to be analysed as a whole fragment with a partitioning strategy as described below.
Phylogenetic reconstructions were performed for each dataset using both non-probabilistic and probabilistic approaches. The non-probabilistic approach, maximum parsimony (MP), was performed using MEGA with bootstrapping method (n = 200) considering all sites informative (missing data non-excluded) and with the Close-Neighbor-Interchange (CNI) search method on Random Trees. The probabilistic approaches were Maximum Likelihood (ML) and Bayesian inference (BI). To define model parameters, the best-fit model of nucleotide substitutions was calculated with JModelTest v.0.1.1  using the selection of the Akaike information criterion (AIC). The general time reversible (GTR) + I + Γ model was indicated as the best-fit model for the cox1, cox2 and cytb considered separately. The GTR + Γ model was the best-fit one for the ITS1 and the ITS2 considered separately whereas the Hasegawa Kishino and Yano (HKY) + Γ model was indicated for the combined ITS1+ITS2 dataset. The three mtDNA genes as a combined dataset followed the GTR + Γ model.
ML phylogenetic analyses were carried out with PhyML v3.0  according to the respective best-fit model. The transition/transversion ratio, the Γ parameter and if necessary, the proportion of invariant sites were estimated. The starting tree was determined by BioNJ analysis and the branch was tested by the Shimodaira-Hasegawa-like branch test.
Phylogenetic analyses were also carried out under BI using MrBayes v3.1.2 . To increase the fit of evolutionary models with data, we used partitioned analyses, which allow subsets of the data to evolve under distinct models and parameters [57, 58]. Three partitioning strategies were defined a priori: strategy P1, corresponding to an unpartitioned analysis; strategy P2, implementing one partition for the mitochondrial genes and one partition for the nuclear markers (ITS1+ITS2); and strategy P3, which implements a partition for each genetic marker cox1, cox2, cytb and ITS1+ITS2. For each partitioning strategy, independent runs of 2,000,000 generations were carried out applying appropriate models of evolution to each partition. A total number of 5000 of the saved trees were discarded and the remaining 15,000 trees were used to construct the BI trees. Clade posterior probabilities (cpp) estimates were used to assess the robustness of tree nodes. The best-fit partitioning strategy was determined by comparing the Bayes factors (BF) estimated as twice the difference of harmonic means.
Molecular phylogeny analysis
Heterogeneity tests on nucleotide frequency were not significant for all the genes except for cox2 (χ2 = 607.43, df = 330, P < 0.001). The heterogeneity of nucleotide frequency for cox2 investigated by codon position showed homogeneity on codons 1+2 (χ2 = 160.65, df = 330, P = 1), and heterogeneity on codon 3 (χ2 = 1672.44, df = 330, P < 0.001) with high transition saturation rate. The cox2 gene was therefore analysed only on codon positions 1+2 (Fig. 1). The saturation plot for the 5.8S gene showed transversion information saturation indicating that this region is highly conserved and uninformative; this locus was removed from the analysis. The exclusion of indels from alignments of the ITS1 and ITS2 datasets produced short fragments of 221 bp and 127 bp, respectively. The two ITS regions were analysed independently and concatenated (Fig. 1). No saturation of the phylogenetic signal was observed for the cox1 and cytb dataset.
Phylogenetic reconstructions were carried out on four datasets comprising cox1, cox2 (codons 1+2), cytb and ITS1+ITS2, representing 498-bp, 352-bp, 507-bp and 348-bp fragments, respectively. A combined dataset was also subjected to analysis with all these datasets giving a concatenated 1705 bp long fragment.
As tree reconstructions using MP and ML resulted in similar topologies and identical species clade composition regardless of the dataset used, only results produced using ML are presented (Fig. 2). Although similarity in topologies with ML were observed, clades representing species groups were not strongly supported (bootstrap values less than 50%) in MP reconstruction, except for the cytb marker for which the clade Imicola group was supported by a 92% bootstrap value (data not shown).
Maximum likelihood reconstruction
Figure 2 represents the ML reconstructions for the four datasets analysed. Three main clades were identified representing species respectively placed into the Obsoletus, Imicola and Grahamii groups. The analysis of the ITS1+ITS2 region showed a well-supported basal topology and these three groups were well supported with SH-Like values above 91%. Intra-group nodes were quite well supported for the Imicola and Grahamii groups but not for the Obsoletus group (not shown as this latter clade is collapsed to provide a better insight of topology).
The Imicola group clade was well supported in each analysis although two clades were consistently observed for C. bolitinos Meiswinkel (Fig. 2). The Obsoletus group clade was supported by cox1, cytb and ITS1+ITS2 but the cox2 analysis differed as it included C. dewulfi in this group (Fig. 2). The basal part of the tree topologies of the cox1 dataset showed lower support than the other datasets.
Bayesian inference on combined dataset
Comparison of the Bayes factor obtained from the three distinct partitioning strategies on the combined data run independently favours the strategy P2. No topology incongruence was detected between well-supported nodes on separate ML analysis and the combined analyses for the three BI strategies. Moreover, the combined analysis showed a well-supported topology of both deep and terminal nodes; the strategy P2 was therefore considered as the best estimate of the phylogeny of subgenus Avaritia based on our dataset. Figure 3 shows the BI tree resulting from the phylogenetic analysis of the combined dataset according to the strategy P2, i.e. one partition for the three mitochondrial genes and one partition for the two nuclear markers.
With the exception of the position of C. dewulfi, C. jacobsoni, C. brevipalpis and C. kibatiensis, the major part of MP, BI and ML topologies were congruent. In the resulting topology (Fig. 3), most nodes were supported by cpp-values above 91%. The Obsoletus, Imicola and Grahamii groups were all supported by cpp-values of 99–100%. Furthermore, the BI analysis provided support for the existence of the Boophagus group by a cpp-value of 91%. The monophyletic status of the Obsoletus and Imicola groups was strongly supported with cpp-values of 100% and 99%, respectively. Within the Imicola group, the recently described species C. asiana Bellis  was highly supported (cpp of 100%) as very close to the Australian species C. brevitarsis. The clade containing C. orientalis Macfie, C. fulvus Sen & Das Gupta, C. tainanus Kieffer and C. obscurus Tokunaga & Murachi was not strongly supported (cpp of 65%). The placement of C. gulbenkiani with either C. tororoensis or C. brevipalpis was not supported in this study.
To our knowledge, this is the first systematic study of Culicoides (Avaritia) based on a five-marker phylogenetic approach and including species from all biogeographical regions. Among the six genetic markers studied, the mitochondrial cox2 and cytb markers both showed strong phylogenetic signal to separate closely related species but were less informative for basal topology. The combined non-coding region ITS1+ITS2 showed to be useful to describe basal tree topologies but 5.8S was entirely uninformative. The cox1 region supported both basal topology and the discrimination of closely related species. cox2, newly tested for intra-subgeneric systematic research, provided informative knowledge of in-group relationships for closely related clades.
Phylogenetic relationships within subgenus Avaritia
Combined phylogenetic information from several distinct genes has provided some insight on the relationships within the subgenus and between species groups.
With the exception of C. kanagai, the species studied which have the combination of cell r2 dark to apex and hairy eyes, including C. actoni, C. grahamii and C. pusillus, were found to be phylogenetically clustered in the Grahamii group. Despite its closer morphological similarity to species of the Grahamii group, our molecular analysis placed C. kanagai closer to C. tororoensis and C. gulbenkiani than to the Grahamii group clade. Culicoides tororoensis and C. gulbenkiani, however, differ morphologically from C. kanagai as they are large species, lack interfacetal hairs and have three pale spots in the third radial cell. In the Oriental region, C. hui has this latter feature but differs as cell r2 is pale apically. Both Wirth & Hubert  and Meiswinkel  placed C. hui but not C. tororoensis or C. gulbenkiani, into the Orientalis group but none of these species were placed close to C. orientalis in any of our phylogenetic reconstructions. Further investigations to unravel the relationships within the Orientalis group would be necessary including species such as C. hui, C. flavipunctatus Kitaoka, C. dumdumi Sen & Das Gupta, C. brosseti Vattier & Adam, C. dubitatus Kremer, Rebholtz-Hirtzel & Delécolle and C. trifasciellus. Bakhoum et al.  in their comprehensive phylogenetic study of the Afrotropical species, included C. trifasciellus and the putative new species Culicoides sp. #20 into the Orientalis group. The phylogenetic data of species morphologically affiliated to the Orientalis group sensu Meiswinkel  are currently sparse and a comprehensive integrative taxonomic study of this group is required.
Morphological feature variations are often observed in the genus Culicoides especially for species with large spatial distribution. Similarly, genetic variations were observed in our dataset for species present across the Oriental and Australasian regions (C. actoni, C. fulvus, C. brevipalpis and C. jacobsoni) and could be indicative of cryptic diversity or on-going speciation. Gopurenko et al.  reported similar cryptic diversity for C. actoni, C. brevipalpis and C. jacobsoni while C. fulvus appeared to be genetically uniform across Australasia and East Asia. Similarly, our study showed genetic diversity between C. bolitinos specimens from South Africa and La Reunion/Madagascar which is consistent with observations of recent studies [43, 59]. These C. bolitinos populations showed genetic distances similar to that between C. brevitarsis and C. asiana supporting the existence of a cryptic species of C. bolitinos. Further studies incorporating specimens from populations across the full distribution of this species are required to clarify the status of this species.
Species of Andicola group are endemic to high altitudes from 1600 to 3200 m in the central and western ranges of the Andes [27, 28]. Due to their rarity and difficulty to access, we were unable to secure any specimens belonging to Andicola group for inclusion in this study. Consequently, the monophyly and placement of this group within the subgenus remains untested.
The status of the Obsoletus group is supported by molecular analysis to include the following 8 species: C. abchazicus Dzhafarov, C. alachua Jamnback & Wirth, C. chiopterus (Meigen), C. montanus Shakirzjanova, C. obsoletus, C. sanguisuga (Coquillett), C. scoticus Downes & Kettle and C. sinanoensis. Culicoides dewulfi has a wing pattern consistent with members of the Obsoletus group but has male genitalia more similar to those in species of the Imicola group which has led to conflicting opinions about its placement within this group [10, 34]. In this study, C. dewulfi was clearly phylogenetically separate from any group, including the Obsoletus and Imicola groups, which agrees with the conclusions of other studies [15, 60]. Evidence of cryptic species diversity within C. obsoletus has been reported previously from Sweden , Switzerland  and the UK  but was not supported in our analysis of specimens from France, Norway, Spain and the USA. Culicoides gornostaevae Mirzaeva, a species recorded from Russia to Japan, was recently recorded in Norway, Poland and Sweden  but the latter study did not investigate the cox1 marker and whether these specimens morphologically identified as C. gornostaevae fit with one of the genetic forms of C. obsoletus remain to be investigated.
The monophyly of the Imicola group within subgenus Avaritia was well supported in the present analyses, in agreement with previous reports [22, 35]. The placement of all species into the Imicola group by Meiswinkel  and Bellis et al. , i.e. C. asiana, C. bolitinos, C. brevitarsis, C. imicola, C. kwagga, C. loxodontis Meiswinkel, C. miombo Meiswinkel, C. nudipalpis Delfinado, C. pseudopallidipennis Clastrier and C. tuttifrutti Meiswinkel, Cornet & Dyce, were supported in the present study. Based on the geographical origin of the specimens studied, we therefore report here C. asiana as a new record for Vietnam and C. imicola as a new record for Taiwan.
In their comprehensive study of the Culicoides from Southeast Asia, Wirth & Hubert  mention the morphological similarity between the Actoni group, comprising the Oriental and Australian species C. actoni and C. minimus, and the Pusillus group of Wirth et al. . Their assessment combined with our phylogenetic results suggest merging the Actoni and Pusillus groups into the Grahamii group, this name having precedence as C. grahamii is the oldest described member of the group. The placement of C. minimus, C. pusilloides and C. impusilloides remains untested as these were not included in this study but their previous placement into the Actoni and Pusillus groups based on morphological evidence suggests they are related and so we place them tentatively into the Grahamii group.
Despite somewhat relatively weak support, the Boophagus group of Dyce et al.  (= Suzukii group of Meiswinkel et al. ) including C. boophagus Macfie, C. suzukii Kitaoka sensu Lien et al.  and C. wadai appears to be a valid group. Morphologically, we define this group as having a quadrate dark area at the tip of vein M2 surrounded by a pale area. However, the illustrations of C. suzukii provided by Kitaoka show the dark area covering the vein M2 being continuous to the dark quadrate area on tip of the vein M2 [45, 65, 66] which differs to the specimen illustrated by Lien et al. . The specimen from Taiwan examined by us is morphologically consistent with C. suzukii sensu Lien et al. , and until the discrepancies between these interpretations of this species are clarified, the position of “true” C. suzukii sensu Kitaoka  within the Boophagus group remains unknown.
First mentioned by Dyce et al.  and Dyce , the Jacobsoni group was composed of two species, one of which is undescribed, until an integrative taxonomic study incorporating cox1 and CAD analysis by Gopurenko et al.  revealed the existence of a further four undescribed species making a total of six species. These authors also provided evidence for the monophyly of the Jacobsoni group. Even though our study has included only two of the cryptic species reported by Gopurenko et al. , our results would provide some support to the monophyly of the Jacobsoni group.
The combined use of five distinct molecular markers belonging to nuclear (ITS1 and ITS2) and mitochondrial DNA (cox1, cox2 and cytb) provided meaningful insights in relation to taxonomic studies and provided direction for future studies. Newly studied for Culicoides phylogeny, the cox2 gene appeared to be useful to investigate closely related species. On the opposite, the 5.8S marker was highly conserved and uninformative. The groups mentioned in this study represented monophyletic clades and the taxonomic rearrangement proposed was based on phylogenetic evidence. Within subgenus Avaritia, the Obsoletus group was sister to all other groups included in this study. The Imicola group was phylogenetically robust, morphologically well defined, and a cryptic diversity within C. bolitinos remains to be investigated. Similarly, cryptic diversity was identified in this study within C. actoni, C. brevipalpis, C. fulvus and C. jacobsoni, which opens an entire field of investigation. The Actoni, Pusillus and Grahamii groups have been merged and some support was provided for the Boophagus and Jacobsoni groups. The monophyly and composition of the Orientalis group remains unclear and further investigations would be necessary to improve its definition. This study at a global scale, has allowed a better definition of the subgenus Culicoides (Avaritia) and as such it constitutes a step forward towards a global reassessment of the systematic and phylogenetics of the genus Culicoides.
Availability of data and materials
The datasets used during the current study are available from the corresponding author upon reasonable request.
Borkent A. World species of biting midges (Diptera: Ceratopogonidae). Belmont University Ceratopogonid Web Page. 2015;241. http://campus.belmont.edu/cienews/cie.html. Accessed 17 Sep 2019.
Harrup LE, Bellis GA, Balenghien T, Garros C. Culicoides Latreille (Diptera: Ceratopogonidae) taxonomy: current challenges and future directions. Infect Genet Evol J Mol Epidemiol Evol Genet Infect Dis. 2015;30:249–66.
Glukhova VM. The subgeneric classification of the genus Culicoides Latreille, 1809 (Diptera, Ceratopogonidae), including morphological characters of the larva [in Russian]. Parazitol Sb. 1977;27:112–8.
Root FM, Hoffman WA. The north American species of Culicoides. Am J Epidemiol. 1937;25:150–76.
Edwards FW, Oldroyd H, Smart J. British blood-sucking flies. London: British Museum; 1939.
Khalaf K. The speciation of the genus Culicoides (Diptera, Heleidae). Ann Entomol Soc Am. 1954;47:34–51.
Glick JI. Culicoides Biting midges (Diptera: Ceratopogonidae) of Kenya. J Med Entomol. 1990;27:85–195.
Khamala CM, Kettle DS. The Culicoides Latreille (Diptera: Ceratopogonidae) of East Africa. Trans R Ent Soc Lond. 1971;123:1–95.
Blanton FS, Wirth WW. The sandflies (Culicoides) of Florida (Diptera: Ceratopogonidae). In: Blanton FS, Wirth WW, editors. Arthropods of Florida and Neighbouring Land Areas. Gainesville: Florida Department of Agriculture and Consumer Services; 1979. p. 204.
Campbell JA, Pelham-Clinton EC. A taxonomic review of the British species of Culicoides Latreille (Diptera: Ceratopogonidae). Proc R Soc Edinb. 1960;68:181–302.
Meiswinkel R. Adult characters defining and separating the Imicola and Orientalis species complexes of the subgenus Avaritia Fox, 1955 (Culicoides, Diptera: Ceratopogonidae). Vet Ital. 2004;40:345–51.
Gomulski LM, Meiswinkel R, Delécolle JC, Goffredo M, Gasperi G. Phylogeny of the subgenus Culicoides and related species in Italy, inferred from internal transcribed spacer 2 ribosomal DNA sequences. Med Vet Entomol. 2006;20:229–38.
Linton YM, Mordue Luntz AJ, Cruickshank RH, Meiswinkel R, Mellor PS, Dallas JF. Phylogenetic analysis of the mitochondrial cytochrome oxidase subunit I gene of five species of the Culicoides imicola species complex. Med Vet Entomol. 2002;16:139–46.
Perrin A, Cetre-Sossah C, Mathieu B, Baldet T, Delecolle J-C, Albina E. Phylogenetic analysis of Culicoides species from France based on nuclear ITS1-rDNA sequences. Med Vet Entomol. 2006;20:219–28.
Schwenkenbecher JM, Mordue AJ, Piertney SB. Phylogenetic analysis indicates that Culicoides dewulfi should not be considered part of the Culicoides obsoletus complex. Bull Entomol Res. 2009;99:371–5.
Fox I. A catalogue of the bloodsucking midges of the Americas (Culicoides, Leptoconops and Lasiohelea) with keys to the subgenera and neartic species. P R Univ J Agr. 1955;39:214–85.
Wirth WW, Hubert AA. The Culicoides of southeast Asia (Diptera: Ceratopogonidae). Mem Am Entomol Inst. 1989;44:1–508.
Yu Y-X, Liu J-H, Liu G-P, Liu Z-J, Hao B-S, Yan G, et al. Ceratopogonidae of China, Insecta, Diptera [in Chinese]. Beijing, China: Military Medical Science Press; 2005.
Dyce AL, Bellis GA, Muller MJ. Pictural atlas of Australiasian Culicoides wings (Diptera: Ceratopogonidae). Canberra: ABRS; 2007.
Boorman J. Taxonomic problems in Culicoides of southwest Asia, in particular of the Arabian peninsula. In: Service MW, editor. Biosystematics of Haematophagous Insects Systematics Association, Special Volume No. 37. Oxford: Clarendon Press; 1988. p. 271–82.
Purse BV, Carpenter S, Venter GJ, Bellis G, Mullens BA. Bionomics of temperate and tropical Culicoides midges: knowledge gaps and consequences for transmission of Culicoides-borne viruses. Annu Rev Entomol. 2015;60:373–92.
Meiswinkel R, Gomulski LM, Delécolle JC, Goffredo M, Gasperi G. The taxonomy of Culicoides vector complexes—unfinished business. Vet Ital. 2004;40:151–9.
Wirth WW, Blanton FS. Biting midges of the genus Culicoides from Panama (Diptera : Heleidae). Proc U S Natl Mus. 1959;109:237–482.
Chang QQ, Jiang XH, Liu GP, Li XF, Hou XH. A species checklist of the subgenus Culicoides (Avaritia) in China, with a description of a new species (Diptera, Ceratopogonidae). ZooKeys. 2017;706:117–35.
Meiswinkel R, Linton YM. Afrotropical Culicoides Latreille (Diptera: Ceratopogonidae): morphological and molecular description of a novel fruit-inhabiting member of the Imicola Complex, with a re-description of its sister species C. (Avaritia) pseudopallidipennis Clastrier. Cimbebasia. 2003;1:37–79.
Wirth WW, Lee VH. New species of Culicoides from high altitudes in the Colombian Andes (Diptera: Ceratopogonidae). Proc U S Natl Mus. 1967;124:1–22.
Rodriguez MC, Wirth WW. A new species of man-biting Culicoides from the high Andes in Colombia (Diptera: Ceratopogonidae). Fla Entomol. 1986;69:311–4.
Spinelli GR, Borkent A. New species of central America Culicoides Latreille (Diptera: Ceratopogonidae) with a synopsis of species from Costa Rica. Proc Entomol Soc Wash. 2004;106:361–95.
Wirth WW, Dyce AL, Spinelli GR. An atlas of wing photographs, with a summary of the numerical characters of the neotropical species of Culicoides (Diptera: Ceratopogonidae). Contrib Am Entomol Inst. 1988;25:71.
Bellis G, Dyce A, Gopurenko D, Yanase T, Garros C, Labuschagne K, et al. Revision of the Culicoides (Avaritia) Imicola complex Khamala & Kettle (Diptera: Ceratopogonidae) from the Australasian region. Zootaxa. 2014;3768:401–27.
Dyce AL. Biogeographic origins of species of the genus Culicoides (Diptera: Ceratopogonidae) from the Australasian Region. IIni Arbovirus Res Aust. 2001;8:133–40.
Carpenter S, Wilson A, Mellor PS. Culicoides and the emergence of bluetongue virus in northern Europe. Trends Microbiol. 2009;17:172–8.
Beer M, Conraths FJ, van der Poel WHM. ’Schmallenberg virus’—a novel orthobunyavirus emerging in Europe. Epidemiol Infect. 2013;141:1–8.
Kremer M. Contribution à l’étude du genre Culicoides Latreille particulièrement en France. In: Chevalier P, editor. Paris: Encyclopédie Entomologique Série A; 1965.
de Santana Souza D, Marinoni L, Laura Monné M, Gómez-Zurita J. Molecular phylogenetic assessment of the tribal classification of Lamiinae (Coleoptera: Cerambycidae). Mol Phylogenet Evol. 2020;145:106736.
Craig JM, Kim LY, Tagliacollo VA, Albert JS. Phylogenetic revision of Gymnotidae (Teleostei: Gymnotiformes), with descriptions of six subgenera. PLoS ONE. 2019;14:e0224599.
Matsumoto Y, Yanase T, Tsuda T, Noda H. Species-specific mitochondrial gene rearrangements in biting midges and vector species identification. Med Vet Entomol. 2009;23:47–55.
Dallas JF, Cruickshank RH, Linton YM, Nolan DV, Patakakis M, Braverman Y, et al. Phylogenetic status and matrilineal structure of the biting midge, Culicoides imicola, in Portugal, Rhodes and Israel. Med Vet Entomol. 2003;17:379–87.
Mathieu B, Perrin A, Baldet T, Delécolle J-C, Albina E, Cêtre-Sossah C. Molecular identification of Western European species of obsoletus complex (Diptera: Ceratopogonidae) by an internal transcribed spacer-1 rDNA multiplex polymerase chain reaction assay. J Med Entomol. 2007;44:1019–25.
Gomulski LM, Meiswinkel R, Delécolle JC, Goffredo M, Gasperi G. Phylogenetic relationships of the subgenus Avaritia Fox, 1955 including Culicoides obsoletus (Diptera: Ceratopogonidae) in Italy based on internal transcribed spacer 2 ribosomal DNA sequences. Syst Entomol. 2005;30:619–31.
Matsumoto Y, Yanase T, Tsuda T, Noda H. Characterization of internal transcribed spacer (ITS1)-ITS2 region of ribosomal RNA gene from 25 species of Culicoides biting midges (Diptera: Ceratopogonidae) in Japan. J Med Entomol. 2009;46:1099–108.
Gopurenko D, Bellis GA, Yanase T, Wardhana AH, Thepparat A, Wang J, et al. Integrative taxonomy to investigate species boundaries within Culicoides (Diptera: Ceratopogonidae): a case study using subgenus Avaritia from Australasia and Eastern Asia. Vet Ital. 2015;51:345–78.
Bakhoum MT, Labuschagne K, Huber K, Fall M, Mathieu B, Venter G, et al. Phylogenetic relationships and molecular delimitation of Culicoides Latreille (Diptera: Ceratopogonidae) species in the Afrotropical region: interest for the subgenus Avaritia. Syst Entomol. 2018;43:355–71.
Glukhova VM. Culicoides (Diptera, Ceratopogonidae) of Russia and adjacent lands. Dipterol Res. 2005;16:75.
Kitaoka S. Descriptions of four new species and the hitherto unknown males of four species of the Culicoides (Diptera, Ceratopogonidae) from Amami-Oshima, Japan. Natl Inst Anim Health Q (Tokyo). 1973;13:211–9.
Lien JC, Weng MH, Lin CC. A revision of the genus Culicoides in Taiwan. Part II Subgenus Avaritia (Diptera: Ceratopogonidae). J Taiwan Mus. 1998;51:21–48.
Meiswinkel R, Baylis M. Morphological confirmation of the separate species status of Culicoides (Avaritia) nudipalpis Delfinado, 1961 and C. (A.) imicola Kieffer, Diptera: Ceratopogonidae. Onderstepoort J Vet Res. 1913;1998(65):9–16.
Wirth WW, Marston N. A method for mounting small insects on microscope slides in Canada balsam. Ann Ent Soc Am. 1968;61:783–4.
Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994;22:4673–80.
Katoh K, Asimenos G, Toh H. Multiple alignment of DNA sequences with MAFFT. Methods Mol Biol. 2009;537:39–64.
Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. MEGA5: Molecular Evolutionary Genetics Analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 2011;28:2731–9.
Xia X, Xie Z. DAMBE: software package for data analysis in molecular biology and evolution. J Hered. 2001;92:371–3.
Gouy M, Guindon S, Gascuel O. SeaView version 4: A multiplatform graphical user interface for sequence alignment and phylogenetic tree building. Mol Biol Evol. 2010;27:221–4.
Posada D. jModelTest: phylogenetic model averaging. Mol Biol Evol. 2008;25:1253–6.
Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel O. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 30. Syst Biol. 2010;59:307–21.
Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19:1572–4.
Brandley MC, Schmitz A, Reeder TW. Partitioned Bayesian analyses, partition choice, and the phylogenetic relationships of scincid lizards. Syst Biol. 2005;54:373–90.
Nylander JA, Ronquist F, Huelsenbeck JP, Nieves-Aldrey JL. Bayesian phylogenetic analysis of combined data. Syst Biol. 2004;53:47–67.
Desvars A, Grimaud Y, Guis H, Esnault O, Allène X, Gardès L, et al. First overview of the Culicoides Latreille (Diptera: Ceratopogonidae) livestock associated species of Reunion Island. Indian Ocean. Acta Trop. 2015;142:5–19.
Augot D, Mathieu B, Hadj-Henni L, Barriel V, Zapata Mena S, Smolis S, et al. Molecular phylogeny of 42 species of Culicoides (Diptera, Ceratopogonidae) from three continents. Parasite. 2017;24:23.
Ander M, Troell K, Chirico J. Barcoding of biting midges in the genus Culicoides: a tool for species determination. Med Vet Entomol. 2013;27:323–31.
Wenk CE, Kaufmann C, Schaffner F, Mathis A. Molecular characterization of Swiss Ceratopogonidae (Diptera) and evaluation of real-time PCR assays for the identification of Culicoides biting midges. Vet Parasitol. 2012;184:258–66.
Barber J, Harrup LE, Silk R, Veronesi E, Gubbins S, Bachanek-Bankowska K, et al. Blood-feeding, susceptibility to infection with Schmallenberg virus and phylogenetics of Culicoides (Diptera: Ceratopogonidae) from the United Kingdom. Parasit Vectors. 2018;11:116.
Kirkeby C, Dominiak P. Culicoides (Avaritia) gornostaevae Mirzaeva, 1984 (Diptera: Ceratopogonidae) - a possible vector species of the Obsoletus group new to the European fauna. Parasit Vectors. 2014;7:445.
Kitaoka S. Six new species and males of some species of Japanese Culicoides (Diptera: Ceratopogonidae). Natl Inst Anim Health Q (Tokyo). 1980;20:11–22.
Kitaoka S. Japanese Culicoides (Diptera: Ceratopogonidae) and keys for species (I). Bull Natl Inst Anim Health. 1984;87:73–89.
Cêtre-Sossah C, Baldet T, Delécolle J-C, Mathieu B, Perrin A, Grillet C, et al. Molecular detection of Culicoides spp and Culicoides imicola, the principal vector of bluetongue (BT) and African horse sickness (AHS) in Africa and Europe. Vet Res. 2004;35:325–37.
For making available numerous specimens and other useful material, we are greatly indebted to J. Anderson, T. Baldet, G. Bellis, F. Biteau-Coroller, E. Bois-Kuentz, P. Boland, G. Bousquet, N. Brodskaïa, M. C. Carrasquilla, P. Catez, Mr. Costet-Cherifi, C. Cotteaux, P. Decroocq, M. Demol, A. Desvars, M. Djerbal, I. Dusfour, A. Gerry, V. Glukhova, M. Goffredo, J. Gosset, H. Guis, C. Herisson, N. Hunt, M. Keane, M. Pizard, F. Le Pont, K. Labuschagne, J. C. Lien, R. Meiswinkel, B. Mullens, S. Mutel, C. Ninio, N. Pagès, F. Pagès, M. Patakakis, C. Paupy, A. Przhiboro, N. Rahola, V. Robert, S. Rondeau, F. Schaffner, F. Simard, D. A. Swanson, L. Tantely, M. H. Weng. The authors would like to thank Glenn Bellis and the anonymous reviewer for their useful comments, which eventually have led to greatly improve this manuscript.
This study was funded by the National Entomological Surveillance System of the French Ministry in charge of agriculture.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Mathieu, B., Garros, C., Balenghien, T. et al. A phylogenetic analysis of the biting midges belonging to Culicoides Latreille (Diptera: Ceratopogonidae) subgenus Avaritia using molecular data. Parasites Vectors 13, 243 (2020). https://doi.org/10.1186/s13071-020-04111-4