Characterization of two cryptic species, Culicoides stigma and C.parroti (Diptera: Ceratopogonidae), based on barcode regions and morphology

Authors


ABSTRACT

Biting midges of the genus Culicoides (Diptera: Ceratopogonidae) are insect vectors of economically important veterinary diseases such as African horse sickness, bluetongue, and Schmallenberg virus. The identification of Culicoides based on morphological features can be difficult. Three species of biting midges, Culicoides nubeculosus, C. stigma, and C. parroti have emerged in the laboratory from mud collected around watering troughs on a farm in northern France. Emerging Culicoides were characterized morphologically and molecularly using molecular markers. The closely related species C. stigma and C.parroti showed highly divergent sequences for both mitochondrial (cytochrome B and cytochrome oxidase I) and ribosomal DNA first internal transcribed spacer. A RFLP based on a single restriction using the same enzyme (HaeIII) for both cytochrome C oxidase I and cytochrome B is proposed to identify these species.

INTRODUCTION

Biting midges of the genus Culicoides Latreille 1809 (Diptera: Ceratopogonidae) are among the world's smallest haematophagous flies, measuring from 1 to 3 mm in size, and are the vectors of orbiviruses such as bluetongue virus (BTV), African horse sickness virus, or Schmallenberg virus (SBV). Outbreaks of BTV and SBV have been recorded since 2006 and 2011, respectively, in France and have caused severe losses in cattle and sheep. However, the identification of Culicoides based on morphological features can be difficult. To solve this problem of identification, several techniques have been developed, including morphometric tools (Pagès and Sarto 2005, Pagès et al. 2009, Augot et al. 2010, Muñoz-Muñoz et al. 2011), use of molecular markers such as Cytochrome oxydase submit I (COI) of the mitochondrial DNA (mtDNA) (Nolan et al. 2007); first and second internal transcribed spacers (ITS1 and ITS2) of the ribosomal DNA (rDNA) (Cêtre-Sossah et al. 2004, Perrin et al. 2006, Gomulski et al. 2006), and an Interactive Identification Key (IIKC) based on morphological items (Mathieu et al. 2012). This database includes 60 descriptions (i.e., wings, spermathecae, eyes, mandible/maxille, cibarial armature, posterior pharynx armature, palpus, antennae, legs). Success rates differed significantly between users and between levels of experience and ranged from 35.1% to 81.1% (Mathieu et al. 2012). The use of COI barcodes as a tool for species identification of biting midges can differentiate 95% of species studied (Ander et al. 2012). Misidentification of disease vectors has important epidemiological implications. Thus, the knowledge of which species could act as disease vectors, as well as their correct identification, remains essential to assess the real risk for disease transmission into disease-free areas.

In France, two very closely related species, Culicoides parroti Kieffer, 1922 and Culicoides stigma Meigen, 1818 belonging to the subgenus Monoculicoides, show overlapping characters. These two species live in sympatry in the country. Their morphological specific identification is difficult for non-specialists using a microscope and impossible based on wing patterns using a stereomicroscope. The goal of the present study was to morphologically identify Culicoides parroti, C. stigma, and C. nubeculosus, that emerged in the laboratory from mud collected in the French Ardennes, and to provide molecular tools for their identification. From these three species, only C. nubeculosus has been proven to biologically transmit BTV to date (Jennings and Mellor 1988). Three different molecular markers were tested for this purpose: i) rDNA ITS1; ii) the mtDNA COI serving as ‘gold standard' for DNA barcode used in routine for Culicoides studies, and iii) a part of the cytochrome b (cyt b) of the mtDNA. The latter was chosen because of its usefullness at the taxonomic level for many other groups of insects, such as phlebotomine sand flies, and because it is commonly used in our laboratory (Moin-Vaziri et al. 2007, Randrianambinintsoa et al. 2012). The suitability of Cytb and COI sequences to provide the basis of a species-diagnostic polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assay was also assessed.

MATERIALS AND METHODS

Study site and selection of specimens

In 2008 over a four-month period, mud and dung were collected in and around a farm (49°53’77″N, 4°19’36″E) in the Ardennes. Soil samples rich in organic matter were randomly collected early in the morning around water troughs used by cattle and sheep in the field. These samples were put into plastic boxes that were then labelled. Samples were transferred to the laboratory and monitored in individual netted cages in order to detect emerging Culicoides. The laboratory temperature ranged from 18° to 21° C and the flow of water into the plastic boxes was kept constant. The cages were examined daily and adult Culicoides were transferred into small vials containing 96% ethanol. The head, wings, and genitalia of individual biting midges were cut off in a drop of ethanol, cleared in boiling Marc-André solution, and mounted between slide and cover slide. The corresponding thorax for each specimen was stored in a vial at −20° C before DNA extraction.

Three species (C. stigma, C. parroti, and C. nubeculosus) emerged from the mud and were morphologically identified and separated using their wing patterns according to the key of Delécolle1. Because of the similarity of the morphological characters of their wings, males and females were identified as C.stigma / C.parroti by specific genital characters (Figure 1).

Figure 1.

Drawing of parameres and spermathecae of two closely related species of Culicoides in Northern France. Culicoides parroti: males (D187, D184, D186) and females (D1, D4, D27); Culicoides stigma: males (D192, D6, D152) and females (D77, D12, D193).

Molecular analyses

Genomic DNA was extracted from 14 individuals Culicoides specimens (six C. parroti, six C. stigma, and two C. nubeculosus) according to the manufacturer's instructions (Qiagen). We used the primers CulitotalF (5’-GACGCTTATTAATATAGTTCC3’) and CulitotalR (5’-TGCGGTCTTCATCGACCCAT3’) designed by Cêtre-Sossah et al. 2004 and Gomulski et al. 2006, respectively. Their combination amplifies a fragment including ITS1, 5.8S, and ITS2. The primers Culitotal F and Culitotal R were used with the following conditions: an initial denaturation step at 94° C for 3 min, followed by 35 cycles of denaturation at 94° C for 1 min, annealing at 56° C for 1min and extension at 68° C for 1 min, and final extension at 68° C for 10 min.

COI was amplified using the primers C1J1718 and C1N2191 as indicated by Dallas et al. (2003). PCR cycling conditions were: an initial denaturation step at 95° C for 15 min, then five 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. A part of the cytchrome b gene was amplified using the primers N1N-PDR and C3B-PDR as detailed by Bounamous et al. (2008). PCR cycling conditions were: an initial denaturation step at 94° C for 3 min, followed by five cycles of denaturation at 94° C for 30 s, annealing at 45° C for 90 s, and extension at 86° C for 60 s, and then 35 cycles of denaturation at 94° C for 30 s, annealing at 51° C for 90 s, and extension at 86° C for 60 s and a final extension at 68° C for 10 min.

Direct sequencing of both DNA strands was performed with the primers used for PCR. Sequence alignments were done utilizing ClustalW software (Thompson et al. 1994). The DNA sequence-based analyses were performed using the neighbor-joining (NJ) method (Kimura 2-parameter) with MEGA software version 3.1 (Kumar et al. 2004). For Cytb, only one sequence is available in data base: C. arakawae from Japan (GenBank under accession NC009809). For ITS, we have included C. parroti from France (AY861153), C. nubeculosus from the United Kingdom (AJ417982) and C. puncticollis from France (AY861158), and C. imicola from Israel (JN408479) to our sampling. For COI, we have added C. nubeculocus from Sweden and Denmark (JQ620128 and JQ683275, respectively), C. puncticollis from Denmark (JQ683314), C. riethi from Denmark (JQ683337, JQ683336), C. stigma from Sweden (JQ620229), and C. imicola from Spain (AF080540). C. imicola have also been processed as an outgroup in ITS1 (Perrin et al. 2006) and COI (Ander et al. 2012), respectively.

Sites for restriction enzymes were predicted for COI and Cytb sequences of C. parroti and C. stigma using CLC DNA Workbench software version 5.2 (http://www.clcgenomicsworkbench.comabout.com/). A panel of restriction enzymes was tested. HaeIII provide an original digestion pattern per species for both mtDNA markers and was selected. PCR-RFLP assays were performed in a 30μl total volume reaction mix, containing 10 μl of PCR product (from PCR vials), 1 μl of HaeIII, and 2 μl of supplied buffer (Fermentas, Germany). PCR products were digested for 20 min at 37° C. The digested samples were separated by electrophoresis in a 3% agarose gel to produce DNA fragments and were sized by comparison with markers 50 pb ladder and 100 pb ladder (Cliniscience, France).

RESULTS

Mud samples were collected near watering troughs, within 30 m of livestock. A total of 115 adult Culicoides specimens made up of 67 females and 48 males emerged. They belonged to three Monoculicoides species: Culicoides stigma (10♀, 8♂); C. parroti (9♀, 9♂), and C. nubeculosus (48♀, 31♂). The date of emergence ranged from four to 36 days for C. parroti, five to 57 days for C. stigma, and 29 to 35 days for C. nubeculosus (Table 1). No other species of Culicoides emerged from the mud during the sampling period.

Table 1. Sampling species in the Ardennes département (France).
SpeciesNo. samplesSexDate of mud collectionDate of emergenceNumber of days
Between collection and emergence
Culicoides parrotiD184M12/02/200820/02/20088
D186M12/02/200820/02/20088
D187M12/02/200820/02/20088
D27F12/02/200820/03/200836
D1F06/03/200810/03/20084
D4F06/03/200810/03/20084
Culicoides stigmaD192M12/02/200810/04/200857
D6M06/03/200813/03/20087
D152M02/04/200807/04/20085
D12F12/02/200813/03/200829
D193F12/02/200810/04/200857
D77F02/04/200807/04/20085
Culicoides nubeculosusD16F12/02/200813/02/200829
D179M02/04/200807/05/200835

The length of PCR products amplified ranged approximately 372 to 537 bp according to the primers. The sequence size, including primer, used to detect sites for restriction enzymes is 574 bp for COI and 544 bp for Cytb. Molecular comparisons after alignment (including Genbank sequences) were based on 400 bp for the ITS-1, 558 bp for the Cyt b, and 472 bp for the COI (including gaps). The sequences have been deposited into GenBank under the accession numbers: for ITS1: C. nubeculosus (KF178258- KF178259), C. parroti (KF178260- KF178265), C. stigma (KF178266- KF178271); for COI: C. nubeculosus (KF178272- KF178273); C. parroti (KF178274- KF178278); C. stigma (KF178279- KF178284); for Cytb: C. nubeculosus (KF178285- KF178286), C. parroti (KF178287- KF178292), C. stigma (KF178293- KF178298). The sequences were compared using the pairwise distance between each group (Table 2) including sequences previously deposited in Genbank.

Table 2. Estimation of pairwise distance species C. parroti, C. stigma, C. nubeculosus, C. puncticollis, C. imicola, C. riethi, and C. arakawae for ITS1 region of the rDNA, the COI, and Cytb domain of the mtDNA, respectively.
  ITS1  COI  Cytb
  1234  12345  123
1C. parroti    1C. parroti     1C. nubeculosus   
2C. stigma0.0082   2C. stigma0.137    2C. parroti0.173  
3C. nubeculosus0.1240.090  3C. nubeculosus0.1690.192   3C. stigma0.1470.157 
4C. puncticollis0.1080.0760.043 4C. puncticollis0.1540.1920.140  4C. arakawae0.2120.1710.204
5C. imicola0.2510.2720.2610.2735C.imicola0.2200.2120.2470.198      
      6C. riethi0.1470.1520.1430.1370.216     

Analysis of ITS1 sequences

The lengths of ITS1 segments were 372–375 bp for C. nubeculosus, 383–423 bp for C. stigma, and 396–415 bp for C. parroti. The genetic distance pairwise within species showed 99.7%, 99.9%, and 100% homology within C. nubeculosus, C. stigma, and C. parroti, respectively, whereas distance pairwise computed between species showed 87.6% homology between C. nubeculosus and C. parroti, 91% between C. nubeculosus and C. stigma, and 91.8% between C. stigma and C. parroti. The NJ analysis suggested that C. stigma is closer to both C. puncticollis and C. nubeculosus than to C. stigma (Figure 2).

Figure 2.

Neighbor-joining trees based on mtDNA (Cytochrome b (a), Cytochrome oxidase I (b)) and first internal transcribed spacer rDNA (c) for sequences of Culicoides nubeculosus, C. parroti, and C. stigma from northern France. Bootstrap values (1,000 replicates) are given on the branches.

aAnalysis of COI sequences

The lengths of COI segments were 513–529 bp for C. nubeculosus, 502–512 bp for C. parroti, and 498–524 bp for C. stigma. The genetic distance pairwise within species showed 99.6%, 99.8%, and 99.4% homology within C. nubeculosus, C. stigma, and C. parroti, respectively, whereas distance pairwise computed between species showed 83.1% homology between C. nubeculosus and C. parroti, 80.8% between C. nubeculosus and C. stigma, and 86.3% between C. stigma and C. parroti. The NJ analysis suggested that C. stigma and C. parroti are the two closest species. On another branch, the species C. riethi C. puncticollis and C. nubeculosus are grouped (Figure 2).

Analysis of cytochrome b sequences

The lengths of Cytb segments were 516–519 bp for C. nubeculosus, 517–537 bp for C. parroti, and 513–525 bp for C. stigma. The genetic distance pairwise within species showed 99.8%, 99.7%, and 99.9% homology within C. nubeculosus, C. stigma, and C. parroti, respectively, while distance pairwise computed between species showed 82.8% homology between C. nubeculosus and C. parroti, 85.3% between C. nubeculosus and C. stigma, and 84.3% between C. stigma and C. parroti. The NJ analysis suggested that C. stigma is the sister group of C. nubeculosus (Figure 2).

Although C. parroti and C. stigma males are closely related morphologically, the mitochondrial and ribosomal molecular markers strongly distinguish the former from the latter (Figure 2). The membership of each sample in the various branches was strongly supported by bootstrap values.

Species-diagnostic restriction enzymes sites

Recognition site for the restriction enzyme (HaeIII) was identified in the COI and Cytb sequences. It distinguishes C. parroti from C. stigma. The observed profiles are in agreement with the predicted profiles for COI sequence: C. parroti (469/105 bp) and C. stigma (574 bp); and for Cyb b sequence: C. parroti (544 bp) and C. stigma (226/318 bp).

DISCUSSION

In our study, pupae of C. nubeculosus, C. stigma, and C. parroti coexisted on the same site rich in organic matter, confirming the observations of Uslu and Dik 2006. Among the several genes of interest for the identification of species, the COI and Cytb genes are of interest for the taxonomy of closely related species. COI is the most commonly sequenced marker for Culicoides barcoding. To our knowledge, Cyt b has never been used before for specific identification of Culicoides. It enabled us to distinguish two closely related species of Culicoides. It yielded readily usable sequences (without sequences containing mixtures of two bases at sites) and could serve as a candidate for standardized molecular identification of Culicoides in the future. Species assigned to the subgenus Monoculicoides by Ander et al. 2012 are confirmed by our study, showing that C. stigma is closely related to C. parroti as correlated by morphology.

The coding regions (mtDNA) and non-coding internal transcribed spacer (ITS) sequences are useful tools for phylogenetic studies. ITS1 sequences have been shown to be good phylogenetic markers (Perrin et al. 2006). Our study on Monoculicoides subgenera confirms the position of C. nubeculosus, C. puncticollis, and C. parroti reported by Perrin et al. (2006). Interestingly, the molecular analysis of ITS1 has grouped C. stigma with C. puncticollis and C. nubeculosus into the same cluster, whereas C. parroti remains outside of the clade. COI provides the grouping of specimens within their specific branch, but the relationships between the branches is different. Regarding COI sequences, C. stigma and C. parroti are the most closely related species. Hybridization between closely related species of insects is known and common for many insects like Phlebotomine sandflies (Pesson et al. 2004). In Culicoides, it has been suspected in natura for C. impunctatus (Ritchie et al. 2004). In our study, the comparison of the mtDNA and rDNA markers sequenced did not suggest any introgression between C. stigma and C. parroti found in sympatry.

We applied COI-RFLP and Cytb-RFLP as tools for the identification of Culicoides species. This technique is rarely used for the identification of Culicoides (Linton et al. 2002, Dallas et al. 2003). The characterization of cryptic species could be obtained using PCR-RFLP and its utilization could be more common in the future.

The identification of Culicoides based on morphological features is difficult and several cryptic species and species complexes are present. With the IKC (Mathieu et al. 2012), the successful identifications were achieved in an average of 6.6 steps with three steps for C. parroti and four for C. stigma. No database is available for identification of Culicoides males. Our study, combining morphological and molecular identification of Culicoides specimens, is thus important for a better understanding of the systematics of biting midges, which are vectors of diseases, and also to initiate molecular ecological studies.

Acknowledgments

This work is part of the project @SPEED-ID “Accurate SPEciEs Delimitation and Identification of eukaryotic biodiversity using DNA markers” proposed by F-BoL, the French Barcode of life initiative, and part of the the grant BI4I (Barcoding insects for identification). We thank Sylvette Gobert for proofreading this manuscript.

  1. 1

    Delécolle, J.C. Nouvelle contribution a l’étude systématique et iconographique des espèces du genre Culicoides (Diptera: Ceratopogonidae) du Nord-Est de la France. PhD thesis. Université Louis Pasteur de Strasbourg, UER Sciences Vie et Terre, 1985.

Ancillary