- Top of page
- Materials and methods
Culicoides species belonging to the Obsoletus complex (Diptera: Ceratopogonidae) have been indicated as primary bluetongue (BT) vectors in many European countries and their possible involvement in the maintenance and overwintering of BT viruses has been suggested, even in regions where Culicoides imicola Keiffer is the main vector. The Obsoletus complex includes two predominant taxa, Culicoides obsoletus (Meigen) and Culicoides scoticus Downes & Kettle. However, the role played by each species in the epidemiology of BT is still unknown. Taxonomic identification is mainly based on the morphology of male genitalia and the lack of other reliable diagnostic features makes the screening of trap-collected vector populations, mainly females, particularly difficult. Although molecular markers have facilitated species identification, little information is yet available on the biology, abundance and population dynamics of the two taxa. The aim of this work was to investigate the genetic profile and temporal distribution of C. obsoletus and C. scoticus by using isozyme electrophoresis applied to adult midges, collected weekly at two selected farms in southern Sardinia. A total of nine enzyme loci were analysed and five of them provided diagnostic allozyme markers (Hk, Mdh, Pgi, Idh-1 and Idh-2). Nei's genetic distance between the two taxa was in the range of other well-separated taxa (D = 1.792), supporting their status as true species. Culicoides scoticus represented almost 61% of the 562 specimens analysed; its genetic structure was characterized by a very low level of intra-population variation (mean heterozygosity He = 0.019) and higher genetic divergence between populations (FST = 0.0016) than in C. obsoletus. The latter species had significantly more heterozygotes (He = 0.123), a higher percentage of polymorphic loci, and no inter-population differentiation (FST≅0). We suggest that different biological and ecological constraints, such as breeding habitat requirements, may contribute to shaping the genetic profiles of C. scoticus and C. obsoletus. However, enough gene flow was maintained between populations of each species as no spatial and temporal structuring was sustained by Fisher's exact probability test (P > 0.5). The seasonal distributions of C. scoticus and C. obsoletus only partially overlapped: both species were mainly found early in the year, when the main vector, C. imicola, was present in low numbers, and peaked in abundance in April and May. Culicoides scoticus was predominant until May, decreased rapidly in the following months and increased again in winter, whereas C. obsoletus decreased more slowly and was still present in early summer. Consequently, C. scoticus may be a good candidate for playing a role in the transmission and maintenance of BT virus in Sardinia, as well as in other Mediterranean countries, during the months of late winter and early spring when the seroconversion of sentinel animals is still occurring in the absence of the main vector.
- Top of page
- Materials and methods
In the last decade European countries have experienced recurrent outbreaks of bluetongue (BT), an arboviral disease which affects ruminants and can cause severe epidemics, particularly in some breeds of sheep (Mellor & Wittmann, 2002; Saegerman et al., 2008). Because of its economic impact, which also results from limitations to the international trade in animals, BT was included in the former List A of epizootic diseases by the Office International des Épizooties (OIE).
The main vector of BT in the Mediterranean basin is Culicoides imicola Keiffer (Diptera: Ceratopogonidae), but other Culicoides species, mainly belonging to the Obsoletus and Pulicaris complexes, have been indicated as effective and important vectors in regions and countries where C. imicola is present at low frequencies or even absent, such as Cyprus, Bulgaria, Greece and central and southern Italy (Mellor & Pitzolis, 1979; Mellor & Wittmann, 2002; Caracappa et al., 2003; Torina et al., 2004; De Liberato et al., 2005; Savini et al., 2005). More recently, BT has spread to central and northern Europe, outside the present geographical range of C. imicola, confirming the importance of these indigenous Culicoides species as main BT vectors (Saegerman et al., 2008; Wilson & Mellor, 2009). During the 2006 outbreak in Germany, the Obsoletus complex, which accounted for 90% of the midges analysed, was, in fact, the only one found positive for BT virus (BTV) (serotype 8) (Mehlhorn et al., 2007).
The Obsoletus complex includes at least three species, Culicoides obsoletus (Meigen), Culicoides scoticus Downes & Kettle and Culicoides montanus Shakirjanova. An unidentified species A, present in mainland Italy, has also been identified based on DNA sequence analysis (Gomulski et al., 2005). However, the taxonomic and phylogenetic relationships within the complex are still debated (Meiswinkel et al., 2004; Gomulski et al., 2005; Mathieu et al., 2007; Nolan et al., 2007; Kiehl et al., 2009; Schwenkenbecher et al., 2009). Morphological identification is, indeed, limited by the availability of clearly distinctive features. Attempts have been made to find reliable diagnostic characters in females (Pagès & Sarto I Monteys, 2005), but, so far, identification is still accomplished mainly through the analysis of male genitalia, thus making field studies on trap-collected Culicoides (mainly females) particularly difficult.
However, despite the increasing importance of members of the Obsoletus complex as primary BT vectors in several European countries, information on the biology and population dynamics of each species within the complex is still scanty and their roles in the epidemiology of BT are yet to be clarified. In fact, most of the papers published so far have dealt with the complex as a whole because of the difficulty in the morphological identification of individual specimens.
In Sardinia, BT was first reported in August 2000, since then various BTV serotypes (1, 2, 4, 16 and 8) have circulated, causing recurrent outbreaks (Office International des Épizooties, 2009). Culicoides imicola is the main BT vector on the island (Goffredo et al., 2003), whereas species of the Obsoletus complex are widespread, but less abundant (Goffredo et al., 2004; Pili et al., 2006). Two species of the complex, C. obsoletus and C. scoticus, were identified based on the analysis of male genitalia (Pili et al., 2006), but their possible involvement in the transmission and maintenance of BTV when C. imicola is less abundant or even absent is still unknown.
The aim of this work was to investigate the genetic profiles of C. obsoletus and C. scoticus by using isozyme electrophoresis applied to adult midges collected in two farms in southern Sardinia. A number of wild-caught Culicoides males were identified to species level using morphological and molecular markers, and were used to search for diagnostic allozyme markers at eight loci. Large samples of females collected during 2004 were then electrophoretically tested to gain insight into the population structure of each species, its relative abundance and temporal and spatial distributions.
- Top of page
- Materials and methods
In the course of the 2-year survey, a total of 6223 specimens were classified as belonging to the Obsoletus complex, corresponding to 5.1% of all Culicoides midges (121 990) collected in 2003 and 2004; this is a low percentage compared with the >46.5% of the catch accounted for by the main BT vector, C. imicola. However, if we consider only collections from January to May, this percentage increases to 28.3%, whereas that for C. imicola decreases to only 8.3% of the total catch. In both years, major peaks in abundance of the Obsoletus complex, measured as maximum catch per month, were recorded in April and May; the frequency of these midges decreased consistently over the summer and increased again during the last months of the year.
Differences in monthly catches were observed between sites and years: collections were more abundant in the eastern coastal locality of Muravera (MV), with peaks in the range of thousands (Fig. 1), than in the western inland site (VP), where the number of adult midges never exceeded a few hundred (Fig. 2). Higher numbers of midges were trapped during 2004 than in the previous year, coinciding with the end of a long period of drought (Servizio Agrometereologico della Sardegna, 2004).
Figure 1. Maximum monthly catches of Culicoides obsoletus s.l. () and Culicoides imicola () in Muravera (logarithmic scale).
Download figure to PowerPoint
Figure 2. Maximum monthly catches of Culicoides obsoletus s.l. () and Culicoides imicola () in Villaperuccio (logarithmic scale).
Download figure to PowerPoint
Culicoides obsoletus and C. scoticus were the only species of the Obsoletus complex identified by characteristic features of the genitalia in 47 trap-collected males. Species status was confirmed by PCR amplification of ITS2 species-specific sequences. Morphological and molecular cross-identification was the basis for searching diagnostic allozyme markers by gel electrophoresis using the same specimens.
Eight loci were identified by seven enzyme systems and diagnostic alleles were found at four of them: Hk, Idh-1, Idh-2 and Mdh. The Hk locus was monomorphic for alternative alleles in the two species, whereas Idh-1, Idh-2 and Mdh were polymorphic and the frequency of the most common allele was >0.78 (Table 1). More than 98% of individuals within each species were also homozygous for alternative alleles at locus Pgi (Table 1). Pgm was polymorphic but less informative than the other loci because allele frequencies did not differ significantly between populations of C. scoticus from VP and C. obsoletus from VP and MV (Fisher's test, P > 0.07). Nevertheless, overall genotype frequencies, analysed using Fisher's exact probability test, were significantly different between taxa at all loci, including Pgm (P = 0.0019). The other two loci, Ak and Gpd, were monomorphic for the same allele in both species.
Table 1. Allelic frequencies in Culicoides scoticus and Culicoides obsoletus populations in Muravera and Villaperuccio.
|Locus||Allele|| C. scoticus || C. obsoletus |
| Hk || ||(200)||(127)||(120)||(113)|
| || 100 ||0.000||0.000||1.000||1.000|
| || 95 ||1.000||1.000||0.000||0.000|
| Idh-1 || ||(220)||(117)||(89)||(87)|
| || 138 ||0.000||0.000||0.006||0.006|
| || 111 ||0.000||0.000||0.899||0.856|
| || 106 ||0.005||0.004||0.000||0.000|
| || 97 ||0.000||0.000||0.096||0.138|
| || 94 ||0.986||0.991||0.000||0.000|
| || 80 ||0.009||0.004||0.000||0.000|
| Idh-2 || ||(232)||(119)||(123)||(109)|
| || 176 ||0.000||0.000||0.000||0.005|
| || 137 ||0.000||0.000||0.782||0.794|
| || 125 ||0.000||0.008||0.000||0.000|
| || 118 ||0.000||0.000||0.210||0.188|
| || 109 ||0.000||0.000||0.008||0.014|
| || 107 ||0.004||0.000||0.000||0.000|
| || 103 ||0.985||0.988||0.000||0.000|
| || 70 ||0.011||0.004||0.000||0.000|
| Mdh || ||(174)||(108)||(112)||(98)|
| || 119 ||0.000||0.000||0.005||0.000|
| || 88 ||0.000||0.000||0.032||0.005|
| || 78 ||0.003||0.000||0.000||0.000|
| || 64 ||0.000||0.000||0.963||0.995|
| || 57 ||0.997||1.000||0.000||0.000|
| Pgi || ||(239)||(128)||(123)||(111)|
| || 183 ||0.002||0.000||0.000||0.000|
| || 146 ||0.992||0.996||0.008||0.005|
| || 100 ||0.006||0.004||0.992||0.991|
| || 56 ||0.000||0.000||0.000||0.005|
| Pgm || ||(177)||(93)||(120)||(96)|
| || 97 ||0.011||0.033||0.054||0.073|
| || 84 ||0.986||0.962||0.946||0.927|
| || 72 ||0.003||0.004||0.000||0.000|
Single females, collected in 2004 and morphologically identified as belonging to the Obsoletus complex, were tested at the eight loci. Between 87 and 239 individuals were analysed per locus, and a total of 562 individuals were included in the final analysis. Based on the electrophoretic pattern, 342 specimens were identified as C. scoticus and 220 as C. obsoletus, indicating the prevalence of the first species in the study area. Ak and Gpd were non-informative and were excluded from the following analysis.
The genetic structure of C. scoticus was characterized by a very low level of variation, with mean expected heterozygosity (He) ranging from 0.018 to 0.020 in the two populations of MV and VP (Table 2); most individuals were homozygous at all loci. More variability was observed in C. obsoletus populations, expressed by a higher frequency of heterozygotes (He≅0.12) and percentage of polymorphic loci (50%).
Table 2. Genetic structure of Culicoides scoticus and Culicoides obsoletus populations in Muravera and Villaperuccio.
|Species||Pop.||Specimens/locus, mean (SE)||Alleles/locus, mean (SE)||Polymorphic loci, %*||Observed heterozygosity, mean (SE)||Expected heterozygosity, mean (SE)|
| C. scoticus ||MV||203 (10.8)||2.5 (0.3)||0/50||0.016 (0.005)||0.018 (0.005)|
| ||VP||120.3 (2.2)||2.2 (0.4)||0/33||0.018 (0.009)||0.020 (0.011)|
| C. obsoletus ||MV||109.3 (6.2)||2.3 (0.3)||50/67||0.086 (0.036)||0.121 (0.053)|
| ||VP||102.3 (4.2)||2.5 (0.4)||50/50||0.111 (0.050)||0.125 (0.058)|
Deviation from Hardy–Weinberg equilibrium was assessed in the MV populations of both species because of an excess of homozygotes mainly at the Idh loci (P < 0.02). However, when all loci where considered, Hardy–Weinberg equilibrium was rejected only for C. obsoletus from MV (P = 0.0004), whereas VP populations of both species were in equilibrium at all loci (P > 0.6). Heterozygote deficit contributed to positive FIS values obtained at most loci in both species. Overall FIS across all loci was 0.12 for C. scoticus and 0.20 for C. obsoletus, indicating deviation from panmixia and the presence of inbreeding. Genetic differentiation within each species was not significant (P > 0.5) and corresponded to FST values of 0.0016 in C. scoticus and −0.0015 in C. obsoletus. The negative FST obtained in the latter species reflects only the computation method employed and should be interpreted as a zero value (i.e. a complete lack of differentiation between populations). In both species, genotypic frequencies did not change significantly (P > 0.06) during the year (January–December 2004, excluding months in which not enough data were available).
Divergence between the two taxa, estimated using Fisher's exact probability test, was significant for all loci in all pairwise comparisons (overall P across loci <0.000001), including Pgm (for this locus P≤ 0.01). Average Nei's genetic similarity (I) and distance (D) between the taxa were 0.167 and 1.792, respectively. If the two monomorphic loci Ak and Gpd were included in the analysis, the similarity index increased to 0.387.
Using the electrophoretic identification of C. scoticus and C. obsoletus in monthly collections of adult midges, seasonal differences in the relative abundance of each species were analysed (Table 3). Although both species were mainly collected in the winter–spring period, C. scoticus was prevalent in the samples during March–May and dropped to a few specimens in the following months, whereas C. obsoletus was always less frequent at both sites, except in June and July (Table 3).
Table 3. Number of Culicoides scoticus and Culicoides obsoletus identified by diagnostic allozyme markers and total monthly catches of Obsoletus complex.
| C. scoticus || C. obsoletus ||Total catch|| C. scoticus || C. obsoletus ||Total catch|
If species ratios obtained from the analysed samples were related to the total number of Obsoletus complex midges collected during 2004, both species appeared to peak in abundance in the same period (April in MV and May in VP). Culicoides scoticus remained prevalent until May, rapidly disappeared in the following months and increased again in winter, whereas C. obsoletus decreased more slowly and was still present in June and July (Fig. 3).
Figure 3. Distribution of Culicoides scoticus () and Culicoides obsoletus () in total monthly catches in (A) Muravera and (B) Villaperuccio.
Download figure to PowerPoint