Fig 1.
Geographical range of the four BTV competent Culicoides vector species in Australia (modified from [40]) and locations from which samples of Culicoides were collected for this study: C. brevitarsis (from Kalumburu, Western Australia (WA); Douglas-Daly and Katherine, Northern Territory (NT); Lismore, Grafton, Bellingen, and Paterson, New South Wales (NSW)); C. marksi (from Moree, NSW; Murray Bridge, South Australia (SA)); C. pallidothorax and C. bundyensis (also from Kalumburu, WA); C. henryi (from Lismore, NSW); and C. bunrooensis (from Denman, NSW). Culicoides brevitarsis and C. marksi mtDNA haplotypes (respectively labelled ‘Cbrev’ and ‘Cmarksi’) identified from all sampling sites are also indicated. C. brevitarsis and C. marksi individuals analysed per site are indicated by ‘n’. Kalumburu sampling dates are 29-Dec-08 ‘A’; 25-Jun-09 ‘B’.
Table 1.
Culicoides samples used in this study including known virus vectored and taxonomic status.
Table 2.
Primers for mtDNA COI regions and exon-primed intron-crossing (EPIC) primers for ribosomal protein nuclear genes used to infer population structure of Culicoides species.
Fig 2.
Culicoides brevitarsis mtDNA haplotype network based on 835 bp of mtDNA COI gene sequence.
2 Culicoides brevitarsis mtDNA COI haplotype network (835bp) from seven sites across Australia (Kalumburu, WA; Douglas-Daly and Katherine, NT; Lismore, Grafton, Bellingen, and Paterson, NSW). The total number of C. brevitarsis surveyed was n = 70. The most common haplotype is Cbrev-01 (it includes n = 52 individuals from all seven sites), followed by Cbrev-02 (n = 7) from Lismore, Grafton, and Paterson. Haplotypes Cbrev-03, Cbrev-04, and Cbrev-05 are represented by two individuals each; all remaining haplotypes are represented by one individual each. The haplotype network was generated using TCS version 1.21 [62], and then refined manually. The numbers of substitutions separating haplotypes are indicated by short bars.
Fig 3.
Culicoides marksi mtDNA haplotype network based on 547 bp of mtDNA COI partial gene sequence.
Culicoides marksi mtDNA COI haplotype network (547bp) from Moree (NSW) and Murray Bridge (SA). The total number of C. marksi samples surveyed was 16. Haplotypes Cmarksi-01, and Cmarksi-03, …, Cmarksi-09 are from Moree, while Cmarksi-10, …, Cmarksi-16 are from Murray Bridge. The haplotype network was obtained using TCS version 1.21, and then refined manually. Apart from Cmarksi-01, which was represented by two individuals, the other haplotypes were represented by one individual each. The numbers of substitutions separating haplotypes are indicated by short bars.
Table 3.
AMOVA results of Australian Culicoides brevitarsis populations based on mtDNA and nuDNA markers.
Table 4.
Australian Culicoides brevitarsis population mtDNA COI pairwise FST estimates and associated P values ± SE.
Table 5.
Australian Culicoides brevitarsis population pairwise FST estimates and associated P-values inferred from nuDNA markers.
Table 6.
Nuclear DNA marker statistics for Australian C. brevitarsis populations.
Fig 4.
Culicoides brevitarsis population structure as inferred from nuDNA (EPIC-PCR and microsatellite DNA) markers using STRUCTURE 2.3.
Culicoides brevitarsis population structure as inferred using the Bayesian clustering algorithm implemented in the program STRUCTURE 2.3 [71], based on nuclear DNA (four EPIC and two microsatellite) markers. Bar graphs are from K = 2 from 50,000 burn-in cycles and 500,000 runs. Populations are numbered from northern to southern regions, and combining (Kalumburu A and B (WA)) as one population (1), Douglas-Daly (2) and Katherine (3) from NT, and Lismore (4), Grafton (5), Bellingen (6), and Paterson (7) from NSW.
Fig 5.
PP-plots for P values distributions from matched-pairs tests of symmetry for mtDNA COI sequence data sets.
PP-plots with the distributions of P values from the matched-pairs tests of symmetry for the following data sets: amino acids, first codon sites, second codon sites, and third codon sites. A J-shaped distribution of dots, with more than 5% of the dots below 0.05 (i.e., the horizontal line) indicates violation of the assumption of evolution under globally SRH conditions. The PP-plot for third codon sites represents a clear example of violation of this phylogenetic assumption.
Fig 6.
Maximum Likelihood phylogenetic tree including bootstrap values of various Culicoides species from this study.
Maximum Likelihood tree, with bootstrap values for some of the internal edges inferred using the UFBoot method. The tree was inferred under the TN+I+G4 model of evolution. Samples are marked using their species name, geographical origin (AU: Australia; JAP: Japan), haplotype name (when available, listed in brackets), and a number (whenever needed, to distinguish some of the sequences). Australian Culicoides are within blue boxes, and known BTV competent Asian/South East Asian Culicoides hosts are in red boxes. The scale bar measures evolutionary distance in terms of average nucleotide substitutions per site. The tree was rooted using four species of Anopheles. The ML tree inferred under the TN+G4 model of evolution differed from this one at seven internal edges, but the overall conclusion remained the same (see text).