Mosquito samples

The Rimini laboratory strain (F35) was used for the construction of the microsatellite-enriched libraries. This strain was established from mosquitoes collected in Rimini (Emilia Romagna) at the Centro Agricoltura Ambiente ‘G. Nicoli’ (Crevalcore, Italy). Three of the nine wild samples previously described, Ban Rai, Cesena and Brescia, were also screened to assess the variability of the newly developed microsatellites markers.

ITS2 marker characterization

DNA was extracted according to Baruffi et al. [42] from individual females of the population samples. The extracted DNA was resuspended in TE (10 mM Tris–HCl, pH 8, 1 mM EDTA) and its concentration was quantified using a Nanodrop ND-1000 spectrophotometer (Nano-drop Technologies Inc., Wilmington, DE, USA). The ribosomal DNA ITS2 region was amplified from the DNA samples using the universal primers 5.8Sf 5′-atcactcggctcgtggatcg-3′ and 28Sr 5′-atgcttaaatttagggggtagtc-3′ [43]. These primers anneal to highly conserved sequences in the 5.8S and 28S rDNA genes flanking the ITS2 region. Reactions were performed in a volume of 25 μl with approximately 3 ng DNA, 1.5 mM MgCl₂, Reaction buffer (10 mM Tris–HCl pH 9, 50 mM KCl), 200 µM dNTP, 10 pmol of each primer and 1 unit Taq DNA polymerase (Invitrogen). Amplifications were performed using an Eppendorf MasterCycler with the following cycle conditions: an initial denaturing step at 94°C for 2 min; twenty-five cycles of denaturation at 94°C for 20 s, annealing at 54°C for 15 s, extension at 72°C for 25 s, followed by a final extension at 72°C for 5 min.

The PCR products were separated on 1.5% agarose gels, stained with ethidium bromide and photographed under ultraviolet light. Amplification products were ligated into the TOPO 2.1 vector and transformed into competent One Shot cells using the TOPO TA cloning kit (Invitrogen). DNA inserts were sequenced on both strands.

ITS2 data analysis

The boundaries of the ITS2 region were determined by comparison with the available Ae. albopictus sequences in GenBank. The ITS2 sequences were aligned using MUSCLE 3.8 [44]. Nucleotide diversity and gene diversity parameters were determined using Arlequin 3.5.1.2 [45]. Geographic structure in the ITS2 data set was investigated by spatial analyses of molecular variance implemented in SAMOVA 1.0 [46]. This approach identifies groups of populations that are genetically and geographically homogeneous and maximally differentiated from each other. The method requires the a priori definition of the number of groups (K) of populations that exist and generates F-statistics (F _SC, F _ST and F _CT) using an AMOVA approach. Different numbers of groups (K) were tested, and a simulated annealing procedure permitted the identification of the composition of each of the K groups that maximizes the F _CT index (proportion of total genetic variance due to differences between groups). The program was run for two to eight groups (K = 2 to K = 8) each time with the simulated annealing process repeated 100 times, starting each time with a different partition of the population samples into the K groups. A phylogenetic analysis was also performed including the unique ITS2 sequences or haplotypes identified in the nine population samples by Arlequin 3.5.1.2 [45]. The haplotype sequences together with an Ae. flavopictus ITS2 outgroup sequence (GenBank accession number: AF353559.1) were aligned using MUSCLE 3.8 [44]. The Kimura-2-parameter and Γ distribution model indicated by the Akaike Information Criterion (AIC = 3064) and the Bayesian Information Criterion (BIC = 3899), was used to perform Maximum Likelihood analyses in MEGA 5.2.2 [47]. The reliability of the resulting haplotype tree was determined by 1000 bootstrap replications. The mid-point rooted tree was drawn using FigTree v1.4 (http://tree.bio.ed.ac.uk/software/figtree/).

Microsatellite isolation and characterization

Total genomic DNA was extracted individually from 64 mosquitoes (40 males and 24 females) from the Rimini strain using the method reported by Baruffi et al. [42]. The DNA was subsequently pooled and submitted to Genetic Identification Services, GIS (Chatsworth, CA, USA, http://www.genetic-id-services.com) for library construction. The library was enriched for four different di- and tri-nucleotide microsatellite motifs, namely (CA)n, (GA)n, (AAC)n, and (ATG)n. Insert DNA from individual clones was amplified by PCR following GIS guidelines. The PCR products were purified using QIAquick columns (Qiagen) and sequenced using the DYEnamic ET Terminator Cycle Sequencing Kit (Amersham Bioscience) with the universal M13 primers (M13f 5′-aggaaacagctatgaccatg-3′ and M13r 5′-acgacgttgtaaaacgacgg-3′). Clones with inserts less than 200 bp were not considered for sequencing. Sequencing products were resolved on the Applied Biosystems model 377 DNA Sequencer (Applied Biosystems). The resulting sequences were screened for the presence of microsatellite motifs using WebSat [48]. All the clone sequences were checked for redundancy using BLASTn [49].

Microsatellite sequences that shared one or both the 5′/3′ flanking regions in common were discarded to exclude potential multi-locus microsatellites families. The sequences that were found to be unique within the library were subjected to BLASTn analysis against the nr database to identify eventual hits with previously described sequences. Primer pairs were designed for the flanking regions of the identified SSR sequences using Primer 3 [50]. Preliminary PCR screenings were performed on 20 individuals (10 males and 10 females) from the Rimini strain to validate the primer efficiencies in single SSR locus amplifications. DNA amplifications were performed on a Mastercycler gradient (Eppendorf). The PCR reaction mixture consisted of 50 ng genomic DNA, 1x PCR buffer, 1.5 mM MgCl₂, 270 µM dNTPs (Invitrogen), 1 U Taq polymerase (Invitrogen), and 10 μM of each primer, one of which was 5′ labelled with a fluorescent dye, in a final volume of 15 μl. PCR cycling conditions were 94°C for 3 min, 30 cycles of 94°C for 30 s, 57-60°C for 30 s, and 72°C for 30 s, followed by a final extension step of 72°C for 10 minutes. Aliquots of PCR products were separated by electrophoresis on 2% agarose gels stained with ethidium bromide, and visualized under UV light. Each PCR product was then diluted 1:10 in ddH₂O water and 1 μl of this dilution was added to 9 μl of a mixture of deionized formamide and GeneScan-500 ROX size standard (Applied Biosystems). After denaturation for 4 min at 94°C the fragments were resolved on an ABI PRISM 310 Genetic Analyzer (Applied Biosystems) with the GENESCAN software package (Applied Biosystems). To avoid genotyping errors, the program TANDEM [51] was used to automate binning of the microsatellite allele length variants.

Microsatellite data analysis

The parameters of intra- and inter-population genetic diversity, such as observed allele size, number of actual alleles (n_a), number of effective alleles (n_e), number of private alleles (n_p) and AMOVA were computed using GenALEx 6.5 [52]. Observed and expected heterozygosity (H_O and H_E, respectively), gene diversity (H_S) and F _ST values were computed using MICROSATELLITE ANALYSER (MSA) V.4.05 [53]. Linkage disequilibrium, the frequency of null alleles (A_n) and the deviations from Hardy-Weinberg expectations were computed using GENEPOP version 4.2 [54,55]. For loci with fewer than five alleles, an exact test of Hardy-Weinberg proportions was performed. For loci with five alleles or more, an unbiased estimate of the exact probability was obtained using the Markov chain method of Guo & Thompson [56] for each combination of locus and population. The allelic Polymorphic Information Content (PIC) was derived using CERVUS [57].

ITS2 sequence variability

ITS2 haplotype phylogeny and population distribution

The phylogenetic analysis based on the 52 ITS2 haplotype sequences identified in the nine geographic populations resulted in a Maximum Likelihood tree with a log likelihood of −1428.3 (Figure 1). The ITS2 sequences form a compact cluster with very few nodes supported by bootstrap values greater than 50%. Highly related sequences are distributed across populations from different geographical regions; some evidence of clustering of tightly related sequences is present in Ban Rai, Phato (Thailand) and to a lesser degree in St. Pierre (Réunion).

The majority (47of 52) of the haplotypes are present in only one population. Only three haplotypes are shared by two populations (haplotype 18 in Pavia and St. Denis, haplotype 21 in Pavia and St. Pierre, and haplotype 39 in Cesena and Modena). Haplotype 23 is shared by three Italian populations (Pavia, Brescia and Modena), and haplotype 13 is present in four populations (Phato, Modena, St. Denis and St. Pierre).

SSR: Isolation and characterization

Out of a total of 109 sequenced clones obtained from the Ae. albopictus library, 92 contained tandem DNA repeats with at least 5 repeat units. BLASTn analyses showed that one or both flanking regions were shared in 42 sequences and they were thus excluded from the subsequent characterization. Within the remaining 50 sequenced clones, five harboured more than one tandem DNA repeat, resulting in the identification of 55 microsatellite motifs. Among these, trinucleotide motifs were the most abundant (28), followed by dinucleotide (22), mononucleotide (3), heptanucleotide (1) and hendecanucleotide motifs (1). Among the trinucleotides motifs, CAA was highly abundant (17) followed by TGA (10). Among the dinucleotide motifs, TG was predominant (14) followed by CT (8). Thirty-seven of these microsatellite motifs contained perfect repeat arrays. Twelve microsatellites contained imperfect repeats and the other six microsatellites contained compound repeats.

Evaluation of microsatellite polymorphism in ancestral and adventive populations