Email updates

Keep up to date with the latest news and content from BMC Evolutionary Biology and BioMed Central.

Open Access Research article

From speciation to introgressive hybridization: the phylogeographic structure of an island subspecies of termite, Reticulitermes lucifugus corsicus

Thomas Lefebvre12, Nicolas Châline13, Denis Limousin1, Simon Dupont1 and Anne-Geneviève Bagnères1*

Author Affiliations

1 I.R.B.I., CNRS UMR 6035, Université François Rabelais de Tours, Faculté des Sciences et Techniques, Parc de Grandmont, 37200 Tours, France

2 L.E.S.T., IRD UMR 137, Université Paris XII – Val de Marne, 94010 Créteil cedex, France

3 L.E.E.C., CNRS UMR 7153, Université Paris 13, 99 avenue J.B. Clément, 93430 Villetaneuse, France

For all author emails, please log on.

BMC Evolutionary Biology 2008, 8:38  doi:10.1186/1471-2148-8-38


The electronic version of this article is the complete one and can be found online at: http://www.biomedcentral.com/1471-2148/8/38


Received:31 May 2007
Accepted:4 February 2008
Published:4 February 2008

© 2008 Lefebvre et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Although much research has been carried out into European Reticulitermes taxonomy in recent years, there is still much discussion about phylogenetic relationships. This study investigated the evolution from intra- to interspecific phylogeny in the island subspecies Reticulitermes lucifugus corsicus and threw new light on this phenomenon. An integrative approach based on microsatellites and mitochondrial and nuclear DNA sequences was used to analyze samples taken from a wide area around the Tyrrhenian sea and showed how the subspecies evolved from its origins to its most recent form on continental coasts.

Results

According to mitochondrial phylogeny and molecular clock calculations, island and continental taxa diverged significantly by vicariance in the Pleistocene glacial period. However, more recently, numerous migrations, certainly human-mediated, affected the structure of the populations. This study provided evidence of direct hybridization and multiple introgressions which occurred in several hybrid areas. Analysis using STRUCTURE based on microsatellite data identified a population in Provence (France) which differed considerably (Fst = 0.477) from populations on the island of Corsica and in Tuscany in the Italian peninsula. This new population, principally distributed in urban areas, is highly heterogeneous especially within the ITS2 regions where homogenization by concerted evolution does not appear to have been completed.

Conclusion

This study provides an unusual picture of genetic interaction between termite populations in the Tyrrhenian area and suggests that more attention should be paid to the role of introgression and human impact on the recent evolution of European termites.

Background

The entomologist P. Rossi [1] described the first subterranean termite species in Tuscany in 1792. The Reticulitermes lucifugus species was first called Termes lucifugum and for a long time was the only recognized species in Europe. Since then, extensive studies of the Reticulitermes genus in this region have shown the existence of six sibling species and a Tyrrhenian subspecies [2]. The precise classification of taxonomic groups is still of current interest and a matter of priority firstly for economic reasons and the need to define urban pests and secondly, of more fundamental importance, to reveal more information about the phylogenetic and biogeographical relationships in the genus [3-10].

Reticulitermes lucifugus corsicus, a subspecies of a European subterranean termite [2], is a species that may help to explain current issues. Although it was established that these termites lived exclusively on islands, recent studies have found the species on the mainland, in Tuscany (Uva et al. found the species in two urban areas [8] and Marini and Montovani found it in one urban area [4]) and in Provence [this study]. The hitherto unsuspected presence of the species on the continent raises certain questions about the characteristics specific to the Corsican termite, in particular its actual taxonomic status and the way in which it migrated to the continent. Its distribution across Tuscany and interaction with Italian termites need to be studied.

R. lucifugus corsicus has still not been clearly defined and is known from subsidiary descriptions in general publications [2,11,12]. As it is closely related to the Italian subspecies and as there is no evidence of reproductive isolation between the two taxa, R. lucifugus corsicus has been classified as an endemic subspecies of the islands of Corsica and Sardinia. Biometric data clearly puts the Corsican termites closer to continental R. lucifugus and only quantitative differences can be noted from chemical data – defensive secretions of the soldiers [13] and cuticular hydrocarbon proportions [2,14]. Only the yellow color of the tibia of reproductives is a significant morphological indicator of their Corsican origin, distinguishing them from R. lucifugus [2]. The best identification method remains the comparison of DNA sequences [3,4,8], as mitochondrial or nuclear haplotypes clearly distinguish the two taxa.

These characteristic differences probably developed after isolation on the islands and confirm that the taxon is endemic. However, it is difficult to date accurately at what stage R. lucifugus corsicus diverged from the Italian R. lucifugus. The geological history of the area is very complex [15-17]. The main hypothesis put forward by Plateaux and Clément [18], later supported by Uva et al. [8] and Luchetti et al. [5] is that the speciation event dates from the last glaciation period with a classic scenario of vicariance and post-glacial colonization. Any questions about the biogeographical history of the island subspecies can be answered by studying the development of termite colonies on the continent. This sympatric existence, in which hybridization is possible, can be considered as a real window on evolutionary processes [19,20] in the Reticulitermes genus. This area contains populations of termites that could possibly shed light on the past and recent evolution of the species complex, from the first speciation events to possible hybridization. More than two centuries after Rossi's discovery, Tuscany could once more be a region that may provide useful information for the study of European termites.

This phylogeographic study in the Tyrrhenian area reveals how the island Reticulitermes lucifugus corsicus termite evolved, from its origins to its recent appearance in continental coastal areas. An integrative approach was used to clarify the phylogenetic relationship with the Italian R. lucifugus termite and to detect any hybrid forms, using various genetic markers [21], such as mitochondrial and nuclear DNA sequences, and several microsatellite regions. These markers have different mutation rates and can, therefore, provide information about evolutionary events on different time scales.

Results

Phylogeny based on mitochondrial genes

The COII sequences of 60 samples (except Vitrolles B; see Table 1 and Figure 1) were aligned and no length polymorphism was detected. Forty-four of the 677 base pairs (bp) were variable and 35 were parsimony-informative. Phylogenetic analyses performed on 54 samples either with maximum parsimony (MP) (not shown; CI = 0.928, RI = 0.988) or with Neighbor-Joining (NJ) methods (Figure 2) gave the same topology. Two major clades were supported by high bootstrap values of 95% and 100% respectively. They correspond to the lucifugus (L) and corsicus (C) taxa matching sequences already in Genbank [4,5] and used for this study (Table 1). Twenty-one private substitutions distinguished the first two groups (p-distance = 0.0347, Table 2). The Sicilian sequences (Palermo and Agrigento) form a third clade (S) with the same degree of divergence from the L and C clades (Table 2). Twenty-two nucleotides separated the consensus sequence of the L clade, 20 compared to Rosarno (Calabre, South Italy) and 26 compared to Viareggio (Tuscany, Italy).

thumbnailFigure 1. Map showing the location of samples in the Tyrrhenian region. Square points correspond to urban sites and round points to natural sites. Numbers refer to Table 1.

thumbnailFigure 2. Phylogenetic tree based on the Neighbor-Joining method, linearized to show the time of divergence. The scale bar represents the time scale (Kyr) according to the geological calibration proposed by Luchetti et al. (2004). Bootstrap values are given above branches. Clades were scored on the right. C = R. l. corsicus. L = R. lucifugus. S = "sicilian" R. lucifugus. Purple brackets followed by an asterisk mark colonies found in unexpected clades.

Table 1. Observed haplotypes, Genbank references and collection locations of samples used in this study.

Table 2. Genetic distance and diversity of clades defined by the phylogenetic analyses on COII sequences.

Several haplotypes were identified in the two main clades in this study, supported by bootstraps higher than 60%. Therefore, with only two subgroups (C1 and C2) found in many different localities, clade C had lower genetic variability than clade L (sequence diversity within clade C = 0.001750 vs. clade L = 0.003148, Table 2). The haplotype C2 was the most common. It was found on all the Tyrrhenian islands (including Elba) as well as in several samples from Provence and Tuscany which, a priori, were geographically non-concordant (Vitrolles, Marseille-O, Uccelina, Follonica, Giannella). Similarly, the other Corsican haplotype (C1) found in the Marseille-C sample could possibly be a continental introduction or introgression. Within clade L, termite colonies from Bastia were also included in an unexpected group from Tuscany (L3) and all the French colonies were linked with colonies from Lazio, South Rome (L2 and some samples of L1).

Nuclear sequence analysis

The nuclear sequences (ITS2), which generally evolve more slowly than the organelle genome [22,23], were far less variable with only 3 nucleotides of the 372 that differentiate the two taxa (Figure 3). Colonies from Follonica, Gianella, Uccelina (Tuscany, Italy) and Valle d'Alesani (Corsica, France) were included in the 'lucifugus' clade, which was contrary to their mitochondrial data. This was also the case for the Castiglioncello colony (Tuscany, Italy), which had a 'Corsican' nuclear haplotype. This lack of concordance between nuclear and cytoplasmic genotypes could be an example of the asymmetrical introgression already described by Avise et al [24].

thumbnailFigure 3. Chromatograms and ITS2 sequences of Provencal hybrid colonies. The three diagnostic nucleotides are shown to distinguish between the taxa. The letters 'R', 'Y' and 'S' indicate the presence of mixed forms, as found on upper chromatograms.

The nuclear data shows evidence of hybridization. Both taxa (Marseille-A: R. lucifugus and Marseille-C: R. l. corsicus) were found in Marseille (Provence, France) as well as a colony with a mix of both haplotypes for the same individual (Marseille-O, Figure 3). The sequence electropherogram (Figure 3) clearly revealed an overlap of the two peaks corresponding to the private substitution of each taxon at each variable site. The absence of noise in the areas round these peaks excluded the possibility of artefacts [25]. This was not an isolated case in Provence (Vitrolles-A). Other colonies in Provence had even more complex forms, either with a crossing-over type mutation on the third diagnostic locus (Marseille-B, Martigues, Plan de Cuques, Venelles) or the insertion of a microsatellite repeat (Vitrolles-B) (Figure 3).

Microsatellite analysis

Five microsatellite loci were analyzed. Heterozygosis was often low (Mean Ho = 0.16) with a mean of 3.83 alleles per locus (Additional file 1). These microsatellite regions were identified in phylogenetically distant Reticulitermes species [26] and seemed to have lower diversity in the Transtyrrenian specific complex. However, analysis of this data using structure [27] revealed significant differences between the populations. It can be difficult to determine the population structure of termites in a sample [28,29] and so the number of populations was estimated using our own biological appreciation [30]. Using the probable estimated values ln P (X|K) and the graphic results (Figure 4), the data could be structured in 3 to 4 populations. Histograms were usually incoherent from K = 7. For K = 2, STRUCTURE application distinguished both taxa. The Corsican termites from Valle d'Alesani could be classified according to their nuclear sequences (ITS2) among the R. lucifugus. Uccelina-0 (Q2.moy = 0.524), and Follonica (Q2.moy = 0,159) in Tuscany may both originate from an admixture between subspecies. The second run (K = 3) showed a new population nested in the R. lucifugus population in Provence, France. The Bastia colony (Corsica) formed a separate cluster for K = 4 and particularly for K = 5. The Marseille-C colony was partly assigned to this Bastia group (Q4.moy = 0,237). Finally, by over-estimating K, the program attributed colonies to populations. This could indicate genetic variability within the R. lucifugus peninsular group. However, this illustrated the homogeneity in the populations in the island samples (from Valle d'Alesani to Ponte Leccia, and Elba), colonies that had apparently been introduced (Bastia) and colonies in Provence (from St Rémy to Salon de Provence, and Marseille A).

Additional file 1. Analyzed microsatellite sequences characteristics and variability. N = number of individuals, Na = number of polymorph alleles and Ho = mean heterozygosis rate observed for each locus in the R. lucifugus and R. l. corsicus colonies.

Format: XLS Size: 15KB Download file

This file can be viewed with: Microsoft Excel ViewerOpen Data

thumbnailFigure 4. Estimated population structure using microsatellite genotyping. Analyses were performed using the admixture model provided by STRUCTURE software (Pritchard et al. 2000). A thin vertical line represents individuals of a colony. Each is probabilistically assigned to K populations (colored segments). Bar plotting was performed for K from 2 to 6. The change in the estimated likelihood values lnP(X|K) is given for each step.

The results obtained using factorial correspondence analysis (FCA) also showed a similar distribution into 4 populations (Figure 5). Axis 1 (26.28%) distinguished R. lucifugus from R. l. corsicus. A hybrid colony (Uccelina) also appeared between the two groups. Axes 2 (12.42%) and 3 (10.49%) separated R. lucifugus colonies into two populations: one from Italy and one from Provence. Close to these clusters, the Bastia colony formed a separate population which was much more similar to the population in Italy. Finally, the lucifugus Italian group included colonies originating from unexpected geographical regions such as Valle d'Alesani (Corsica), Vitrolles-A and Marseille-O (Provence). Based on these results, the genetic differentiation between the three populations was calculated using F-statistics (Table 3). It appears that all the populations are highly distant to each other (Fst > 0.25). The population related to R. l. corsicus is the most differentiated population (mean Fst = 0,534), but, among R. lucifugus taxon, the Provencal population is also highly separated to the Italian population (Fst = 0.265).

thumbnailFigure 5. Factorial correspondence analysis from microsatellite data. The point clouds corresponding to genotyped colonies were assigned to different populations according to their assembling in space. Some colonies have been marked; those whose presence is contradictory to their geographical position were marked with an asterisk.

Table 3. F-statistics (Fst) between different clusters defined by FCA for all populations from microsatellite data. These results refer to figure 5.

Discussion

Past evolution: Origins of endemism

The mitochondrial phylogeny shows that the Tyrrhenian, Sicilian and mainland populations of Reticulitermes form three clearly distinct clades with a high level of divergence. The genetic distances between the island clades (C, S) and the mainland clade (L) are similar (mean distance ≈ 0.0350, Table 2). This suggests that the island populations (C, S) have a comparable evolutionary history. Today, very little is known about Sicilian populations of Reticulitermes. It would be interesting to investigate the level of endemism of these populations to see whether they could be considered as a subspecies, as has already been suggested [5].

The large number of mutations that have accumulated between the three clades shows a long-term separation without any interbreeding. The different hypotheses suggesting that all present-day European Reticulitermes taxa started to diverge from different southern Ice Age refugia [2,5,7,8,12,18] confirmed the postglacial colonization routes described by Taberlet et al. [31] and Hewitt [20]. More recently, Luchetti et al. suggested that Corsican and Sicilian populations differentiated later from the peninsular populations during the Last Glacial Maximum (LGM) period (18 kyr to 12 kyr) [32]. This time of divergence seems relatively recent in comparison to species-level differences observed in other taxa from these areas [20]. Some endemic taxa of the island of Corsica would have started to diverge from 0.01 to 0.08 Myr for mammals [33-35] and from 1.2 to 2.7 Myr for insects [36,37]. Many Corsican taxa may also have originated from the Messinian period (5 Myr) and earlier [38,39]. To justify the high rate of substitution seen in European Reticulitermes populations, Luchetti et al. argued that the social and geographical structure of populations could significantly increase the mtDNA substitution rate [32,40]. But the calibration of the molecular clock used by the authors and based on a geological event could also be questionable. Indeed, several marine regressions occurred in the early Pleistocene [15] which would have led to many junctions-disjunctions between the Tyrrhenian islands and the mainland. In our view, it is so possible that the present-day European Reticulitermes taxa have older origins than previously thought.

It is, however, difficult to determine whether dispersion or vicariance play a greater role in cladogenic events [41,42]. The geological history of the Transtyrrhenian area is highly complex [16,17] and does not support any particular theory. As mentioned above, many species of the Tyrrhenian area have different evolutionary and biogeographical histories. The possibility of an insular colonization from Tuscany after post-glacial expansion through the Italian peninsula has also been suggested [32]. Furthermore, the high level of genetic homogeneity (mitochondrial, nuclear and also microsatellite loci) observed in termite populations in Corsica could be the result of a founding effect initiated by a migratory and colonizing event. The endemic origin of R. l. corsicus is also a possibility, with ancestral origins from an isolated population in an Ice Age refugium. Thus, each taxon would have been separated from taxa from different refuges (in South Italy, in Sicilia, and somewhere in Corsica or in Sardania). This could explain why the three clades have almost the same number of substitutions (about 20 nucleotides) and consequently the same divergence time. As the Corsican mitotype is closer to the Calabran (19 nucleotides) than to the Tuscan (23 nucleotides), the populations in Corsica and Sardinian may have started to diverge before recolonization of Italy.

The expansion on the continent is confirmed by mitochondrial and microsatellite data. Colonization was slow and may have lasted several thousand years. At the mitotype level, this expansion through the Italian peninsula generated a South/North cline, already reported by Uva et al. [8]. The expansion rate was probably the same as for the Iberian Reticulitermes, as the divergence times are similar [6]. Moreover, subterranean termites disperse over short distances in natural conditions. For these reasons, Italian recolonization by R. lucifugus could correspond to a 'phalanx' expansion mode [43]. The microsatellite data (Figure 4) shows a clear allelic diversity, specific to continental populations (island-continental model [44]). However, with the exception of the Provencal area, the colonies form a homogeneous entity. This unusual structure could be attributed to a bottleneck event during the post-glacial colonization, particularly when crossing the Ligurian Alps. The mitochondrial haplotypes, similar to those in Rome could also suggest a more recent introduction, perhaps human-mediated, as this route was used by several civilizations.

Recent evolution: Introductions, hybridization and introgression

In the more recent past, since 2000 BC, human civilization developed considerable marine traffic in the Tyrrhenian region. It is therefore highly probable that colonies of Reticulitermes were introduced into unexpected zones by human transport [2]. This kind of event has already been documented – R. lucifugus in Bordeaux (France) (personal communication Bagnères.), R. grassei in Devon (UK) ([45] and personal communication Bagnères), R. santonensis in Hamburg (Germany) and Santiago (Chile) [2], and R. urbis in France and Italy ([7] and Bagnères et al., unpublished results). Three colonies -Bastia, Viareggio and Marseille-C- appear to have been introduced. Of these, the colony in Bastia is of particular interest as it forms an entity, neither entirely Provencal (Figure 4, run 2) nor entirely Italian (Figure 4, runs 1 and 3). The two other introduced colonies may have crossed with autochtonous populations, as shown by traces of allochtonous genes in the microsatellite data.

More generally, migrations in the Tyrrhenian region have created different hybrid zones. They resemble the bimodal zones described by Jiggins and Mallet [46] and are characteristic of a quasi-absence of direct intermediate forms. Only the Uccelina-O colony has F1-type hybrids. Nevertheless, there are signs of varying degrees of hybridization in colonies such as the presence of unexpected genotypes.

Other authors [4,5,8] suggested that some hybrid zones were situated in natural areas in the Tuscany pine-woods. This study did in fact find some introgressive forms in Tuscan colonies – Parco del Uccelina, Reserve of Follonica, Castiglioncello and Gianella – with conflicting mitochondrial and nuclear haplotype data (Table 4). Although most of these Tuscan colonies have an allochtonous mitotype, introgression seems to act in both directions. Indeed, termites in Castiglioncello have Corsican nuclear sequences with an Italian mitotype. The fixation of allochtonous haplotypes can occur through backcrossing to the parental populations [21,47] or can be helped by biased sex-ratios already observed in Reticulitermes lucifugus [48,49].

Table 4. Description of genetic particularities observed in some localities for all markers considered.

On the opposite shore, the persistence of Corsican microsatellite alleles in the Follonica sample (Tuscany) and the presence of microsatellite alleles characteristic of Tuscany along the East coast of Corsica (Solenzara, Valle d'Alesani) could also indicate relatively recent hybridizations. In general, these traces of hybridization tend to diminish and even disappear by backcross with local populations.

In Southern France, the Provence region may be a specific new hybrid zone. Among the multiple copies of the ITS2 gene, some colonies (Figure 3) have a mix of haplotypes from the two origins, several forms of which come from DNA recombination. However, no signs of hybridization can be detected with microsatellite markers. Perhaps a higher number of microsatellite markers is required to improve detection of hybrid individuals [50]. The numerous variants in the sample may result from mechanisms such as unequal crossing-over and gene conversion during the concerted evolution, which may have homogenized the number of rDNA copies [51,52]. ITS sequence homogenization by concerted evolution has already been observed in a species of subterranean termites Coptotermes gestroi (Rhinotermitidae) by Jenkins et al [53]. The concerted evolution might currently be slow [54], and in the end slower than microsatellite loci homogenization by backcross or by genetic drift. This might explain why no apparent signs of hybridization are found in Provence. Nevertheless, current results do not exclude the possibility that these different forms come from an ancestral polymorphism, but the presence of both Corsican mitotype and Italian nuclear haplotypes in the Vitrolles A colony (Table 4) clearly supports the theory of a hybrid origin of ITS2 mixed forms.

The Provence region appears quite different from Tuscany and Corsica. Firstly, there is a distinct population entity, which is different from the Italian R. lucifugus and R. l. corsicus groups (Figures 4 and 5). Secondly, most of the colonies sampled were collected from urban areas (Figure 1). Thirdly, there are genetic particularities (mixed forms for ITS genes) not found anywhere else in the data (Figure 3). This raises the questions of whether these characteristics are correlated and whether the urban environment influences the termites' adaptation process and the evolution of hybrids.

Conclusion

Reticulitermes lucifugus corsicus is one of the best-described subspecies of termite, defined by its distribution (Corsica, Sardinia, Elba and some continental localities), by various taxonomic criteria (genetic, morphological and chemical) and also by this complex phyletic evolution (see also the subspecies of Reticulitermes speratus [10]). Its divergence goes back to cladogenic events during the Ice Age. However, owing to many migratory events, the evolution of transtyrrhenian populations of termites has been more complex. The study carried out in the area revealed the existence of three very different populations (FST = 0.477, Table 3): one population on the Italian peninsula (Tuscany), a second in the Tyrrhenian islands (Corsica) and a third in Provence. With the Sicilian phylum, they form strong cladistic units, constituting a complex of species within R. lucifugus. Although hybrids occur, it seems that systematic homogenization of the genomes occurs subsequently. With a few intermediate forms and even some introgressive forms, these taxa may maintain an identity signature, and may finally be on the way to speciation.

Nevertheless, evidence of introgression in hybrid zones could influence how Reticulitermes adapt. Allowing horizontal transfers of the genetic material, better adapted forms could appear in some small populations [21]. The colonies sampled in this study are small and patchily distributed in pinewood areas, especially in anthropized environments and could, therefore, fulfill the conditions necessary for such adaptation. From this point of view, it is important to consider the Reticulitermes as urban pests and study how they are adapting to cities – as for urban colonies in Provence. After thousands of years of speciation under paleogeographical constraints, human impact [2,5,55,56]-urbanism, maritime transport, and accidental infestations-seems to be the determining element in recent Reticulitermes evolution.

Methods

Sampling

Various Reticulitermes samples from the Tyrrhenian region were analyzed. Some samples had been collected recently in Corsica (France) and Tuscany (Italy), others came from old collections that had been kept in a laboratory (Uva 1999) or from ad hoc donations. Of the 61 collection sites detailed in Table 1, 42 were in natural areas near pine-woods and 19 were from infested urban sites. A transect was taken in Uccelina (Tuscany). A hundred specimens from each colony were stored in 96% alcohol for genetic analyses. Further information on the samples is given in Table 1 and the sites are shown on a map in Figure 1.

DNA extraction, amplification and sequencing

Total DNA was extracted from a single termite preserved in alcohol, using a modified version of the method described by Kocher et al. [57]. PCR amplification was performed using a Biometra 96 T1 thermal cycler or a Stratagene robocycler Gradiant 96. Different programs and amplification conditions were used depending on the types of primers used for sequencing or microsatellite analyses (Additional file 2).

Additional file 2. List of primers and PCR profiles used for amplification and sequencing.

Format: XLS Size: 16KB Download file

This file can be viewed with: Microsoft Excel ViewerOpen Data

Two types of sequencing commonly used to study the phylogeny of the Reticulitermes genus were used – mitochondrial sequences – cytochrome oxydase II gene (COII) [3,6,58]- and nuclear sequences – the ITS2 region (internal transcribed spacer) of the ribosomal DNA [6,45,58] (Additional file 2). Amplified DNA was purified using the Quiagen-PCR purification kit and sequenced in both directions. Each sequence generated by this study was deposited in GenBank database (the accession numbers are given in Table 1).

Phylogenetic analysis

Multiple consensus sequences were aligned using the Clustal W algorithm [59] from the BIOEDIT 7.0.1 sequence editor [60] and corrected manually. Phylogenetic analyses of this data were performed using MEGA 3.0 [61]. DNA sequences were first analyzed using the neighbor-joining (NJ) method [62] on Kimura 2-parameter distances [63]. Maximum parsimony (MP) analyses were then conducted using a heuristic search method with 100 repeats and tree bisection-reconnection (TBR) with all characters unordered and unweighted. R. santonensis was chosen as the outgroup. Node support was estimated by searching on 1000 non parametric bootstrap replicates [64].

Microsatellite genotyping

Ten termites from 36 Corsican and Tuscan colonies (including the Uccelina transect) were genotyped at six microsatellite loci previously isolated in the species R. flavipes [65] and R. santonensis [26](Additional file 1). PCR products were separated by electrophoresis on 6% polyacrylamide gels run on a LI-COR 4000 L sequencer. Alleles were scored using Gene Profiler 4.03 (Scanalytics, Inc.).

A factorial correspondence analysis (FCA) was then performed using genetix 4.05.2 software to classify the populations in different clusters [66]. genepop was used to calculate Fst values between the different clusters defined by the AFC [67]. STRUCTURE was then used for the genetic identification of clustered populations, allogenous introductions and potential admixed colonies [27]. This software uses a model-based clustering method to infer population structure using multiloci genotypes. Each individual in the colony was probabilistically assigned to K populations. Runs of various lengths were performed with different numbers of genetic clusters (K), testing all values of K from 2 to 7. Because the populations could potentially hybridize, the admixture model [27] provided by the software was used.

Authors' contributions

TL did most of the sampling. TL, NC, DL and SD did the DNA sequencing and microsatellite genotyping. TL and NC analyzed the data and wrote the paper. AGB conceived and coordinated the study. All the authors read and approved the final manuscript.

Acknowledgements

We are grateful to G Landrin from the DDE13, towns in the Provence region, PCOs and students for their help in providing samples. We wish to thank P Uva for his assistance, particularly during the trip to Tuscany with TL, and the Logli family for looking after TL' during his trip to Corsica in 2004. We also wish to thank S Dronnet, M Kutnik and JP Christides for their scientific and technical assistance, N Bierne and M Zimmermann for their comments on the manuscript. We are especially gratefull to T Tebby and C Jouzeau for help in English formatting. CNRS provided financial support for this study and AGB received a grant from the "Association des Villes pour la Lutte contre les Termites".

References

  1. Rossi P: Termes lucifugum.

    Mantissa Insectorum. Tome 1 1792, Plate 5:107. OpenURL

  2. Clément J-L, Bagnères A-G, Uva P, Wilfert L, Quintana A, Reinhard J, Dronnet S: Biosystematics of Reticulitermes termites in Europe: morphological, chemical and molecular data.

    Insectes Soc 2001, 48:202-215. Publisher Full Text OpenURL

  3. Austin JW, Szalanski AL, Uva P, Bagnères A-G, Kence EA: A comparative genetic analysis of the subterranean Termite genus Reticulitermes.

    Ann Entomol Soc Am 2002, 95:753-760. Publisher Full Text OpenURL

  4. Marini M, Montovani B: Molecular relationships among European samples of Reticulitermes (Isoptera, Rhinotermitidae).

    Mol Phylogenet Evol 2002, 22:171-191. Publisher Full Text OpenURL

  5. Luchetti A, Trenta M, Montovani B, Marini M: Taxonomy and phylogeny of north mediterranean Reticulitermes termites (Isoptera, Rhinotermitindae): a new insight.

    Insectes soc 2004, 51:117-122. Publisher Full Text OpenURL

  6. Kutnik M, Uva P, Brinkworth L, Bagnères A-G: Phylogeography of two European Reticulitermes (Isoptera) species: the Iberian refugium.

    Mol Ecol 2004, 13:3099-3113. PubMed Abstract | Publisher Full Text OpenURL

  7. Uva P, Clément J-L, Austin JW, Aubert J, Zaffagnini V, Quintana A, Bagnères A-G: Origin of a new Reticulitermes termite inferred from mitochondrial DNA data.

    Mol Phylogenet Evol 2004, 30:344-353. PubMed Abstract | Publisher Full Text OpenURL

  8. Uva P, Clément J-L, Bagnères A-G: Colonial and geographical variations in agonistic behaviour, cuticular hydrocarbons and mtDNA of Italian populations of Reticulitermes lucifugus (Isoptera, Rhinotermitidae).

    Insectes Soc 2004, 51:163-170. Publisher Full Text OpenURL

  9. Nobre T, Nunes L, Eggleton P, Bignell DE: Distribution and genetic variation of Reticulitermes (Isoptera: Rhinotermitidae) in Portugal.

    Heredity 2006, 96:403-409. PubMed Abstract | Publisher Full Text OpenURL

  10. Park YC, Kitade O, Schwarz M, Jim JP, Kim W: Intraspecific molecular phylogeny, genetic variation and phylogeography of Reticulitermes speratus (Isoptera: Rhinotermitidae).

    Mol Cells 2006, 21:89-103. PubMed Abstract | Publisher Full Text OpenURL

  11. Clément J-L: L'agression inter et intraspécifique des espèces française du genre Reticulitermes.

    R Acad Sci 1978, 286:351-354. OpenURL

  12. Plateaux L, Clément JL: La séparation récente des termites Reticulitermes du complexe lucifugus.

    Rev Fac Sci Tunis 1984, 3:179-206. OpenURL

  13. Quintana A, Reinhard J, Faure R, Uva P, Bagnères A-G, Massiot G, Clément JL: Geographic variation in terpenoid composition of European Reticulitermes termite defensive secretion.

    J Chem Ecol 2003, 29:639-652. PubMed Abstract | Publisher Full Text OpenURL

  14. Uva P: Studio comportamentale e variazione degli idrocarburi cuticolari nelle popolazioni italiane di Reticulitermes lucifugus (Isoptera: Rhinotermitidea). In Tesi di Laurea. La Sapienza University, Roma; 1999. OpenURL

  15. Zunino M, Zullini A: Biogeografia-la dimensione spaziale dell'evoluzione. 2nd edition. Casa Editrice Ambrosiana; 1995.

  16. De Jong H: In search of historical biogeographic patterns in the western Mediterranean terrestrial fauna.

    Biol J Linnean Soc 1998, 65:99-164. OpenURL

  17. Palmer M, Cambefort Y: Evidence for reticulate paleogeography: beetle diversity linked to connection-disjunction cycles of the Gibraltar strait.

    J Biogeography 2000, 27:403-416. Publisher Full Text OpenURL

  18. Clément J-L: Spéciation des Reticulitermes Européens (Isoptères). In Thèse d'Etat. Pierre and Marie Curie University; 1981. OpenURL

  19. Harrison RG: Hybrid zone and the evolutionary process. New York: Oxford University Press; 1993. OpenURL

  20. Hewitt GH: Speciation, hybrid zones and phylogeography – or seeing genes in space and time.

    Mol Ecol 2001, 10:537-549. PubMed Abstract | Publisher Full Text OpenURL

  21. Dowling TE, Secor CL: The Role of Hybridization and Introgression in the Diversification of Animals.

    Ann Rev Ecol Syst 1977, 28:593-619. Publisher Full Text OpenURL

  22. Vawter L, Brown WM: Nuclear and Mitochondril DNA comparisons reveal extreme rate variation in the molecular clock.

    Science 1986, 234:194-196. PubMed Abstract | Publisher Full Text OpenURL

  23. Morgan JAT, Blair D: Relative merits of nuclear ribosomal internal transcribed spacers and mitochondrial CO1 and ND1 genes for distinguishing among Echinostma species (Trematoda).

    Parasitology 1998, 116:289-297. PubMed Abstract | Publisher Full Text OpenURL

  24. Avise JC, Arnold J, Ball RM, Bermingham E, Lamb T, Neigel JE, Reeb CA, Saunders NC: Intraspecific Phylogeography: The Mitochondrial DNA Bridge Between Population Genetics and Systematics.

    Ann Rev Ecol Syst 1987, 18:489-522. OpenURL

  25. Harpke D, Peterson A: Non-concerted ITS evolution in Mammillaria (Cactaceae).

    Mol Phylogenet Evol 2006, 41:579-593. PubMed Abstract | Publisher Full Text OpenURL

  26. Dronnet S, Bagnères A-G, Juba TR, Vargo EL: Polymorphic microsatellite loci in the European subterranean termite, Reticulitermes santonensis Feytaud.

    Mol Ecol Notes 2004, 4:127-129. Publisher Full Text OpenURL

  27. Pritchard JK, Stephens P, Donnelly P: Inference of population structure using multilocus genotype data.

    Genetics 2000, 155:945-959. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  28. Documentation for STRUCTURE software: Version 2 [http://pritch.bsd.uchicago.edu] webcite

  29. Falush D, Stephens M, Pritchard JK: Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies.

    Genetics 2003, 164:1567-1587. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  30. Semon M, Nielsen R, Jones MP, McCouch SR: The population structure of African cultivated rice Oryza glaberrima (Steud.): evidence for elevated levels of linkage disequilibrium caused by admixture with O. sativa and ecological adaptation.

    Genetics 2005, 169:1639-1647. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  31. Taberlet P, Fumagalli L, Wust-Saucy AG, Cosson JF: Comparative Phylogeography and postglacial colonization routes in Europe.

    Mol Ecol 1998, 7:453-464. PubMed Abstract | Publisher Full Text OpenURL

  32. Luchetti A, Marini M, Montovani B: Mitochondrial evolutionary rate and speciation in termites: data on European Reticulitermes taxa (Isoptera, Rhinotermitindae).

    Insectes soc 2005, 52:218-221. Publisher Full Text OpenURL

  33. Hmwe SS, Zachos FE, Eckert I, Lorenzini R, Fico R, Hartl GB: Conservation genetics of the endangered red deer from Sardinia and Mesola with further remarks on the Phylogeography of Cervus elaphus corsicanus.

    Biol J Linn Soc 2006, 88:691-701. Publisher Full Text OpenURL

  34. Pierpaoli M, Riga F, Trocchi V, Randi E: Species distinction and evolutionary relationships of the Italian hare (Lepus corsicanus) as described by mitochondrial DNA sequencing.

    Mol Ecol 1999, 8:1805-1817. PubMed Abstract | Publisher Full Text OpenURL

  35. Lebarbenchon C, Poitevin F, Montgelard C: Genetic variation of the weasel (Mustela nivalis) in Corsica based on mitochondrial control region sequences.

    Mamm Biol 2006, 71:164-171. Publisher Full Text OpenURL

  36. Allegrucci G, Toscido V, Sbordoni V: Molecular phylogeography of Dolichopoda cave crickets (Orthoptera, rhaphidophoridae): A scenario suggested by mitochondrial DNA.

    Mol Phylogenet Evol 2005, 37:153-164. PubMed Abstract | Publisher Full Text OpenURL

  37. Ribera I, Bilton DT, Vogler P: Mitochondrial DNA phylogeography and population history of Meladema diving beetles on the Atlantic Islands and in the Mediterranean basin (Coleoptera, Dytiscidae).

    Mol Ecol 2003, 12:153-167. PubMed Abstract | Publisher Full Text OpenURL

  38. Aubert J, Barascud B, Descimon H, Michel F: Ecology and genetics of interspecific hybridization in the swallowtails, Papilio hospiton Géné and P. machaon L., in Corsica (Lepidoptera: papilionidae).

    Biol J Linn Soc 1997, 60:467-492. OpenURL

  39. Ketmaier V, Argano R, Caccone A: Phylogeography and molecular rates of subterranean aquatic Stenasellid Isopods with peri-Tyrrhenian distribution.

    Mol Ecol 2003, 12:547-555. PubMed Abstract | Publisher Full Text OpenURL

  40. Hoelzer GA, Wallman J, Melnick DJ: The effects of social structure, geographical structure, and population size on the evolution of mitochondrial DNA: II. Molecular clocks and the lineage sorting period.

    J Mol Evol 1998, 4:21-31. Publisher Full Text OpenURL

  41. Ronquist F: Dispersal-Vicariance Analysis: A new approach to the Quantification of Historical Biogeography.

    Syst Biol 1997, 46:195-203. Publisher Full Text OpenURL

  42. Noonan BP, Chippindale PT: Dispersal and vicariousness: the complex evolutionary history of boid snakes.

    Mol Phylogenet Evol 2006, 40:347-358. PubMed Abstract | Publisher Full Text OpenURL

  43. Nichols RA, Hewitt GM: The genetic consequences of long distance dispersal during colonisation.

    Heredity 1994, 72:312-317. OpenURL

  44. Barton NH: Natural selection and random genetic drift as causes of evolution on Islands. In Evolution on Islands. Edited by Grant P. Oxford University Press; 1998:102-173. OpenURL

  45. Jenkins TM, Dean RE, Verkerk R, Forshler BT: Phylogenetic analyses of two mitochondrial genes and one nuclear intron region illuminate European subterranean termite (Isoptera: Rhinotermitidae) gene flow, taxonomy, and introduction dynamics.

    Mol Phylogenet Evol 2001, 20:286-293. PubMed Abstract | Publisher Full Text OpenURL

  46. Jiggins CD, Mallet J: Bimodal hybrid zones and speciation.

    Trends Ecol 2000, 15:250-255. Publisher Full Text OpenURL

  47. Arnold ML: Natural Hybridization as an Evolutionary Process.

    Annu Revi Ecol Syst 1992, 23:237-261. Publisher Full Text OpenURL

  48. Herfs A: Der Schwarmflug von Reticulitermes lucifugus Rossi.

    Z Angew Entomol 1951, 33:69-77. OpenURL

  49. Grassi B, Sandias A: Costituzione e sviluppo della società dei termitidi.

    Atti Accad Gioena Sci Nat Catania 1893, 6:1-75. OpenURL

  50. Vähä J-P, Primmer CR: Efficiency of model-based Bayesian methods for detecting hybrid individuals under different hybridization scenarios and with different number of loci.

    Mol Ecol 2006, 15:63-72. PubMed Abstract | Publisher Full Text OpenURL

  51. Li WH: Concerted evolution of multigene families. In Molecular Evolution. Edited by Li WH. New Ed: Sinauer assoc Inc; 1997:309-334. OpenURL

  52. Álvarez I, Wendel JF: Cryptic Interspecific Introgression and Genetic Differentiation within Gossypium aridum (Malvaceae) and its Relatives.

    Evolution 2006, 60:505-517. PubMed Abstract OpenURL

  53. Jenkins TM, Jones SC, Lee C-Y, Forshler BT, Chen Z, Lopez-Martinez G, Gallagher NT, Brown G, Neal M, Thistleton B, Kleinschmidt S: Phylogeography illuminates maternal origins of exotic Coptotermes gestroi (Isoptera: Rhinotermitidae).

    Mol Phylogenet Ecol 2007, 42:612-621. Publisher Full Text OpenURL

  54. Wörheide G, Nichols SA, Goldberg J: Intragenomic variation of the rDNA internal transcribed spacers in sponges (Phylum Porifera): implications for phylogenetic studies.

    Mol Phylogenet Evol 2004, 33:81-830. Publisher Full Text OpenURL

  55. Gay FJ: A world review of introduced species of termites.

    Bull Commonw Sci Ind Res Org Australia, Melbourne 1967, 286:1-88. OpenURL

  56. Aber A, Fontes LR: Reticulitermes lucifugus (Isoptera, Rhinotermitidae), a pest of wooden structures, is introduced into the South American continent.

    Sociobiol 1993, 21:335-339. OpenURL

  57. Kocher TD, Thomas WK, Meyer A: Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers.

    Proc Nati Acad Sci USA 1989, 86:6196-6200. Publisher Full Text OpenURL

  58. Roy V, Demanche C, Livet A, Harry M: Genetic differentiation in the soil-feeding termite Cubitermes sp. affinis subarquatus: occurence of cryptic species revealed by nuclear and mitochondrial markers.

    BMC Evol Biol 2006, 6:102. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text OpenURL

  59. 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.

    Nucl Acids Res 1994, 22:4673-4680. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  60. Hall TA: Bioedit: a user-friendly biological sequence alignment, editing and analysis program for windows 95/98/NT.

    Nucleic Acids Symp Ser 1999, 41:95-98. OpenURL

  61. Kumar S, Tamura K, Nei M: MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment.

    Brief Bioinformatics 2004, 5:150-163. PubMed Abstract | Publisher Full Text OpenURL

  62. Saitou N, Nei M: The neighbour-joining method: a new method for reconstructing phylogenetic trees.

    Mol Biol Evol 1987, 4:406-425. PubMed Abstract | Publisher Full Text OpenURL

  63. Kimura M: A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences.

    J Mol Evol 1980, 16:111-120. PubMed Abstract | Publisher Full Text OpenURL

  64. Hillis DM, Bull JJ: An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis.

    Syst Biol 1993, 42:182-192. Publisher Full Text OpenURL

  65. Vargo EL: Polymorphism at trinucleotide microsatellite loci in the subterranean termite Reticulitermes flavipes.

    Mol Ecol Notes 2000, 9:817-829. OpenURL

  66. Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F: GENETIX 4.05, logiciel sous Windows TM pour la génétique des populations. Laboratoire Génome, Populations, Interactions, CNRS UMR 5171, Université de Montpellier II (France); 2004. PubMed Abstract | Publisher Full Text OpenURL

  67. Raymond M, Rousset F: GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism.

    J Heredity 1995, 86:248-249. OpenURL