Acessibilidade / Reportar erro

Phylogeographical analysis of neotropical Rhagoletis (Diptera: Tephritidae): did the Andes uplift contribute to current morphological differences?

Análisis filogeográfico de Rhagoletis neotropicales (Diptera: Tephritidae): ¿ha contribuido el levantamiento de los Andes a explicar las actuales diferencias morfológicas?

Abstracts

Neotropical Rhagoletis species are arranged in four groups: nova,psalida,striatella and ferruginea, which include 18 species. On both sides of the Andes, the evolution of morphological differences among these groups has been suggested to be related to the Andes uplift process. In order to test this hypothesis, a phylogenetic analysis of morphological and molecular data was performed. The results suggest that: 1) Neotropical species of Rhagoletis constitute a separate group from Paleartic and North American species, with the only exception being a member of the striatella group having a certain association with the northern species. 2) Neotropical species seem to form a monophyletic clade, although statistical support for this is weak. 3) The split of South American Rhagoletis from other groups was dated at 4.333 million years ago, which is before the emergence of a continuous landbridge between Central and South América. 4) Within species distributed in South América, morphological and molecular data were coincident, placing species of the ferruginea group separate from the other Neotropical Rhagoletis. 5) The divergence of the ferruginea group from the other groups was dated at 3.882 million years ago, which is before the last uplift of the Andes. These results suggest that diversification of the ferruginea,psalida and nova groups, on each side of the Andes, was the result of a vicariant separation followed by dispersal and isolation processes. Thus, these results support the hypothesis that the Andes uplift has played an important role in Neotropical Rhagoletis diversification.

Fruit fly; diversification; phylogeny; South América


Las especies de Rhagoletis neotropicales han sido agrupadas en cuatro grupos: nova,psalida,striatella y ferruginea, constituyendo 18 especies. Se han descrito diferencias morfológicas entre estas especies a ambos lados de la cordillera de los Andes que podrían relacionarse con el proceso de levantamiento cordillerano. En este trabajo se evalúa esta hipótesis usando análisis filogenético de atributos morfológicos y moleculares. Los resultados muestran que: a) las especies Neotropicales de Rhagoletis constituyen un grupo separado de las especies Palearticas y Norteamericanas, con la excepción de un miembro del grupo striatella el cual presenta cierta asociación con las especies Norteamericanas; 2) Las especies Neotropicales parece conformar un clado monofilético; 3) La separación de los grupos Sudamericanos de otros grupos fue estimada en 4.333 millones de años antes del presente, proceso anterior a la emergencia del puente de tierra entre América Central y Sudamérica; 4) Dentro de las especies con distribución Sudamericana, los caracteres morfológicos y moleculares coinciden en ubicar algunas especies del grupo ferruginea separadas de especies de Rhagoletis Neotropicales. 5) La separación del grupo ferruginea fue estimada en 3.882 millones de años antes del presente, evento que precede al último levantamiento de los Andes. La diversificación de los grupos ferruginea,psalida y nova a uno y otro lado de la cordillera de los Andes parece responder inicialmente a un proceso vicariante y posteriores eventos de dispersión y aislamiento. Estos resultados sugieren que el levantamiento de los Andes habría participado en los patrones de diversificación de las Rhagoletis Neotropicales.

Mosca de la fruta; diversificación; filogenia; Sudamérica


SYSTEMATICS, MORPHOLOGY AND PHYSIOLOGY

Phylogeographical analysis of neotropical Rhagoletis (Diptera: Tephritidae): did the Andes uplift contribute to current morphological differences?

Análisis filogeográfico de Rhagoletis neotropicales (Diptera: Tephritidae): ¿ha contribuido el levantamiento de los Andes a explicar las actuales diferencias morfológicas?

Claudio C. RamírezI; Marcela SalazarI; R. Eduardo PalmaII; Cecilia CorderoI; Luís Meza-BassoI

IInstituto de Biología Vegetal y Biotecnología, Univ. de Talca, Casilla 747, Talca, Chile; clramirez@utalca.cl, m_salazar@utalca.cl, lcordero@utalca.cl, mezabas@utalca.cl

IIDepto. de Ecología and Centro de Estudios Avanzados en Ecología y Biodiversidad (CASEB), Pontificia Universidad Católica de Chile, Alameda 340, Santiago 6513677; epalma@bio.puc.cl

ABSTRACT

Neotropical Rhagoletis species are arranged in four groups: nova,psalida,striatella and ferruginea, which include 18 species. On both sides of the Andes, the evolution of morphological differences among these groups has been suggested to be related to the Andes uplift process. In order to test this hypothesis, a phylogenetic analysis of morphological and molecular data was performed. The results suggest that: 1) Neotropical species of Rhagoletis constitute a separate group from Paleartic and North American species, with the only exception being a member of the striatella group having a certain association with the northern species. 2) Neotropical species seem to form a monophyletic clade, although statistical support for this is weak. 3) The split of South American Rhagoletis from other groups was dated at 4.333 million years ago, which is before the emergence of a continuous landbridge between Central and South América. 4) Within species distributed in South América, morphological and molecular data were coincident, placing species of the ferruginea group separate from the other Neotropical Rhagoletis. 5) The divergence of the ferruginea group from the other groups was dated at 3.882 million years ago, which is before the last uplift of the Andes. These results suggest that diversification of the ferruginea,psalida and nova groups, on each side of the Andes, was the result of a vicariant separation followed by dispersal and isolation processes. Thus, these results support the hypothesis that the Andes uplift has played an important role in Neotropical Rhagoletis diversification.

Key words: Fruit fly, diversification, phylogeny, South América

RESUMEN

Las especies de Rhagoletis neotropicales han sido agrupadas en cuatro grupos: nova,psalida,striatella y ferruginea, constituyendo 18 especies. Se han descrito diferencias morfológicas entre estas especies a ambos lados de la cordillera de los Andes que podrían relacionarse con el proceso de levantamiento cordillerano. En este trabajo se evalúa esta hipótesis usando análisis filogenético de atributos morfológicos y moleculares. Los resultados muestran que: a) las especies Neotropicales de Rhagoletis constituyen un grupo separado de las especies Palearticas y Norteamericanas, con la excepción de un miembro del grupo striatella el cual presenta cierta asociación con las especies Norteamericanas; 2) Las especies Neotropicales parece conformar un clado monofilético; 3) La separación de los grupos Sudamericanos de otros grupos fue estimada en 4.333 millones de años antes del presente, proceso anterior a la emergencia del puente de tierra entre América Central y Sudamérica; 4) Dentro de las especies con distribución Sudamericana, los caracteres morfológicos y moleculares coinciden en ubicar algunas especies del grupo ferruginea separadas de especies de Rhagoletis Neotropicales. 5) La separación del grupo ferruginea fue estimada en 3.882 millones de años antes del presente, evento que precede al último levantamiento de los Andes. La diversificación de los grupos ferruginea,psalida y nova a uno y otro lado de la cordillera de los Andes parece responder inicialmente a un proceso vicariante y posteriores eventos de dispersión y aislamiento. Estos resultados sugieren que el levantamiento de los Andes habría participado en los patrones de diversificación de las Rhagoletis Neotropicales.

Palabras Clave: Mosca de la fruta, diversificación, filogenia, Sudamérica

The genus Rhagoletis Loew has been studied mainly because of its economic importance as an agricultural pest to fruits in many countries, and because it is considered a model of sympatric speciation processes. Within Rhagoletis, species with Neotropical distribution have been arranged into four groups: 1) the nova group (R. penela Foote, R. nova (Schiner), R. conversa (Brèthes), R. tomatis Foote, R. lycopersella Smyth, R. willinki Aczél and Rhagoletis brncici Frias); 2) the psalida group (R. psalida Hendel, R. rhytida Hendel and R. metallica Hendel); 3) the striatella group (R. striatella Wulp, R. macquartii (Loew), R. jamicensis Foote, R. triangularis Hernández & Frías, R. solanophaga Hernández & Frías and R. nicaraguensis Hernández & Frías); and 4) the ferruginea group (R. ferruginea Hendel, R. adusta Foote and R. blanchardi Aczél) (sensu Foote 1981, Hernández-Ortiz & Frías 1999). Although there is little information regarding host affiliation, most of these species are mainly associated with the Solanaceae (Bush 1966, Foote 1981, Frías 1992). Members of the South American genus Rhagoletis were described to form a coherent monophyletic group, although their relationships are still a matter of study (Smith et al. 2006).

Morphological differences among Neotropical Rhagoletis distributed east and west of the Andes have been described (Frías 1992). All Rhagoletis species located east of the Andes (with the exception of R. willinki) contain a complete posterior apical crossband which extends from the sub-apical band to the wing margin, whereas species located west of the Andes lack this band or have it present as an elongated spot (Fig. 1). Similarly, species east of the Andes have a spherical spermatheca, which differs from those having a rodlike spermatheca on the west side of the Andes (Fig. 1). Therefore, it is likely that the Andes uplift may have played a certain role in the diversification of Neotropical Rhagoletis, although no independent evidence has corroborated this. It is worth noting that the Andes have played some role in the generation of different races of the butterfly Heliconius erato L., where a basal split between groups of races from the east and west of the Andes seems to have occurred at the beginning of the Pleistocene by vicariant separation (Brower 1994). A similar process could have taken place in Neotropical Rhagoletis species.


The aim of this study was to examine the phylogeography (sensu Avise 2000) of Neotropical Rhagoletis species. In particular, the aim was to test the hypothesis regarding the possible role of the Andes uplift as a diversifying mechanism. First, phylogenetic affinities among Neotropical Rhagoletis were studied based on a set of morphological data. Second, a phylogenetic analysis with estimations of divergence times of clades based on available mitochondrial cytochrome c oxidase subunit II (COII) DNA sequences was performed. In both cases, sequences of North American and Palearctic Rhagoletis species were also included in order to study their relationship with the Neotropical species. Finally, a dendrogram based on the similarity of their geographical ranges (biogeogram) was built to estimate the geographic relationship of the Neotropical Rhagoletis species

Material and Methods

Molecular data. We examined available COII sequences from sixteen species and obtained new sequences for two more species (Table 1). The outgroup species Ceratitis capitata (Wiedemann) was chosen on its taxonomic position within the Tephritidae (Smith & Bush 1997, McPheron & Han 1997). New sequences for R. nova and R. tomatis were obtained by isolating DNA and following an optimized protocol for alcohol-preservation of individual flies (Han & McPheron 1997). The primers used for PCR amplification were: CITOX-5: 5'ATG-ACA-ACA-TGA-GCT-GCC-CTT-GGC-CTT-CAA3' (forward) and CITOX-3: 5'TGA-ATT-TAC-TCT-ATT-TTT-AAT-TCA-TTT-AAT3' (reverse) (Gasparich et al. 1995). PCR was performed in 25 µl reaction volumes containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 6.7 mM MgCl2, 0.25 mM each of dATP, dCTP, dGTP and dTTP, 0.8 µM of each primer, 1.5 units of Taq polymerase (Life Technologies) and 25 to 100 ng of template DNA. After the initial denaturation (95°C, 3 min), PCR was run for 40 cycles consisting of a 94°C denaturation step (1 min), a 48°C annealing step (1 min) and a 72°C elongation step (2 min) in a MJ Research, PTC200 thermocycler. At the end of the run, a final cycle of 94°C (1 min), 50°C (1 min) and 72°C (9 min) was carried out. Finally, samples were maintained at 4°C and stored at -20°C until digestions or cloning could be performed. Five µl aliquots of each PCR product were loaded onto a 1.5% agarose gel in TAE buffer (Sambrook et al. 1989), and electrophoresis was carried out at 7 volts cm µ1. Amplified products were visualized under UV using ethidium bromide. The sizes of the resulting DNA fragments were estimated by comparison with commercial size markers (100 bp and 1 kbp DNA ladders, Life Technologies, USA). Bands were excised from gels and DNA was extracted using a Wizard PCR purification kit (Promega, USA). Purified PCR products were ligated into the pGEM5-T cloning vector (Promega, USA) as described by Salazar et al. 2002. DNA sequencing of the inserts in pGEM5-T was performed by the Departamento de Ciencias Ecológicas, Pontificia Universidad Católica de Chile. Sequencing data for the COII sequence could not be obtained for R. penela because of technical difficulties.

Distribution of the species belonging to either side of the Andes was based on records as appeared in Foote (1981). Highest mountain peaks describe an imaginary line dividing the Andes in two sides to which species can be assigned to (Fig. 1). A special case involved the specie R . psalida and R. rhytida of the psalida group, with distribution in the higher Altiplano, but no records and the east of Andes.

Morphological data. A total of 19 characteristics based on adult external morphology were obtained from the literature (Table 2). Character states were recorded as absent (0) or present (1), and also as multistate (e.g. 0, 1, 2). Eighteen Rhagoletis species plus the outgroup species C. capitata were included. Main sources of information used here were the studies of Foote (1981), Hernández-Ortiz & Frías (1999) and the database of Carroll et al. (2005).

Phylogenetic analyses of morphologcal data. Phylogeny reconstruction was performed using maximum parsimony as implemented by PAUP (version 4.0b10, Swofford 2002). Hierarchical structure in the data was assessed by partitioning tail permutation test (PTP test; Faith & Cranston 1991) as implemented in PAUP. Parsimonious trees were found through a heuristic search strategy using TBR branch swapping. Branch support was based on calculating bootstrap values (10.000 replicates and with rearrangement limit of 10.000.000 rearrangements per addition sequence) and by estimating Bremer decay values using Autodecay 5.0 (Eriksson 2001).

Phylogenetic analyses of molecular data. COII sequences were aligned with ClustalW (Gibson et al. 1996). Sequences were analyzed as unordered, excluding those which were phylogenetically uninformative. Partitioning tail permutation (PTP; Faith & Cranston 1991) as implemented in PAUP was used to evaluate non-random hierarchical structure in the data (heuristic searches with 1.000 replicates).

A COII phylogenetic hypothesis was generated using neighbour-joining tree (NJ), maximum parsimony (MP) and maximum-likelihood (ML) analysis. These approaches were chosen because best phylogenetic hypotheses are most likely to be reliable when congruent evidences are obtained independently (Barkman et al. 2000). Neighbour-joining analysis (Saitou & Nei 1987) was conducted using the Tajima-Nei distance to account for the nucleotide bias in animal mtDNA (Tajima & Nei 1984), implemented in MEGA3 (Kumar et al. 2004). Maximum parsimony analysis to find equally parsimonious trees was performed using heuristic search with TBR branch swapping with PAUP. For ML analysis, Modeltest version 3.04 (Posada & Crandall 1998) with the Akaike information criterion (AIC) was used to select an appropriate nucleotide substitution model. The best model fitting the data was the GTR+G+ I. Thus, ML was further performed with a heuristic search strategy using TBR branch swapping in PAUP.

Branch support was accomplished by calculating non-parametric bootstrap values. Support for NJ analysis was performed with 10.000 replicates. For MP analysis, 10.000 replicates using branch-swapping by TBR and 10 random addition sequence replicates were performed. For ML, 500 replicates with initial tree were obtained via random addition, also using branch-swapping by TBR with GTR+G+I substitution model. Support for branches in MP analysis was also investigated by extracting Bremer decay values which were computed with Autodecay 5.0 (Eriksson 2001) and the PAUP search parameters: hsearch addseq = random nreps = 100. For all topologies, confidence values lower than 50% are not shown.

A Bayesian analysis of molecular data was carried out to estimate the time of divergence of the Neotropical Rhagoletis by using the software BEAST v1.4.6 (Drummond and Rambaut, 2003). Two types of analyses were preformed: one using a known constant mutation rate (strict clock) and other with an unknown mutation rate (relaxed clock). For the strict clock analysis, the COII mean substitution rate per million years was set to a value of 2.1 following the estimation provided by Brower (1994). Both analyses were preformed either with or without prior information of a previously reported divergence time of a given clade. In the case where dating was included, Yule process of speciation according to the time of divergence of R. pomonella from a Mexican ancestor, which has been estimated at 1.57 million years ago (Feder et al. 2003), was included. Analyses included five independent runs of MCMC for 1.0 × 106 steps sampling every 1000 under the GTR model. Estimations of divergence times and their corresponding errors (lower and upper 95% highest posterior densities) were viewed using the software Tracer v1.1.1 (Drummond & Rambaut 2003). To obtain a maximum clade credibility tree, trees were summarized with the software TreeAnnotator 1.4.6 (http://tree.bio.ed.ac.uk/software/treeannotator/), and the final tree topology was viewed with FigTree v.1.1.2 (http://tree.bio.ed.ac.uk/software/figtree/).

Geographic relationship of the Neotropical Rhagoletis species. To estimate the geographic relationship of Neotropical Rhagoletis species with distribution in South América, a dendrogram based on the similarity of their geographical ranges (a biogeogram) was constructed. Distribution ranges for Rhagoletis species were obtained from the literature, mainly from the studies of Foote (1981) and Hernández-Ortiz and Frías (1999). A map of South América (1:400 km) was divided into squares of 1.5° longitude and latitude, and each square was numbered. This map was overlayed on a map of the distribution of each species, and the number of squares in which each species was present was recorded. Thus, a matrix of Euclidean distances between Rhagoletis species was constructed on the basis of their presence or absence in each square. The dendrogram was constructed using cluster analysis of this matrix using Ward's Method implemented in SPSS statistical software.

Results

Morphological phylogeny. Significant value of PTP test (P < 0.01) suggests the presence of a phylogenetic signal in morphological data, supporting further analysis. All characters were parsimony-informative. The heuristic search for a maximum parsimony solution resulted in only one parsimonious tree (length = 56, HI= 0.518, CI = 0.482, RI = 0.663, excluding uninformative characters; Fig. 2). This tree contained 15 resolved nodes of which many were strongly supported by decay values and in some nodes bootstrap values showed more resolution


Molecular phylogeny. The significant value of PTP test (P < 0.01) indicates that COII sequence data have a cladistic structure. This gene sequence resulted in 687 nucleotides in all Rhagoletis including the outgroup C. capitata. Of the 687 characters obtained for COII sequences, 253 (36.8%) were variable, with 159 of these being phylogenetically informative, and 48 at first or second codon positions. The base composition was not different across taxa (χ2 = 18.7 (d.f.=51), P = 0.99). Fig. 3 shows frequencies of transversions and transitions for COII, indicating a bias for transitions (ti/tv = 3.78).


Neighbour-joining tree (Fig. 4), maximum parsimony (Fig. 5), maximum-likelihood analysis (Fig. 6) and Bayesian analysis using a strict clock (Fig. 7) of COII resulted in rather similar topologies. MP analysis of the aligned data set yielded up to five parsimonious trees (length = 590, HI = 0.459, CI = 0.540, RI = 0.498, excluding uninformative characters; Fig. 5 shows the strict consensus tree). ML analysis resulted in five best trees with a natural log likelihood score of -3596.69. This tree was similar in topology to the maximum parsimony tree, except for the position of R. cerasi and for the node separating the species of the ferruginea group from those of the nova group. Unlike the other analyses, the Bayesian phylogeny placed R. tomatis as the ancestor of nova and psalida groups.





Genetic divergence. The pattern of sequence divergence estimated with Beast is shown in Fig. 7. The times of divergence estimated with strict and relaxed clock were very similar, either with or without calibrating using a known divergence time. For instance, the split between the South American and North American Rhagoletis was dated at 4.333 Mya (with lower and upper 95% highest posterior densities of 3.837 and 4.836 Mya, respectively; estimated as a strict clock without priors). The split between the ferruginea group and the nova and psalida groups was estimated to occur at 3.882 Mya (with lower and upper 95% HPD of 3.417 and 4.376 Mya). Within the ferruginea group, the split between R. ferruginea and R. blanchardi was dated to occur 1.917 Mya (with lower and upper 95% HPD of 1.437 and 2.393 Mya). In the case of the psalida and nova groups, divergence of estimation overlapped to a large extent, with R. tomatis diverging from the other species 1.85 Mya (with lower and upper 95% HPD of 1.467 and 2.22 Mya), followed by the divergence of R. psalida from the rest of the species occurring 1.772 Mya (with lower and upper 95% HPD of 1.41 and 2.129 Mya) and the divergence of R. lycopersella occurring 1.643 My ago (with lower and upper 95% HPD of 1.275 and 2.015 Mya). Within this group, the most recent divergence was dated for the species R. nova and R. conversa, estimated to have occurred only 75,210 year ago (with lower and upper 95% HPD of 2.580 and 178,600 years ago).

Geographic relationship. Cluster analysis with Ward's method identified six main groups of Rhagoletis species according to the degree of similarity of their geographical ranges (Fig. 7). Most of the Rhagoletis species with east Andes distribution were grouped together (top cluster in Fig. 7). These species belonged to a wide range of groups including psalida,striatella,ferruginea and nova groups. Contrastingly, species with distribution to the west of the Andes were joined in another cluster (bottom cluster in Fig.7), clusters that included only species of nova group. Of this last group, only R. willinki has been recorded at the east of the Andes.

Discussion

Molecular data suggest that the Neotropical Rhagoletis formed a single lineage, apart from the North American and the Paleartic Rhagoletis species, although the resolution was poor. For instance, NJ, MP, and ML did not verify bootstrap or Bremer support for separating both groups. A similarly low resolution was found in a previous phylogeny for the North American Rhagoletis with the same sequence, which only included two Neotropical species (Smith & Bush 1997). Using the 16S ribosomal RNA gene, McPheron & Han (1997), also found low resolution for the position of the Neotropical species therein (R. striatella and R. conversa). As mentioned by Smith & Bush (1997), mitochondrial COII data seems insufficient to resolve these distant relationships. Nevertheless, the present analysis, with the inclusion of the other six Neotropical Rhagoletis species, was able to establish relationships between the Neotropical Rhagoletis groups. Bayesian estimation of phylogenetic relationships found better support for the separation of Neotropical Rhagoletis from other relatives (Fig. 7).

The morphological analysis suggests that the species of striatella group are basal to the nova and the psalida species, with no support by the bootstrap and Bremer values (Fig. 2). Differently, NJ, MP, ML, and Bayesian analysis for the molecular data found that R. striatella showed more affinities with the Paleartic Rhagoletis than with species from a southern distribution, although its position within the Paleartic Rhagoletis was also variable depending on the analysis performed. NJ and MP placed R. striatella with R. berberidis and separated them from the other Paleartic Rhagoletis, while ML and Bayesian analysis placed R. striatella at the most basal position relative to the North American and the Paleartic species, although this was not well supported by the bootstrap values. In any case, R. striatella never joined with the Neotropical Rhagoletis. This is consistent with the fact that R. striatella has the most northern distribution within the striatella group, which is also one of the more widely distributed groups in continental América (Hernández-Ortiz & Frías 1999).

Host-affiliation information for this group is quite poor, though it is known that larvae of R. striatella can develop in fruits of Physalis species (Solanaceae). All other species of the striatella group, and also all other Neotropical Rhagoletis species, breed mainly on Solanaceae (Foote 1981). These findings suggest that, despite the host affinities of the striatella group with groups distributing in South America, this group seems to have more affinity for the Paleartic and the North American species. Distribution of R. macquartii of the striatella group in South América may have been the result of an invasion and later speciation of a striatella ancestor that dispersed from Central América to South América. Thus, considering striatella as a separated group from the South American groups, the mitochondrial COII data suggest that the ferruginea,psalida and nova groups form a monophyletic clade. Monophylia of the South American Rhagoletis is very apparent in NJ and MP, but not in ML analysis, with only slight bootstrap support in NJ analysis. Molecular analysis of Carpomyina subtribe performed by Smith et al. (2006) also found that South American Rhagoletis forms a monophyletic group. It is worth noting that in our analysis using a strict clock without setting prior information of a known divergence time for a clade, the split of R. pomonella and R. tabellaria was found to have occurred 1.398 Mya (with lower and upper 95% HPD of 0.9854 and 1.806 Mya, respectively), which is the range where divergence of R. pomonella from its Mexican ancestor was estimated (1.57 Mya) by Feder et al. (2003).

The split between the North American and Neotropical Rhagoletis was dated at about 4.333 Mya. The lower limit of confidence of this estimation is about the time of the emergence of a continuous landbridge between Central and South América, which occurred in the Pliocene about 3.5 Mya (Keigwin 1982). Considering that North American species, including the striatella group, possess more derived morphological features with respect to homologous structure in the Neotropical taxa (Bush 1966, Smith & Bush 1997), diversification of North American Rhagoletis species presumably occurred from a northern colonization/dispersion of a southern taxa throughout the existing land connection. Given that the only Neotropical taxon having an affinity with North American species was R. striatella, it is likely that one extinct dispersing taxon from striatella group was the ancestor of North American groups. Resolving phylogenetic relationships between all striatella group species and their affinities with North and South American species would contrast this hypothesis.

Morphological and molecular data were coincident, placing species of the group ferruginea separating from other Neotropical Rhagoletis (Fig 2, 4, 5, 6 and 7). Biogeogram also showed a geographical association between ferruginea group by placing together R. blanchardi and R. ferruginea separated from species with distributions to the west of the Andes (Fig. 8). Since the last uplift of the Andes occurred in the Plio-Pleistocene period (about 5 to 1.8 million Mya), and the split between ferruginea groups and other Rhagoletis groups occurred 3.882 Mya, it seems reasonable that this was due to a vicariant process elicited by the Andes uplift, separating species between those currently distributed to the east and west of the Andes. In the case of those species with distributions to the west side of the Andes, molecular data (NJ, MP and ML analysis) placed R. psalida in a basal position in relation to the nova species, which suggest that nova species originated from a psalida member. Smith et al. (2006) also found similar relationships when studying Carpomyina subtribe, demonstrating that analysis of DNA sequences of alleles at an anonymous nuclear locus indicated that R. psalida was also a sister to R. conversa and R. nova, with R. lycopersella being the sister taxon to this clade. On the other hand, since psalida species distribution currently overlaps with R. lycopersella (biogeogram in Fig. 8), which is the species of the nova group with a more northern distribution, it is likely that a primitive psalida member gave rise to R. lycopersella by sympatric speciation. Thereafter, a population of R. lycopersella may have experienced southern expansion, evolving into R. tomatis.


Bayesian analysis produced a different picture (Fig. 7). In this last case, R. tomatis is an ancestor of R. psalida, which is an ancestor of R. lycopersella, which in turn may have originated from R. nova and R. conversa. In addition, the climatic and biogeographic changes occurring as a consequence of the loss of continuity of subtropical vegetation in South América (Villagrán & Hinojosa 1997) may have caused the isolation of southern populations of R. tomatis or R. lycopersella, which subsequently may have given rise to the rest of the nova species. The present overlapping distribution of R. nova and R. conversa (Fig. 8), and the close phylogenetic affinity found herein, support the notion that these species have undergone a recent divergence. Because different host plants have been described for these species, namely Solanum muricatum (Aiton) for R. nova, and Physalis sp., Solanum tomatillo and Solanum nigrum for R. conversa (Foote 1981; Frías 1992, 2001; Salazar et al. 2002), a process of sympatric speciation by host shift could have promoted this divergence.

The lack of COII sequence data for R. penela and R. willinki specimens did not allow us to establish their phylogenetic relationship with other nova species. However, morphological and biogeographic data clearly show that R. penela and R. willinki are closely related. Frías (1992) suggested that R. willinki originated by dispersal of a population of R.conversa, a process that could have been facilitated by the low altitude of Andes at the latitude of Neuquen, where R. willinki are distributed. The present study confirms the close association between these species.

Our results support the notion that the Andes have played an important role during Neotropical Rhagoletis diversification. A vicariant process, promoted by the Andes uplift during the Pleistocene, was presumably followed by a succession of dispersal and isolation events during the Holocene that affected the divergence pattern on both sides of the Andes.

Acknowledgments

We gratefully acknowledge Dr. J.J Smith for facilitating the manuscript on Carpomyina subtribe. We also acknowledge funding from Gobierno Regional VI-VII and Regional Programs of the Chilean National Science and Technology Foundation (CONICYT), FNDR VII Región, Universidad de Talca through the DIAT-Biology Program and Anillo-PBCT ACT38.

Received 06/VIII/07. Accepted 01/VII/08.

  • Avise, J.C. 2000. Phylogeography: The history and formation of species. Harvard University Press, Harvard, 447p.
  • Barkman, T.J., G. Chenery, J.R. McNeal, J. Lyons-Weiler, W.J. Ellisens, G. Moore, A.D. Wolfe & C.W. dePamphilis. 2000. Independent and combined analyses of sequences from all three genomic compartments converge on the root of flowering plant phylogeny. Proc. Natl. Acad. Sci. USA. 97: 13166-13171.
  • Bush, G.L. 1966. The taxonomy, cytology, and evolution of the genus Rhagoletis in North America (Diptera, Tephritidae). Bull. Mus. Comp. Zool. 134: 431-562.
  • Brower, A.V.Z. 1994. Rapid morphological radiation and convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proc. Natl. Acad. Sci. USA 91: 6491-6495.
  • Carroll, L.E., A.L. Norrbom, M.J. Dallwitz & F.C. Thompson. 2005. 2004 onwards. Pest fruit flies of the world larvae. Version: 13th April 2005. ftp://delta-intkey.com/ffl/index.htm
  • Drummond, A.J. & A. Rambaut. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 7: 214.
  • Eriksson, T. 2001. AutoDecay ver. 5.0 (program distributed by the author). Bergius Foundation, Royal Swedish Academy of Sciences, Stockholm.
  • Faith, D.P. & P.S. Cranston. 1991. Could a cladogram this short have arisen by chance alone? On permutation tests for cladistic structure. Cladistics 7:1-28.
  • Feder, J.L., S.H. Berlocher, J.B. Roethele, H. Dambroski, J.J. Smith, W.L. Perry, V. Gavrilovic, K.E. Filchak, J. Rull & M. Aluja. 2003. Allopatric genetic origins for sympatric host-plant shifts and race formation in Rhagoletis Proc. Natl. Acad. Sci. USA 100: 10314-10319.
  • Foote, R.H. 1981. The genus Rhagoletis Loew South of the United States (Diptera: Tephritidae). U S Dep. Agric. Tech. Bull. 1607.
  • Frías, D.A. 1992. Genética, ecología y evolución de las especies chilenas del género Rhagoletis (Diptera: Tephritidae). Acta Entomol. Chil. 17: 211-223.
  • Frias, D.A. 2001. Genetics and morphological differences of immature stages of two races of Rhagoletis conversa (Brethes) (Diptera: Tephritidae) associated to Solanum plants: Geographic distribution and possible sympatric origin of a new species. Rev. Chil. Hist. Nat. 74: 73-90.
  • Gasparich, G.E., W.S. Sheppard, H.-Y. Han, B.A. McPheron & G.J. Steck. 1995. Analysis of mitochondrial DNA and development of PCR-based diagnostic molecular markers for Mediterranean fruit fly (Ceratitis capitata) populations. Insect Mol. Biol. 4: 61-67.
  • Gibson, T., D. Higgins & J. Thompson. 1996. ClustalW. EMBL, Heidelberg, Germany and EMBL/EBI, Hinxton, UK.
  • Han, H.-Y. & B.A. McPheron. 1997. Molecular phylogenetic study of Tephritidae (Insecta: Diptera) using partial sequences of the mitochondrial 16S ribosomal DNA. Mol. Phylogenet. Evol. 7: 17-32.
  • Hernández-Ortiz, V. & D. Frías. 1999. A revision of the striatella species group of the genus Rhagoletis (Diptera: Tephritidae). Insecta Mundi 13: 11-20.
  • Keigwin, L. 1982. Isotopic paleoceanography of the Caribbean and East Pacific: Role of Panama uplift in late Neogene time. Science 217: 350-353.
  • Kumar, S., K. Tamura, & M. Nei. 2004. MEGA3: Integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief. Bioinform. 5: 150-163.
  • McPheron, B.A. & H.-Y. Han. 1997. Phylogenetic analysis of North American Rhagoletis (Diptera: Tephritidae) and related genera using mitochondrial DNA sequence data. Mol. Phylogenet. Evol. 7: 1-16.
  • Posada, D. & K.A. Candall. 1998. MODELTEST: Testing the model of DNA substitution. Bioinformatics 14: 817-818.
  • Saitou, N. & M. Nei. 1987. The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol. Phylogenet. Evol. 4: 406-425.
  • Salazar, M., C. Theoduloz, A. Vega, F. Poblete, E. González, R. Badilla & L. Meza-Basso. 2002. PCR-RFLP identification of endemic Chilean species of Rhagoletis (Diptera: Tephritidae) attacking Solanaceae. Bull. Entomol. Res. 92: 337-41.
  • Sambrook, J., E.F. Fritsch & T. Maniatis. 1989. Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1626p.
  • Smith, J.J. & G.L. Bush. 1997. Phylogeny of the genus Rhagoletis (Diptera: Tephritidae) inferred from DNA sequences of mitochondrial cytochrome oxidase II. Mol. Phylogenet. Evol. 7: 33-43.
  • Smith, J.J., M.A. Jaycox, M. Smith, & G.L. Bush. 2006 Analysis of mitochondrial DNA and morphological characters in the subtribe Carpomyina (Diptera: Tephritidae). Israel J. Entomol. 35-36: 317-340.
  • Swofford, D.L. 1998. PAUP*, Phylogenetic analysis using parsimony (*and other methods), Version 4.Sinauer Associates, Sunderland, MA.
  • Tajima, F. & M. Nei. 1984. Estimation of evolutionary distance between nucleotide sequences. Mol. Biol. Evol. 1: 269-285.
  • Villagrán, C. & L.F. Hinojosa. 1997. Historia de los bosques del sur de Sudamérica, II: Análisis fitogeográfico. Rev. Chi. Hist. Nat. 70: 241-267.

Publication Dates

  • Publication in this collection
    16 Jan 2009
  • Date of issue
    Dec 2008

History

  • Received
    06 Aug 2007
  • Accepted
    01 July 2008
Sociedade Entomológica do Brasil Sociedade Entomológica do Brasil, R. Harry Prochet, 55, 86047-040 Londrina PR Brasil, Tel.: (55 43) 3342 3987 - Londrina - PR - Brazil
E-mail: editor@seb.org.br