Phylogenetic relationships in the Drosophila fasciola species subgroup (Diptera, Drosophilidae) inferred from partial sequences of the mitoc
http://www.100md.com
《遗传学和分子生物学》
IDepartamento de Genetica, Faculdade de Medicina de Ribeiro Preto, Universidade de So Paulo, Ribeiro Preto, So Paulo, Brazil
IIDepartamento de Biologia, Faculdade de Filosofia Ciências e Letras de Ribeiro Preto, Universidade de So Paulo, Ribeiro Preto, So Paulo, Brazil
ABSTRACT
The repleta group is one of the largest and most complex species groups in the genus Drosophila, representing an important biological model for the elaboration of evolutionary and biogeographical hypotheses on the American Continent. It is divided into six subgroups, of which the fasciola subgroup is the only one with most of its species found in the humid forests of Central and South America. With the aim of understanding the origin and evolution of the fasciola subgroup, and consequently adding information about the repleta group, we analyzed partial sequences of the mitochondrial cytochrome oxidase subunit I (COI) gene of nine Drosophila species (D. carolinae, D. coroica, D. ellisoni, D. fascioloides, D. ivai, D. moju, D. onca, D. rosinae and D. senei). The phylogenetic hypotheses obtained neither defined the relationships between the subgroups of the repleta group nor defined all the interspecific relationships in the fasciola subgroup. We found that the species D. carolinae, D. coroica, D. onca, D. rosinae and D. senei comprised a clade in which D. carolinae, D. onca and D. senei appeared together at a polytomy while D. fascioloides and D. ellisoni comprised another clade with a high bootstrap value.
Key words: fasciola subgroup, molecular phylogeny, mtDNA, systematics, repleta group.
The Drosophila repleta species group, endemic to the American Continent, is one of the largest and most complex species groups in the genus Drosophila, because of which it represents an excellent biological model for speciation studies. This species group contains about one hundred described species (Vilela, 1983; Rafael and Arcos, 1989; Vilela and Bchli, 1990; Tidon-Sklorz and Sene, 1995a, 2001; Etges et al., 2001; Bchli and Vilela, 2002) divided into the D. fasciola, D. inca, D. hydei, D. mercatorum, D. mulleri and D. repleta subgroups based on chromosome inversions and morphological characters (Wasserman, 1960, 1982, 1992; Vilela, 1983; Rafael and Arcos, 1989). The monophyly of the repleta group is defined on morphological (Throckmorton, 1975, 1982; Vilela, 1983), cytological (Wasserman, 1992) and molecular data (Tatarenkov and Ayala, 2001) but Durando et al. (2000) have suggested that the repleta group is not monophyletic, stating that '...this inference [of monophyly] may be the result of poor resolution at the base of the phylogenetic tree'.
The repleta group is widely distributed throughout the American Continent where it is mostly found in semiarid regions with open vegetation (Vilela, 1983; Vilela et al., 1983; Tidon-Sklorz and Sene, 1995b). The Drosophila species in the hydei, mercatorum and repleta subgroups are mostly generalists, while those in the inca and mulleri subgroups use cacti as breeding sites (Pereira et al., 1983; Rafael and Arcos, 1989).
The fasciola subgroup comprises 21 species (Vilela, 1983; Vilela and Bchli, 1990; Bchli and Vilela, 2002), of which 13 species (D. carolinae, D. coroica, D. ellisoni, D. fascioloides, D. ivai, D. mapiriensis, D. moju, D. mojuoides, D. onca, D. papei, D. querubimae, D. rosinae and D. senei) have been recorded in Brazil. Regarding the geographical distribution of the fasciola subgroup over the American Continent, five species (D. ellisoni, D. fasciola, D. moju, D. mojuoides and D. pictura) are found in South and Central America, ten species (D. carolinae, D. coroica, D. fascioloides, D. ivai, D. mapiriensis, D. onca, D. papei, D. querubimae, D. rosinae and D. senei) in South America, four species (D. linearepleta, D. paraguttata, D. pictilis and D. prorepleta) in Central America and two (D. fulvalineata and D. hermionae) in North America (Vilela, 1983; Vilela and Bchli, 1990; Bchli and Vilela, 2002).
The species of the fasciola subgroup inhabit mostly humid forests but little is known about the substrates with which they are associated except for the following: D. moju is known to feed on fallen blossoms and drier fruits (Pipkin, 1965); D. fasciola feeds and breeds on the fallen flowers of Aphelandra micans (Acanthaceae), Erythrina berteroana (Fabaceae), Heliconia latispatha (Heliconiaceae), Aroid sp. (Araceae) and Coffea arabica fruits (Rubiaceae) (Pipkin, 1965; Pipkin et al., 1966); D. fulvalineata was collected on fleshy fungi (Patterson and Wheeler, 1942); D. onca breeds in cladodes of decaying epiphytic cacti (Rhipsalis sp.) (Sene et al., 1977); D. rosinae emerged from cladodes of decaying columnar cacti (Cereus sp.) (Tidon-Sklorz and Sene, 1995b). The last two examples suggest that species of the fasciola subgroup, even though inhabiting forests, may be associated with cacti. It has already been noted that D. rosinae associates with cacti of the genus Cereus and Opuntia (Vilela, 1983; Tidon-Sklorz and Sene, 1995b) and our work described in the present paper concerns D. rosinae specimens collected from dunes and rocky shores where Cereus and Opuntia were growing.
Most of the studies regarding the establishment of a phylogeny for the fasciola subgroup have been cytological (Wasserman, 1962, 1982, 1992; Kuhn et al., 1995; Diniz and Sene, 2004). The existence of the 3c inversion in both the mulleri complex of the mulleri subgroup and in species of the fasciola subgroup supports the hypothesis of a common ancestry, with the fasciola subgroup stemming from the mulleri subgroup (Wasserman, 1982, 1992). However, Diniz and Sene (2004) recently showed that the fixed 3c inversion of the mulleri complex is not the same as the one observed in the fasciola subgroup, suggesting that the name of the inversion in the mulleri complex should be changed from 3c to 3u. This suggests that the ancestral chromosomal composition of the fasciola subgroup (Primitive VII) derived from the Primitive I, the ancestral sequence of the repleta group (Wasserman, 1960; Wasserman, 1992) (Figure 1). Furthermore, Diniz and Sene (2004) proposed two new species complexes within the fasciola subgroup, the fascioloides complex made up of D. ellisoni and D. fascioloides that share the 2d3 inversion, two chromosomal centric fusions (2-4F and 3-5F) and karyotype similarity regarding the X chromosomes (Kuhn et al., 1995) (Figure 1) and the onca complex, containing D. carolinae and D. onca which both share the 4f2 inversion (Figure 1). Previously, chromosomal data had suggested two complexes, the pictilis complex (D. pictilis and D. pictura) and the moju complex (D. moju, D. mojuoides and D. paraguttata) (Wasserman, 1992) (Figure 1).
There are two hypotheses about the origin and diversification of the repleta species group. One hypothesis suggests that the ancestral populations of the fasciola subgroup were forest inhabitants that colonized the dry areas and led to the development of the species of the other repleta group subgroups (Throckmorton, 1975, 1982; Morais et al., 1995; Tatarenkov and Ayala, 2001), while the other hypothesis considers that the fasciola subgroup is derived from species adapted to dry areas which later invaded forests (Wasserman, 1962, 1992) but this hypothesis was based on the sharing of the 3c inversion among species in the fasciola subgroup and the mulleri complex.
Many questions concerning the evolutionary relationships among the species in the fasciola subgroup remain obscure. To help resolve this we analyzed partial sequences of the mitochondrial cytochrome oxidase subunit I (COI) gene of nine species of the fasciola subgroup (Table 1) with the aim of trying to understand the origin and evolution of the this subgroup and consequently adding to the current knowledge on the repleta group.
The species analyzed were collected using closed traps (Tidon-Sklorz and Sene, 1988) containing banana, orange and yeast. Flies from the fasciola subgroup were identified by morphological analysis of the aedeagus (male genitalia) (Vilela, 1983), males being directly identified and females indirectly by their male offspring.
We used D. melanogaster (NC_001709) and D. yakuba (NC_001322) as the outgroup in the phylogenetic analysis and D. hydei (hydei subgroup), D. mercatorum (mercatorum subgroup), D. repleta (repleta subgroup) and D. serido (mulleri subgroup) were used to provide information on the relationships between the fasciola subgroup and the other subgroups of the repleta group.
Genomic DNA was extracted from a single male of each species using the Wizard Genomic DNA Purification kit (Promega) and the polymerase chain reaction (PCR) was used to amplify COI gene fragments using the primers TY-J-1460 - 5'TACAATTTATCGCCTAACTTC AGCC3' (Simon et al., 1994) and 2191 - 5'TTGATTTTTT GGTCATCCAGAAGT3' (De Brito et al., 1998). The following amplifications conditions were used: 94 °C for 1 min 30 s followed by 25 cycles of 40 s at 94 °C, 40 s at 46 °C and 2 min at 72 °C. The PCR products were purified with the ExoSAP-IT kit (Amersham Biosciences Part of GE Healthcare) and sequenced directly. Sequencing reactions were prepared with the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems) and reaction products sequenced on an ABI Prism 377 Automated DNA Sequencer.
Forward and reverse sequences of each taxa were compared, corrected and edited using the Chromas 2.3 program (http://www.technelysium.com.au/chromas_lite.html) and multiple sequence alignments were generated using the ClustalW 1.8 program (Thompson et al., 1994). The presence of saturation in the data was checked using Dambe 4.2.13 program (Xia and Xie, 2001).
Phylogenetic reconstruction was performed in a maximum parsimony (MP) framework using the PAUP4.0b10 program (Swofford, 2002) with the following settings: search type = heuristic; addition sequences = random; number of replicates = 100; and branch swapping = Tree Bisection Reconnection (TBR). Support at each node was assessed using bootstrap proportions (Felsenstein, 1985) with 1,000 replicates.
Reyes et al. (1999) and Bull et al. (2003) have suggested likely saturation problems for third mitochondrial codon positions and explored the use of a priori differential weighting of molecular characters. Bull et al. (2003) obtained the best resolution of the maximum parsimony analyses when third mtDNA codon positions were weighted by 0.5 and 0.1 with transitions excluded entirely. Based on this we used two weighting methods, one based on equal weighting of all informative characters and another omitting the transitions (zero weight) and assigning weights of 0.5 for the third codon positions with all other characters weighted equally.
Partial 624 nucleotide sequences from the COI gene were obtained from all the species studied except D. repleta. The first base of the sequences obtained corresponded to position 1,510 of the mitochondrial DNA of D. yakuba (NC_001322), the first codon from this base corresponding to the thirteenth amino acid of the COI gene. No gaps were found but there were 193 variable characters (including the outgroup species) and 137 parsimony informative characters. The third codon positions were the most variable, whereas the second codon positions were the most conserved in this COI gene region. There was an excess of T (38.9%) and A (28.9%) over G (16.7%) and C (15.5%). The saturation test (Figure 2) suggested likely saturation for transitions.
Phylogenetic analysis with equal weighting of all informative characters resulted in a single parsimony tree with consistency index (CI) of 0.450, retention index (RI) of 0.308 and a total length of 547 steps (Figure 3A). This parsimony tree contained a poorly resolved clade (repleta group). This polytomy can be explained by the reduced amount of phylogenetic information contained in the data due to the high level of saturation suggested for transitions (Figure 2) and/or for third codon positions. Multilocus sequence analysis or the inclusion of more fasciola subgroup taxa could provide enough information for a robust hypotheses. But inclusion of taxa could prove difficult since the strains might not be readily available.
On the basis of the likely saturation for transitions (Figure 2) and for third mtDNA codon positions (Allard and Carpenter, 1996; Reyes et al., 1999; Bull et al., 2003) we carried out a second maximum parsimony analysis with all transitions excluded (133 parsimony informative characters) and the third codons given a 0.5 weighting. This analysis resulted in a single parsimony tree with CI of 0.411, RI of 0.404 and a length of 137.5 steps (Figure 3B) with better resolution than the cladogram obtained with equal weighting of the characters (Figure 3A).
The cladograms obtained support monophyly of the repleta group (Figure 3) when tested with D. melanogaster and D. yakuba (not a strong test), which is in agreement with other studies in the literature (Throckmorton, 1975, 1982; Wasserman, 1992; Tatarenkov and Ayala, 2001). On the other hand, Durando et al. (2000) suggest that the repleta group is not monophyletic, although these authors cautioned that this inference '...may be the result of poor resolution at the base of the phylogenetic tree'.
The repleta and mercatorum subgroups appeared as sister taxa only in the maximum parsimony analysis with equal character weighting, but support for this clade was low with a bootstrap value of only 58% (Figure 3A). This relationship has previously been suggested by morphological (Throckmorton, 1975), cytological (Wasserman, 1992) (Figure 1) and molecular data (Durando et al., 2000; Tatarenkov and Ayala, 2001). The relationships among the other subgroups in the repleta group were not resolved by our analysis, supporting previous work by Throckmorton (1975), Wasserman (1992) and Durando et al. (2000).
The phylogeny with equal character weighting (Figure 3A) did not suggest monophyly of the fasciola subgroup. We used the nonparametric Templeton test (Templeton, 1983) function of the PAUP program to test if this departure from the conventional notion of the fasciola subgroup relationships was significantly supported by the COI gene data. This analysis showed significant differences (p = 0.0001) between the constrained topology monophyly of the fasciola subgroup (522 steps) and the unconstrained topology (547 steps) (Figure 3A), with the constrained topology explaining the data significantly better than the unconstrained topology. Therefore, the Templeton test showed that the data do not reject monophyly of the fasciola subgroup. The phylogeny obtained using differential weightings (Figure 3B) suggested monophyly of the fasciola subgroup, but again, support for this clade was low with a 53% bootstrap and the Templeton test indicated no significant differences (p = 0.6844) between the topology of the Figure 3B (137.5 steps) and the constrained topology of monophyly of the fasciola subgroup (136.5 steps). Thus, our COI gene data neither rejected nor supported the concept of monophyly of the fasciola species subgroup.
The phylogenies did not define all the interspecific relationships in the fasciola subgroup. In both of the hypotheses described above D. ellisoni and D. fascioloides appear as sister species (Figure 3), which is in agreement with the cytogenetic data produced by Diniz and Sene (2004) (Figure 1). In the differentially weighted phylogeny (Figure 3B) the species D. carolinae, D. coroica, D. onca, D. rosinae and D. senei formed a single clade with a 63% bootstrap in which D. carolinae, D. onca and D. senei formed a nested polytomy with a 98% bootstrap. However, the equally weighted cladogram (Figure 3A) indicated that D. onca, D. carolinae and D. senei comprised one clade in which D. carolinae and D. senei appeared as sister species, contradicting the morphological data suggesting that D. carolinae and D. onca are sister taxa (Vilela, 1983) but which shows no relationship between D. carolinae and D. senei. The relationships between the other species were not defined in the equally weighted cladogram. Our results are in agreement with theories on the fascioloides and onca complexes proposed by Diniz and Sene (2004). It is possible that D. senei also belongs to the onca complex, but additional cytological data are necessary to confirm this hypothesis.
Because the relationships between the subgroups were not defined our results show no evidence that the ancestral populations of the repleta group inhabited forests (Morais et al., 1995; Throckmorton, 1975, 1982; Tatarenkov and Ayala; 2001) or arid areas (Wasserman, 1962, 1992).
However, the hypothesis that the fasciola subgroup represents the oldest lineage of the repleta group is supported by data produced by Throckmorton (1982) suggesting that the species groups of the virilis-repleta radiation, closely related to the repleta group (castanea, canalinea, dreyfusi and mesophragmatica groups), occur in forests and that, morphologically, the fasciola subgroup is the least derived repleta group subgroup. Furthermore, Morais et al. (1995) conducted ecological studies of Drosophila yeast communities associated to species of the fasciola subgroup and produced data suggesting that this subgroup represents the lineage from which the South American species of the repleta group evolved. More recently, Tatarenkov and Ayala (2001) used the nuclear genes Ddc and amd to construct a molecular phylogeny of the virilis-repleta radiation which strongly indicates that the fasciola subgroup is the earliest offshoot in the repleta group. It thus seems that, as with the fasciola subgroup, the ancestral repleta group probably inhabited forests.
While supporting monophyly of the repleta group, our data clearly does not resolve the phylogenetic relationships in the fasciola subgroup, possibly due to saturation in relation to the mitochondrial gene used to construct the phylogeny (Allard and Carpenter, 1996; Reyes et al., 1999; Durando et al., 2000; Bull et al., 2003). Research using nuclear genes could help elucidate the phylogenetic relationships in the fasciola subgroup and between this and the other subgroups of the repleta group.
Acknowledgments
We thank Dr. Reinaldo O.A. de Brito for providing representative hydei, mercatorum, mulleri and repleta subgroups COI gene sequences and also the Universidade de So Paulo (USP), Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq), Coordenao de Pessoal de Nivel Superior (CAPES) and the Fundao de Amparo a Pesquisa do Estado de So Paulo (FAPESP) for financial support and a personal fellowship from Fundao de Amparo a Pesquisa do Estado de So Paulo (FAPESP) proc. 02/02195-9 for E.C.C. Silva-Bernardi. We also thank two anonymous referees for their helpful comments.
REFERENCES
Allard MW and Carpenter JM (1996) On weighting and congruence. Cladistics 12:183-198.
Bchli G and Vilela CR (2002) Six new species of Drosophila (Diptera, Drosophilidae) from Parque Nacional Henri Pittier, Venezuela. Mitt Schweiz Ent Ges 75:223-243.
Bull NJ, Schwarz MP and Cooper SJB (2003) Phylogenetic divergence of the Australian allodapine bees (Hymenoptera, Apidae). Mol Phylogenet Evol 27:212-222.
De Brito RA, Manfrin MH and Sene FM (1998) Use of nuclear and mitochondrial data to investigate evolution and phylogenetic relationships of the repleta group (genus Drosophila) radiation. Genet Mol Biol 21(suppl):258.
Diniz NM and Sene FM (2004) Chromosomal phylogeny of the Drosophila fasciola species subgroup revisited (Diptera, Drosophilidae). Genet Mol Biol 27:561-566.
Durando CM, Baker RH, Etges WJ, Heed WB, Wasserman M and DeSalle R (2000) Phylogenetic analysis of the repleta species group of the genus Drosophila using multiple sources of characters. Mol Phylogenet Evol 16:296-307.
Etges WJ, Armella MA, O'Grady PM and Heed WB (2001) Two new species of Drosophila (Diptera, Drosophilidae) in the repleta group from Mexico. Ann Entomol Soc Am 94:16-20.
Felsenstein J (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39:783-791.
Kuhn GCS, Diniz NM, Alves MAR and Sene FM (1995) Constituio cariotipica de especies do subgrupo fasciola do grupo repleta do gênero Drosophila. Rev Bras Genet 18(suppl):282.
Morais PB, Rosa CA, Hagler AN and Mendona-Hacler LC (1995) Yeast communities as descriptors of habitat use by the Drosophila fasciola subgroup (repleta group) in Atlantic rain forests. Oecologia 104:45-51.
Patterson JT and Wheeler MR (1942) Description of new species of the subgenera Hirtodrosophila and Drosophila. Univ Texas Publ 4213:67-109.
Pereira MAQR, Vilela CR and Sene FM (1983) Notes on breeding and feeding sites of some species of the repleta group of the genus Drosophila (Diptera, Drosophilidae). Cienc Cult 35:1313-1319.
Pipkin SB (1965) The influence of adult and larval food habits on population size of neotropical ground-feeding Drosophila. Am Midl Nat 74:1-27.
Pipkin SB, Rodriguez RL and Leon J (1966) Plant host specificity among flower-feeding Neotropical Drosophila (Diptera, Drosophilidae). Am Nat 100:135-156.
Rafael V and Arcos G (1989) Subgrupo inca, un nuevo subgrupo del grupo repleta, con descripcion de Drosophila huancavilcae N.SP. (Diptera, Drosophilidae). Evol Biol 3:233-243.
Reyes SG, Cooper SJB and Schwarz MP (1999) Species phylogeny of the bee genus Exoneurella Michener (Hymenoptera, Apidae, Allodapini): Evidence from molecular and morphological data sets. Ann Entomol Soc Am 92:20-29.
Sene FM, Paganelli CHM, Pedroso LG, Garcia E and Palombo CR (1977) Local natural de criao de Drosophila onca Dobzhansky & Pavan, 1943. Cienc Cult 29(suppl):716.
Simon C, Frati F, Beckenbach A, Crespi B, Liu H and Flook P (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann Entomol Soc Am 87:651-701.
Swofford DL (2002) PAUP. Phylogenetic Analysis Using Parsimony (and Other Methods), Version 4. Sinauer Associates, Sunderland, Massachusetts.
Tatarenkov A and Ayala FJ (2001) Phylogenetic relationships among species groups of the virilis-repleta radiation of Drosophila. Mol Phylogenet Evol 21:327-331.
Templeton AR (1983) Phylogenetic inference from restriction endonuclease cleavage site maps with particular reference to the evolution of humans and the apes. Evolution 37:221-244.
Thompson JD, Higgins DG and Gibson TJ (1994) CLUSTALW: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673-4680.
Throckmorton LH (1975) The phylogeny, ecology and geography of Drosophila. In: King RC (ed) Handbook of Genetics. Plenum Press, New York, pp 421-459.
Throckmorton LH (1982) Pathways of evolution in the genus Drosophila and the founding of the repleta group. In: Barker JSF and Starmer WT (eds) Ecological Genetics and Evolution: The Cactus-Yeast-Drosophila Model System. Academic Press, New York, pp 33-48.
Tidon-Sklorz R and Sene FM (1988) A trap that retains and keeps Drosophila alive. Drosoph Inf Serv 67:89.
Tidon-Sklorz R and Sene FM (1995a) Drosophila seriema n. sp.: New member of the Drosophila serido (Diptera, Drosophilidae) superspecies taxon. Ann Entomol Soc Am 88:139-142.
Tidon-Sklorz R and Sene FM (1995b) Fauna of Drosophila (Diptera, Drosophilidae) in the northern area of the "Cadeia do Espinhao", states of Minas Gerais and Bahia, Brazil: Biogeographical and ecological aspects. Iheringia, Ser Zool 78:85-94.
Tidon-Sklorz R and Sene FM (2001) Two new species of the Drosophila serido sibling set (Diptera, Drosophilidae). Iheringia, Ser Zool 90:141-146.
Vilela CR (1983) A revision of the Drosophila repleta species group (Diptera, Drosophilidae). Rev Bras Entomol 27:1-114.
Vilela CR and Bchli G (1990) Taxonomic studies on Neotropical species of seven genera of Drosophilidae (Diptera). Mitt Schweiz Ent Ges 63(suppl):1-332.
Vilela CR, Pereira MAQR and Sene FM (1983) Preliminary data on the geographical distribution of Drosophila species within morphoclimatic domains of Brazil: II. The repleta group. Cienc Cult 35:66-70.
Wasseman M (1960) Cytological and phylogenetic relationships in the repleta group of the genus Drosophila. Proc Nat Acad Sci 46:842-859.
Wasserman M (1962) Cytological studies of the repleta group of the genus Drosophila: VI. The fasciola subgroup. Univ Texas Publ 6205:119-134.
Wasserman M (1982) Evolution of the repleta group. In: Ashburner M, Carson HL and Thompson JN (eds) The Genetics and Biology of Drosophila. v. 3b. Academic Press, London, pp 61-139.
Wasserman M (1992) Cytological evolution of the Drosophila repleta species group. In: Krimbas CB and Powell JR (eds) Drosophila Inversion Polymorphism. CRC Press, Boca Raton, pp 455-555.
Xia X and Xie Z (2001) DAMBE: Data analysis in molecular biology and evolution. Journal of Heredity 92:371-373.(Erica C.C. Silva-Bernardi)
IIDepartamento de Biologia, Faculdade de Filosofia Ciências e Letras de Ribeiro Preto, Universidade de So Paulo, Ribeiro Preto, So Paulo, Brazil
ABSTRACT
The repleta group is one of the largest and most complex species groups in the genus Drosophila, representing an important biological model for the elaboration of evolutionary and biogeographical hypotheses on the American Continent. It is divided into six subgroups, of which the fasciola subgroup is the only one with most of its species found in the humid forests of Central and South America. With the aim of understanding the origin and evolution of the fasciola subgroup, and consequently adding information about the repleta group, we analyzed partial sequences of the mitochondrial cytochrome oxidase subunit I (COI) gene of nine Drosophila species (D. carolinae, D. coroica, D. ellisoni, D. fascioloides, D. ivai, D. moju, D. onca, D. rosinae and D. senei). The phylogenetic hypotheses obtained neither defined the relationships between the subgroups of the repleta group nor defined all the interspecific relationships in the fasciola subgroup. We found that the species D. carolinae, D. coroica, D. onca, D. rosinae and D. senei comprised a clade in which D. carolinae, D. onca and D. senei appeared together at a polytomy while D. fascioloides and D. ellisoni comprised another clade with a high bootstrap value.
Key words: fasciola subgroup, molecular phylogeny, mtDNA, systematics, repleta group.
The Drosophila repleta species group, endemic to the American Continent, is one of the largest and most complex species groups in the genus Drosophila, because of which it represents an excellent biological model for speciation studies. This species group contains about one hundred described species (Vilela, 1983; Rafael and Arcos, 1989; Vilela and Bchli, 1990; Tidon-Sklorz and Sene, 1995a, 2001; Etges et al., 2001; Bchli and Vilela, 2002) divided into the D. fasciola, D. inca, D. hydei, D. mercatorum, D. mulleri and D. repleta subgroups based on chromosome inversions and morphological characters (Wasserman, 1960, 1982, 1992; Vilela, 1983; Rafael and Arcos, 1989). The monophyly of the repleta group is defined on morphological (Throckmorton, 1975, 1982; Vilela, 1983), cytological (Wasserman, 1992) and molecular data (Tatarenkov and Ayala, 2001) but Durando et al. (2000) have suggested that the repleta group is not monophyletic, stating that '...this inference [of monophyly] may be the result of poor resolution at the base of the phylogenetic tree'.
The repleta group is widely distributed throughout the American Continent where it is mostly found in semiarid regions with open vegetation (Vilela, 1983; Vilela et al., 1983; Tidon-Sklorz and Sene, 1995b). The Drosophila species in the hydei, mercatorum and repleta subgroups are mostly generalists, while those in the inca and mulleri subgroups use cacti as breeding sites (Pereira et al., 1983; Rafael and Arcos, 1989).
The fasciola subgroup comprises 21 species (Vilela, 1983; Vilela and Bchli, 1990; Bchli and Vilela, 2002), of which 13 species (D. carolinae, D. coroica, D. ellisoni, D. fascioloides, D. ivai, D. mapiriensis, D. moju, D. mojuoides, D. onca, D. papei, D. querubimae, D. rosinae and D. senei) have been recorded in Brazil. Regarding the geographical distribution of the fasciola subgroup over the American Continent, five species (D. ellisoni, D. fasciola, D. moju, D. mojuoides and D. pictura) are found in South and Central America, ten species (D. carolinae, D. coroica, D. fascioloides, D. ivai, D. mapiriensis, D. onca, D. papei, D. querubimae, D. rosinae and D. senei) in South America, four species (D. linearepleta, D. paraguttata, D. pictilis and D. prorepleta) in Central America and two (D. fulvalineata and D. hermionae) in North America (Vilela, 1983; Vilela and Bchli, 1990; Bchli and Vilela, 2002).
The species of the fasciola subgroup inhabit mostly humid forests but little is known about the substrates with which they are associated except for the following: D. moju is known to feed on fallen blossoms and drier fruits (Pipkin, 1965); D. fasciola feeds and breeds on the fallen flowers of Aphelandra micans (Acanthaceae), Erythrina berteroana (Fabaceae), Heliconia latispatha (Heliconiaceae), Aroid sp. (Araceae) and Coffea arabica fruits (Rubiaceae) (Pipkin, 1965; Pipkin et al., 1966); D. fulvalineata was collected on fleshy fungi (Patterson and Wheeler, 1942); D. onca breeds in cladodes of decaying epiphytic cacti (Rhipsalis sp.) (Sene et al., 1977); D. rosinae emerged from cladodes of decaying columnar cacti (Cereus sp.) (Tidon-Sklorz and Sene, 1995b). The last two examples suggest that species of the fasciola subgroup, even though inhabiting forests, may be associated with cacti. It has already been noted that D. rosinae associates with cacti of the genus Cereus and Opuntia (Vilela, 1983; Tidon-Sklorz and Sene, 1995b) and our work described in the present paper concerns D. rosinae specimens collected from dunes and rocky shores where Cereus and Opuntia were growing.
Most of the studies regarding the establishment of a phylogeny for the fasciola subgroup have been cytological (Wasserman, 1962, 1982, 1992; Kuhn et al., 1995; Diniz and Sene, 2004). The existence of the 3c inversion in both the mulleri complex of the mulleri subgroup and in species of the fasciola subgroup supports the hypothesis of a common ancestry, with the fasciola subgroup stemming from the mulleri subgroup (Wasserman, 1982, 1992). However, Diniz and Sene (2004) recently showed that the fixed 3c inversion of the mulleri complex is not the same as the one observed in the fasciola subgroup, suggesting that the name of the inversion in the mulleri complex should be changed from 3c to 3u. This suggests that the ancestral chromosomal composition of the fasciola subgroup (Primitive VII) derived from the Primitive I, the ancestral sequence of the repleta group (Wasserman, 1960; Wasserman, 1992) (Figure 1). Furthermore, Diniz and Sene (2004) proposed two new species complexes within the fasciola subgroup, the fascioloides complex made up of D. ellisoni and D. fascioloides that share the 2d3 inversion, two chromosomal centric fusions (2-4F and 3-5F) and karyotype similarity regarding the X chromosomes (Kuhn et al., 1995) (Figure 1) and the onca complex, containing D. carolinae and D. onca which both share the 4f2 inversion (Figure 1). Previously, chromosomal data had suggested two complexes, the pictilis complex (D. pictilis and D. pictura) and the moju complex (D. moju, D. mojuoides and D. paraguttata) (Wasserman, 1992) (Figure 1).
There are two hypotheses about the origin and diversification of the repleta species group. One hypothesis suggests that the ancestral populations of the fasciola subgroup were forest inhabitants that colonized the dry areas and led to the development of the species of the other repleta group subgroups (Throckmorton, 1975, 1982; Morais et al., 1995; Tatarenkov and Ayala, 2001), while the other hypothesis considers that the fasciola subgroup is derived from species adapted to dry areas which later invaded forests (Wasserman, 1962, 1992) but this hypothesis was based on the sharing of the 3c inversion among species in the fasciola subgroup and the mulleri complex.
Many questions concerning the evolutionary relationships among the species in the fasciola subgroup remain obscure. To help resolve this we analyzed partial sequences of the mitochondrial cytochrome oxidase subunit I (COI) gene of nine species of the fasciola subgroup (Table 1) with the aim of trying to understand the origin and evolution of the this subgroup and consequently adding to the current knowledge on the repleta group.
The species analyzed were collected using closed traps (Tidon-Sklorz and Sene, 1988) containing banana, orange and yeast. Flies from the fasciola subgroup were identified by morphological analysis of the aedeagus (male genitalia) (Vilela, 1983), males being directly identified and females indirectly by their male offspring.
We used D. melanogaster (NC_001709) and D. yakuba (NC_001322) as the outgroup in the phylogenetic analysis and D. hydei (hydei subgroup), D. mercatorum (mercatorum subgroup), D. repleta (repleta subgroup) and D. serido (mulleri subgroup) were used to provide information on the relationships between the fasciola subgroup and the other subgroups of the repleta group.
Genomic DNA was extracted from a single male of each species using the Wizard Genomic DNA Purification kit (Promega) and the polymerase chain reaction (PCR) was used to amplify COI gene fragments using the primers TY-J-1460 - 5'TACAATTTATCGCCTAACTTC AGCC3' (Simon et al., 1994) and 2191 - 5'TTGATTTTTT GGTCATCCAGAAGT3' (De Brito et al., 1998). The following amplifications conditions were used: 94 °C for 1 min 30 s followed by 25 cycles of 40 s at 94 °C, 40 s at 46 °C and 2 min at 72 °C. The PCR products were purified with the ExoSAP-IT kit (Amersham Biosciences Part of GE Healthcare) and sequenced directly. Sequencing reactions were prepared with the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems) and reaction products sequenced on an ABI Prism 377 Automated DNA Sequencer.
Forward and reverse sequences of each taxa were compared, corrected and edited using the Chromas 2.3 program (http://www.technelysium.com.au/chromas_lite.html) and multiple sequence alignments were generated using the ClustalW 1.8 program (Thompson et al., 1994). The presence of saturation in the data was checked using Dambe 4.2.13 program (Xia and Xie, 2001).
Phylogenetic reconstruction was performed in a maximum parsimony (MP) framework using the PAUP4.0b10 program (Swofford, 2002) with the following settings: search type = heuristic; addition sequences = random; number of replicates = 100; and branch swapping = Tree Bisection Reconnection (TBR). Support at each node was assessed using bootstrap proportions (Felsenstein, 1985) with 1,000 replicates.
Reyes et al. (1999) and Bull et al. (2003) have suggested likely saturation problems for third mitochondrial codon positions and explored the use of a priori differential weighting of molecular characters. Bull et al. (2003) obtained the best resolution of the maximum parsimony analyses when third mtDNA codon positions were weighted by 0.5 and 0.1 with transitions excluded entirely. Based on this we used two weighting methods, one based on equal weighting of all informative characters and another omitting the transitions (zero weight) and assigning weights of 0.5 for the third codon positions with all other characters weighted equally.
Partial 624 nucleotide sequences from the COI gene were obtained from all the species studied except D. repleta. The first base of the sequences obtained corresponded to position 1,510 of the mitochondrial DNA of D. yakuba (NC_001322), the first codon from this base corresponding to the thirteenth amino acid of the COI gene. No gaps were found but there were 193 variable characters (including the outgroup species) and 137 parsimony informative characters. The third codon positions were the most variable, whereas the second codon positions were the most conserved in this COI gene region. There was an excess of T (38.9%) and A (28.9%) over G (16.7%) and C (15.5%). The saturation test (Figure 2) suggested likely saturation for transitions.
Phylogenetic analysis with equal weighting of all informative characters resulted in a single parsimony tree with consistency index (CI) of 0.450, retention index (RI) of 0.308 and a total length of 547 steps (Figure 3A). This parsimony tree contained a poorly resolved clade (repleta group). This polytomy can be explained by the reduced amount of phylogenetic information contained in the data due to the high level of saturation suggested for transitions (Figure 2) and/or for third codon positions. Multilocus sequence analysis or the inclusion of more fasciola subgroup taxa could provide enough information for a robust hypotheses. But inclusion of taxa could prove difficult since the strains might not be readily available.
On the basis of the likely saturation for transitions (Figure 2) and for third mtDNA codon positions (Allard and Carpenter, 1996; Reyes et al., 1999; Bull et al., 2003) we carried out a second maximum parsimony analysis with all transitions excluded (133 parsimony informative characters) and the third codons given a 0.5 weighting. This analysis resulted in a single parsimony tree with CI of 0.411, RI of 0.404 and a length of 137.5 steps (Figure 3B) with better resolution than the cladogram obtained with equal weighting of the characters (Figure 3A).
The cladograms obtained support monophyly of the repleta group (Figure 3) when tested with D. melanogaster and D. yakuba (not a strong test), which is in agreement with other studies in the literature (Throckmorton, 1975, 1982; Wasserman, 1992; Tatarenkov and Ayala, 2001). On the other hand, Durando et al. (2000) suggest that the repleta group is not monophyletic, although these authors cautioned that this inference '...may be the result of poor resolution at the base of the phylogenetic tree'.
The repleta and mercatorum subgroups appeared as sister taxa only in the maximum parsimony analysis with equal character weighting, but support for this clade was low with a bootstrap value of only 58% (Figure 3A). This relationship has previously been suggested by morphological (Throckmorton, 1975), cytological (Wasserman, 1992) (Figure 1) and molecular data (Durando et al., 2000; Tatarenkov and Ayala, 2001). The relationships among the other subgroups in the repleta group were not resolved by our analysis, supporting previous work by Throckmorton (1975), Wasserman (1992) and Durando et al. (2000).
The phylogeny with equal character weighting (Figure 3A) did not suggest monophyly of the fasciola subgroup. We used the nonparametric Templeton test (Templeton, 1983) function of the PAUP program to test if this departure from the conventional notion of the fasciola subgroup relationships was significantly supported by the COI gene data. This analysis showed significant differences (p = 0.0001) between the constrained topology monophyly of the fasciola subgroup (522 steps) and the unconstrained topology (547 steps) (Figure 3A), with the constrained topology explaining the data significantly better than the unconstrained topology. Therefore, the Templeton test showed that the data do not reject monophyly of the fasciola subgroup. The phylogeny obtained using differential weightings (Figure 3B) suggested monophyly of the fasciola subgroup, but again, support for this clade was low with a 53% bootstrap and the Templeton test indicated no significant differences (p = 0.6844) between the topology of the Figure 3B (137.5 steps) and the constrained topology of monophyly of the fasciola subgroup (136.5 steps). Thus, our COI gene data neither rejected nor supported the concept of monophyly of the fasciola species subgroup.
The phylogenies did not define all the interspecific relationships in the fasciola subgroup. In both of the hypotheses described above D. ellisoni and D. fascioloides appear as sister species (Figure 3), which is in agreement with the cytogenetic data produced by Diniz and Sene (2004) (Figure 1). In the differentially weighted phylogeny (Figure 3B) the species D. carolinae, D. coroica, D. onca, D. rosinae and D. senei formed a single clade with a 63% bootstrap in which D. carolinae, D. onca and D. senei formed a nested polytomy with a 98% bootstrap. However, the equally weighted cladogram (Figure 3A) indicated that D. onca, D. carolinae and D. senei comprised one clade in which D. carolinae and D. senei appeared as sister species, contradicting the morphological data suggesting that D. carolinae and D. onca are sister taxa (Vilela, 1983) but which shows no relationship between D. carolinae and D. senei. The relationships between the other species were not defined in the equally weighted cladogram. Our results are in agreement with theories on the fascioloides and onca complexes proposed by Diniz and Sene (2004). It is possible that D. senei also belongs to the onca complex, but additional cytological data are necessary to confirm this hypothesis.
Because the relationships between the subgroups were not defined our results show no evidence that the ancestral populations of the repleta group inhabited forests (Morais et al., 1995; Throckmorton, 1975, 1982; Tatarenkov and Ayala; 2001) or arid areas (Wasserman, 1962, 1992).
However, the hypothesis that the fasciola subgroup represents the oldest lineage of the repleta group is supported by data produced by Throckmorton (1982) suggesting that the species groups of the virilis-repleta radiation, closely related to the repleta group (castanea, canalinea, dreyfusi and mesophragmatica groups), occur in forests and that, morphologically, the fasciola subgroup is the least derived repleta group subgroup. Furthermore, Morais et al. (1995) conducted ecological studies of Drosophila yeast communities associated to species of the fasciola subgroup and produced data suggesting that this subgroup represents the lineage from which the South American species of the repleta group evolved. More recently, Tatarenkov and Ayala (2001) used the nuclear genes Ddc and amd to construct a molecular phylogeny of the virilis-repleta radiation which strongly indicates that the fasciola subgroup is the earliest offshoot in the repleta group. It thus seems that, as with the fasciola subgroup, the ancestral repleta group probably inhabited forests.
While supporting monophyly of the repleta group, our data clearly does not resolve the phylogenetic relationships in the fasciola subgroup, possibly due to saturation in relation to the mitochondrial gene used to construct the phylogeny (Allard and Carpenter, 1996; Reyes et al., 1999; Durando et al., 2000; Bull et al., 2003). Research using nuclear genes could help elucidate the phylogenetic relationships in the fasciola subgroup and between this and the other subgroups of the repleta group.
Acknowledgments
We thank Dr. Reinaldo O.A. de Brito for providing representative hydei, mercatorum, mulleri and repleta subgroups COI gene sequences and also the Universidade de So Paulo (USP), Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq), Coordenao de Pessoal de Nivel Superior (CAPES) and the Fundao de Amparo a Pesquisa do Estado de So Paulo (FAPESP) for financial support and a personal fellowship from Fundao de Amparo a Pesquisa do Estado de So Paulo (FAPESP) proc. 02/02195-9 for E.C.C. Silva-Bernardi. We also thank two anonymous referees for their helpful comments.
REFERENCES
Allard MW and Carpenter JM (1996) On weighting and congruence. Cladistics 12:183-198.
Bchli G and Vilela CR (2002) Six new species of Drosophila (Diptera, Drosophilidae) from Parque Nacional Henri Pittier, Venezuela. Mitt Schweiz Ent Ges 75:223-243.
Bull NJ, Schwarz MP and Cooper SJB (2003) Phylogenetic divergence of the Australian allodapine bees (Hymenoptera, Apidae). Mol Phylogenet Evol 27:212-222.
De Brito RA, Manfrin MH and Sene FM (1998) Use of nuclear and mitochondrial data to investigate evolution and phylogenetic relationships of the repleta group (genus Drosophila) radiation. Genet Mol Biol 21(suppl):258.
Diniz NM and Sene FM (2004) Chromosomal phylogeny of the Drosophila fasciola species subgroup revisited (Diptera, Drosophilidae). Genet Mol Biol 27:561-566.
Durando CM, Baker RH, Etges WJ, Heed WB, Wasserman M and DeSalle R (2000) Phylogenetic analysis of the repleta species group of the genus Drosophila using multiple sources of characters. Mol Phylogenet Evol 16:296-307.
Etges WJ, Armella MA, O'Grady PM and Heed WB (2001) Two new species of Drosophila (Diptera, Drosophilidae) in the repleta group from Mexico. Ann Entomol Soc Am 94:16-20.
Felsenstein J (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39:783-791.
Kuhn GCS, Diniz NM, Alves MAR and Sene FM (1995) Constituio cariotipica de especies do subgrupo fasciola do grupo repleta do gênero Drosophila. Rev Bras Genet 18(suppl):282.
Morais PB, Rosa CA, Hagler AN and Mendona-Hacler LC (1995) Yeast communities as descriptors of habitat use by the Drosophila fasciola subgroup (repleta group) in Atlantic rain forests. Oecologia 104:45-51.
Patterson JT and Wheeler MR (1942) Description of new species of the subgenera Hirtodrosophila and Drosophila. Univ Texas Publ 4213:67-109.
Pereira MAQR, Vilela CR and Sene FM (1983) Notes on breeding and feeding sites of some species of the repleta group of the genus Drosophila (Diptera, Drosophilidae). Cienc Cult 35:1313-1319.
Pipkin SB (1965) The influence of adult and larval food habits on population size of neotropical ground-feeding Drosophila. Am Midl Nat 74:1-27.
Pipkin SB, Rodriguez RL and Leon J (1966) Plant host specificity among flower-feeding Neotropical Drosophila (Diptera, Drosophilidae). Am Nat 100:135-156.
Rafael V and Arcos G (1989) Subgrupo inca, un nuevo subgrupo del grupo repleta, con descripcion de Drosophila huancavilcae N.SP. (Diptera, Drosophilidae). Evol Biol 3:233-243.
Reyes SG, Cooper SJB and Schwarz MP (1999) Species phylogeny of the bee genus Exoneurella Michener (Hymenoptera, Apidae, Allodapini): Evidence from molecular and morphological data sets. Ann Entomol Soc Am 92:20-29.
Sene FM, Paganelli CHM, Pedroso LG, Garcia E and Palombo CR (1977) Local natural de criao de Drosophila onca Dobzhansky & Pavan, 1943. Cienc Cult 29(suppl):716.
Simon C, Frati F, Beckenbach A, Crespi B, Liu H and Flook P (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann Entomol Soc Am 87:651-701.
Swofford DL (2002) PAUP. Phylogenetic Analysis Using Parsimony (and Other Methods), Version 4. Sinauer Associates, Sunderland, Massachusetts.
Tatarenkov A and Ayala FJ (2001) Phylogenetic relationships among species groups of the virilis-repleta radiation of Drosophila. Mol Phylogenet Evol 21:327-331.
Templeton AR (1983) Phylogenetic inference from restriction endonuclease cleavage site maps with particular reference to the evolution of humans and the apes. Evolution 37:221-244.
Thompson JD, Higgins DG and Gibson TJ (1994) CLUSTALW: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673-4680.
Throckmorton LH (1975) The phylogeny, ecology and geography of Drosophila. In: King RC (ed) Handbook of Genetics. Plenum Press, New York, pp 421-459.
Throckmorton LH (1982) Pathways of evolution in the genus Drosophila and the founding of the repleta group. In: Barker JSF and Starmer WT (eds) Ecological Genetics and Evolution: The Cactus-Yeast-Drosophila Model System. Academic Press, New York, pp 33-48.
Tidon-Sklorz R and Sene FM (1988) A trap that retains and keeps Drosophila alive. Drosoph Inf Serv 67:89.
Tidon-Sklorz R and Sene FM (1995a) Drosophila seriema n. sp.: New member of the Drosophila serido (Diptera, Drosophilidae) superspecies taxon. Ann Entomol Soc Am 88:139-142.
Tidon-Sklorz R and Sene FM (1995b) Fauna of Drosophila (Diptera, Drosophilidae) in the northern area of the "Cadeia do Espinhao", states of Minas Gerais and Bahia, Brazil: Biogeographical and ecological aspects. Iheringia, Ser Zool 78:85-94.
Tidon-Sklorz R and Sene FM (2001) Two new species of the Drosophila serido sibling set (Diptera, Drosophilidae). Iheringia, Ser Zool 90:141-146.
Vilela CR (1983) A revision of the Drosophila repleta species group (Diptera, Drosophilidae). Rev Bras Entomol 27:1-114.
Vilela CR and Bchli G (1990) Taxonomic studies on Neotropical species of seven genera of Drosophilidae (Diptera). Mitt Schweiz Ent Ges 63(suppl):1-332.
Vilela CR, Pereira MAQR and Sene FM (1983) Preliminary data on the geographical distribution of Drosophila species within morphoclimatic domains of Brazil: II. The repleta group. Cienc Cult 35:66-70.
Wasseman M (1960) Cytological and phylogenetic relationships in the repleta group of the genus Drosophila. Proc Nat Acad Sci 46:842-859.
Wasserman M (1962) Cytological studies of the repleta group of the genus Drosophila: VI. The fasciola subgroup. Univ Texas Publ 6205:119-134.
Wasserman M (1982) Evolution of the repleta group. In: Ashburner M, Carson HL and Thompson JN (eds) The Genetics and Biology of Drosophila. v. 3b. Academic Press, London, pp 61-139.
Wasserman M (1992) Cytological evolution of the Drosophila repleta species group. In: Krimbas CB and Powell JR (eds) Drosophila Inversion Polymorphism. CRC Press, Boca Raton, pp 455-555.
Xia X and Xie Z (2001) DAMBE: Data analysis in molecular biology and evolution. Journal of Heredity 92:371-373.(Erica C.C. Silva-Bernardi)