|
Mem Inst Oswaldo Cruz, Rio de Janeiro, Vol. 96(3) Apr. 2001, pp. 403-405 Courtship Song Genes and Speciation in Sand Flies SG Oliveira*, M Bottecchia, LGSR Bauzer, NA Souza**, RD Ward***, CP Kyriacou****, AA Peixoto/+ Departamento de Bioquímica e Biologia Molecular **Departamento de Entomologia, Instituto Oswaldo Cruz- Fiocruz, Av. Brasil 4365, 21045-900 Rio de Janeiro, RJ, Brasil *Departamento de Biologia Animal e Vegetal, Universidade Estadual do Rio de Janeiro, Rio de Janeiro, Brasil ***The Centre for Applied Entomology and Parasitology, School of Life Sciences, University Keele, UK ****Department of Genetics, University of Leicester, Leicester, UK +Corresponding
author. Fax: +55-21-590.3495. E-mail: apeixoto@gene.dbbm.fiocruz.br
Received 13 November 2000 Code Number: oc01062 Lutzomyia longipalpis (Lutz & Neiva, 1912) (Diptera: Psychodidae: Phlebotominae) is a vector of visceral leishmaniasis in the Americas and it might represent a complex of sibling species. Reproductive isolation between closely related species often involves differences in courtship behaviour. cacophony (cac) and period (per) are two Drosophila genes that control features of the "lovesong" males produce during courtship that has been implicated in the sexual isolation between closely related species. We are using gene fragments from L. longipalpis' homologues of these two genes to study the speciation process in this putative species complex. Key words: Lutzomyia longipalpis - cacophony - period - polymorphism Genes controlling aspects of reproductive behaviour are among the best molecular markers in the study of speciation in insect vectors. In Drosophila, males vibrate their wings during courtship producing a kind of "lovesong". This acoustic signal is potentially involved in the reproductive isolation between closely related species (Kyriacou & Hall 1982, 1986, Ritchie et al. 1999). Among the genes controlling this "lovesong" (Hall 1994), cacophony (cac) and period (per) are particularly interesting as they control features of the song that are known to vary between species (see below). cac codes for a voltage-gated calcium channel a-1 subunit, a large protein that forms the pore of the channel. cac is a very large (over 45kb) and complex gene that presents alternative splicing and RNA editing of its transcript (Smith et al. 1996, 1998a, b, Peixoto et al. 1997). It is also the site of song (cacS), visual (night-blind-A) and lethal (L13) mutations. In vertebrates, voltage-gated calcium channels are divided in many different classes according to their electrophysiological characteristics, pharmacology, sequence similarities and tissue distribution (Stea et al. 1995). cac, also known as Dmca1A, shows homology to vertebrate calcium channel classes A, B and E. The song produced by its mutant allele cacS is characterized by longer interpulse-intervals (IPIs) and pulses that contain more cycles than normal (Schilcher 1977, Kulkarni & Hall 1987, Peixoto & Hall 1998). These are two features that commonly show differences among various Drosophila species (e.g. Hoikkala & Lumme 1987, Ritchie & Gleason 1995). The clock gene per controls a different feature of the "lovesong" of Drosophila. D. melanogaster and related species (Kyriacou & Hall 1980, Demetriades et al. 1999) present cyclic variation in their mean IPI. These song rhythms are species-specific and they have been implicated as one of the signals females used to recognise conspecific males (Kyriacou & Hall 1982, 1986, Ritchie et al. 1999). Because per controls the differences in song rhythms that seem to contribute to the reproductive isolation between D. melanogaster and its sibling species D. simulans (Wheeler et al. 1991), it has been called a "speciation gene" (Coyne 1992). Acoustic communication is also a feature of the courtship behaviour of Lutzomyia longipalpis (Ward et al. 1988), a vector of visceral leishmaniasis in the Americas and a putative complex of sibling species (Ward et al. 1988, Lanzaro et al. 1993, Alexander et al. 1998, Uribe 1999). Using PCR with degenerated primers, we cloned and sequenced gene fragments from L. longipalpis' putative homologues of per and cac. In the case of per, the fragment is nearly 1 Kb long and includes a small intron (~50 bp). It encodes the end of the PAS protein dimerisation domain and the Thr-Gly repeat region. The cac fragment is ~270 bp and encodes the IVS6 domain of the channel. It also includes an intron (~100 bp). Based on these first sequences we designed more specific oligos for this species and we are now using them to study the molecular variation and differentiation among populations of the L. longipalpis complex in Brazil. Figure shows the first results we obtained with the analysis of the IVS6 region of cac in the population of Natal (State of Rio Grande do Norte). Length variation was observed among PCR fragments amplified from different individuals. Sequencing of these fragments revealed a number of differences due to insertion/deletion events and point mutations within the intron in this region of L. longipalpis' cac (Figure). Preliminary analysis of a number of other sand flies from Natal and from the population of Lapinha (Lagoa Santa, State of Minas Gerais) confirms that this intron of cac is highly polymorphic and indicates that the two populations are quite differentiated. For example, the Table shows the frequencies of haplotypes associated with two major deletions found in this intron. Significant differences are observed between the populations of Natal and Lapinha (X2 = 10.26; d.f. = 2; p = 0.0059). The haplotype associated with deletion 1 has a frequency of 36.4% (8/22) in Natal but has yet to be observed in Lapinha, while in this latter population the haplotype associated with deletion 2 is fairly common (30%; 6/20), although it was not found in Natal. We also started the analysis of the molecular variation in a region of the per gene fragment of L. longipalpis in the populations of Lapinha and Jacobina (State of Bahia). The first results also indicate that these two populations are quite differentiated. We sequenced a 266 bp fragment in 35 individuals (17 of Lapinha and 18 of Jacobina). Of a total of 33 polymorphic sites found so far, only six are shared between the two populations. The calculated Fst for the data is 0.3911 (Hudson et al. 1992, Rozas & Rozas 1999). This figure predicts only 0.39 migrants per generation (Nm), a value nearly ten times smaller than the ones predicted by isoenzyme loci (Mukhopadhyay et al. 1998, Mutebi et al. 1999, Azevedo et al. 2000). The differentiation between the Natal, Lapinha and Jacobina populations at the molecular level is supported by the first results obtained with the analysis of the "lovesongs" produced by L. longipalpis males from these very same populations (Souza & Peixoto, unpublished), and raises the question of whether they represent different sibling species. Cryptic speciation is one of the most interesting phenomena in evolution and its occurrence in phlebotomine sand flies has important epidemiological consequences (Lanzaro & Warburg 1995). The results obtained so far with the analysis of the molecular variation in cac and per suggest that courtship song genes will be very useful to enhance our knowledge of the speciation process in the L. longipalpis complex. ACKNOWLEDGEMENTS To Reginaldo Brazil (CPqRR, Belo Horizonte, Brazil), Iraci Duarte (FNS, Natal, Brazil) and Italo Sherlock (CPqGM, Salvador, Brazil) for helping us in obtaining sand fly samples. To Paulo Roberto de Amoretty and Robson Costa da Silva for their technical assistance. REFERENCES
This work was supported by the Wellcome Trust with additional support from Faperj, Fiocruz and CNPq. Copyright 2001 Fundacao Oswaldo Cruz Fiocruz |
|