THE REPRODUCTIVE RELATIONSHIPS OF DROSOPHILA SECHELLIA WITH D. MAURITIANA,D. SIMULANS,AND D. MELANOGASTER FROM THE AFROTROPICAL REGION

Hybridization tests among the four sibling species of the Drosophila melanogaster complex were made to determine the reproductive status of the recently discovered D. sechellia (which is endemic to a few islands and islets of the Seychelles archipelago) with regard to its three close relatives, D. mauritiana (endemic to Mauritius) and Afrotropical strains of the two cosmopolitan species D. melanogaster and D. simulans. Interstrain variation in the ability to hybridize with other species was also analyzed for D. melanogaster and D. simulans. D. mauritiana and D. simulans appear to be more weakly isolated from each other than either species is from D. sechellia. A striking unilateral mating success is observed in the cross of D. sechellia with D. simulans. The most extreme isolation is between D. melanogaster and its three siblings. Variation in the ability of strains to hybridize is observed in heterospecific crosses between D. simulans and either D. melanogaster or D. mauritiana.     

RACE FORMATION,SPECIATION, AND INTROGRESSION WITHIN DROSOPHILA SIMULANS,D. MAURITIANA,AND D. SECHELLIA INFERRED FROM MITOCHONDRIAL DNA ANALYSIS

Mitochondrial DNA cleavage maps from three chromosomally homosequential species Drosophila simulans, D. mauritiana, and D. sechellia, were established for 12 restriction enzymes. One isofemale strain was studied in D. sechellia (se), 13 in D. simulans, and 17 in D. mauritiana: in the last two species, respectively, three (siI, II, and III) and two (maI and II) cleavage morphs were found. The evolutionary relationships based on mtDNA cleavage map comparisons show that the maI and se mtDNAs are internal branches of the phylogenetic tree of the D. simulans mtDNA. D. mauritiana and D. sechellia species appear to be derived from a population of D. simulans which carried an ancestral form of the current siI mtDNA type. In addition, two cleavage morphs (siIII [only present in D. simulans from Madagascar] and maI) appeared to be identical, although found in different species. We present a speculative interpretation of data on biogeography and hybridization which is consistent with the hypothesis of a recent introgression of mitochondrial DNA of D. simulans from Madagascar into D. mauritiana.     

Variability on the dot chromosome in the Drosophila simulans clade

A recent study suggested that recent nuclear gene introgression between Drosophila simulans and D. mauritiana may have obscured efforts to estimate the phylogeny of the species of the D. simulans clade, which includes these two species and D. sechellia. Here, we report sequence variation of an intron of the eyeless gene in this species group. This gene should introgress freely between these species because it is not linked to any known barriers to gene exchange. We have also reevaluated levels of sequence divergence among species in this clade, noting differences between loci in regions of low recombination (as in all chromosome 4 loci) relative to other loci. Overall, none of the data analyzed were consistent with recent introgression exclusively between D. simulans and D. mauritiana.     

The evolutionary genetics of thehobo transposable element in theDrosophila melanogaster complex

Hobo elements are a family of transposable elements found inDrosophila melanogaster and its three sibling species:D. simulans, D. mauritiana andD. sechellia. Studies inD. melanogaster have shown thathobo may be mobilized, and that the genetic effects of such mobilizations included the general features of hybrid dysgenesis: mutations, chromosomal rearrangements and gonadal dysgenis in F1 individuals. At the evolutionary level somehobo-hybridizing sequences have also been found in the other members of themelanogaster subgroup and in many members of the relatedmontium subgroup. Surveys of older collected strains ofD. melanogaster suggest that completehobo elements were absent prior to 50 years ago and that they have recently been introduced into this species by horizontal transfer. In this paper we review our findings and those of others, in order to precisely describe the geographical distribution and the evolutionary history ofhobo in theD. melanogaster complex. Studies of the DNA sequences reveal a different level of divergence between the groupD. melanogaster, D. simulans andD. mauritiana and the fourth speciesD. sechellia. The hypothesis of multiple transfers in the recent past into theD. melanogaster complex from a common outside source is discussed.     

Insertion sites of the transposable elementmariner are fixed in the genome ofDrosophila sechellia

Summary The abundance of the transposable elementmariner differs dramatically in the genomes of the closely related speciesDrosophila simulans, D. mauritiana, D. sechellia, andD. melanogaster. Natural populations ofD. simulans andD. mauritiana have 1–10 and 20–30 copies per diploid genome, respectively, and the insertion sites are polymorphic. The element has not been found inD. melanogaster. In this paper we show thatD. sechellia, a species endemic to the Seychelles Islands, contains only twomariner elements that are at fixed sites in the genome. One element, inserted in chromosome 2R at 51A1–2, contains three deletions and is missing much of the 3 end. The other element, inserted in chromosome 3L at 64A10–11, is the full length of 1286 bp. Although the sequence of the full-length element is polymorphic in populations ofD. sechellia, at least some of the sequences are closely related to elements fromD. simulans andD. mauritiana that are known to be active. However, judging from the progeny of crosses betweenD. sechellia andD. simulans, the biological activity of the full-lengthD. sechellia element appears to be low, either because of the nucleotide sequence of the element or because of its position in the genome, or both.     

Comparative copulation anatomy of the Drosophila melanogaster species complex (Diptera: Drosophilidae)

The Drosophila melanogaster species complex consists of four species: D. melanogaster, D. simulans, D. sechellia and D. mauritiana. To identify these closely related species, researchers often examine the male genitalia, especially species‐specific shapes of the posterior process, as the most reliable and easily observable character. However, compared to genetic aspects, the evolutionary significance of the posterior process and other genital parts remains largely unexplained. By comparing genital coupling among these species, we revealed that the posterior processes, which are hidden under the female abdominal tergite VII when genital coupling is established, mesh with different parts of the intersegmental membrane between the tergite VIII and the oviscapts and that this membrane region broadens in a species‐specific manner. Furthermore, in D. simulans and D. sechellia, this membrane region is likely to incur wounds from the sharply pointed tip of the posterior process. On the basis of the use and functions of these and other genital parts, we discuss possible evolutionary forces underlying the diversification of genitalia in this group.     

Association Between Levels of Coding Sequence Divergence and Gene Misregulation in <Emphasis Type="Italic">Drosophila</Emphasis> Male Hybrids

Previous studies have shown widespread conservation of gene expression levels between species of the Drosophila melanogaster subgroup as well as a positive correlation between coding sequence divergence and expression level divergence between species. Meanwhile, large-scale misregulation of gene expression level has been described in interspecific sterile hybrids between D. melanogaster, D. simulans, D. mauritiana, and D. sechellia. Using data from gene expression analysis involving D. simulans, D. melanogaster, and their hybrids, we observed a significant positive correlation between protein sequence divergence and gene expression differences between hybrids and their parental species. Furthermore, we demonstrate that underexpressed misregulated genes in hybrids are evolving more rapidly at the protein sequence level than nonmisregulated genes or overexpressed misregulated genes, highlighting the possible effects of sexual and natural selection as male-biased genes and nonessential genes are the principal gene categories affected by interspecific hybrid misregulation. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users. Carlo G. Artieri and Wilfried Haerty contributed equally to this publication.     

Relationships within the melanogaster species subgroup of the genus Drosophila (Sophophora)

The polytene chromosomes of two new species of Drosophila, D. sechellia and D. orena, both members of the melanogaster species subgroup, are described. The chromosomes of D. sechellia, a species endemic to certain islands in the Seychelles, are homosequential with those of D. simulans and D. mauritiana. The chromosomes of D. orena, a species from the mountains of west Africa, are very similar to those of D. erecta. We discuss the interrelationships of the eight known species of the melanogaster species subgroup, based upon an analysis of their chromosome banding patterns.     

Sexual Isolation Between Drosophila Melanogaster, D. Simulans and D. Mauritiana: Sex and Species Specific Discrimination

The sexual isolation among the related species Drosophila melanogaster, D. simulans and D. mauritiana is asymmetrical. While D. mauritiana males mate well with both D. melanogaster and D. simulans females, females of D. mauritiana discriminate strongly against males of these two species. Similarly, D. simulans males mate with D. melanogaster females but the reciprocal cross is difficult. Interspecific crosses between several populations of the three species were performed to determine if (i) males and females of the same species share a common sexual isolation genetic system, and (ii) males (or females) use the same genetic system to discriminate against females (or males) of the other two species. Results indicate that although differences in male and female isolation depend on the populations tested, the isolation behaviour between a pair of species is highly correlated despite the variations. However, the rank order of the isolation level along the populations was not correlated in both sexes, which suggests that different genes act in male and female sexual isolation. Neither for males nor for females, the isolation behaviour of one species was paralleled in the other two species, which indicates that the genetic systems involved in this trait are species-pair specific. The implications of these results are discussed. This revised version was published online in August 2006 with corrections to the Cover Date.     

VARIATIONS IN CUTICULAR HYDROCARBONS AMONG THE EIGHT SPECIES OF THE DROSOPHILA MELANOGASTER SUBGROUP

In addition to protecting against desiccation, some of the hydrocarbons of the waxy cuticle have previously been shown to be mating pheromones in Drosophila melanogaster and D. simulans. Therefore, cuticular hydrocarbons were compared among the eight species in the D. melanogaster subgroup. For the two cosmopolitan species and several geographic strains that were studied, all males are quite similar with very abundant monoenes. The major compound in most cases is 7-tricosene. Only three exceptions were found: D. sechellia, and the Afrotropical strains of D. melanogaster and D. simulans. A significant sexual dimorphism exists in three species: D. melanogaster, D. erecta, and D. sechellia. Greater variation was observed in females than in males. D. erecta is singular in the production of long-chain molecules (31–33 carbons). Only three species (D. melanogaster, D. erecta, and D. sechellia) produce diene in significant amounts. Such products, especially 7,11-heptacosadiene, are known to act as aphrodisiacs for D. melanogaster males. In the five other species, females show only quantitative differences from males, generally with 7-tricosene as the most abundant compound. This compound is an aphrodisiac for D. simulans males. Some species such as D. yakuba, D. teissieri, D. orena, D. mauritiana, and the Seychelles strain of D. simulans are almost identical in the chemical composition of cuticular hydrocarbons. In contrast, important variations are observed between geographic populations of D. melanogaster and D. simulans.     

CYTOLOGICAL CHARACTERIZATION OF PREMEIOTIC VERSUS POSTMEIOTIC DEFECTS PRODUCING HYBRID MALE STERILITY AMONG SIBLING SPECIES OF THE DROSOPHILA MELANOGASTER COMPLEX

In accordance with Haldane's rule, hybridizations between species of the Drosophila simulans clade produce fertile females but sterile males. In this study, a comprehensive characterization was undertaken on the six types of F₁ males that were the result of the crosses between D. simulans, D. sechellia, and D. mauritiana. With the use of light and electron microscopy, it was shown that while each particular hybrid genotype exhibited a specific sterility phenotype, these phenotypes fell into two distinct classes. The two hybrid genotypes that possessed D. mauritiana X-chromosomes contained spermatogenic defects that caused arrests in premeiotic spermatogenic stages. The other four F₁ hybrids possessed postmeiotic spermatogenic defects. Nonsynchronous cell divisions, underdeveloped mitochondrial derivative-axonemal associations, and microtubule abnormalities were common to all of these hybrids. Each particular postmeiotically defective hybrid genotype demonstrated characteristically distinct profiles in sperm bundle number in addition to characteristic spermiogenic arrests in the furthest developed spermatids. These results in species hybrids contrast with the absence of significant differences in spermatogenic characters between species of this clade. In addition, by utilizing an attached-X cross, we investigated the influence of maternal effects and cytoplasmic factors on the sterility of D. simulans F₁ hybrids and found none. However, we discovered a strain of D. simulans (2119) that caused a large shift in sterility from postmeiotic to premeiotic when crossed to D. sechellia. This suggests that D. simulans is polymorphic for genes involving premeiotic and postmeiotic sterility and that the two types of sterilities between species may have a simple genetic basis.     

Incompatibility and competitive exclusion of genomic segments between sibling Drosophila species

The extent and nature of genetic incompatibilities between incipient races and sibling species is of fundamental importance to our view of speciation. However, with the exception of hybrid inviability and sterility factors, little is known about the extent of other, more subtle genetic incompatibilities between incipient species. Here we experimentally demonstrate the prevalence of such genetic incompatibilities between two young allopatric sibling species, Drosophila simulans and D. sechellia. Our experiments took advantage of 12 introgression lines that carried random introgressed D. sechellia segments in different parts of the D. simulans genome. First, we found that these introgression lines did not show any measurable sterility or inviability effects. To study if these sechellia introgressions in a simulans background contained other fitness consequences, we competed and genetically tracked the marked alleles within each introgression against the wild-type alleles for 20 generations. Strikingly, all marked D. sechellia introgression alleles rapidly decreased in frequency in only 6 to 7 generations. We then developed computer simulations to model our competition results. These simulations indicated that selection against D. sechellia introgression alleles was high (average s = 0.43) and that the marker alleles and the incompatible alleles did not separate in 78% of the introgressions. The latter result likely implies that most introgressions contain multiple genetic incompatibilities. Thus, this study reveals that, even at early stages of speciation, many parts of the genome diverge to a point where introducing foreign elements has detrimental fitness consequences, but which cannot be seen using standard sterility and inviability assays.     

USING HITCHHIKING GENES TO STUDY ADAPTATION AND DIVERGENCE DURING SPECIATION WITHIN THE DROSOPHILA MELANOGASTER SPECIES COMPLEX

Several studies of intraspecific and interspecific DNA sequence variation from Drosophila loci have revealed a pattern of low intraspecific variation from genomic regions of low recombination. The mechanisms consistently invoked to explain these patterns are the selective sweep of advantageous mutations together with genetic hitchhiking of linked loci. To examine the effect of selective sweeps on genetic divergence during speciation, we studied two loci in different genomic regions thought to be subject to selective sweeps. We obtained DNA sequences from 1.1kb pair portions of the fourth chromosome locus cubitus interruptus Dominant (ci^D) and from the asense locus near the telomere of the X chromosome. At ci^D, we found very low variation among multiple lines of Drosophila mauritiana and D. sechellia. This finding is consistent with an earlier report of very low variation in D. melanogaster and D. simulans at ci^D and supports the conclusion of selective sweeps and genetic hitchhiking on the nonrecombining fourth chromosome. The pattern of variation found at asense suggests that a selective sweep has occurred recently at the tip of the X chromosome in D. simulans, but not in D. melanogaster or D. mauritiana. The data from ci^D and asense are compared with data from three X chromosome loci (period, zeste, and yolk protein 2) that experience normal levels of recombination. By examining estimated genealogies and the rates at which different classes of mutations have accumulated, we conclude that selective sweeps are common occurrences on the fourth chromosome but less common near the tip of the X chromosome. An interesting pattern of low variation at ci^D among D. simulans, D. mauritiana, and D. sechellia suggests that a selective sweep may have occurred among these forms even after divergence into separate species had begun.     

Hybrid lethal systems in theDrosophila melanogaster species complex

Lethal phases of the hybrids betweenDrosophila melanogaster and its sibling species,D. simulans are classified into three types: (1) embryonic lethality in hybrids carryingD. simulans cytoplasm andD. melanogaster X chromosome, (2) larval lethality in hybrids not carryingD. simulans X, and (3) temperature-sensitive pupal lethality in hybrids carryingD. simulans X. The same lethal phases are also observed when either of the two other sibling species,D. mauritiana orD. sechellia, is employed for hybridization withD. melanogaster. Here, we describe genetic analyses of each hybrid lethality, and demonstrate that these three types of lethality are independent phenomena. We then propose two models to interpret the mechanisms of each hybrid lethality. The first model is a modification of the conventional X/autosome imbalance hypothesis assuming a lethal gene and a suppressor gene are involved in the larval lethality, while the second model is for embryonic lethality assuming an interaction between a maternal-effect lethal gene and a suppressor gene.     

Development time and pupation behavior in theDrosophila melanogaster subgroup (Diptera: Drosophilidae)

This study is an in-depth analysis of intersexual, intraspecific, and interspecific variability in larvopupal developmental time, pupation site preference, and larval and pupal survival of a number of isofemale lines of the speciesDrosophila mauritiana, D. melanogaster, D. sechellia, D. simulans, D. teissieri, andD. yakuba. There was no significant sex differences in pupation height, but females eclosed significantly earlier than males in all species. In addition, the suggestion of a strong negative correlation between larval developmental time and pupation height could not be confirmed in this study. The hypothesis that differences in pupation height provide a basis for niche partitioning between closely related species with overlapping distributions was tested by three planned orthogonal contrast analyses of variance. First, the two speciesD. teissieri andD. yakuba, with largely overlapping distribution, were significantly different in pupation height. Second, the two allopatric, nonoverlapping island speciesD. mauritiana andD. sechellia did not significantly differ in pupation height. However, the absence of a significant difference in the final contrast between the two cosmopolitan speciesD. melanogaster andD. simulans, which are often found together, makes us cautious to accept the hypothesis.     

Microsatellite variation suggests a recent fine-scale population structure of <Emphasis Type="Italic">Drosophila sechellia</Emphasis>, a species endemic of the Seychelles archipelago

Drosophila sechellia is closely related to the cosmopolitan and widespread model species, D. simulans. This species, endemic to the Seychelles archipelago, is specialized on the fruits of Morinda citrifolia, and harbours the lowest overall genetic diversity compared to other species of Drosophila. This low diversity is associated with a small population size. In addition, no obvious population structure has been evidenced so far across islands of the Seychelles archipelago. Here, a microsatellite panel of 17 loci in ten populations from nine islands of the Seychelles was used to assess the effect of the D. sechellia’s fragmented distribution on the fine-scale population genetic structure, the migration pattern, as well as on the demography of the species. Contrary to previous results, also based on microsatellites, no evidence for population contraction in D. sechellia was found. The results confirm previous studies based on gene sequence polymorphism that showed a long-term stable population size for this species. Interestingly, a pattern of Isolation By Distance which had not been described yet in D. sechellia was found, with evidence of first-generation migrants between some neighbouring islands. Bayesian structuring algorithm results were consistent with a split of D. sechellia into two main groups of populations: Silhouette/Mahé versus all the other islands. Thus, microsatellites suggest that variability in D. sechellia is most likely explained by local genetic exchanges between neighbouring islands that have recently resulted in slight differentiation of the two largest island populations from all the others.     

ASYMMETRIC MISMATCH IN STRAIN‐SPECIFIC GENITAL MORPHOLOGY CAUSES INCREASED HARM TO DROSOPHILA FEMALES

Although several evolutionary forces have been proposed to contribute to genital morphological diversification, it is unclear which might act early during the evolution of novel structural traits. We test the hypothesis that mismatch between interacting male and female secondary sexual structures gives rise to increased harm to females, consistent with the outcome predicted from a history of sexual conflict. We mate Drosophila sechellia females to males from a collection of D. mauritiana–D. sechellia interspecific genetic introgression lines that possess quantitative morphological variation in the posterior lobe of the genital arch, an external genital structure that can cause wounds to the female abdomen during mating. We find that males with smaller posterior lobes, and those that possess lobes with similarities in shape to D. mauritiana, cause more severe wounding compared to either D. sechellia males with strain‐specific morphologies or introgression males that possess larger lobes or lobes with more pronounced D. sechellia features. These results suggest a possible history of sexual conflict during the evolution of the posterior lobe in D. sechellia, but also suggest a potential contribution of divergence in sensory recognition mechanisms to posterior lobe evolution.     

Population transcriptomics: insights from <Emphasis Type="Italic">Drosophila simulans,</Emphasis><Emphasis Type="Italic">Drosophila sechellia</Emphasis> and their hybrids

Sequence differentiation has been widely studied between populations and species, whereas interest in expression divergence is relatively recent. Using microarrays, we compared four geographically distinct populations of Drosophila simulans and a population of Drosophila sechellia, and interspecific hybrids. We observed few differences between populations, suggesting a slight population structure in D. simulans. This structure was observed in direct population comparisons, as well as in interspecific comparisons (hybrids vs. parents, D. sechellia vs. D. simulans). Expression variance is higher in the French and Zimbabwean populations than in the populations from the ancestral range of D. simulans (Kenya and Seychelles). This suggests a large scale phenomenon of decanalization following the invasion of a new environment. Comparing D. simulans and D. sechellia, we revealed 304 consistently differentially expressed genes, with striking overrepresentation of genes of the cytochrome P450 family, which could be related to their role in detoxification as well as in hormone regulation. We also revealed differences in genes involved in Juvenile hormone and Dopamine differentiation. We finally observed very few differentially expressed genes between hybrids and parental populations, with an overrepresentation of X-linked genes.     

THE GENETICS OF AN ISOLATING MECHANISM BETWEEN TWO SIBLING SPECIES OF DROSOPHILA

Matings between Drosophila simulans females and males of the sibling species D. mauritiana are of abnormally short duration. These rapid matings interrupt the transfer of sperm, leading to substantial reproductive isolation in interspecific as compared to intraspecific copulations. Genetic analysis of this behavior shows that it is influenced much more by the male than the female genotype, with genes from D. simulans being dominant. In males, shortened copulation is caused by interspecific divergence at a minimum of three loci, with one gene on each of the major chromosomes. This is an underestimate of the true number of loci affecting the trait, which could be much larger. The two autosomes have the largest effect, whereas that of the X chromosome is much smaller. The genetic architecture of copulation duration and the larger effect of male than female genotype suggest that females can detect and discriminate against differences in male genitalia.     

Variable post‐zygotic isolation in Drosophila melanogaster/D. simulans hybrids

The study of hybrid inviability reveals cryptic divergence between the genetic interactions that maintain stable phenotypes in the pure species . We characterized the effects of natural variation on the penetrance of hybrid inviability phenotypes in crosses between Drosophila melanogaster and two species of the D. simulans subcomplex, D. simulans and D. sechellia. Using a panel of wild‐caught lines, we studied the levels of genetic variance present in D. simulans and D. sechellia affecting prezygotic and post‐zygotic isolation in hybridizations with D. melanogaster females. We observed extensive variability in the viability of hybrid individuals, dependent on the genotype of the parents, suggesting that intraspecific natural variation manifests directly in hybrid phenotypes. Furthermore, we found that genetic background significantly affects the penetrance of a well‐studied determinant of hybrid inviability: the interaction between Hmr^mel–Lhr^sim. Our results suggest that hybrid inviability – and reproductive isolation generally – can be modified by polymorphisms at multiple loci segregating within the parental species. Just as the penetrance of most mutant phenotypes can be modified by the genetic background within the pure species, the penetrance of hybrid inviability phenotypes is highly influenced by the parental genotypes.