Information

Have we ever observed two drosophila lineages that evolved reproductive isolation in labs?


Background

The standard definition of species refers to the concept of reproductive isolation. If two lineages are found to be reproductively isolated, then we consider these two lineages to belong to different species. My question concerns evolved reproductive isolation in Drosophila sp. following labs due to experimental evolution.

Question

Have we ever demonstrated that two Drosophila sp. lineages that could initially interbreed (in nature or in labs) evolved through artificial selection (and drift and mutations) in labs to finally not be able to interbreed anymore either due pre- or post- zygotic isolation (see wiki)? Or, in other words, have we ever demonstrated that two drosophila lineages evolved to become different species (reproductive isolation definition) in labs experiments?

If not, have we ever observed some partial reproductive isolation such as inbreeding depression for example?


Note: This question is motivated by @LotusBiology that could not receive the answers she/he was waiting for because he/she failed to ask questions that are possibly answerable! So I wanted to ask this question that somehow addresses this question he/she asked (now on-hold)

For more information about the concept of species, please have a look at How could humans have interbred with Neanderthals if we're a different species?


Rice and Salt$^1$ bred fruit flies for 35 generations and from one line of flies created two groups that were isolated from each other reproductively. They could not interbreed because they no longer bred in the same environment. Depending on one's definition of 'species' this could be a case of artificial speciation.

$^1$Rice WR, Salt GW (1988), Speciation via disruptive selection on habitat preference: experimental evidence". The American Naturalist 131 (6): 911-917.


Diane Dodd's experiments on Drosophila pseudoobscura would be another example of lab-based speciation.

http://www.jstor.org/stable/2409365?__redirected

To summarise - four populations each adapted to feeding on a starch-based diet and a maltose-based diet were evolved in the lab to test effects on mating preferences; compared to what is expected by random, using chi-squared tests, 11 of 16 combinations showed greater isolation than expected by chance - strong evidence for assortative mating.

An alternate link to the PDF of the paper is here http://www.sulfide-life.info/mtobler/images/stories/readings/dodd%201989%20evolution.pdf


Reference:

Dodd, Diane M. B. (September 1989). "Reproductive isolation as a consequence of adaptive divergence in Drosophila pseudoobscura". Evolution 43 (6): 1308-1311. doi:10.2307/2409365.


Reproductive Isolation

Uniform Selection (Mutation-Order Speciation)

Reproductive isolation might also evolve during a process of mutation-order speciation, defined as the evolution of reproductive isolation by the fixation of different advantageous mutations in separate populations experiencing similar selection pressures, that is, uniform selection. In essence, different populations find different genetic solutions to the same selective problem. In turn, the different genetic solutions (i.e., mutations) are incompatible with one another, causing reproductive isolation. During ecological speciation, different alleles are favored between two populations. By contrast, during mutation-order speciation, the same alleles are favored in both populations, but divergence occurs anyway because, by chance, the populations do not acquire the same mutations or fix them in the same order. Divergence is therefore stochastic, but the process involves selection, and, thus, is distinct from random genetic drift. Selection can be ecologically based under mutation-order speciation, but ecology does not favor divergence as such, and an association between ecological divergence and reproductive isolation is not expected. How might mutation-order speciation arise? Sexual selection may cause mutation-order speciation if reproductive isolation evolves by the fixation of alternative advantageous mutations in different populations living in similar ecological environments.


Furfural

Genotoxicity

Furfural was tested negative in Drosophila heritable translocation assay and in multiple strains of Salmonella reverse mutation tests except for a mild positive with TA100. However, positive results were reported in several in vitro assays with mammalian cells in the absence of metabolic activators. These include thymidine kinase gene mutation in mouse lymphoma, chromosomal aberrations in Chinese Hamster Ovary and V79 cells, and spindle fiber damage in cultured human blood cell. Furfural was found to induce breakage in calf thymus DNA in vitro by reacting primarily with the adenine–thymine base pairs. In addition, furfural induced nonspecific DNA damage in rec-assay with Bacillus subtilis and sister chromatid exchange (SCE) in human lymphocytes in the absence of metabolic activation.

In vivo, intraperitoneal injection did not result in SCE and chromosomal aberrations in mouse bone marrow cells. However, ras oncogene activation was found in liver tumors of furfural-treated B6C3F1 mice.


RESULTS

Evolutionary EST screen identifies candidate female reproductive genes:

We constructed a cDNA library from dissected D. simulans female reproductive tracts minus ovaries. Ovaries were excluded because they express a diverse array of transcripts important for embryonic development, and we wished to enrich our cDNA library for candidate molecules expressed in, or secreted from, reproductive epithelia. We performed a differential hybridization screen of our cDNA library with 32 P-labeled cDNA made from whole adult male D. simulans. Low and nonhybridizing clones were selected for further analysis to enrich the collection of ESTs to be analyzed for those with predominant expression in female reproductive tracts (although transcripts expressed at low levels in both sexes are still present). It is important to note the possibility that not all proteins important in reproduction are female specific or enriched. As such, our approach may have screened out some non-sex-specific genes whose products, in females, interact with male proteins. However, the need to screen out abundant general molecules like actin, tubulin, etc., made it critical to include this step in our screen. We selected 960 clones for sequencing. Of these, we were able to obtain sequence reads of 𾄀 bp for 908 clones. These were used for further analyses.

The 908 ESTs corresponded to 526 independent genes. We focused on genes predicted to encode extracellular or cell surface molecules, since they could potentially be receptors or binding partners for Acps or sperm or be involved in male-independent extracellular processes. We used a bioinformatics approach to identify genes encoding proteins with a predicted secretory signal sequence and/or transmembrane domains. The identification of a signal sequence relies on a correct prediction of the first coding exon. Since initial exons are notoriously difficult to predict (D avuluri et al. 2001) and some proteins have internal secretory signals, we also included genes containing one or more predicted transmembrane regions. Thirty-five encoded proteins with a predicted signal sequence and transmembrane domain, 75 had just a predicted signal sequence, and 59 had predicted transmembrane domains but no predicted signal sequence.

Several male reproductive proteins show the molecular signature of adaptive evolution, and several hypotheses to account for that rapid evolution would predict a similar pattern for the female proteins with which they interact. We thus incorporated evolutionary information into our screen by deriving our ESTs from D. simulans, which allows comparison of them to their putative orthologs in the completed D. melanogaster genome. We then calculated the rate of synonymous (silent, dS) and nonsynonymous substitution (amino acid replacement changes, dN) using maximum-likelihood methods (G oldman and Y ang 1994). The average dN/dS ratio of the 461 protein-coding ESTs is 0.15 ± 0.25 (with the average dN being 0.013 and dS being 0.091).

The signature of adaptive evolution is a dN/dS ratio significantly exceeding 1, as equal numbers of nonsynonymous and synonymous substitutions, normalized to the number of possible nonsynonymous and synonymous changes in the gene, are expected under strict neutrality. Our goal here, however, is to use a genomic screen to identify candidate genes that have been subjected to positive selection, possibly at only a small subset of their codons. For example, mammalian egg coat proteins (ZP proteins) have an overall dN/dS ratio of 𢏀.5, but upon detailed analysis incorporating variation in the dN/dS ratio between sites using maximum likelihood (Y ang et al. 2000) it can be demonstrated that these genes are subjected to positive selection (S wanson et al. 2001b) with a class of codons having a dN/dS ratio > 1. We therefore surveyed the literature for articles utilizing the method of Y ang et al. (2000) for detecting adaptive evolution through analysis for variation in the dN/dS ratio between sites. We have plotted the proportion of genes with evidence of positive selection in relation to their overall dN/dS ratio in Figure 1 . At a dN/dS ratio Ϡ.5, 19 of 20 genes analyzed showed statistical evidence for adaptive evolution, suggesting this may be a reasonable value to identify candidate genes that may have been subjected to adaptive evolution. The genes in Figure 1 that fall between a dN/dS ratio of 0.3𠄰.5 also include a high proportion that show statistical evidence for adaptive evolution upon closer examination; however, these may be overrepresented in our analyses due to the lack of reports detailing negative results (and thus they are not included in our analysis). The genes, references, and summary information are available as online supplementary material at http://www.genetics.org/supplemental/ for the 70 genes analyzed in Figure 1 . Although only 25% of the 70 genes reported failed to show statistical evidence for adaptive evolution in subsequent PAML analysis, the proportion of genes under positive selection is surely overestimated due to the lack of reports that failed to detect adaptive evolution. Nonetheless, genes with an overall dN/dS ratio Ϡ.5 are more likely to have been subjected to adaptive evolution and are thus good candidates for further study. In our EST screen, 27 out of the total of 461 protein-coding genes have dN/dS ratios Ϡ.5 ( Figure 2 ), including eight of the candidate receptor proteins (containing signal sequences and/or transmembrane regions; Table 1 ).

Analysis of 70 genes, from published research articles on detecting adaptive evolution by analysis for variation in the dN/dS ratio between sites by the method of Y ang et al. (2000). Additional information and references can be found as online supplementary material at http://www.genetics.org/supplemental/.

Plot of dN vs. dS for the 461 D. simulans ESTs that matched protein-coding regions of D. melanogaster genes. The solid line is the neutral expectation of dN/dS = 1. The dashed line is the cutoff of dN/dS = 0.5 used to identify candidate genes that may have been subjected to positive selection.

TABLE 1

Classification of ESTs based upon evolutionary and bioinformatics analyses

ClassificationNo. genesNo. cdsdSdNdN/dSNo. with
dN/dS > 0.5
SS and TM0.1020.0090.14
SS0.1110.0230.23
TM0.0990.0150.13
All candidates combined1691520.1050.0170.17
Noncandidates3573090.0840.0110.1419
All5264610.0910.0130.1527

SS, signal sequence; TM, transmembrane region; All candidates combined, those with SS and/or TM domains; Noncandidates, lacked TM and/or SS domains; All, all genes identified in the EST screen; No. cds, the number of ESTs containing protein-coding sequence.

Some of the genes identified by this female reproductive tract evolutionary EST approach have predicted ORF sequences consistent with likely functions for Drosophila reproductive proteins. Sixteen predicted peptidases and eight predicted protease inhibitors were found. At least two Drosophila male seminal fluid proteins that are transferred to females undergo proteolytic cleavage (M onsma et al. 1990; B ertram et al. 1996), and in at least one case this cleavage is dependent on contributions from the female as well as the male (P ark and W olfner 1995). Although the nature of the female contribution is unknown, it could involve proteases (to cleave) and protease inhibitors (to confine cleavage to appropriate sites in the protein) such as the predicted ones identified here. Additionally, there are 47 different proteins with putative transporter activity and 11 different putative signal tranducer genes that could be involved in regulating the mated female's physiology ( Table 2 ). For example, it has been hypothesized that a transporter moves the Acp70a (sex peptide) from the reproductive tract to the hemolymph, where it binds receptors in the nervous system of the female (D ing et al. 2003). Finally, there are several genes predicted to be involved in defense or immunity. These candidates are all prime targets for functional analyses. A summary of the molecular functions based upon the gene ontology classification (A shburner et al. 2000) is provided in Table 2 . Details of all genes identified in our screen can be found as online supplementary material at http://www.genetics.org/supplemental/.

TABLE 2

Gene Ontology Functions

Molecular functionNo. from 526
independent genes
Unclassified184
Catalytic activity148
Binding
Transport activity
Structural molecule
Enzyme regulator
Transcription regulator
Signal transducer
Translation regulator ਉ
Chaperone activity ਉ
Antioxidant activity ਄
Motor activity ਃ
Defense/immunity protein ਂ
Unknown ਂ
Cell adhesion ਁ
Apoptosis regulator ਁ
Protein tagging ਁ

Divergence and polymorphism studies demonstrate adaptive evolution:

The evolutionary EST approach utilized here (isolating ESTs from one organism and comparing to the completed genome of a close relative; S wanson et al. 2001a) is aimed at identifying candidate genes for further tests for adaptive evolution. Each individual prediction of adaptive evolution needs to be independently verified. To test if any of the candidate genes identified herein have actually been subjected to positive selection, we performed a polymorphism survey of nine of the genes from D. melanogaster and D. simulans isofemale lines isolated from Maryland ( Table 3 ) and divergence analyses on five of the same genes in five to eight Drosophila species ( Table 4 ). Genes were chosen on the basis of predicted extracellular localization of the protein they encode and/or overall dN/dS ratio Ϡ.5. For the polymorphism survey, we analyzed the frequency spectrum (i.e., analysis of proportion of alleles at high vs. low frequencies) of the polymorphisms for departures from equilibrium neutral expectations (A quadro 1997). In particular, we analyzed for an excess of rare alleles (i.e., singletons; T ajima 1989; F u and L i 1993) or an excess of high-frequency-derived polymorphisms (F ay and W u 2000). Either pattern could have resulted from a recent selective sweep or a population bottleneck. To maximize the power of our statistical tests, we focused our analyses on intron regions, which should maximize variation within and between species under neutrality. We ruled out any genome-wide confounding effects, such as demographics (e.g., population bottleneck), on these statistics, since three loci ( Table 3 ) and additional unpublished studies of these samples (C. F. A quadro , unpublished results) conform to equilibrium neutral expectations. We performed polymorphism surveys for nine loci and found evidence for selective sweeps at six of these loci ( Table 3 ), suggesting the recent action of positive selection at or near these genes. Our results are bolstered by finding evidence for recent selective events using multiple statistics that utilize different regions of the frequency spectrum (i.e., high and low frequency). The genes under positive selection by this analysis include two putative proteases, a predicted transmembrane receptor, and three genes with unknown function.

TABLE 3

Polymorphism survey identifies positive selection in several candidate genes

D. melanogasterD. simulans
GenebpGO functionEST
dN/dS
RationaleNπTaj. DF&L DF&W HNπTaj. DF&L DF&W H
CG16705822Protease0.30SS270.0070.10.5𢄠.4110.0110.00.5𢄢.1
CG17108731Unknown0.36SS310.002𢄡.6*𢄣.0*0.80.004𢄡.9*𢄠.3𢄧.0*
CG4928750TM receptor0.03TM340.002𢄡.30.2𢄤.1*0.013𢄠.7𢄤.3*𢄠.1
CG10200716Unknown1.26SS, dN/dS230.010𢄡.9*𢄢.9*𢄤.40.0080.3𢄠.11.0
CG16707830Unknown1.38SS, TM, dN/dS190.003𢄡.7*𢄣.3*1.20.004𢄡.2*𢄠.1𢄢.7*
CG7415859Protease0.05Function190.001𢄡.8*𢄢.2*0.60.007𢄠.7𢄡.3𢄠.34
CG8827753Protease0.03SS340.0080.61.30.9120.019𢄠.30.23.2
CG11390793Ligand carrier0.04SS250.007𢄠.4𢄡.20.40.005𢄠.2𢄠.91.1
CG3066788Protease0.17SS, TM130.007𢄠.3𢄠.3𢄤.2*110.0130.20.03.1

bp, number of base pairs sequenced; GO function, gene ontology function (A shburner et al. 2000); EST dN/dS, dN/dS ratio from EST screen; N, number of individuals sequenced; π, nucleotide diversity; Taj. D, Tajima's D, F&L D, Fu and Li's D with outgroup; F&W H, Fay and Wu's H. *P < 0.05. Rationale indicates why the gene was investigated: SS, signal sequence; TM, transmembrane; dN/dS, dN/dS > 0.5; and/or Function, predicted function.

TABLE 4

Detection of positive selection by maximum-likelihood analysis

M0 vs. M3 M7 vs. M8
GeneGO functionSpeciesdN/dSpsωpsωM8 vs. M8A:
probability
CG4928TM receptorere, eug, lut, mel, pse, sim, tei0.00.06**0.240.370.0
CG10200Unknownere, eug, mel, sim, tei, yak0.40.20***1.460.061.80.34
CG16707Unknownere, mel, sim, tei, yak0.50.09**5.50.09*5.5π.01
CG7415Proteaseere, lut, mel, sim, tei, yak0.10.01***1.20.140.3
CG3066Proteaseere, eug, lut, mel, pse, sim, tei, yak0.10.04***1.80.03*2.0π.05

GO function, gene ontology (A shburner et al. 2000) function; Species, species from the set D. erecta (ere), D. eugracilis (eug), D. lutescens (lut), D. melanogaster (mel), D. pseudoobscura (pse), D. simulans (sim), D. teissieri (tei), and D. yakuba (yak); dN/dS, estimate of dN/dS assuming no rate heterogeneity; M0 vs. M3, M3 parameter estimates of dN/dS in the highest site class (ω) and the proportion of sites (ps) estimated to belong to that class; M7 vs. M8, M8 free parameter estimate of dN/dS (ω) and proportion of sites (ps) estimated to belong to that class; M8 vs. M8A: probability, probability that the dN/dS in model 8 is significantly ϡ. *P < 0.05; **P < 0.01; ***P < 0.001.

For the divergence studies, we sequenced from several additional Drosophila species five of the genes identified from our polymorphism analysis as having evidence for a recent selective sweep in D. melanogaster and/or D. simulans. We then analyzed the sequence data using maximum-likelihood methods (N ielsen and Y ang 1998; Y ang et al. 2000) to detect variation in the dN/dS ratio between sites. Divergence analyses were not performed on CG17108 due to the biased amino acid and codon usage seen in this gene, which may induce errors in parameter estimations using codon models. Whereas the polymorphism-based tests are capable of detecting recent selection in a single species, the divergence analyses can detect repeated episodes of positive selection on the same codons in several species. A significant result using these latter methods suggests that a subset of codons in a gene has been subjected to positive selection in several species. We find evidence of variation in the dN/dS ratio for all five genes using the discrete model M3. Four of these genes have a class of sites with dN/dS > 1. These four genes are still considered as only candidates for adaptive evolution since using a discrete model with three classes of dN/dS ratios compared to a single overall average dN/dS ratio is not a robust test of adaptive evolution (S wanson et al. 2001b) and should be considered as only a test for variable dN/dS ratios between sites. Using a more refined test with a beta distribution of dN/dS for “neutral” or functionally constrained codons that covers the interval 0𠄱, we find evidence of positive selection acting upon a subset of codons for two of the five genes studied ( Table 4 ). In both cases the sites in this extra class have dN/dS ratios significantly ϡ, since a model (M8) with a freely estimated extra class is significantly better than a model where the extra class has a dN/dS ratio fixed at 1 (M8A; Table 4 ). One gene (CG3066) is a predicted trypsin-like serine protease. Several of the codons inferred to be under positive selection in this gene lie within the predicted trypsin catalytic domain. Furthermore, several putatively selected codons lie in the predicted clip domain, which may be involved in protein-protein interactions (J iang and K anost 2000). The second gene (CG16707) does not belong to any predicted functional class.


Discovery of the Crystal-Stellate System and Their Structural Organization

The crystal-Stellate genetic system was discovered by studying the testes of D. melanogaster males with a missing Y chromosome (X/0) using the phase contrast microscopy. Crystalline aggregates of star-like and needle-like shape were found in the nuclei and cytoplasm of primary spermatocytes in these testes (Meyer et al., 1961). It was later shown that the testes of X/0 males also exhibited defects in the condensation and segregation of meiotic chromosomes, such as frequent chromosome non-disjunctions, and X/0 males were sterile (Lifschytz and Hareven, 1977 Hardy et al., 1984).

Now it is established that the crystal-Stellate genetic system contains several interacting loci mapping to the X and Y chromosomes. The Y chromosome of D. melanogaster is completely heterochromatic and contains only a few genes, mainly responsible for male fertility (Charlesworth, 2001 Hoskins et al., 2002 Chang and Larracuente, 2019). The first uncovered locus of the crystal-Stellate system was mapped to the h11 region of the mitotic prometaphase map of the Y chromosome. The loss of this locus or even its partial deletion was found to be sufficient for the crystal accumulation in spermatocytes (Hardy et al., 1984). Thus, the locus was named crystal (cry), but later it was renamed to Suppressor of Stellate [Su(Ste)] (Hardy et al., 1984). Along with the generation of crystalline aggregates in the testes of males with a deficiency in the cry locus (X/Y cry 1 ), similar defects of chromosome condensation and segregation with the X/0 male testes were found (Palumbo et al., 1994).

The components of the system include two Stellate (Ste) loci, one of which resides in euchromatic cytolocation 12E1-2 of the X chromosome, whereas the other is mapped to pericentromeric heterochromatin of the X chromosome (the h26 region of the mitotic prometaphase map) (Hardy et al., 1984 Livak, 1984 Palumbo et al., 1994 Tulin et al., 1997). Molecular analysis revealed that the crystal and Stellate loci consist of multiple homologous tandemly repeated sequences (Livak, 1984, Figure 1 ). The severity of meiotic abnormalities, abundance and shape of crystals in the cry 1 testes have been shown to depend on the Ste allele (Livak, 1984 Palumbo et al., 1994). The low-copy Ste + allele contains a small number of Stellate repeats (15� copies) and leads only to the appearance of little needle-like aggregates, weak meiotic disturbances and reduced male fertility, whereas the high-copy Ste allele (150� copies) leads to the formation of a multitude of crystals, visible under phase contrast as star-shaped entities, severe meiotic defects and complete sterility. Non-disjunction of the XY- and 2nd chromosomes, fragmented chromatin, and chromatin bridges have been found among the intrinsic meiotic defects in the testes of cry 1 males. However, in the examined natural and laboratory lineages of D. melanogaster, the Ste+ alleles significantly predominate over the Ste ones (Palumbo et al., 1994). The severity of male fertility defects and the degree of meiotic disorders are associated with the number of Stellate copies and independent from the ratio of euchromatic and heterochromatic Stellate repeats. The boundary for fertility is considered to be 50� Stellate copies the presence of more copies in the genome leads to complete male sterility (Palumbo et al., 1994). Stellate genes are expressed in the testes of cry 1 males as 750 nt polyadenylated transcripts (Livak, 1990), and their abundance is proportional to the number of repeats in both Stellate loci (Palumbo et al., 1994). In the cry 1 testes Stellate transcripts from both loci are translated generating small proteins of about 17� kDa, which have homology with the regulatory β-subunit of protein kinase CK2, CK2β (Livak, 1990 Bozzetti et al., 1995 Egorova et al., 2009 Olenkina et al., 2012b). Stellate proteins, products of the heterochromatic and euchromatic clusters, possess high intra-cluster homogeneity, having minor differences in amino acid sequence between themselves and slightly different electrophoretic mobility (Olenkina et al., 2012b). Immunostaining of the cry 1 testes with anti-Stellate antibodies reveals that Stellate is main or the only component of crystalline aggregates (Bozzetti et al., 1995 Egorova et al., 2009, Figures 2A,D ). In wild-type flies, Stellate gene expression is strongly suppressed and no Stellate proteins are detected ( Figure 2B ).

General scheme of Stellate and Su(Ste) repeats. Promoters are indicated by a blue color bar, introns are depicted by green lines, intergenic spacers are depicted by gray lines. Stellate gene contains an ORF (brown color bar) and two introns (green lines). An individual Su(Ste) repeat carries the region homologous to the Stellate ORF (brown color bar), Y-specific region (orange bar) and an insertion of the defective hoppel transposon (violet bar) flanked by inverted repeats (not shown) in the promoter. Start sites of sense transcription of Stellates and Su(Ste) and multiple starts of antisense Su(Ste) transcription within the body of hoppel are indicated by black arrows [modified from Aravin et al. (2001)].

Distribution of derepressed Stellate protein in the testes of D. melanogaster. (A𠄼) Internal confocal slices of stained testis preparation of cry 1 males (A,C) and wild-type control (B). Testes were immunofluorescently stained with anti-Stellate (green) and anti-lamin (red) antibodies, chromatin was stained with DAPI (cyan). Anti-lamin staining indicates nuclear membrane position. (A,C) Diffuse Stellate signals in the nuclei (arrows in A) and bright needle-like and dot-like crystalline Stellate aggregates mainly in the cytoplasm are seen in spermatocytes of cry 1 males. (C) The nuclei of mature spermatocytes. (D) 3D reconstruction of the stained testis preparation of cry 1 males. (A𠄼) are reproduced from Figure 2 in Egorova et al. (2009). (D) is reproduced from Figure 2 in Kibanov et al. (2013) by permission of Elsevier (Licenses ## 4913121387410 and 4913131090753).

The organization of the Su(Ste) locus has also been studied in detail. According to previously published data in most laboratory strains of D. melanogaster the number of Su(Ste) repeats comprises about 80 copies, whereas in natural populations, strains with 240 repeats were found (Lyckegaard and Clark, 1989 Balakireva et al., 1992 McKee and Satter, 1996). However, recent Y chromosome assembly using the Iso1 strain of D. melanogaster with improved annotation of both protein-coding genes and repeats contains more than 500 Su(Ste) copies (Chang and Larracuente, 2019). The size of a typical complete Su(Ste) repeat is about 28 000 nt. It consists of three main parts: the region homologous to Stellate gene, the AT-rich region specific for the Y chromosome, and the insertion of transposable element hoppel (1360) into the promoter sequence ( Figure 1 ). Su(Ste) repeats are transcribed and processed to polyadenylated mRNAs, however, unlike Stellate transcripts, they contain numerous frameshift mutations due to point mutations and deletions (Balakireva et al., 1992 Shevelyov, 1992). Translation products of Su(Ste) repeats are not detected. The insertion of a defective transposon hoppel copy is responsible for antisense transcription of Su(Ste) repeats (Aravin et al., 2001).


Testing the Role of Reproductive Isolation in Speciation Dynamics

Because reproductive isolation can be quantified, it is possible to directly test whether it is a rate-limiting control on taxonomic speciation rates ( Rabosky & Matute, 2013). All else being equal, lineages that evolve reproductive isolation more quickly should be characterized by faster rates of speciation. As a thought experiment, consider two distinct species, X and Y, such that X belongs to a clade of organisms that can evolve reproductive isolation rapidly, and Y belongs to a clade where reproductive isolation evolves slowly. Suppose that a geological event splits both species X and Y into two populations: X1 and X2, and Y1 and Y2. After an equivalent amount of time has elapsed, populations X1 and X2 would show greater reproductive isolation than populations Y1 and Y2. If the rate at which reproductive isolation evolves is the rate-limiting control on speciation rates, then the lineage to which species X belongs should, over long timescales, speciate more rapidly than the lineage of Y. If another factor is the rate-limiting control on speciation rates, then the realized rate of speciation will be independent of the rate at which reproductive isolation evolves.

This logic forms the basis of a statistical test for the contribution of any form of reproductive isolation to macroevolutionary speciation rates. One can quantify the rate at which particular components of reproductive isolation evolve in different clades or lineages ( Fig. 3) and test whether variation in the rate of evolution of reproductive isolation predicts speciation rates. Possible relationships between these quantities are shown in Figure 4. The key advantage of this approach is that it avoids assumptions about the presumed effects of particular organismal traits on the evolution of reproductive isolation ( Coyne & Orr, 2004) and estimates parameters of the process directly. This test has been applied to birds and to drosophilid flies aiming to test whether the rate at which lineages acquire postzygotic genetic incompatibilities (e.g. alleles that cause interspecies hybrids to be sterile or inviable) is associated with speciation rates. Although individual clades of both birds and flies varied with respect to the rate at which they evolved at least one component of reproductive isolation, this variation was unrelated to taxonomic speciation rates ( Rabosky & Matute, 2013). However, the results reported by Rabosky & Matute (2013) should be interpreted with circumspection, given uncertainties in quantifying speciation rate variation and the rate at which reproductive isolation evolves. For example, the biology of intrinsic reproductive isolation in drosophilid flies has been studied by dozens of researchers over much of the past century, generating perhaps the highest-resolution dataset on reproductive isolation for any group of organisms ( Yukilevich, 2012). However, our understanding of taxonomic speciation rates in the drosophilidae is poor: indeed, it is possible that hundreds or thousands of distinct drosophilid taxa remain to be described ( Markow & O'Grady, 2006). Such taxonomic inadequacy has implications for the speciation rates used by Rabosky & Matute (2013). Similarly, our analyses of avian postzygotic isolation were largely based on a single compilation of avian hybrids ( Gray, 1958) and we had no direct information on premating isolation for birds.

Pairwise postzygotic isolation from interspecific crosses of birds as a function of the pairwise genetic distance between them. Results are shown for two major clades (pheasants, Phasianidae parrots, Psittacidae). For a given level of genetic divergence, pheasants show greater levels of postzygotic isolation than the parrots, indicating that this sort of reproductive isolation accumulates more quickly in pheasants than in parrots. If intrinsic postzygotic isolation (hybrid inviability and sterility) is the dominant control on speciation rates, pheasants should have faster rates of speciation than parrots. Note that relationships are bounded at 0 (all hybrid offspring fully viable and fertile) and 1 (no offspring produced, or all offspring sterile). Lines show fitted linear relationships between reproductive isolation and genetic distance for each clade. Data are from Price & Bouvier (2002) and Gray (1958) analyses are from Rabosky and Matute (2013). For this pair of clades, speciation rates are faster in the clade with faster rates of evolution of reproductive isolation (pheasants: speciation = 0.26 lineages Myr –1 parrots: speciation = 0.22 lineages Myr –1 ). However, across all birds, these quantities appear to be unrelated.

Pairwise postzygotic isolation from interspecific crosses of birds as a function of the pairwise genetic distance between them. Results are shown for two major clades (pheasants, Phasianidae parrots, Psittacidae). For a given level of genetic divergence, pheasants show greater levels of postzygotic isolation than the parrots, indicating that this sort of reproductive isolation accumulates more quickly in pheasants than in parrots. If intrinsic postzygotic isolation (hybrid inviability and sterility) is the dominant control on speciation rates, pheasants should have faster rates of speciation than parrots. Note that relationships are bounded at 0 (all hybrid offspring fully viable and fertile) and 1 (no offspring produced, or all offspring sterile). Lines show fitted linear relationships between reproductive isolation and genetic distance for each clade. Data are from Price & Bouvier (2002) and Gray (1958) analyses are from Rabosky and Matute (2013). For this pair of clades, speciation rates are faster in the clade with faster rates of evolution of reproductive isolation (pheasants: speciation = 0.26 lineages Myr –1 parrots: speciation = 0.22 lineages Myr –1 ). However, across all birds, these quantities appear to be unrelated.

Some possible relationships between the rate at which lineages evolve reproductive isolation and their rate of speciation. A, direct correspondence, where the evolution of reproductive evolution shows a one-to-one relationship with the macroevolutionary rate of speciation. In this scenario, the evolution of reproductive isolation is the exclusive determinant of macroevolutionary speciation dynamics. B, offset/dampened relationship, where the rate of evolution of reproductive isolation is the dominant control on speciation rates, although speciation rates are lower than predicted by the rate of evolution of reproductive isolation alone. This relationship implies that many populations evolving reproductive isolation fail to persist through deep time. C, decoupled, such that reproductive isolation shows no predictive relationship with macroevolutionary speciation rates. This scenario suggests that reproductive isolation is not the rate-limiting control on the rate of speciation. Adapted from Rabosky (2013).

Some possible relationships between the rate at which lineages evolve reproductive isolation and their rate of speciation. A, direct correspondence, where the evolution of reproductive evolution shows a one-to-one relationship with the macroevolutionary rate of speciation. In this scenario, the evolution of reproductive isolation is the exclusive determinant of macroevolutionary speciation dynamics. B, offset/dampened relationship, where the rate of evolution of reproductive isolation is the dominant control on speciation rates, although speciation rates are lower than predicted by the rate of evolution of reproductive isolation alone. This relationship implies that many populations evolving reproductive isolation fail to persist through deep time. C, decoupled, such that reproductive isolation shows no predictive relationship with macroevolutionary speciation rates. This scenario suggests that reproductive isolation is not the rate-limiting control on the rate of speciation. Adapted from Rabosky (2013).

Coyne & Orr (2004) distinguished between two temporal aspects of the speciation process: the ‘biological speciation interval’ (BSI), or the waiting time between the origin of new reproductively isolated lineages, and the ‘transition time for biological speciation’, or the amount of time required for strong reproductive isolation to evolve once the evolution of isolation has begun. The biological speciation rate is simply the inverse of the biological speciation interval (1/BSI). Coyne & Orr (2004) suggested that there is little reason to expect equivalence between transition times and biological speciation intervals. However, the rate at which reproductive isolation evolves can still be the rate-limiting step on speciation rates even if transition times are much shorter (or longer) than BSIs. For example, the occurrence of partial intrinsic postzygotic isolation between populations might trigger reinforcement, such that complete prezygotic isolation evolves rapidly in response to maladaptive hybridization ( Servedio & Noor, 2003 Matute, 2010). As such, the rate-limiting step on taxonomic speciation rates can still be the rate at which the initial postzygotic isolation arises, even if it is the subsequent evolution of premating isolation that ultimately drives speciation to completion. This process would potentially be testable by developing more sophisticated modelling frameworks that enable researchers to distinguish between lineage-specific differences in the rate at which any measurable reproductive isolation arises (e.g. duration of the lag phase Mendelson, Inouye & Rausher, 2004) from the rate at which strong reproductive isolation arises.

A desirable feature of the approach illustrated in Figure 4 is that it provides a fairly direct test of the contribution of reproductive isolation to taxonomic speciation rate. As such, the approach can be contrasted with phylogenetic comparative methods for identifying correlations between specific organismal traits and diversification rates. Numerous studies have found at least some association between traits and diversification rates ( Coyne & Orr, 2004 Jablonski, 2008 Ng & Smith, 2014). Such associations can arise if the traits under consideration increase the rate at which reproductive isolation evolves ( Panhuis et al., 2001 Coyne & Orr, 2004), to which we might add: ‘provided that the rate of evolution of reproductive isolation is the rate limiting step on taxonomic speciation rates’. However, demonstration that a particular trait is correlated with taxonomic speciation rate does not necessarily imply that the underlying mechanism involves the effects of the trait on reproductive isolation, even if we assume that the trait influences reproductive isolation. Because of the complex ways in which traits can influence metapopulation dynamics (Levin, 2000), we should be cautious in assuming that any particular traits (e.g. sexual dichromatism in animals, floral characteristics, etc.) influence species richness through their effects on reproductive isolation.

It is also important to recognize the limitations of the approach illustrated in Figure 4. The lack of a relationship between a particular control (e.g. intrinsic postzygotic reproductive isolation) and speciation rates should not be interpreted as evidence that the control is irrelevant to speciation. It simply means that the control does not determine the rate at which speciation occurs the control may nonetheless be an integral part of the speciation process. Furthermore, observing that one component of reproductive isolation fails to predict speciation rates provides no information about the importance of other forms of reproductive isolation for speciation rates. Finally, the quality of the available phylogenetic, taxonomic, and reproductive isolation data limit the use of this framework in practice.

Other conceptual tools may provide insight into the role of splitting and persistence controls on speciation rates. The protracted speciation model (Etienne & Rosindell, 2012) is an important theoretical framework for understanding how the origination, extinction, and persistence of incipient species influence the shapes of phylogenetic trees. Recently, Etienne et al. ( 2014) developed a form of the protracted speciation model that could be fitted to phylogenetic datasets, potentially enabling researchers to estimate parameters associated with both the rate of incipient species formation and the time required for successful speciation. Conceivably, extensions of this general framework may be developed into a formal test of the relative importance of these and other controls on taxonomic speciation rates.


Discussion

We describe a new model of the evolution of reproductive isolation of hybrid populations, a first step towards hybrid speciation. Unlike previous models of hybrid speciation, our model does not assume positive selection on hybrid genotypes or inbreeding, but rather deterministic selection against hybrid incompatibilities in randomly mating hybrid populations. With moderate selection (i.e. s≤0.2) on two or more incompatibility pairs in an allopatric hybrid population, reproductive isolation from both parental species emerges with

50% (or higher) probability. Hybrid reproductive isolation also evolves frequently with substantial levels of ongoing migration between hybrids and parental species (4Nm < 20 each parent).

Another striking result of our simulations is the speed with which reproductive isolation evolves between hybrids and parental species. Depending on parameters, reproductive isolation can emerge in fewer than 100 generations with moderate selection (S3 Text). The idea that hybrid speciation can occur rapidly has been supported by experimental results [14, 63, 64] and to some extent by previous models of hybrid speciation [9, 14]. Our model suggests that simple selection on incompatibilities in hybrid populations could also lead to rapid reproductive isolation on timescales much faster than expected for allopatric speciation due to the accumulation of neutral BDM incompatibilities. Given that epistatic incompatibilities are common, our results on the probability and speed of isolation suggest that this process may frequently occur in hybrid populations.

Previous empirical work has emphasized the importance of ecological differentiation between hybrid and parental populations or positive selection on hybrid genotypes as a route to hybrid reproductive isolation [6, 8–10, 12, 63, 65]. The novel finding of our simulations is that reproductive isolation evolves readily in hybrid populations without positive selection on hybrids. However, the two are not mutually exclusive and ecological factors, which have been shown to underlie several cases of hybrid speciation [6, 8, 63], may complement selection on genetic incompatibilities to further strengthen reproductive isolation. For example, in Helianthus, a combination of chromosomal rearrangements and novel hybrid phenotypes are important in distinguishing hybrid and parental species [6, 66]. Like other models ([9, 14]), our model predicts that isolation between hybrids and parental species is inherently weaker than isolation between the two parental species. We propose that fixation of incompatibilities could be a crucial step in initially limiting gene flow between hybrids and parental species, allowing for the development of other isolating mechanisms. For example, theoretical work predicts that reinforcement can develop even when selection against gene flow is moderate [67–70].

Previous models of hybrid speciation have incorporated species-specific inversions that are assumed to be underdominant. Under this “underdominant inversion” model, hybrid populations can fix for novel inversion combinations, resulting in isolation between hybrid and parental species [15]. Simulation results under this model have suggested that inbreeding [14] or positive selection on hybrid genotypes [9, 14] is important for the evolution of hybrid reproductive isolation. However, past simulation efforts focused on hybrids in a tension zone, either with no spatial isolation from parental species [14] or with high migration rates from parental species [17]. To investigate the dynamics of the underdominant inversion model in situations where migration is more restricted, we simulate the underdominant inversion model in an isolated hybrid swarm scenario that is similar to our epistatic incompatibility model (S7 Text). Interestingly, we find that isolation evolves frequently under this model even without positive selection (

40% of simulations, see S7 Text). These results show that, in hybrid-dominated populations, the inversion model has similar behavior to our model of selection against negative epistatic interactions (S7 Text). Which mechanism of isolation is more prevalent in hybrid populations will depend on the frequency of hybrid incompatibilities of each type. Empirical evidence suggests that while underdominance can be a common isolating mechanism in plants (reviewed in [21]), negative epistatic interactions may be a more common mechanism of reduced hybrid fitness in animals [24].

It is important to note several factors that may influence how common our epistatic interactions model of hybrid speciation will be in natural populations. First, our model assumes that hybrids are abundant in a population and, while this appears to be reasonably common (see S6 Text S9 Table), this is clearly not a feature of all hybrid zones. We also note that our model only represents fitness in terms of genetic incompatibilities and that hybrid populations can have lower fitness as a result of ecological or sexual selection. For example, in our simulations, we assumed random mating between hybrids and parentals. But when parental species exert negative sexual selection against hybrids, hybrid populations are significantly more likely to be outcompeted by parentals (S10 Table). There is substantial variation in the mating preferences of parentals for hybrids [71]. In two species of cyprinidontiform fishes, male and female parentals mate readily with hybrids [45, 72, 73], while mice discriminate against them [74]. This suggests that the likelihood of this process will depend in part on the biology of the hybridizing species.

An additional consideration is that hybrid reproductive isolation is most likely to evolve during a particular window of divergence between parental species. When the fitness of hybrid populations is low (i.e. corresponding to high levels of divergence between parental species), they are more prone to extinction or displacement by parentals (S6 Fig., S5 Text). This suggests that the evolution of hybrid reproductive isolation through this mechanism is most likely to occur in a period of evolutionary divergence during which species have accumulated some hybrid incompatibilities but have not diverged to the point at which hybrids are largely inviable. The most detailed work characterizing genetic incompatibilities has been between Drosophila species, where hybrids generally have substantially reduced fitness compared to parents [56, 57, 75]. Hybrids between several other species studied to date, however, are affected by fewer incompatibilities or incompatibilities of weaker effects [26, 55, 59, 76–79]. Such groups may be more likely to form hybrid populations, and should be the focus of future empirical research. In addition, even species that currently have strong isolation may have historically produced hybrid populations, though investigating ancient hybrid speciation by the mechanism we describe would be challenging. This is because if parental and hybrid lineages have diverged substantially since the time of initial hybridization it may not be possible to determine whether or not incompatibilities were initially derived from parental genomes.

It is interesting to note that reduced frequency of reproductive isolation with increasing selection on hybrids can be mitigated to some extent by an increase in the total number of hybrid incompatibility pairs. In our simulations, we see a positive relationship between the number of interactions and the probability of developing reproductive isolation, and a negative relationship between the total strength of selection on hybrids and the probability of developing reproductive isolation (Figs. 3 and S6). This tradeoff suggests that reproductive isolation can evolve between hybrid and parental populations even when the fitness of hybrids is low (as in Figs. 3, 4, and S6, keeping in mind that extinction occurs frequently when hybrid fitness is nearly zero).

Similarly, our model is sensitive to skewed initial admixture proportions, but increasing the number of hybrid incompatibility pairs increases the probability that skewed hybrid populations will be isolated from both parental species by at least one incompatibility (S7 Fig.). For example, with two incompatibility pairs, the probability of isolation from both parental species in an ancestry-skewed population (65% parent 1) was 7% while with four incompatibility pairs the probability rose to 15%. In addition, because discrete populations in a cline often span a range of admixture proportions (e.g. [80–82]), it is likely that some hybrid populations will fall in the range where we predict that isolation can evolve. On the other hand, our results show that high levels of migration (as might be observed in continuous clines) can prevent isolation future research should investigate the dynamics of this process in a range of hybrid zone structures.

Finally, our model assumes that coevolving incompatibilities or BDM incompatibilities arising from adaptive evolution frequently occur between species. Accumulating evidence suggests that incompatibilities arising from coevolution may be common [30, 36, 83–86]. For example, in marine copepods, coevolution between cytochrome c and cytochrome c oxidase results in a reciprocal breakdown of protein function in hybrids [86]. In addition, the fact that many known incompatibility genes involve sexual conflict, selfish genetic elements, or pathogen defense suggests an important role for coevolution in the origin of incompatibilities [36, 83, 87, 88]. Our model also applies to BDM incompatibilities that arise due to within-lineage adaptation, assuming that the fitness advantage of the derived alleles is not dependent on the parental environment. It is currently unknown whether incompatibilities are more likely to be neutral or adaptive. Though there is evidence for asymmetric selection on many hybrid incompatibilities [28, 29, 89], neutrality has not been established in these cases. Anecdotal evidence supports the idea that adaptive incompatibilities are common, since many of the genes underlying hybrid incompatibilities identified so far show evidence of positive selection within lineages [90], but the relative frequency of adaptive and neutral BDM incompatibilities awaits answers from further empirical research. Intriguingly, theoretical work also suggests that neutral BDM incompatibilities are unlikely to persist if there is gene flow between species [32].

The patterns predicted by our model are testable with empirical approaches. A large number of studies have successfully mapped genetic incompatibilities distinguishing species [25, 26, 41, 56, 57, 79, 91]. Ancestry at these sites can be determined in putative hybrid species, and the relative contribution of parental-derived incompatibilities to reproductive isolation can be determined experimentally. For some species, it may be possible to evaluate the dynamics of incompatibilities relative to the genetic background in experimentally generated hybrid swarms [92]. We predict that many hybrid populations exhibiting postzygotic isolation from parental species will have fixed incompatibility pairs for each parental species. Several cases of hybrid speciation report reduced fitness of offspring between parental and hybrid species consistent with the mechanism described here [6, 16, 53, 93] and are promising cases for further empirical research. Strikingly, a recent study on Italian sparrows concludes that reproductive isolation between parental and hybrid species is partly due to the fixation of parental-derived incompatibilities [94].

An intriguing implication of our model is that independently formed hybrid populations between the same parental species can develop reproductive isolation from each other. The likelihood of this outcome increases with the number of incompatibility pairs. In sunflowers, empirical studies of ecologically-mediated hybrid speciation have identified multiple hybrid species derived from the same parental species [95]. It is interesting to note that selection against hybrid incompatibilities could generate the same pattern in replicate hybrid populations. In fact, this mechanism could generate a species phylogeny pattern similar to that expected from an adaptive radiation, with multiple closely related species arising in a relatively short evolutionary window. This finding is striking because our model does not invoke adaptation and suggests that non-adaptive processes (i.e. selection against incompatibilities) could also explain clusters of rapidly arising, closely-related species.


MATERIALS AND METHODS

CDNA library preparation:

Total RNA was purified by the guanidinium isothiocyanate/CsCl method (M ac D onald et al. 1987) from 600 female reproductive tracts minus ovaries (oviducts, uterus, parovaria, spermathecae, and seminal receptacle) that had been dissected from D. simulans of mixed aged adult flies from a bottle culture. mRNA was purified using QIAGEN (Valencia, CA) oligotex spin columns. Oligo(dT)-primed cDNA was synthesized using superscript reverse transcriptase and cloned into the pCMV-Sport6 vector (Invitrogen, San Diego). We did not perform in-solution subtractive hybridization or normalize the cDNA library because these methods typically result in truncated cDNAs, and we desired full-length cDNA for our evolutionary comparisons. The resulting library contained 130,000 CFUs, of which 99% were recombinant. The average insert size was 1.2 kb. Two sets of probes were utilized for differential hybridization. First, oligo(dT)-primed first-strand male cDNA was prepared from mixed age and mating status whole adult male D. simulans flies using Bethesda Research Laboratories (Gaithersburg, MD) superscript II reverse transcriptase incorporating 32 P-labeled dCTP and then denatured at 65° for 30 min in 0.3 m NaOH. Second, a random-primed probe was generated from a mixture of RT-PCR products from the three female yolk protein genes from D. melanogaster: YP1, YP2, and YP3 (B arnett et al. 1980). These genes were screened out of the library since yolk protein RNAs are abundantly expressed in the fat body, which is associated with the reproductive tract (B arnett et al. 1980) (they are also expressed in the ovary, which was removed). Hybridization was for 18 hr at 65° in 5× SSPE, 5× Denhardt's, 0.5% SDS, 0.2 mg/ml salmon sperm DNA. Final washes were at 65°, 0.1× SSPE for 10 min. Sequencing was from QIAGEN purified plasmid DNA using ABI big dye terminator sequencing chemistry analyzed on an ABI 3100 automated sequencer. EST sequences are deposited in GenBank under accession nos. CO391819, CO392724, CO408479, and CO408480.

Polymorphism survey:

DNA was extracted using the PureGene DNA isolation kit from isofemale lines of D. melanogaster and D. simulans previously collected by C. Aquadro in Beltsville, Maryland. To maximize the power of our statistical tests, we focused our analyses on intron regions, which should maximize variation within and between species under neutrality. PCR primers and conditions are available as online supplementary material at http://www.genetics.org/supplemental/. PCR products were diluted eightfold with water and sequenced directly using ABI big dye terminator sequencing chemistry and analyzed on an ABI 3100 automated sequencer. Sequences are deposited in GenBank under accession nos. AY665365, AY665366, AY665367, AY665368, AY665369, AY665370, AY665371, AY665372, AY665373, AY665374, AY665375, AY665376, AY665377, AY665378, AY665379, AY665380, AY665381, AY665382, AY665383, AY665384, AY665385, AY665386, AY665387, AY665388, AY665389, AY665390, AY665391, AY665392, AY665393, AY665394, AY665395, AY665396.

Divergence study:

We assessed DNA sequence divergence among five to eight increasingly divergent species of Drosophila for five genes. For each we used either all or overlapping subsets of the following species: D. erecta, D. eugracilis, D. lutescens, D. melanogaster, D. pseudoobscura, D. simulans, D. teissieri, and D. yakuba (detailed in results ). We used two tree topologies [differing only in the placement of D. erecta (K o et al. 2003)] and the results were consistent. The two topologies were: (pseudoobscura, lutescens, (eugracilis, (erecta, ((teissieri, yakuba), (melanogaster, simulans))))) and (pseudoobscura, lutescens, (eugracilis, ((erecta, (teissieri, yakuba)), (melanogaster, simulans)))). Sequences for D. melanogaster and D. pseudoobscura were obtained from public databases (http://genome.ucsc.edu/). Stocks for the other species (except our own D. simulans) were obtained from the Drosophila Species Stock Center in Tucson, Arizona. Since the analyses are based upon coding regions, we amplified the coding sequence from cDNA. Total RNA was extracted from mixed-age females using Trizol Reagent (Invitrogen). Random decamer primed cDNA was synthesized using MMLV-Reverse Transcriptase (Ambion, Austin, TX). Primers were designed in conserved regions of the genes of interest, which were identified by aligning the D. melanogaster gene sequences with their tblastn best hits in the genome of D. pseudoobscura. PCR primers and conditions are available as online supplementary material at http://www.genetics.org/supplemental/. PCR products were purified using the QIAquick PCR purification kit (QIAGEN) and sequenced using an ABI 3700 sequencer (Macrogen). Sequences are deposited in GenBank under accession nos. AY665365, AY665366, AY665367, AY665368, AY665369, AY665370, AY665371, AY665372, AY665373, AY665374, AY665375, AY665376, AY665377, AY665378, AY665379, AY665380, AY665381, AY665382, AY665383, AY665384, AY665385, AY665386, AY665387, AY665388, AY665389, AY665390, AY665391, AY665392, AY665393, AY665394, AY665395, AY665396.

Evolutionary and bioinformatic analyses:

The D. simulans EST sequences were aligned against the D. melanogaster predicted coding sequences, and the alignment was used to calculate dN/dS ratios using the maximum-likelihood methods (G oldman and Y ang 1994) implemented in the program PAML (Y ang 2000). Assessment of the significance of excess dN over dS was determined as follows. dN and dS were estimated as two free parameters by maximum likelihood (L1). The likelihood was also calculated for the null model having dN equal to dS (L0). The negative of twice the difference in the log-likelihood obtained from these two models (−2[log(L0) − log(L1)]) was compared to the chi-square distribution with 1 d.f. For the polymorphism survey, Tajima's D (T ajima 1989), Fu and Li's D (F u and L i 1993), and Fay and Wu's H (F ay and W u 2000) were calculated using DnaSP4.0 (R ozas and R ozas 1999). Significance was determined by coalescent simulations with R (recombination) estimated from the data by the method of H udson (1987). These three statistics for polymorphism data analyze the frequency of alleles (frequency spectrum) within the sample. The departures from neutrality include an excess of rare alleles (T ajima 1989 F u and L i 1993) or an excess of high-frequency-derived alleles (F ay and W u 2000). These specific departures are expected to be associated with recent selection acting at or near a locus. During a selective sweep, in the presence of recombination, linked variation is dragged toward fixation, resulting in an excess of high-frequency-derived mutations in regions flanking the target of selection. The fixation of the favored variant results in the elimination of polymorphism at sites immediately surrounding the selected site (size of region is dependent upon recombination and the strength of selection). As new mutations occur in this region after the sweep and drift upward in frequency, there is an initial excess of rare alleles since every new mutation produces a new allele. The time to return to an equilibrium frequency distribution is a function of the population size and can be quite slow for large populations.

For the divergence analyses, we used PAML (Y ang 2000) to calculate the likelihood of a neutral model where no codons could have a dN/dS ratio > 1 (L0) and compared it to the likelihood of a model in which a subset of sites could have a dN/dS ratio > 1 (L1) (Y ang and B ielawski 2000). The negative of twice the difference in the log-likelihood obtained from these two models (−2[log(L0) − log(L1)]) was compared to the chi-square distribution with degrees of freedom equal to the difference in number of estimated parameters. Variation in the dN/dS ratio between sites was modeled using both discrete (PAML models M0 and M3) and β-(PAML models M7 and M8) distributions. We consider the comparison of model M0 and M3 to be a test for variation in the dN/dS ratio between sites and not a robust test of adaptive evolution. The comparison of M7 and M8 is a robust test of adaptive evolution. To determine if the dN/dS ratio significantly exceeds 1, we compared the M8 model to the likelihood of a model (M8A) with the additional proportion of sites fixed at a dN/dS ratio of 1 (S wanson et al. 2003). Details of the distributions and test statistics can be found in Y ang et al. (2000). Signal sequences were predicted using the program SignalP (http://www.cbs.dtu.dk/services/SignalP-2.0/ N ielsen et al. 1997). Transmembrane regions were predicted using the TMHMM methods (S onnhammer et al. 1998), using the TMHMM server (http://www.cbs.dtu.dk/services/TMHMM-2.0/).


RESULTS

The mean flowering-initiation dates (±1 SE) at the two different sites were as follows: dry site: I. fulva—March 30 (±1.4 days) BCIF—April 10 (±0.4 days) F1—April 14 (±0.7 days) BCIB—April 23 (±0.4 days) I. brevicaulis—May 2 (±0.4 days) wet site: I. fulva—March 30 (±1.4 days) BCIF—April 9 (±0.4 days) F1—April 11 (±0.7 days) BCIB—April 22 (±0.4 days) I. brevicaulis—May 1 (0.7 days). Results of two-way ANOVA indicated that there was a main effect for “cross type,” no main effect for “site,” and no “site × cross type” interaction (Table 1). Examining the significant main effect (cross type) further, posthoc Tukey HSD tests accounting for multiple comparisons revealed that all cross types were statistically different from one another with respect to flower initiation (corrected P-values <0.05 for all 10 comparisons).

Results of a two-way analysis of variance on “date of first flower” conducted in two separate field sites

We tested whether the flowering-date phenotypes of F1 and reciprocal backcross hybrids deviated significantly from an additive model of genetic inheritance. Under a purely additive model, the expected mean flowering date of F1 hybrids would be the mean flowering date of both pure-species parents, and F1 hybrids did deviate from this additive model, but only with marginal significance (linear contrasts, F = 2.91, 1 d.f., P = 0.088, Figure 2). For BCIF hybrids, given strict additivity, the expected mean flowering date of BCIF hybrids would be 0.75 × the mean flowering date of I. fulva + 0.25 × the mean flowering date of I. brevicaulis. The BCIF hybrids did in fact deviate significantly from this null additive model, with BCIF hybrids flowering only 3 days earlier than F1 hybrids, on average, but 11.5 days later, on average, than pure I. fulva species (linear contrast, F = 4.44, 1 d.f., P = 0.035, Figure 2). The expected mean flowering date of BCIB hybrids, given a purely additive genetic model of inheritance, would be 0.75 × the mean flowering date of I. brevicaulis + 0.25 × the mean flowering date of I. fulva, and the BCIB hybrids did not significantly deviate from this model (linear contrast, F = 0.063, 1 d.f., P = 0.802, Figure 1).

Observed vs. expected mean flower initiation dates (±2 SE) for I. fulva, I. brevicaulis, F1, BCIF, and BCIB hybrids. The diagonal line denotes the expected mean flowering date under an additive model of gene action.

Quantitative trait locus analysis:

Using CIM followed by refinement with MIM, we identified 17 QTL in the BCIF mapping population that affected flowering time in one or more field habitats or greenhouse years (Table 2, Figure 3). In this backcross population, LG1 possessed four QTL: three of the QTL had negative effects, meaning that the introgressed I. brevicaulis alleles caused flowering time to occur earlier. The confidence intervals of these three QTL were overlapping, suggesting that the genes causing earlier flowering times could be the same across the three separate habitats (greenhouse season 2002, wet and dry field sites Table 2, Figure 3). A fourth QTL was detected at the topmost portion of the linkage group, and this QTL had a positive effect, meaning that the introgressed I. brevicaulis alleles caused flowering time to occur later (in the dry field site Table 2, Figure 3). LG12 was the only other linkage group in the BCIF mapping population that revealed a QTL with negative effect (i.e., introgressed I. brevicaulis alleles caused flowering to initiate earlier in the dry field site). In contrast, a QTL that caused a later flowering time in the wet field site was also detected on LG12. These two QTL, given their opposite effects, are unlikely to be due to the same genes. The remaining 11 QTL detected in the BCIF mapping population caused flowering to initiate later. These loci included single QTL (i.e., affecting flowering time in only one of the four greenhouse or field habitats) found on LG2, LG6, LG8, LG9, and LG13 and two overlapping QTL (affecting flowering time in two of the four greenhouse or field habitats) found on LG5, LG7, and LG11. Four epistatic interactions were detected between QTL using MIM methodologies: between QTL 2 and 3 in the 2002 greenhouse study between QTL 2 and 3 and 3 and 4 in the dry site and between QTL 1 and 6 in the wet site (Table 2). In all instances, epistatic interactions between two alleles resulted in flowering time occurring later than expected, given a purely additive model.

Linkage map of dominant I. brevicaulis IRRE retrotransposon display markers segregating in the F1 hybrid used to produce BCIF hybrids. Significant QTL for flowering phenology are denoted (with 2-LOD confidence intervals) to the right of the linkage groups. Red bars represent regions where introgressed I. brevicaulis alleles caused flowering to initiate earlier, while blue bars represent regions where introgressed I. brevicaulis alleles caused flowering to initiate later. QTL analyses were performed during the two greenhouse years 2002 and 2003 as well as in two field plots (dry and wet) in 2006.

QTL for flowering initiation in two field sites (dry and wet) and in two greenhouse seasons (2002 and 2003) detected in a BC1 (BCIF) population

In the BCIB mapping population, we utilized CIM followed by MIM and found a total of 15 QTL that affected flowering time (Table 3, Figure 4). Only 4 of the 15 detected QTL had positive effects (i.e., caused later flowering when I. fulva alleles were present). Two of these QTL were detected in the 2002 greenhouse study and were located on LG6 and LG15. The other two positive QTL were detected in the wet field site and were located on LG9 and LG13. All other QTL in the BCIB mapping population were negative i.e., they caused earlier flowering when introgressed I. fulva alleles were present. Two overlapping, negative QTL were located on LG2 and were detected in the 2002 greenhouse sample and the dry field site. Another two overlapping QTL were located on LG7 and were detected in the 2002 and 2003 greenhouse studies. The remaining negative QTL were found singly (i.e., affected flowering time in only one of the four greenhouse or field habitats) on LG4, LG5, LG10, LG16, LG17, LG20, and LG21. Three epistatic interactions were detected in the BCIB mapping population: between QTL 2 and 3 and between QTL 4 and 5 in the 2002 greenhouse study and between QTL 1 and 4 in the wet field site (Table 3). The epistatic interactions detected in the greenhouse study caused an earlier flowering time when both I. fulva alleles at the different QTL were present. In contrast, the epistatic interaction detected in the wet site caused flowering time to occur later than expected in comparison to a purely additive model.

Linkage map of dominant I. fulva IRRE retrotransposon display markers segregating in the F1 hybrid used to produce BCIB hybrids. Significant QTL for flowering phenology are denoted (with 2-LOD confidence intervals) to the right of the linkage groups. Red bars represent regions where introgressed I. fulva alleles caused flowering to initiate earlier, while blue bars represent regions where introgressed I. fulva alleles caused flowering to initiate later. QTL analyses were performed during the two greenhouse years 2002 and 2003 as well as in two field plots (dry and wet) in 2006.

QTL for flowering initiation in two field sites (dry and wet) and in two greenhouse seasons (2002 and 2003) detected in a BC1 (BCIB) population


Chemical Cues that Guide Female Reproduction in Drosophila melanogaster

Chemicals released into the environment by food, predators and conspecifics play critical roles in Drosophila reproduction. Females and males live in an environment full of smells, whose molecules communicate to them the availability of food, potential mates, competitors or predators. Volatile chemicals derived from fruit, yeast growing on the fruit, and flies already present on the fruit attract Drosophila, concentrating flies at food sites, where they will also mate. Species-specific cuticular hydrocarbons displayed on female Drosophila as they mature are sensed by males and act as pheromones to stimulate mating by conspecific males and inhibit heterospecific mating. The pheromonal profile of a female is also responsive to her nutritional environment, providing an honest signal of her fertility potential. After mating, cuticular and semen hydrocarbons transferred by the male change the female’s chemical profile. These molecules make the female less attractive to other males, thus protecting her mate’s sperm investment. Females have evolved the capacity to counteract this inhibition by ejecting the semen hydrocarbon (along with the rest of the remaining ejaculate) a few hours after mating. Although this ejection can temporarily restore the female’s attractiveness, shortly thereafter another male pheromone, a seminal peptide, decreases the female’s propensity to re-mate, thus continuing to protect the male’s investment. Females use olfaction and taste sensing to select optimal egg-laying sites, integrating cues for the availability of food for her offspring, and the presence of other flies and of harmful species. We argue that taking into account evolutionary considerations such as sexual conflict, and the ecological conditions in which flies live, is helpful in understanding the role of highly species-specific pheromones and blends thereof, as well as an individual’s response to the chemical cues in its environment.

This is a preview of subscription content, access via your institution.