This page has been archived and is no longer updated
Genes and speciation
Author: Chung-I Wu
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"What is a species? How do new species come into existence? These questions are some of the most enduring in biology and remain controversial today. Under many commonly accepted species definitions, speciation can be viewed as the process by which two identical populations diverge genetically to the point at which their subsequent merger would not be possi- ble. Species are therefore both genetically distinct and independent. Although distinctness is often observ- able (in morphology, for example), independence usually is not. How do species evolve to become phenotypically distinct? What are the underlying genes? What forces drive their divergence between species? Some insights into these questions might come from studying genes that cause reduced fitness in hybrids that are intermedi- ate in phenotype between two species. The fitness reduction can range from ecological maladaptation or behavioural aberration to inviability or sterility. The loci that underlie such reductions in fitness might be considered ?speciation genes?, which are important in driving the nascent species to become independent genetic entities. Below, we review recent progress in the characterization of speciation genes. Operationally, species are often delineated by distinct phenotypes, such as distinct plumage in birds. It was the introduction of the concept of reproductive isolation (RI) that redirected the emphasis to the independent nature of species 1?4 .According to Mayr 3 ,species are ?groups of interbreeding natural populations that are reproductively isolated from other such groups? 3 .RI therefore refers to the independence of gene pools, among which new mutations and allele frequency changes are not shared. A central question about RI is whether this independence, or non-sharing, should apply to every locus in the genome. Those who argue for the primacy of RI in specia- tion 1,3 are essentially arguing for a ?whole-genome? con- cept 5 (see BOX 1). If we apply the concept of RI to only a portion of the genome, where would we draw the line? Does it make sense to say that 75% of the genome is reproductively isolated? Indeed, in this NEO-DARWINIAN view of RI, genetic changes between species are seen to be so strongly CO-ADAPTED that few genes can be inte- grated into the genome of another species. So, almost all regions in the genome are either part of such a ?cohesive? network or are closely linked to an element in such a network. In either case, gene flow across nascent species boundaries is effectively eliminated 6 . By contrast, the alternative is a genic view of specia- tion, as explained in FIG. 1 (see also REFS 5,7). Although rarely recognized as such, these different perspectives on the genetic architecture of species differences are the gen- esis of the long-running debate on the geographical mode of speciation; that is, whether speciation most commonly occurs when the diverging populations are in allopatry, parapatry or sympatry (see BOX 1). The central issue is whether two populations can evolve into good species while they continue to exchange genes during the process. Under the whole-genome view that gives pri- macy to RI in speciation 3 ,gene flow between diverging populations would have such negative effects on those populations that strict geographical barriers would have to be the prelude to speciation. The alternative genic GENES AND SPECIATION Chung-I Wu* and Chau-Ti Ting ? It is only in the past five years that studies of speciation have truly entered the molecular era. Recent molecular analyses of a handful of genes that are involved in maintaining reproductive isolation between species (speciation genes) have provided some striking insights. In particular, it seems that despite being strongly influenced by positive selection, speciation genes are often non-essential, having functions that are only loosely coupled to reproductive isolation. Molecular studies might also resolve the long-running debate on the relative importance of allopatric and parapatric modes of speciation. NEO-DARWINIAN (SYNTHESIS) The modern theory of evolution that combines both natural selection and population genetics, in which the Darwinian concept of spontaneous variation is explained in terms of mutation and genetic recombination. CO-ADAPTATION Selection by which harmoniously interacting genes accumulate in the gene pool of a population. *Department of Ecology and Evolution, University of Chicago, Chicago, Illinois 60637, USA. ? Department of Life Science, National Tsing Hua University, Hsinchu, Ta iwan 300, Republic of China. Correspondence to C.-I W. e-mail: ciwu@uchicago.edu doi:10.1038/nrg1269 REVIEWS 114 | FEBRUARY 2004 | VOLUME 5 www.nature.com/reviews/genetics NATURE REVIEWS | GENETICS VOLUME 5 | FEBRUARY 2004 | 115 REVIEWS INTROGRESSION The integration of a genomic region from one species into the genome of another species. Even a few percent of the introgressed genome can lead to hybrid incompatibility. Here,we review the exciting new molecular studies that are beginning to clarify the genetic basis of specia- tion. First, we discuss studies that are starting to show how many genes underlie RI and how they interact together. We then go on to summarize some important recent studies that have begun to shed light on the nature of a ?speciation gene? before discussing attempts to assess the population-genetic mode of speciation on a genome-wide scale. The genetic architecture of speciation The number of genes underlying RI and the interactions among them define the genetic architecture of speciation. Studies of Drosophila species have provided the most extensive data on the genetic architecture of RI and, therefore, we focus on these studies (see REFS 9?11 for more detailed reviews). In general, even closely related sibling species of Drosophila show extensive functional-genetic differentiation. For example, standard quantitative trait locus analysis indicates that there are approximately 20 genes that control the difference in the shape of the genital arch between D. simulans and D. mauritiana 12 . view (FIG. 1), in which a subset of genes determines RI or differential adaptation and which allows the possibility of gene flow between nascent species, underlies the concept of parapatric speciation. Although the idea that most speciation events occur in allopatry has dom- inated thinking in the second half of the last century, the genetic evidence has been marginal. Support has come mainly from ecological or biogeographical data (see Howard and Berlocher 8 for a review). Importantly, at the genic level, allopatric speciation provides a well- defined null hypothesis. For the first time, we might have the wherewithal to test this model definitively, owing to the genomic tools that have been developed in the past five years. So, even an apparently ecological issue, such as whether speciation most frequently occurs in allopatry or in parapatry, depends on which of the genic or whole- genome views of speciation is most correct. Speciation research has just crossed the threshold into the molecular era. It is these molecular studies that will ultimately resolve whether the genic view or the whole-genome view provides a more accurate representation of speciation. Box 1 | The concept and classification of reproductive isolation The basic concept Reproductive isolation (RI) is a population-genetic concept that refers to the non-exchange of genes between two species that are in contact with each other. The cessation of gene flow is the result of the genetic properties of the two species in question, and not the result of extrinsic barriers that prevent contact. One common mistake is to equate RI with a reduction in fitness in the hybrids between two forms. As long as some hybrids can still pass genes back and forth between the two forms, even at a very low rate, there is effectively only one gene pool. Furthermore, RI is a whole-genome concept; if two diverging populations still share a common gene pool for some parts of the genomes, these populations cannot be considered to be reproductively isolated (but see Mayr 73 for an alternative view). Pre-mating versus post-mating isolation The mechanisms of RI can be divided into those that act before and those that act after mating. Pre-mating isolation refers to the absence of interspecific hybrids because the two species do not mate owing to ecological or behavioural factors. In post-mating isolation, members of the two species do mate but the hybrids are inviable, sterile or ecologically maladapted. Obviously, pre-mating isolation does not incur as much cost in reproductive efforts as post-mating isolation does. Mechanisms of post-mating isolation can be further divided into those that act before fertilization (pre-zygotic) and those that act after fertilization (post-zygotic). The geographical modes of speciation In allopatric speciation, the diverging populations have to be geographically separated with no gene flow between them for divergence to take place (see also FIG. 3). Parapatric speciation refers to populations that are usually geographically separated, but connected by gene flow, during speciation (see also FIG. 3). Sympatric speciation is the extreme form of parapatric speciation, with almost unrestricted gene flow and few differences in the ecological niches of the diverging populations. The genetic basis of RI There are three broad classes of genetic element that contribute to RI: extra-chromosomal, chromosomal and genic. Extra-chromosomal elements include cytoplasmic symbionts and transposable elements that can cause RI. Chromosomal elements include rearrangements that can cause aberrant meioses in interspecific hybrids. However, the most common mode of chromosomal speciation is polyploidization, in which the genome size is doubled by having two sets of the same genome (autotetraploidy) or by having two sets of different genomes (allotetraploidy). Genic RI is caused by the incompatibilities between the genes of the diverging species (see also BOX 2). The definition of a speciation gene In the genic view of speciation 5 ,speciation genes are those that contribute to RI, often in the form of hybrid inviability, sterility or behavioural aberration. This definition can include genes that cause isolation owing to physiological, behavioural or even ecological factors. For example, an INTROGRESSION hybrid might be viable and fertile but less cold-resistant. The underlying genes would still fail to successfully migrate across the incipient species boundary. 116 | FEBRUARY 2004 | VOLUME 5 www.nature.com/reviews/genetics REVIEWS in RI is an important point to keep in mind. After all, the function of RI genes could not possibly be to steril- ize their carriers. In this way, the analysis of RI is not unlike the study of graft rejection in organ transplanta- tion: the biology of the major histocompatability com- plex, which underlies graft rejection, certainly has much wider and more profound implications than the phe- nomenon of graft rejection itself. In modelling the evo- lution of RI, the practice has been to consider only the RI phenotype without addressing the underlying func- tion (for example, see BOX 2). The reason for such a glar- ing omission is clear ? the identities of speciation genes, and so, their normal functions, have not been known until very recently. Now, there are a handful of studies in which the identities of speciation genes have been shown: each of which we discuss. By definition, a speciation gene is one that can be shown to cause some degree of ecological, sexual or post-mating isolation between young, or even nascent, species. Although there have been other claims of speciation genes being identified, we consider the five that we discuss to be the only studies that have truly identified the molecules involved. There are many excel- lent studies that focus on genes that differ between species and the molecular interactions between them but that do not address their phenotypic effects on the whole organism (for example, see REF. 23). Until these further studies have been done, such genes cannot be classified as speciation genes. Similarly, at least 15 genes control the differences in mating behaviour between two behavioural races in D. melanogaster 13,14 .However, in at least one study of the differences between Drosophila species, the genetics of sexual isolation seem to be much simpler, with only a few loci being involved 15 (see REF. 13 for a discussion of this contrast). The most extensive studies of the genetic architecture that underlies RI have focused on hybrid male sterility between D. simulans and D. mauritiana and hybrid invia- bility between the more distantly related D. simulans and D. melanogaster.To dissect the genetic architecture of RI, such studies attempt to POSITIONALLY CLONE genes that are involved in hybrid incompatibility (see FIG. 2). The num- ber of genes that contribute to hybrid male sterility between D. simulans and D. mauritiana is estimated to be at least 50 when introgressed in the homozygous state 10 .This type of architecture can be characterized as ?weak effect?strong interaction? ? that is, each individ- ual gene has little effect on its own, but in combination, these genes cause reproductive incompatibility 10 . Functional studies of one of these interacting genes (OdsH;see below) have indeed shown that it has an extremely weak effect. Despite the weak individual effects of these genes, it was estimated that the hybrids between these species are still sterilized 15 times over 16 . It is important to point out that, in the above com- parisons between D. simulans and D. mauritiana, there is virtually no hybrid inviability despite extensive hybrid male sterility. The extreme rapidity with which hybrid male sterility has evolved, relative to hybrid inviability or female sterility, has become a hallmark of post-mating isolation. Both the causes and the consequences are important 17,18 .The contrast in the genetic basis of hybrid sterility compared with hybrid inviability is notable not only for the difference in the number of loci involved but also in how the genes interact. Studies of two divergent species in which hybrid inviability has evolved (D. melanogaster and D. simulans) have shed more light on these differences 19 . DEFICIENCY MAPPING and ALLELE COMPLEMENTATION (FIG. 2b) were used to study hybrid male sterility 20,21 and hybrid inviability 19,22 between these species. Intriguingly, the complementation test showed that no single locus rescued fertility of the hybrid male. On the other hand, single genes with a substantial effect seemed common for hybrid inviability (see also below). In summary, with respect to RI between closely- related species, Drosophila studies have shown that genetic differentiation is often extensive. Although other studies have indicated a much simpler genetic architec- ture outside Drosophila,the apparent discrepancy might be the result of a difference in the mapping resolution 10 . The molecular genetics of speciation To understand the molecular basis of RI, three key ques- tions need to be addressed. First, what genes contribute to RI? Second, what are the normal functions of those genes? Third, how did these normal functions diverge among different populations, leading to RI? The fact that the genes that underlie post-mating isolation must have normal functions that are distinct from their role POSITIONAL CLONING The procedure by which we identify and isolate genes on the basis of their location in the genome, involving detailed genetic and physical maps of chromosomes. DEFICIENCY MAPPING Uses chromosomes that have different sections deleted to locate the position of a gene of interest. Without the deficiency, the normal functional gene usually masks the effect of (that is, complements) the defective or foreign copy that we wish to identify. ALLELE COMPLEMENTATION A test of whether a wild-type phenotype can be restored with two given alleles in a diploid genome. a b c d Figure 1 | The genic view of species differentiation. The two horizontal bars represent the genomes of two diverging populations. When they start to differentiate, only a few loci (indicated by black lines) are differentially adapted and genes at such loci are not exchanged between populations (a). Gene flow continues in the rest of the genome (arrows). Although the regions of differential adaptation expand, the amount of gene flow between the two genomes is gradually reduced owing to linkage with such regions (indicated in red/purple) (b,c) until the two populations are completely reproductively isolated and are therefore considered to be separate species (d). Modified with permission from REF. 5 � (2001) Blackwell Science. NATURE REVIEWS | GENETICS VOLUME 5 | FEBRUARY 2004 | 117 REVIEWS transmembrane growth factor of the RECEPTOR TYROSINE KINASE SUPERFAMILY that is important in signal transduc- tion. Its closest homologue in humans is the epidermal growth factor receptor (EGFR) 30 .All the features of Xmrk-2 are consistent with those of the dominant ONCOGENE that causes the melanomas in the hybrid fish. In particular, mutations at the Xmrk-2 locus abolish the Tu phenotype and the overexpression of Xmrk-2 gives rise to a high frequency of tumour formation. In the adjacent genomic region, another EGFR homologue, Xmrk-1, was found in all Xiphophorus fish. Xmrk-1 and Xmrk-2 are therefore duplicated genes. However, Xmrk-1 transcripts can be found in all tissues, whereas Xmrk-2 transcripts are only abundant in the melanomas of the hybrids. Xmrk-2 apparently originated from non-homologous recombination between Xmrk-1 and an adjacent D locus 31 .So, this hybrid locus has the regulatory region from the D locus and most of the coding regions from Xmrk-1.The R gene represses Xmrk-2 as well as the D locus. Another important difference between Xmrk-1 and Xmrk-2 is the two amino-acid replacements in the extracellu- lar domain, which shows ligand-independent acti- vation 32 .So,divergence after gene duplication is important in the differentiation of these species and this might be a common feature of speciation genes (see also below). Although the definition of speciation genes includes those with strong or weak effects on ecological, behav- ioural or physiological differences, most of these initial studies have concentrated on genes that have large effects on physiological characteristics. Four of these five exam- ples focus on post-mating isolation. The final example, which concerns ecological adaptation between behav- ioural races, is therefore of considerable interest. Melanoma formation in Xiphophorus species hybrids (Xmrk-2). Many species in the fish genus Xiphophorus have spots on their skin that are composed of black pig- ment cells. In interspecific hybrids between X. maculatus (platyfish) and X. helleri (swordtail), these spots some- times spontaneously develop malignant melanomas 24?26 . A two-locus Dobzhansky-Muller (DM)-type model (see BOX 2) has been proposed to explain the formation of malignant melanomas. In this model, overexpression of the Tu gene causes these melanomas to form. The second locus involved, called the R gene, is a suppressor that neg- atively controls Tu.The platyfish contains both Tu and R genes, whereas the swordfish contains neither. In the backcross F 2 hybrids, a quarter of the offspring produce melanomas owing to the presence of Tu but the absence of the R gene. The X-linked Tu locus was subsequently mapped to a candidate gene, Xmrk-2 (REFS 27?29). Xmrk-2 encodes a RECEPTOR TYROSINE KINASE SUPERFAMILY One of the important cell-surface receptors that interacts with water-soluble ligands. ONCOGENE A gene that induces uncontrolled cell proliferation. a Fertile Sterile M 5 M 1 M 2 M 3 M 4 M 6 M 7 M 8 M 5 M 1 M 2 M 3 M 4 M 6 M 7 M 8 b Inviable V iable Figure 2 | Positional cloning of hybrid incompatibility genes. a | Recombination mapping: the coloured horizontal bars indicate introgressed chromosomes. The vertical orange bar marks the position of the gene that is responsible for hybrid sterility (for example, the Drosophila gene OdsH). Different sterile and fertile hybrids are typed for markers on the introgressed chromosomes that are known to be polymorphic between the two species (M 1 ?M 8 ). The pattern of marker distribution among the different hybrids allows the location of the hybrid sterility gene to be mapped at a resolution that is dependent on the concentration of informative markers in the region. b | Deficiency mapping: the phenotypes that are associated with various overlapping chromosome deficiencies (indicated by the gaps in the bars) show the position (marked with the vertical orange bar) of the hybrid incompatibility gene (for example, the Drosophila gene Nup96). Viable and inviable hybrids with different chromosome deletions are typed for markers that are known to be polymorphic between the two species (M 1 ?M 8 ). The pattern of marker distribution among the different hybrids allows the location of the hybrid inviability gene to be mapped at a resolution that is dependent on the concentration of informative markers in the region. 118 | FEBRUARY 2004 | VOLUME 5 www.nature.com/reviews/genetics REVIEWS The most direct approach to assess the normal function and the phenotypic effect of a specific gene is to knock it out. Interestingly, the deletion of OdsH from D. melanogaster results in no obvious adverse phenotype 35 .So, at least at this crude level of observa- tion, OdsH is dispensable. However, a more detailed examination showed a subtle effect: males missing OdsH suffer a 40% fertility reduction when they are two days old and mate repeatedly. This fertility reduc- tion lessens to 20% and 8% in the next two days, also under sperm-exhaustion conditions. After five days, the role of OdsH in fertility enhancement vanishes. One interpretation of these findings is that the role of OdsH is to accelerate the maturation of sperm. So, only very young males under sperm-exhaustion con- ditions are affected. Comparative analyses indicate that OdsH was duplicated in the Drosophila lineage from a neuron- expressed gene, unc-4, after it diverged from the mos- quito lineage. Whereas unc-4 in Drosophila has not diverged much in either sequence or expression from the ancestral state in the common ancestor it shares with mouse and C. elegans, OdsH has changed in both sequence and expression. Specifically, OdsH has been evolving away from the unc-4 pattern of embryonic and neuronal expressions to a testicular role 35 . Because OdsH is divergently regulated between D. simulans and D. mauritiana, its expression in the testis of the sterile hybrids is highly misregulated. OdsH transcripts accumulate in very young spermatocytes. The pattern is not observed in either parental species or in the fertile introgression line, which differs from the sterile line by a 3-kb segment of OdsH (REF. 34).The expression of unc-4 in the sterile introgression line is also normal. Therefore, the divergence in the sequence and the expression of OdsH might both contribute to the hybrid sterility. Hybrid inviability in Drosophila species (Hmr). Another classical RI system in Drosophila is the hybrid incompat- ibility between D. melanogaster and D. simulans,two species that have been reproductively isolated more than 2.5 million years. Crosses between these species produce only inviable or sterile hybrids 36?38 .Five muta- tions that could rescue the inviable F 1 hybrid progeny have been found and several were elegantly character- ized for their MATERNAL or ZYGOTIC EFFECTS 39?43 .Given the multi-locus nature of hybrid incompatibility, it was surprising that such hybrid-rescue mutations could be identified. Among the hybrid-rescue mutations, the X-linked Hmr (hybrid male rescue) gene, which rescues the invi- able hybrid males, was mapped and cloned 44,45 . Hmr was identified as a transcription factor in the myeloblas- tosis family 44 .The primary amino-acid sequence of Hmr contains two DNA-binding protein motifs that indicate its role in transcription regulation. There were many amino-acid substitutions between the sibling species in the DNA-binding domains of Hmr. So, this pattern indicates that positive selection might drive the rapid evolution of Hmr.However, the Hmr mutation Xmrk-2 induces tumour formation only in the hybrids in which the R gene is absent, in accordance with the classical DM model of post-mating isolation. In other species, such as the medaka, overexpression of Xmrk-2 also causes embryonic lethality. The constitu- tively expressed Xmrk-2 activates a transcription factor, STAT5, and subsequently upregulates several down- stream targets 33 .So,overall, there is strong circumstan- tial evidence that Xmrk-2 is a speciation gene. However, the multiple alleles at each locus in the natural popula- tions of each species, all of which cause a different degree of hybrid phenotype, require further study. It is possible that some of these alleles are not becoming fixed but rather are simply deleterious mutations in the process of being removed by purifying selection. Those deleterious alleles that are destined to be removed would not contribute to species differentiation. Hybrid male sterility in Drosophila species (OdsH). The Odysseus (OdsH) gene from D. mauritiana causes com- plete male sterility when co-introgressed with the adja- cent segment into D. simulans.Genetic mapping of the male sterility locus (FIG. 2a) allowed the initial identification of OdsH as this RI locus 34 .Recent transgenic studies have con- firmed this identification 35 . OdsH is a homeobox gene from a family of transcription factor-encoding genes that are known to be slowly evolving. Curiously, OdsH has been evolving rapidly within the D. melanogaster subgroup even though its homologues from other species are extremely conservative. MATERNAL EFFECT The effect of the maternal genotype on the phenotype of the offspring, or the zygotes, usually at the embryonic stage (see also zygotic effects). ZYGOTIC EFFECT The effect of the zygotes? own genotype on their own phenotype. Box 2 | Dobzhansky-Muller model of hybrid incompatibility How does genic incompatibility between species evolve without simultaneously causing defects in pure species? A popular explanation is the Dobzhansky-Muller (DM) model of hybrid incompatibility. In the ancestral population, the genotype is AA BB.When the population is split into two, A evolves into a in one population and B evolves into b in the other. a and b are mutually incompatible. As the a?b interaction is not present in the pure species, the evolution of incompatibility is possible. Detailed genetic analysis on hybrid male sterility, however, has shown that hybrid incompatibility often involves conspecific genic interactions as well 9,10,18,20 .The DM model is far too simple in that respect. In the far right (hybrid), the divergence process is indicated by the black double- headed arrows and the incompatibility is indicated by the green double-headed arrow. The relationship between the black and green arrows should be the essence of the evolution of reproductive isolation. A deficiency of the DM model is that it does not consider the divergence process that is indicated by the black arrows but focuses instead on the incompatibility, which is a byproduct of that divergence. Fixation of new mutation Ancestral genotype New mutation Hybrid a B a B A b A b A b A B B BA a A AB B a B A b NATURE REVIEWS | GENETICS VOLUME 5 | FEBRUARY 2004 | 119 REVIEWS double-bond in a long chain of saturated hydrocarbons. Two independent studies have identified the gene that controls the (5,9)/(7,11) difference to be a desaturase gene, desat2 (REFS 48,49).Although CHs often act as con- tact pheromones between sexes, they have also been implicated in ecological adaptations, such as heat or starvation tolerance 50 . The desat2 gene apparently diverts the synthesis of 7,11-heptacosadiene into the 5,9-type. The loss of the promoter in the desat2 gene therefore results in the 7,11- type among the M flies. This observation raises the interesting possibility that loss of function of a gene has a role in this particular case of nascent speciation. The geographical distribution of the two desat2 variants (predominantly desat2 + in Africa and desat2 0 elsewhere) indicates that this strong differentiation must be main- tained by differential selective pressure. An excess of high-frequency nucleotide mutations highlighted the influence of positive selection on the desat2 polymor- phism 49 .Greenberg et al. 17,50 were able to show, by gene knock-out, that the loss of the desat2 gene (as in non- African M flies) results in an increase in cold tolerance and a decrease in starvation tolerance. It is plausible that, in the colder climate, a non-functional desat2 would spread through the cosmopolitan populations. So, this seems to be a case of ecological adaptation and differentiation. An interesting aspect of the Z?M differentiation is the unidirectional sexual isolation between these forms 51 .Zimbabwe females, in the presence of Z and M males, do not mate with M males. (Note that the observation by itself does not indicate male or female choice.) We know that at least seven or eight genes control female or male mating behaviour, respect- ively 13,14 .So, the question is whether desat2 is one of the loci that governs Z females? mating characteris- tics (for example, reduced attractiveness to M males). CH differences have been known to govern females? attractiveness in interspecific crosses 52 .However,it was widely thought that desat2 was not involved in female attractiveness in the Z?M system because Caribbean flies, which carry the African desat2 allele, behave like M flies. Nevertheless, recent observations have shown that, within three African populations, the presence of the African desat2 allele correlates nearly perfectly with Z-femaleness 53 . One possible interpretation of this pattern is that desat2 governs female attractiveness to M males and that the Caribbean population is an anomaly that results from recent admixture between African and North American flies. Although this interpretation seems to contradict the widely-accepted view that D. melanogaster males might not be discriminatory when choosing a mate 54 , new work indicates that M males might not court Z females as ardently as they court M females, espe- cially when the females are not highly receptive (C.-T.T. and C.-I W., unpublished observations). If this is the case, the desat2 gene might be playing a double role in this nascent speciation through differentiation in eco- logical adaptation and, secondarily, through mating preference. that rescued hybrid viability was a P-element insertion in its 5? region that resulted in a reduction in the amount of wild-type transcript. For Hmr to be consid- ered a true ?speciation gene?, it would be necessary to show that the D. simulans and D. melanogaster alleles are functionally divergent in their rescue effect of hybrid viability. A recent transgenic study indicates that this might indeed be the case (D. Barbash, personal communication). Hybrid inviability in Drosophila species (Nup96). Complementation mapping (FIG. 2b) has been used to analyse hybrid inviability between D. melanogaster and D. simulans.High-resolution mapping has allowed a spe- ciation gene, Nup96,to be cloned and characterized 22 . The Nup96 allele from D. simulans causes inviability in the F 1 hybrids if the copy from D. melanogaster is absent. Nup96, which has homologues in yeast, worm and human genomes, encodes a subunit of a nuclear-pore complex, which transports macromolecules between the nucleus and cytoplasm 46 and is therefore essential for viability in flies. An excess of non-synonymous substitutions in Nup96 between D. melanogaster and D. simulans relative to non-synonymous polymorphisms within these species (calibrated against synonymous changes with the MCDONALD AND KREITMAN TEST) indicated that this gene is under positive selection. With the sequences from D. mauritiana and D. yakuba, it was possible to map putative adaptive changes onto an evolutionary tree. Presgraves et al. 22 concluded that the adaptive changes occurred in the distant past, a suggestion that is corrob- orated by the analysis of the extant polymorphisms in D. melanogaster and D. simulans.Had some adaptive changes occurred recently, a reduction in the amount of neutral polymorphism, which might also be accom- panied by a skew towards very low- and/or very high- frequency variants, would have been expected. Neither was observed in Nup96. Not only were Presgraves et al. 22 able to map the Nup96 gene, but they were also able to locate the interact- ing locus in the DM model of hybrid incompatibility to the X chromosome. They did this by switching the source of the X chromosome in the hybrid males. One question to be addressed in the future is whether there are multiple loci on the X chromosome that interact with Nup96. Ecological/behavioural races in Drosophila melanogaster (desat-2). The final example of a proven speciation gene (under our broad definition; see BOX 1) provides a glimpse of the molecular genetics of ecological, and pos- sibly behavioural, isolation. D. melanogaster from central- southern Africa around Zimbabwe and those from the rest of the world (referred to as the Z and M types, respectively) have evolved to become different ecologi- cal/behavioural races. The females of African and cosmopolitan D. melanogaster carry different forms of a specific type of non-volatile CONTACT PHEROMONES. These two forms ? the 5,9-heptacosadiene and 7,11- heptacosadiene forms of the 27-carbon cuticular hydrocarbons (CH) 47 ? differ in the position of a MCDONALD AND KREITMAN TEST A test that contrasts interspecific divergence against intraspecific polymorphism. It is a powerful test to detect excess of non- synonymous substitutions between species. CONTACT PHEROMONES Chemical signals that are transmitted through the direct physical contact of two individuals. Contact pheromones in Drosophila are often sexual signals. 120 | FEBRUARY 2004 | VOLUME 5 www.nature.com/reviews/genetics REVIEWS FIG. 1). As discussed earlier under a neo-Darwinian scheme ? in which extrinsic (for example, geographi- cal) barriers to gene flow are necessary before the onset of speciation ? such a scheme, as depicted in FIG. 3c,d, is improbable. One approach to addressing the question of whether gene flow can occur during speciation (that is, parapatric speciation) was elegantly modelled by J. Hey, J. Wakeley and colleagues 60?63 .This approach considers both diver- gence between species and polymorphism within species. If there has been continual gene flow during speciation, the variance in the numbers of shared polymorphisms and fixed differences would increase above that predicted for speciation that has occurred in strict allopatry (see FIG. 1a,b). Machado et al. 63 used this approach to show that gene flow occurred between D. pseudoobscura and D. simulans during speciation (see also REF. 61). Although an analysis that incorporates within-species polymorphisms is useful for understanding the popula- tion genetics of speciation in Drosophila (for example, Kliman et al. 62 ), it might not always be so useful in other groups. For example, most polymorphisms in humans can actually be traced back to common ancestral alleles that can be dated to no more than a million years ago. Such polymorphisms can tell us nothing about the human/chimpanzee speciation event that took place more than five million years ago. Furthermore, data on within-species polymorphisms in closely-related species are usually not available for many loci. An approach that circumvents these problems focuses on one genome from each species, but looks at data from hundreds of loci, and asks the simple question: is the t value constant across a large number of genes (see FIG. 3)? In principle, t should be reflected in the amount of neutral divergence between the species. Therefore, we would expect greater variation in K s (number of syn- onymous substitutions per synonymous site) when spe- ciation occurs in parapatry (FIG. 3c,d) than when it occurs in allopatry (FIG. 3a,b).An outgroup can be used to cali- brate the divergence between the species and so account for any variation in K s owing to mutation or selection. The residual variation in K s across loci should then be a function of t and 2N e in FIG. 3 (REFS 55,57?59,64).The larger the variation in K s , the larger N e has to be to account for the observation when t is assumed to be a constant. Given the variation in K s between primates, it has been necessary to invoke a large N e for the ancestral human/chimpanzee population ? often much larger than the N e of the extant human populations 65?67 . An alternative explanation for the large among- gene variation in K s is that there is large variation in t. In a new study, the null hypothesis of a constant ratio of t/2N e can be rejected for 347 coding versus 143 intergenic regions compared between humans and chimpanzees (N. Osada and C.-I W., unpublished observations). This indicates that intergenic regions might have remained introgressable across the nascent species boundary between human and chimpanzee long after many coding regions have ceased to be able to introgress. We should note that the null hypothesis that was rejected in these cases was based on the The population genetics of speciation Elucidating the molecular basis of speciation through identifying and studying speciation genes attacks the problem of speciation from the bottom up. At the same time, genomic approaches are also making top-down studies much more powerful. In particular, now that speciation studies are entering the molecular era, the debate on the pervasiveness of allopatric speciation seems resolvable. Testing the allopatric model with genomic data. The allopatric model makes a strong prediction ? that all genes between the same two species have identical species-divergence time (t in FIG. 3a,b). These allopatric genealogies are not identical in all cases because of dif- ferences that were previously present when the species split (2N e in FIG. 3a,b) 55?59 .However, they are distinct from the genealogical histories of two genes that became non-introgressable at different times (FIG. 3c,d;see also a Species 1 Species 2 2N e t b Species 1 Species 2 2N e t c Species 1 Species 2 d Species 1 Species 2 Figure 3 | Geographical models of speciation. The strength of the barrier to gene flow between nascent species is indicated by the intensity of the red shade between the two branches. The black line represents a gene tree as it diverges between the two species. The wider yellow-shaded areas represent the whole genome of the two diverging species and the parent species from which they derived. a,b | The allopatric model of speciation in which the barrier is complete at a fixed time for all genes. The difference in divergence time between the two loci that are depicted was present at the time of allopatric speciation (which is the minimum divergence time between any two homologues in the two species). c,d | The parapatric model of speciation in which there is an extended period of time during which some genes can permeate through the nascent species boundary whereas others cannot. N e , effective population size; t, time since species divergence. NATURE REVIEWS | GENETICS VOLUME 5 | FEBRUARY 2004 | 121 REVIEWS reflective of the entire genome. So, it is the range of genes that are involved during incipient speciation that holds the greatest interest. In all five cases, there is sub- stantial divergence in DNA sequences and, among sev- eral of them, in expression as well. In at least four of the five cases, positive selection has driven the divergence. The results are relevant to the debate on whether RI might have evolved neutrally in the absence of adaptive forces. We propose that, for a gene to diverge in function, it needs to be released from its old functional niche. The release would then allow the diverging species to use the gene differently. We shall refer to this hypothesis as the genetic ?niche-release? hypothesis. There might be two different ways for a gene to experience niche release ? environmental change and gene duplication. Divergence in desat2 is an example of gene divergence that has occurred as a result of environmental change that took place when the flies migrated out of Africa, whereas OdsH and Xmrk are both the result of gene duplica- tion. (Note that this present hypothesis is different from a previous model on gene duplication and RI 72 .) Whichever mechanism is involved, a gene under niche release should be more likely to be functionally dispens- able. Operationally, functional dispensability means that the deletion of the gene would not lead to lethality, sterility or other forms of severe fitness reduction. Such dispensable genes might be prone to diverge in function, often becoming non-functional. Under this definition, both OdsH and desat2 are dispensable. In the classical DM model, the emphasis has always been on the incompatibility interaction (the green double-headed arrow in BOX 2), but we might ask a deeper question ? whether the process of divergence (the black double-headed arrows) and the resultant inter- action are related and, if so, how they are related. Such a linkage seems obvious for Xmrk-2 and Nup96.For exam- ple, in the former, the dark spots, when unregulated, become melanomas. On the other hand, it might not be surprising to find some RI genes for which the normal function and the RI phenotype are only weakly coupled, or even completely unrelated. For example, the deletion of OdsH has a subtle effect on male fertility but, in the appropriate genetic background, the presence or absence of the allele from D. mauritiana determines full fertility versus (nearly) normal fertility. Similarly, the normal function of desat2 might be cold tolerance but a corre- lated response is the change in CH, or the ?perfume? on the females. In either case, the RI phenotype is out of the range of what might have been predicted on the basis of the normal function/phenotype of the speciation gene. In the past five years of limited molecular analysis on speciation, we have learned the identities of several speciation genes. Their biological functions show the molecular bases of species differentiation. These stud- ies should re-focus our attention to the genic basis of speciation. The tenet of speciation study, namely the concept of RI, is fundamentally a whole-genome concept and should be revisited after we have a more comprehensive understanding of genes and their roles in speciation. simplest kind of allopatric speciation. There is a class of models that overlay allopatry on species with a deep population structure. Such models, which are interme- diate between allopatric and parapatric ones, deserve further study. Another recent study also addressed the question of parapatric speciation from a different angle 68 .The authors observed that the K A /K S RATIOS between human and chimpanzee are higher for genes on rearranged chromosomes than on collinear ones, in agreement with the parapatric model 68 .However,a separate analysis showed that the K a /K s pattern is observable, with nearly identical values, between human and orangutan or human and macaque 69 .So, although the observations in this study are meaningful in other respects, given the positive result from the ?negative control?, the interpretation of parapatric speciation cannot be supported 68,70 . Strictly speaking, the discussion in this section is rel- evant to animal studies only. Plant literature is replete with references to hybridization and introgression dur- ing speciation.?Hybridization speciation?, in which a third species is formed by mixing the genetic materials from two parental species 71 , is another important demonstration. What might be the basis of these dis- crepant views between animal and plant literature? It is possible that plant genomes are more modular, such that mixing components from different sources can still make a well-fit plant. If that is true,?allopatric genealogy? (FIG. 3a,b) in plants should often be rejected using the type of analysis described above. Implications and perspectives Speciation is not an easy subject. However, the myth that speciation is ?the mystery of mysteries? or that it is both ?unknown and unknowable? has not helped us to understand the subject. The perceived difficulties with the topic stem largely from a lack of knowledge at the fundamental genic level. For example, how could we hope to understand post-mating isolation when the phenotype that defines the class of genes that we are interested in does not even contain a hint of the original function for which each of these genes evolved? The five cases discussed above represent all those we could find that fulfil the criteria of a ?speciation gene?. Nevertheless, even from this limited set, which is based mainly on one taxonomic group (four of them are from Drosophila), the range of the molecular identity, as well as the under- lying principle, is broad. Three of the five cases are related to transcriptional regulation (Xmrk, OdsH and Hmr), supporting the common postulate that species divergence is regulatory in nature. Moreover, in desat2, which is not a regulatory gene, the change during INCIPIENT SPECIATION is nevertheless in the regulatory region. The nature of speciation genes is also likely to be a function of the age of the speciation event. The range of genes underlying RI between highly differentiated species, such as D. melanogaster and D. simulans, might be different from that between incipient species. As the divergence increases, more genes might contribute to RI and the range might become broader and more evenly K A /K S RATIOS Ratios of non-synonymous substitutions to synonymous substitutions per site. INCIPIENT SPECIATION The initial stage of species formation during which reproductive isolation is only partial. 122 | FEBRUARY 2004 | VOLUME 5 www.nature.com/reviews/genetics REVIEWS 1. Dobzhansky, T. Genetics of the Evolutionary Process (Columbia Univ. Press, New York, 1970). 2. Fisher, R. A. The Genetical Theory of Natural Selection (Clarendon, Oxford, 1930). 3. Mayr, E. Animal Species and Evolution (The Belknap, Cambridge, Massachusetts, 1963). 4. Wilson, A. C., Maxon, L. R. & Sarich, V. M. Two types of molecular evolution. Evidence from studies of interspecific hybridization. Proc. Natl Acad. Sci. USA 71, 2843?2847 (1974). 5. Wu, C.-I. The genic view of the process of speciation. J. Evol. Biol. 14, 851?865 (2001). In this genic view of speciation, RI develops gradually across the genome with some genes being unable to permeate through the nascent species boundary early on, whereas other genes continue to flow between them until complete RI evolves. The view suggests that parapatry is a common mode of speciation. 6. Mayr, E. Integration of genotypes: synthesis. Cold Spring Harb. Symp. Quant. Biol. 20, 327?333 (1955). 7. Wu, C.-I. Gene and speciation. J. Evol. Biol. 14, 889?891 (2001). 8. Howard, D. J. & Berlocher, S. H. Endless Forms (Oxford Univ. Press, New York, 1998). 9. Wu, C.-I. & Hollocher, H. in Endless Forms: Species and Speciation (ed. Berlocher, S.) 339?351 (Oxford Univ. Press, Oxford, 1998). 10. Wu, C.-I. & Palopoli, M. F. Genetics of postmating reproductive isolation in animals. Ann. Rev. Genet. 28, 283?308 (1994). 11. Coyne, J. A. & Orr, H. A. The evolutionary genetics of speciation. Philos. Trans. R. Soc. Lond. B 28, 287?305 (1998). 12. Zeng, Z.-B. et al. Genetic architecture of a morphological shape difference between two Drosophila species. Genetics 154, 299?310 (2000). 13. Ting, C.-T., Takahashi, A. & Wu, C.-I. Incipient speciation by sexual isolation in Drosophila: concurrent evolution at multiple loci. Proc. Natl Acad. Sci. USA 98, 6709?6713 (2001). 14. Takahashi, A. & Ting, C.-T. Genetic basis of sexual isolation in Drosophila melanogaster. Genetia (in the press). 15. Doi, M., Matsuda, M., Tomaru, M., Matsubayashi, H. & Oguma, Y. A locus for female mate discrimination causing sexual isolation in Drosophila. Proc. Natl Acad. Sci. USA 98, 6714?6719 (2001). 16. Tao, Y., Zeng, Z.-B., Li, J., Hartle, D. L. & Laurie, C. C. Genetic dissection of hybrid incompatabilities between Drosophila simulans and D. mauritiana. II. Mapping hybrid sterility loci on the third chromosome. Genetics 164, 1399?1418 (2003). 17. Wu, C.-I. & Davis, A. W. Evolution of postmating reproductive isolation: the composite nature of Haldane?s rule and its genetic bases. Am. Nat. 142, 187?212 (1993). 18. Wu, C.-I., Johnson, N. A. & Palopoli, M. F. Haldane?s rule and its legacy: why are there so many sterile males? Trends Ecol. Evol. 11, 281?284 (1996). 19. Presgraves, D. C. A fine-scale genetic analysis of hybrid incompatibilities in Drosophila. Genetics 163, 955?972 (2003). 20. Sawamura, K., Davis, A. W. & Wu, C.-I. Genetic analysis of speciation by means of introgression into Drosophila melanogaster. Proc. Natl Acad. Sci. USA 97, 2652?2655 (2000). 21. Sawamura, K., Roote, J., Wu, C.-I. & Yamamoto, M. Reassessment of the two-locus Dobzhansky-Muller model of reproductive isolation: extreme genetic complexity underlying hybrid male sterility in Drosophila. Genetics (in the press) . 22. Presgraves, D. C., Balagopalan, L., Abmayr, S. M. & Orr, H. A. Adaptive evolution drives divergence of a hybrid inviability gene between two species of Drosophila. Nature 423, 715?719 (2003). One of the five ?speciation genes? that has been characterized so far is presented in this study. The influence of positive selection on its divergence between species is clearly shown (see main text for detail). 23. Rawson, P. D. & Burton, R. S. Functional coadaptation between cytochrome c and cytochrome c oxidase within allopatric populations of a marine copepod. Proc. Natl Acad. Sci. USA 99, 12955?12858 (2002). 24. Cordon, M. The genetics of a viviparous top-minnow Platpoecilus: the inheritance of two kinds of melanophores. Genetics 12, 253?2283 (1927). 25. Kosswig, C. Uber bastarde der teleostier Platypoecilus und Xiphophorus. Zeitschrift fur induktive Abstammungs und Vererbungslehre 44, 253 (1927). 26. Anders, F. Contributions of the Gordon-Kosswig melanoma system to the present concept of neoplasia. Pigment Cell Res. 4, 7?29 (1991). 27. Schartl, M. A sex chromosomal restriction-fragment-length marker linked to melanoma-determining Tu loci in Xiphophorus. Genetics 119, 679?685 (1988). 28. Wittbrodt, J. et al. Novel putative receptor tyrosine kinase encoded by the melanoma-inducing Tu locus in Xiphophorus. Nature 341, 415?421 (1989). Tu is perhaps one of the first genes that was identified to be incompatible between closely related species. The mechanism of fitness reduction in the hybrids is shown to be tumorigenesis (see main text for detail). 29. Zechel, C., Schleenbecker, U., Anders, A. & Anders, F. v-erbB related sequences in Xiphophorus that map to melanoma determining Mendelian loci and overexpress in a melanoma cell line. Oncogene 3, 605?617 (1988). 30. Wittbrodt, J., Lammers, R., Malitschek, B., Ullrich, A. & Schartl, M. The Xmrk receptor tyrosine kinase is activated in Xiphophorus malignant melanoma. EMBO J. 11, 4239?4246 (1992). 31. Adam, D., Dimitrijevic, N. & Schartl, M. Tumor suppression in Xiphophorus by an accidentally acquired promoter. Science 259, 816?819 (1995). 32. Gomez, A., Wellbrock, C., Gutbrod, H., Dimitrijevic, N. & Schartl, M. Ligand-independent dimerization and activation of the oncogenic Xmrk receptor by two mutations in the extracellular domain. J. Biol. Chem. 276, 3333?3340 (2001). 33. Wellbrock, C. et al. Signalling by the oncogenic receptor tyrosine kinase Xmrk leads to activation of STAT5 in Xiphophorus melanoma. Oncogene 16, 3047?3056 (1998). 34. Ting, C.-T., Tsaur, S. C., Wu, M.-L. & Wu, C.-I. A rapidly evolving homeobox at the site of a hybrid sterility gene. Science 282, 1501?1504 (1998). This study reported the cloning of the OdsH gene of hybrid male sterility. OdsH is an example of a homeobox gene, a family of genes that are generally highly conservative and are therefore unlikely candidates for speciation genes. 35. Sun, S. Functional Analysis of the Hybrid Male Sterility Gene Odysseus in Drosophila. (Univ. Chicago Press, Chicago, 2003). 36. Ashburner, M. Drosophila: A Laboratory Handbook (Cold Spring Harbor Laboratory Press, New York, 1989). 37. Sawamura, K., Yamamoto, M.-T. & Watanabe, T. K. Hybrid lethal systems in the Drosophila melanogaster species complex. II. The zygotic hybrid rescue (zhr) gene of Drosophila simulans. Genetics 133, 307?313 (1993). 38. Hutter, P. Genetics of hybrid inviability in Drosophila. Adv. Genet. 36, 157?185 (1997). 39. Watanabe, T. K. A gene that rescues the lethal hybrids between Drosophila melanogaster and Drosophila simulans. Jpn. J. Genet. 54, 325?331 (1979). 40. Hutter, P. & Ashburner, M. Genetic rescue of inviable hybrids between Drosophila melanogaster and its sibling species. Nature 327, 331?333 (1987). 41. Hutter, P., Roote, J. & Ashburner, M. A genetic basis for the inviability of hybrids between sibling species of Drosophila. Genetics 124, 909?920 (1990). 42. Sawamura, K., Watanabe, T. K. & Yamamoto, M.-T. Hybrid lethal systems in the Drosophila melanogaster species complex. Genetica 88, 175?185 (1993). A detailed description of the various mutations in D. melanogaster that suppress hybird inviabilities. 43. Sawamura, K., Taira, T. & Watanabe, T. K. Hybrid lethal systems in the Drosophila melanogaster species complex. I. The maternal hybrid rescue (mhr) gene of Drosophila simulans. Genetics 133, 299?305 (1993). 44. Barbash, D. A., Siino, D. F., Tarone, A. M. & Roote, J. A rapidly evolving MYB-related protein causes species isolation in Drosophila. Proc. Natl Acad. Sci. USA 100, 5302?5307 (2003). Molecular cloning of one of the suppressor mutations of hybrid inviabilities (see main text for detail). 45. Barbash, D. A., Roote, J. & Ashburner, M. The Drosophila melanogaster hybrid male rescue gene causes inviability in male and female species hybrids. Genetics 154, 1747?1771 (2003). 46. Belgareh, N. et al. An evolutionarily conserved NPC subcomplex, which redistributes in part to kinetochores in mammalian cells. J. Cell Biol. 154, 1147?1160 (2001). 47. Ferveur, J. F., Cobb, M., Boukella, H. & Jallon, J.-M. World-wide variation in Drosophila melanogaster sex pheromone: behavioural effects, genetic bases and potential evolutionary consequences. Genetica 97, 73?80 (1996). 48. Dallerac, R. et al. A delta 9 desaturase gene with a different substrate specificity is responsible for the cuticular diene hydrocarbon polymorphism in Drosophila melanogaster. Proc. Natl Acad. Sci. USA 97, 9449?9454 (2000). 49. Takahashi, A., Tsaur, S. C., Coyne, J. A. & Wu, C.-I. The nucleotide changes governing cuticular hydrocarbon variation and their evolution in Drosophila melanogaster. Proc. Natl Acad. Sci. USA 98, 3920?3925 (2001). 50. Greenberg, A. J., Moran, J. R., Coyne, J. A. & Wu, C.-I. Ecological adaptation during incipient speciation revealed by precise gene replacement. Science 320, 1754?1757 (2003). The first application of gene replacement between nascent species shows the ecological consequence of desaturase divergence, cold tolerance, starvation intolerance and pheromonal changes. Mate choice might be affected by these pheromones. 51. Wu, C.-I. et al. Sexual isolation in Drosophila melanogaster: a possible case of incipient speciation. Proc. Natl Acad. Sci. USA 92, 2519?2523 (1995). 52. Coyne, J. A., Crittenden, A. P. & Mah, K. Genetics of a pheromonal difference contributing to reproductive isolation in Drosophila. Science 265, 1461?1464 (1994). 53. Fang, S., Takahashi, A. & Wu, C.-I. A mutation in the promoter of desaturase 2 is correlated with sexual isolation between Drosophila behavioral races. Genetics 162, 781?784 (2003). 54. Wood, D. & Ringo, J. M. Male mating discrimination in Drosophila melanogaster, D. simulans and their hybrids. Evolution 34, 320?329 (1980). 55. Watterson, G. A. On the number of segregating sites in genetical models without recombination. Theoret. Pop. Biol. 7, 256?276 (1975). 56. Li, W. H. Distribution of nucleotide differences between two randomly chosen cistrons in a finite population. Genetics 85, 331?337 (1977). 57. Gillespie, J. H. & Langley, C. H. Are evolutionary rates really variable? J. Mol. Evol. 13, 27?34 (1979). 58. Hudson, R. R. Properties of a neutral allele model with intragenic recombination. Theoret. Pop. Biol. 23, 183?201 (1983). 59. Takahata, N. Gene diversity in finite populations. Genet. Res. 46, 107?113 (1985). 60. Wakeley, J. & Hey, J. Estimating ancestral population parameter. Genetics 145, 847?855 (1997). 61. Wang, R. L., Wakeley, J. & Hey, J. Gene flow and natural selection in the origin of Drosophila pseudoobscura and close relatives. Genetics 147, 1091?1106 (1997). 62. Kliman, R. M. et al. The population genetics of the origin and divergence of the Drosophila simulans complex species. Genetics 156, 1913?1931 (2000). 63. Machado, C. A., Kliman, R. M., Markert, J. A. & Hey, J. Inferring the history of speciation from multilocus DNA sequence data: the case of Drosophila pseudoobscura and close relatives. Mol. Biol. Evol. 19, 472?488 (2002). 64. Li, W. H. Distribution of nucleotide differences between two randomly chosen cistrons in a finite population. Genetics 85, 331?337 (1977). 65. Takahata, N., Satta, Y. & Klein, J. Divergence time and population size in the lineage leading to modern humans. Theoret. Pop. Biol. 48, 198?221 (1995). 66. Takahata, N. & Satta, Y. Evolution of the primate lineage leading to modern humans: phylogenetic and demographic inference from DNA sequences. Proc. Natl Acad. Sci. USA 94, 4811?4815 (1997). 67. Wall, J. D. Estimating ancestral population sizes and divergence times. Genetics 163, 395?404 (2003). 68. Navarro, A. & Barton, N. H. Chromosomal speciation and molecular divergence-accelerated evolution in rearranged chromosomes. Science 300, 321?324. (2003). 69. Lu, J., Li, W. H. & Wu, C.-I. Comment on ?chromosomal speciation and molecular divergence-accelerated evolution in rearranged chromosomes?. Science 302, 988 (2003). 70. Navarro, A., Marqu�s-Bonet, T. & Barton, N. H. Response to comment on ?chromosomal speciation and molecular divergence-accelerated evolution in rearranged chromosomes?. Science 302, 988 (2003). 71. Rieseberg, L. H. et al. Major ecological transitions in wild sunflowers facilitated by hybridization. Science 301, 1211?1216 (2003). This study recreated new species in sunflowers by hybridization between two good species. The synthetic new species mimicked the naturally-occurring ancient hybridizations. Certain combinations of the parental genomes are favoured by natural selection. Hybridization speciation shows the ?modular? nature of the genome (see reference 5). 72. Lynch, M. & Conery, J. S. The evolutionary fate and consequences of duplicate genes. Science 290, 1151?1155 (2000). 73. Mayr, E. Wu?s genic view of speciation. J. Evol. Biol. 14, 866?867 (2001). Acknowledgements We thank S. Shi and N. Osada for technical help and two anony- mous referees for critical comments. Grants from the National Institutes of Health (USA) to C.-I Wu and from the National Science Council (ROC) to C.-T. Ting are gratefully acknowledged. Competing interests statement The authors declare that they have no competing financial interests. Online links DATABASES The following terms in this article are linked online to: FlyBase: http://www.flybase.org desat2 | Hmr | Nup96 | OdsH | unc-4 Access to this interactive links box is free online. "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment