The origin of species

Ever since Darwin entitled his great book on evolution ‘The origin of species’, this has been a central theme in evolutionary biology. That biodiversity should be composed of fine-grained discontinuities is a remarkable feature of our world. This conference set out very clearly what we know and what we don't know about the origin of these discontinuities. As co-authors of ‘Speciation’ (Coyne & Orr, 2004), the most recent book to summarize and synthesize the field, it was fitting that Jerry Coyne (University of Chicago, Chicago, IL, USA) and H. Allen Orr (University of Rochester, Rochester, NY, USA) gave the opening and closing addresses, respectively, serving as conference bookends. Indeed, they have directly or indirectly provided much of the framework for current research. That is not to say that they were the only intellectual heavyweights in attendance; indeed, so many pillars were present that the meeting resembled the Parthenon. ‘… the Holy Grail is to achieve a genetically detailed understanding of traits that confer reproductive isolation, tied to an understanding of the natural setting in which speciation has occurred’ Following the lead of Coyne and Orr, speciation genetics currently focuses (almost) exclusively on genetics of reproductive isolation. As noted by Coyne, the increasing interest in speciation genetics results from a wide variety of tools and approaches being focused on the problem. Some have taken a ‘genetics first’ approach, and, while purists might argue that ‘it ain't genetics if you don't make a cross’, most allow the dissection of the genetic basis of traits by whatever means, including the ‘quasigenetic’ approaches of genomics, to fall under the genetics umbrella. Taking this approach, loci contributing to isolation are sought via their genetic signatures, with subsequent characterization of gene functions and mode of action. For example, Richard Abbott (University of St Andrews, St Andrews, UK), Daven Presgraves (University of Rochester), and Tom Turner (University of California, Davis, CA, USA) each presented evidence of altered gene expression accompanying speciation. Tom Turner and co-workers have identified three genomic regions that differentiate cryptic taxa known as the ‘M’ and ‘S’ forms in the mosquito Anopheles gambiae (Turner et al., 2005). Further analysis points to a handful of genes that are candidates for causing reproductive isolation between the forms. Following the ‘ecology first’ approach, the genetic basis of traits with a demonstrated role in isolation is sought. Several speakers explored this approach, illustrated in the work of Dolph Schluter (University of British Columbia, Vancouver, BC, Canada) and colleagues on the genetics of reproductive isolation in sticklebacks (Gasterosteus spp.). Regardless of the approach, the Holy Grail is to achieve a genetically detailed understanding of traits that confer reproductive isolation, tied to an understanding of the natural setting in which speciation has occurred. Another contrast represented at the conference was that between broad comparative surveys looking for patterns (a phylogenetic approach) and, at the other extreme, the detailed study of species pairs. The complementarity in these approaches is exemplified by the work of Scott Hodges (University of California Santa Barbara, Santa Barbara, CA, USA) and co-workers on the role of nectar spurs as key innovations broadly in angiosperms and specifically in North American columbines. Theory vs experiment has been another important axis of variation, with both approaches having a seminal influence. Theory can guide the search for isolation genes and interpret the significance of the genes when found. Influential theory includes the Dobzhansky–Müller model, fitness landscapes and Haldane's rule (Turelli & Orr, 2000). At this conference, for example, the work of Maria Servedio (University of North Carolina, Chapel Hill, NC, USA) provided insight into the conditions under which mate preferences could contribute to speciation. Additional theoretical perspectives were provided by Michael Doebeli (University of British Columbia), Sergey Gavrilets (University of Tennessee, Knoxville, TN, USA) and Claus Rueffler (University of Toronto, Toronto, ON, Canada). One very notable feature of speciation genetics as exemplified at this conference is the wide range of organisms represented, and this is to be welcomed. If we are to gain insights into what is general about the mechanisms being discovered, we must have data from a wide range of life forms and life history strategies. Although Drosophila has been the organism par excellence of speciation genetics, this conference was more than a fruit fly fest. Several talks reported experimental results from Drosophila, but the other talks covered a broad range of organisms. Plants were well represented, with a diverse array of species being discussed (Mimulus, Solanum, Aquilegia, orchids, Helianthus, Senecio and Gossypium), as were insects, molluscs, fish, birds and mammals. The greater ability of researchers to tackle organisms outside the classical canon is driven in part by the greater availability of genomic resources. For example, to date 12 Drosophila species have had their genomes completely sequenced. In addition, the mouse genome was released in 2002 and Mimulus and stickleback genomes are imminent. The increasing availability of genomic resources has the happy effect that models are now chosen more for their interesting natural history than exclusively for their generation time, as ‘quasigenetics’ aids genetics. While longer lived organisms are still poorly represented among models, this may have as much to do with practical limitations in their use for field studies and experiments as with the ability to do genetics. It is interesting to note, however, that, while Drosophila, the original animal model system, was well represented, its botanical counterpart Arabidopsis thaliana was absent from oral presentations at this meeting, despite the growing awareness that this species and its close relatives indeed have interesting natural histories (Lexer & Fay, 2005). This highlights the fact that many botanists who study speciation have yet to make full use of this model organism. Interest in the field has been greatly stimulated by the discovery of ‘speciation genes’ (Wu & Ting, 2004). These are genes causing intrinsic postzygotic reproductive isolation between species. Such intrinsic genetic factors were predicted by Dobzhansky (Dobzhansky, 1937) who presented the Dobzhansky–Müller model for speciation based on epistatic interactions between two loci. Four such loci are now well known: odysseus (OdsH) (Sun, Ting & Wu, 2004), Xmrk-2 (Wittbrodt et al., 1989), Hmr (Orr & Irving, 2000) and Nup96 (Presgraves et al., 2003). The hunt is on for new genes and their partners in epistatic sterility. The conference heard directly about some of these speciation genes and possible new ones from Presgraves (University of Rochester) and Orr (University of Rochester). Although few such genes are well characterized, it is tempting to draw generalizations. They tend to be genes with essential but quite ordinary functions, and they tend to evolve rapidly under strong natural selection. There are doubtless huge numbers still to be found, and Orr posed the question of whether finding further genes would amount to an uninteresting repetitive labour or whether real surprises lay ahead. He emphatically predicted surprises and gave possible examples from work in the pipeline. We won't spoil the surprise here; suffice it to say: ‘watch for future developments’. Orr also conjectured that, while intrinsic postzygotic isolation is likely to be driven by intrinsic molecular effects (of the Dobzhansky–Müller type), extrinsic postzygotic isolation and prezygotic isolation will be found to be driven by extrinsic selection (either sexual selection or ecological adaptation). Extrinsic isolating factors were reported on by a number of speakers, including Katie Peichel (Fred Hutchinson Cancer Research Centre, Seattle, WA, USA) and Dolph Schluter (University of British Columbia) for genes influencing mate choice in sticklebacks and Darren Irwin (University of British Columbia) and Kerry Shaw (University of Maryland, College Park, MD, USA) for vocalizations in birds and crickets, respectively. The evolution of the freshwater phenotype in sticklebacks is particularly intriguing, as this appears to have happened numerous times in parallel but utilizing the same allele. It may be that this key allele is present at low levels in marine populations but is frequent enough to be available for freshwater speciation where and when it is advantageous. A similar speculation emerged from the work of Chris Jiggins (University of Edinburgh, Edinburgh, UK), who reported that parallel changes in pigment pattern may be driven by the same allele being utilized in different species, travelling over ‘hybrid bridges’. This sharing of allelic variation across speciation events implies that species boundaries are porous to alleles of high selective advantage, a viewpoint most often associated with the work of Loren Rieseberg (University of British Columbia) on Helianthus. An insight into this was presented by Nolan Kane (Indiana University, Bloomington, IN, USA) who discussed selective sweeps as a mechanism to maintain the integrity of species. This work effectively solves a major problem in species biology: that, although all members of a species can potentially exchange genes, isolated populations at opposite ends of large ranges rarely do so. So how do widespread species maintain their morphological and ecological integrity? Previous work has suggested that selective sweeps may provide the mechanism (Morjan & Rieseberg, 2004). Kane showed, from work on Helianthus, that 2% of a sample of loci had been the subject of recent range-wide selective sweeps and that selection allows rapid fixation even with low migration and gene flow. Based on these very exciting data, it seems that thousands of genes may have been involved in sweeps in the last 10 000 years. The implication is that sweeps stop at species boundaries not because of reproductive isolation (under strong selection rare events would be sufficient to sweep a gene through a species boundary), but instead because the selection coefficient of the allele will drop off in the different genetic background and different ecological circumstances. It seems certain that many exciting discoveries both in the area of speciation genes and in the area of allelic adaptation are around the corner. Loren Rieseberg and his team are to be congratulated on organizing such a stimulating conference, which has provided a freeze-frame snapshot of a fast-moving field.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.