Portal de Periódicos da CAPES

Contemporary and future studies in plant speciation, morphological/floral evolution and polyploidy: honouring the scientific contributions of Leslie D. Gottlieb to plant evolutionary biology

2014; Royal Society; Volume: 369; Issue: 1648 Linguagem: Inglês

10.1098/rstb.2013.0341

1471-2970

Daniel J. Crawford, Jeffrey J. Doyle, Pamela S. Soltis, Pamela S. Soltis, Jonathan F. Wendel,

Plant Reproductive Biology

You have accessMoreSectionsView PDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InRedditEmail Cite this article Crawford Daniel J., Doyle Jeffrey J., Soltis Douglas E., Soltis Pamela S. and Wendel Jonathan F. 2014Contemporary and future studies in plant speciation, morphological/floral evolution and polyploidy: honouring the scientific contributions of Leslie D. Gottlieb to plant evolutionary biologyPhil. Trans. R. Soc. B3692013034120130341http://doi.org/10.1098/rstb.2013.0341SectionYou have accessIntroductionContemporary and future studies in plant speciation, morphological/floral evolution and polyploidy: honouring the scientific contributions of Leslie D. Gottlieb to plant evolutionary biology Daniel J. Crawford Daniel J. Crawford Department of Ecology and Evolutionary Biology, and Biodiversity Institute, University of Kansas, Lawrence, KS 66045, USA [email protected] Google Scholar Find this author on PubMed Search for more papers by this author , Jeffrey J. Doyle Jeffrey J. Doyle L. H. Bailey Hortorium, Department of Plant Biology, Cornell University, Ithaca, NY 14853, USA Google Scholar Find this author on PubMed Search for more papers by this author , Douglas E. Soltis Douglas E. Soltis Department of Biology, University of Florida, Gainesville, FL 17 32611, USA Google Scholar Find this author on PubMed Search for more papers by this author , Pamela S. Soltis Pamela S. Soltis Florida Museum of Natural History, University of Florida, Gainesville, FL 17 32611, USA Google Scholar Find this author on PubMed Search for more papers by this author and Jonathan F. Wendel Jonathan F. Wendel Department of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, IA 50011, USA Google Scholar Find this author on PubMed Search for more papers by this author Daniel J. Crawford Daniel J. Crawford Department of Ecology and Evolutionary Biology, and Biodiversity Institute, University of Kansas, Lawrence, KS 66045, USA [email protected] Google Scholar Find this author on PubMed , Jeffrey J. Doyle Jeffrey J. Doyle L. H. Bailey Hortorium, Department of Plant Biology, Cornell University, Ithaca, NY 14853, USA Google Scholar Find this author on PubMed , Douglas E. Soltis Douglas E. Soltis Department of Biology, University of Florida, Gainesville, FL 17 32611, USA Google Scholar Find this author on PubMed , Pamela S. Soltis Pamela S. Soltis Florida Museum of Natural History, University of Florida, Gainesville, FL 17 32611, USA Google Scholar Find this author on PubMed and Jonathan F. Wendel Jonathan F. Wendel Department of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, IA 50011, USA Google Scholar Find this author on PubMed Published:05 August 2014https://doi.org/10.1098/rstb.2013.03411. IntroductionThis special commemorative issue is dedicated to the life and scientific contributions of Leslie D. Gottlieb. Dr Gottlieb's contributions to our understanding of plant evolution were many and varied, from which we selected some of the most influential to be included in this special issue honouring his career. Gottlieb's studies were largely hypothesis driven, and he invariably spelled out the rationale for the design of his research and how the results could be used to test his hypotheses. His research was of general interest and attracted widespread attention because it focused on central issues in plant evolution. Though Gottlieb used applicable state-of-the-art methods for that time period, the tremendous methodological and technical advances since his work, especially in the areas of molecular biology, genetics and now genomics, make it possible to examine in greater depth the questions that most intrigued him. The papers in this special issue, contributed by leading workers using largely cutting-edge methods, document that Leslie Gottlieb's work was indeed visionary and that it remains as relevant today as it was when published (figure 1). Figure 1. Leslie D. Gottlieb (1936–2012). Photo taken in Ashland, Oregon on 25 May 2009, one day before his 73rd birthday. Photo courtesy of Vera Gottlieb. (Online version in colour.)Download figureOpen in new tabDownload PowerPointIt could be argued that the process of speciation is central to any discussion of plant evolution because the tremendous diversity seen in the green plant clade is the result of the generation of independently evolving lineages. Interest in and research on speciation has been centred on primary divergence and the processes that drive the divergence [1–7]. However, it is now increasingly recognized that hybridization between divergent taxa at the same ploidal level can generate novel, independent lineages, a process termed homoploid hybrid speciation [8,9]. Similarly, recent analyses have convincingly demonstrated that all land plants are 'palaeopolyploid' and that all angiosperm lineages contain multiple rounds of past genome doubling [10], and it has long been recognized that polyploidy is an important mechanism for hybrid speciation within contemporary plant lineages [11]. As a result of its pervasiveness in plant evolution, polyploidy receives considerable attention in this issue, just as it did in Leslie Gottlieb's research.The evolutionary shift from outcrossing to selfing is generally recognized as one of the most frequent transitions in flowering plants [12,13]. Among many other things, it may serve as an isolating mechanism [14–16], and indeed in many plant lineages the origin of self-compatibility with the loss of function of the self-incompatibility locus is associated with speciation [16–19]. In fact, Leslie Gottlieb provided an elegant example of the association of speciation and the transition to selfing in his classic studies in the genus Stephanomeria [20,21]. It had long been recognized that selfing species are less variable genetically at the population level [22], and as discussed by Barrett et al. [23] in this issue, Gottlieb was one of the first to use enzyme electrophoresis to document lower genetic diversity in outcrossing compared with selfing plants [20,24–26].High-throughput DNA sequencing now makes it possible to examine a number of long-standing questions about the factors that drive the transition to selfing and the consequences of its evolution. In this issue, Barrett et al. [23] examine the efficacy of genomic data to test the two most frequently cited hypotheses for the evolution of selfing. The first is the so-called automatic selection hypothesis in which a selfer can contribute outcross pollen as well as fertilize its own ovules (3 : 2 advantage) [27,28]. The second major hypothesis for the evolution of selfing is reproductive assurance, which posits that selfing will be favoured when pollen vectors and/or compatible mates are rare and limit seed production [29,30]. Inbreeding depression is recognized as the most important factor opposing the evolution of selfing.Barrett et al. [23] provide a test, first advanced by Schoen et al. [31], for inferring the importance of the two hypotheses for the origin of selfing. The rationale of Schoen et al. [31] for using molecular data to test the two hypotheses is that reproductive assurance would involve population bottlenecks, and possibly founder events, both resulting in reduced genetic diversity throughout the genome. In contrast to reproductive assurance, with automatic selection, selfing mutants have a 3 : 2 advantage and could spread through a population in the absence of population bottlenecks. Under this scenario, there would not be reduced diversity at unlinked neutral loci. High-throughput sequencing technologies now allow assessment of genome-wide diversity in a comprehensive manner not even dreamed of when Leslie Gottlieb was employing enzyme electrophoresis for measuring diversity at relatively few loci. Using comparative empirical data and simulations, Barrett et al. [23] suggest that available molecular data are not able to test hypotheses for the selective forces or mechanisms that drive the transition to selfing. While they find little evidence of bottlenecks associated with the transition to selfing, they emphasize that this does not rule out the role of reproductive assurance as a driver of the evolution of selfing. Also, it is possible that genetic bottlenecks could occur subsequent to the origin of selfing. A critical factor in using molecular or other data for elucidating factors promoting selfing is studying plants that are in the process of the transition or where a transition from outcrossing to selfing has been very recent. Indeed, Gottlieb [20] emphasized that studying recent divergence is critical for understanding the forces and mechanisms in plant evolution. Barrett et al. [23] highlight the importance of future studies integrating data from demography, genomics and experimental populations for identifying the drivers of mating system shifts, including the role of these shifts in facilitating divergence and speciation.Speciation has traditionally been viewed within a geographical context and the biogeographic/spatial scales on which it may occur continue to be a topic of discussion and debate. As a result, the geographical context of speciation is a component of several papers in this issue, just as it was a topic that permeated many of Leslie Gottlieb's scientific papers. As stated by Coyne and Orr [32, p. 4] 'Among all the scientifically tractable questions about speciation, the most hotly contested concerns its biogeography'. Allopatric speciation has been historically the most widely accepted geographical mode of speciation because divergence between geographically separated populations may occur through time without being retarded or prevented by gene flow and recombination. By contrast, sympatric homoploid speciation has been the most contentious mode of speciation, with arguments against its occurrence centring on several issues. One problem is the challenge of documenting whether divergence occurred in sympatry, or divergence was in fact allopatric with subsequent secondary contact [33–35]. While it is difficult to rigorously falsify the allopatric–secondary contact hypothesis, evidence such as sister group relationship and high genetic similarity between the species in question [20,36], especially when the species occur together at one locality or in a restricted geographical area, combine to argue strongly for in situ divergence [20,33,35]. Leslie Gottlieb was very interested in sympatric speciation. One of his earliest and most highly cited papers [20] contained 'sympatric speciation' in the title and focused on divergence within a local population.As indicated above, another major question regarding sympatric divergence is how it could occur in the face of gene flow from the parental species. Data from molecular markers and ecological studies show the potential for selection to drive divergence in the face of gene flow and recombination [37–41]. In his studies of Stephanomeria, Gottlieb was concerned with how initial divergence could have occurred and how the rarer derivative species could coexist with its progenitor at the same locality. While he identified several factors reducing interspecific gene flow and thus presumably explaining the rarity of natural interspecific hybrids in Stephanomeria, he never succeeded in elucidating the factors (indeed, if they existed) contributing to ecological/habitat divergence between the two species growing intermixed at a single locality [20,21,42,43]. The same issues that intrigued Gottlieb decades ago remain of interest and the focus of contemporary studies.In this issue, Papadopulos et al. [44] continue a series of studies [33,35,39,40] to test the hypothesis of sympatric speciation using plants of Lord Howe Island as a model system. They employ molecular markers to determine genetic structure (and thus infer gene flow) associated with several ecological factors (both individually and in combination), community composition and spatial scale to infer which of these variables may be important in divergence. Papadopulos et al. [44], while providing the appropriate caveats for the approaches they employed, present data supporting ecological divergence as a common factor associated with genetic divergence within species on Lord Howe Island.Future studies of sympatric speciation will focus on fine-scale genetic analyses of the traits, e.g. adaptation to edaphic types, flowering time, selfing, etc. that lead to divergence, even in the face of gene flow [45,46]. These genetic/genomic investigations will be most powerful when combined with field studies such as transplant experiments [47–49]. The ultimate goals of such studies would be to demonstrate that plants with certain phenotypic traits are more fit in a given environment than those with alternative traits, and then eventually to elucidate the genetic basis of the contrasting traits. Species on Lord Howe Island appear to be excellent candidates for these types of intensive studies.Leslie Gottlieb's interest in studies of speciation at the local level via the divergence of populations within the ranges or on the periphery of geographically widespread species extended beyond his classic investigations of Stephanomeria malheurhensis [20,21], and included members of the genera Layia [50] and Clarkia [51]. Those investigations consisted of morphological, biosystematic and electrophoretic approaches. In the case of Layia, the localized species is an edaphic endemic, whereas in Clarkia habitat differences between the two species were not apparent and, as noted earlier for Stephanomeria, were never well characterized.The contribution by Ferris et al. [52] in this issue has a strong geographical component and focuses on population divergence within the range of one species to give rise to a new species. This geographical scale of speciation is receiving considerable attention with the growing appreciation that it may be more prevalent than previously recognized [53,54]. The genus Mimulus (monkeyflowers) has attracted the attention of many researchers and has attained 'model' status for plant evolutionary studies because of its array of ecological, phenotypic and genomic diversity [55]. Within Mimulus, the geographically widespread, morphologically variable, outcrossing and ecologically diverse Mimulus guttatus complex has proved rewarding for studies of local speciation. Within the geographical distribution of the complex, there are localized (commonly on specialized substrates), often selfing, populations that can be delimited morphologically, albeit with some effort in certain cases, and have been recognized as species in many instances [54,55]. These localized species have diverged independently from within the M. guttatus complex and represent multiple progenitor/derivative species pairs [56].Ferris et al. [52] studied the recently described Mimulus filicifolius, a species in the M. guttatus complex. They employed several molecular markers to demonstrate that the species is genetically distinct, and thus presumably not just a minor morphological variant of its closest relative, another localized derivative species in the M. guttatus complex. Ferris et al. [52] then studied components of reproductive isolation. They found edaphic differences between M. filicifolius and its presumed closest relatives. These differences include soil temperature, per cent moisture and saturation point. Molecular markers showed that M. filicifolius, like some other local Mimulus species, is highly self-fertilizing [54]. As discussed earlier, selfing can serve as a barrier to gene flow among populations and facilitate divergence and speciation. Lastly, Ferris et al. [52] demonstrated strong post-zygotic isolation between M. filicifolius and Mimulus laciniatus, another local species and presumably the closest relative of M. filicifolius.The genomic resources available for Mimulus (Phytozome v. 8.0; www.phytozome.net) are providing finer insights into the genetic architecture of traits that drive divergence and speciation. Two recent investigations illustrate the value of genomic tools for examining local speciation, and point to additional approaches that could be applied, for example, to elucidate the genetic architecture of isolating barriers detected by Ferris et al. [52]. One of these studies [57], when taken within the context of prior investigations, provides an elegant example of how ecological and genomic studies may be complementary in examining early stages in local population divergence and incipient speciation. The species Mimulus aurantiacus has red- and yellow-flowered ecotypes, with the former almost exclusively hummingbird pollinated and the latter pollinated primarily by hawkmoths [58,59]. The isolation of the ecotypes is not complete, as hybrid zones occur where species overlap [60], and there are weak intrinsic barriers to gene flow between the two ecotypes [61]. These observations indicate that selection for different pollinators of the two floral colour morphs is an important factor in initial divergence. Therefore, elucidating the genetic architecture of flower colour would provide clues to the genetics of incipient speciation. Streisfeld et al. [57] demonstrated that flower colour differences between the two ecotypes of M. aurantiacus are based on a cis-regulatory mutation in a transcription factor that regulates enzymes in the anthocyanin biosynthetic pathway. The two ecotypes are fixed for alternative alleles specifying the floral colour morphs. Fixation of the alleles has occurred despite gene flow between the two ecotypes at neutral molecular markers, indicating that selection has driven the fixation of the alternative alleles [57].Another recent study involves M. guttatus and one of its derivative species localized on high-copper soil from old mines in California [62]. Prior studies [63,64] suggested that copper tolerance and hybrid lethality are controlled by a single locus because crosses between copper-intolerant and copper-tolerant plants resulted in various levels of lethality in the F1 hybrids. However, high-resolution genome mapping demonstrated copper tolerance and hybrid lethality are controlled by two tightly linked loci and that copper tolerance is not pleiotropic with hybrid inviability [62]. These two recent studies [57,62] demonstrate not only the utility of cutting-edge approaches for elucidating the genetic basis of characters, but, perhaps even more importantly, the enhanced value of the data when applied to interesting evolutionary situations/questions identified and posed by earlier workers using field observations and biosystematics approaches.In contrast to the universal recognition of primary divergence in plant speciation, the importance of hybridization in adaptive evolution and in plant speciation at the same ploidal level has been more debateable [65–67]. With the advent of the Modern Synthesis, there was a general divide between those studying animals and plants, with the former viewing hybrids as maladaptive and hybridization as a final step in the development of reproductive barriers through the process of reinforcement [65]. On the other hand, prominent plant evolutionary biologists, with very few exceptions [66], tended to view hybridization as a creative process [67,68]. While the role of hybridization and the formation of maladaptive hybrids in plants as 'reinforcement' in speciation is of current interest [69], it is also now well accepted that hybridization functions in a creative way in plant evolution [9,11,70].The contribution by Abbott & Brennan [71] focuses on hybrid zones along altitudinal gradients as model systems for studying the evolutionary role of hybridization. Assessing the impact of hybridization involves determining the fitness of hybrids relative to their parents, identifying the traits that affect fitness and elucidating the genetic basis of those traits [72–74]. Hybridization may create novelty, including new species [8,9,71,75,76], and also be a mechanism transferring adaptive traits between species [77,78], and in these two ways be important in plant evolution. Abbott & Brennan [71] emphasize the challenges of understanding the evolutionary impact of hybridization and multi-faceted studies, including field, experimental and genetic/genomic approaches, will be needed in the future to meet those challenges [46–48].Homoploid hybrid speciation, the stabilization of hybrid recombinants (typically between two species) at the same ploidal level, has been the subject of renewed interest and activity over the past two decades, due largely to the work of Rieseberg and collaborators on the sunflower genus Helianthus [8,79–83]. Plant evolutionists have been interested in the possibility of homoploid hybrid speciation for decades, with models proposed [84,85], stabilized hybrids synthesized [86–88], and a handful of reports of naturally occurring hybrid species based on data from morphology and biosystematics [89]. Only about 20 cases of hybrid species in nature have been reported [11] but the number varies depending on what is considered sufficient evidence for documenting hybrid origin in any given case. Whether the difficulty of detection or true rarity are more significant causes of the paucity of well-documented examples of naturally occurring homoploid hybrid species remains an open question, and it could in fact be a combination of the two factors. This is not a trivial issue because rigorous documentation of hybrid origin is obviously the critical first step in any study of homoploid hybrid speciation. Leslie Gottlieb was the first to employ molecular markers to document the occurrence of a naturally occurring stabilized homoploid hybrid species [90]. With genomic resources providing unlimited molecular markers, it will become more straightforward to document hybridization [91].Homoploid hybrid speciation is generally regarded as a form of sympatric speciation [92] initiated by gene exchange between two co-occurring species, and indeed this is likely the common condition. However, whether or not the initial hybridization occurred between sympatric congeners, or via some other mechanisms such as rare dispersal events, is not always easy to discern. Even if hybrids originate initially from crosses between sympatric parents, it does not mean that the stabilized hybrid derivatives would necessarily evolve in sympatry with the parental species [92]. Indeed, one of the classic examples of a homoploid hybrid species, Senecio squalidus [93,94], evolved and stabilized in England following the introduction of material from a hybrid zone in Italy. This is an example of a sympatric origin of a homoploid hybrid species, but with stabilization of the species in allopatry with its parents.Studies of homoploid hybrid speciation will become increasingly multi-faceted. In addition to employing large suites of molecular markers or tools such as genomic resequencing to document hybridization, field and experimental studies of the fitness of parents and hybrids under different habitats and experimental manipulations [95,96] will be expanded. The real frontier will be in using genomic tools for achieving refined insights into the processes facilitating the isolation and stabilization of the hybrid derivatives. One focus will be on intrinsic isolation of parents from hybrids by selection for fertile genetic and chromosomal recombinants of parental karyotypes in the hybrids [81–83]. A second major focus will be elucidating the genetic/genomic basis of adaptive ecological/phenotypic traits in the hybrid species [97]. Resequencing data promise insights into these questions, as well as many others, such as the genomic distributions of parental contributions and the genetic architecture of adaptive traits.Transposable elements (TEs) constitute the predominant portion of most plant genomes [98]. Given their ubiquity, comparative studies of TE activity and proliferation in hybrid species and their progenitors are potentially relevant to understanding the speciation process. Determining their role in the stabilization of hybrid genomes and the divergence of parental and hybrid genomes are areas of interest. The same studies could also reciprocally illuminate the dynamics of TE activation and proliferation in plant genomes. Returning to hybrid sunflowers, an initial study [99] showed that the larger genomes of hybrid species compared with their diploid parental species was due primarily to TEs (more specifically, long terminal repeat retrotransposons). In this issue, Renaut et al. [100] continue studies [101,102] into the dynamics of TEs in two parental Helianthus species and their homoploid hybrid derivative species, and use high-throughput technology to assess transcriptional activity of TEs. The authors found higher mean values of aggregate TE expression phenotypes in the hybrid species compared with the parental species or F1 hybrids between the parents. They detected variation within species in both the aggregate levels and the component elements making up the total levels. The reason or reasons for the variation remain obscure and the authors suggest several possibilities and future studies to address the questions.The role of TEs in hybridization and hybrid speciation is an important question because, as emphasized by Renaut et al. [100], of the necessary balance between the possible negative impact of proliferation of TEs on fitness, on the one hand, and their role in the reorganization and stabilization of hybrid genomes and the origin of phenotypic novelty on the other hand. The fact that all of the hybrid sunflower species have elevated TE proliferation relative to their parents and synthetic F1 hybrids suggests that past bursts of TE activity play some role(s) in the success of the stabilized hybrid derivatives. Identifying the mechanisms that regulate and repress the expansion of TEs in hybrids and the possible phenotypic effects of the resulting insertional mutagenesis will remain an active area of research.Leslie Gottlieb, while disposed to using available molecular methods in his studies, always kept in mind that it is phenotypes and their underlying genetic architecture that are visible to natural selection. Thus, often in collaboration with Vera Ford Gottlieb, he studied the genetic/developmental aspects of both floral and vegetative morphology [103–108]. One example serves as an illustration of how their work stimulated further research. They were intrigued by the genetic basis and ecological/evolutionary significance of petal spot patterns in the flowers of a species of Clarkia [105], a genus they used as a model system for many and varied studies. Subsequent studies have examined the ecological significance [109,110] and genetics [111] of petal spot patterns.The contribution to this issue by Olsen et al. [112] on cyanogenesis (the production of hydrogen cyanide in damaged plant tissue, which in turn functions to reduce the activity of herbivores) in the genus Trifolium (clovers) is illustrative of the multi-faceted studies valued by Gottlieb for understanding the genetic basis of phenotypic traits and the ecological significance of the traits. Olsen et al. [112] delve into not only the genetic architecture of the trait of interest, but also place it in an evolutionary context. Two key questions are whether the same traits have evolved more than once within a clade, and if so, whether the same mechanism was recruited to achieve parallel evolution of the trait. Olsen et al. [112] demonstrate how molecular data can be employed both for dissecting a trait and for tracing the evolutionary history of the trait. They show the power of genetic/genomic data for elucidating the mechanisms for cyanogenesis, but also emphasize that identifying the ecological factors shaping the patterns of occurrence seen in nature is a daunting task. This point is especially well illustrated in the cyanogenic response, which comprises two distinct components, synthesis of cyanogenic glucosides and the subsequent production of hydrogen cyanide. In some Trifolium species, cyanogenic glucosides are produced, but hydrogen cyanide is not. This situation requires the formulation of more complex adaptive hypotheses than the more straightforward situation where the herbivore-deterring hydrogen cyanide is produced. The authors consider several scenarios that have been advanced under which cyanogenic glucosides in the absence of conversion to hydrogen cyanide could be adaptive. The results of Olsen et al. [112], along with the papers they cite, provide an excellent example of the complexity of the ecological component of ecogenomic [113] studies, something that has been emphasized by others [46–48].The diversity of flower structure and colour have long-fascinated biologists, with a large component of the interest centred on the ecological roles of floral attributes in pollination and other aspects of reproductive biology [114]. The papers by Hileman [115] and Wessinger et al. [116] consider floral evolution at different taxonomic scales, the former taking a broad perspective of floral symmetry in angiosperms, and the latter focusing on two species in the genus Penstemon. Gottlieb's interests and published work have significant elements in common with the topics of both papers. As pointed out by Hileman [115], Leslie Gottlieb was interested in floral development before the age of genomics and the evolution of development (evo-devo), and this interest is documented in several of his publications [104–108]. Also, he strived to place floral transitions in an evolutionary context [106,107], although some of his early work predated the routine use of DNA sequences for c