JXB Advance Access originally published online on September 21, 2006
Journal of Experimental Botany 2006 57(13):3505-3516; doi:10.1093/jxb/erl132
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Evolution of Flowers and Inflorescences |
Progress and challenges in studies of the evolution of development
Department of Biology, University of Missouri-St Louis, One University Boulevard, St Louis, MO 63121, USA
* E-mail: tkellogg{at}umsl.edu
Received 10 May 2006; Accepted 20 July 2006
 |
Abstract |
|---|
Top
Abstract
IntroductionPlant evolutionary developmental genetics (EDG) has made considerable progress over the last decade. This is in part due to the accumulation of large amounts of sequence data that have provided robust organismal phylogenies and, increasingly, broad assessments of molecular evolution. Attempts to use primary sequence data to identify genes that have changed function in evolutionary time have not been as successful as initially hoped. The coding sequences of most genes, which are more amenable to statistical analysis than are regulatory sequences, are generally under purifying selection, as would be expected if much evolutionary change is the result of changes in cis-regulatory sequences. Sequence-based analysis of the regulatory sequences themselves remains difficult. Comparative studies of gene expression have been useful to identify genes whose developmental role may have changed in evolutionary time and will be critical to the future development of EDG. Such studies can be used to test hypotheses of gene function. Transformation experiments are often illuminating, but can be hard to interpret, particularly if genes from multiple species are all placed into a single heterologous system such as Arabidopsis. The ideal experiment would be a gene swap or promoter swap between two species, but this awaits development of good transformation systems. The immediate need for EDG is studies of gene expression on a massive scale, far broader than any studies undertaken to date.
Key words: cis-regulatory sequence, evo-devo, phylogeny, selection, taxonomy
|
Introduction |
|---|
The field of evolutionary development genetics (EDG), or evo-devo, has generated a great deal of excitement over the last 10–15 years. Several papers have outlined general questions and approaches for plant evolution and development, and pointed hopefully to the way forward (Kellogg and Shaffer, 1993; Kellogg, 1996, 2004; Baum, 2002). An update is provided here on where we are and where we are going, with some tentative suggestions about why the big questions of evo-devo have yet to be answered. This paper is not intended to be a comprehensive review of the field, and I apologize in advance to the many excellent scientists whose work I have not included. Following Baum (2002), the term EDG is used throughout, instead of evo-devo.
The goal of EDG is to be able to choose a particular internode on a phylogeny, corresponding to an ancestral lineage, and to determine what genetic change(s) occurred within that ancestral lineage during cladogenesis that might have led to the different morphologies of the descendants. The interest in the question increases with the depth of the node in the tree—the genetic events that led to the morphological divergence of monocots and eudicots from their common ancestor are likely to generate more excitement generally than the genetic events that led to the morphological divergence of, for example, Poa alsodes from P. saltuensis (which differ by the pattern of trichomes on the lemma). However, the difficulty in answering the question also increases with the depth of the node in the tree.
It is obvious that for morphological evolution to occur, genetic change must also occur. In other words, genes must change function. However, the literature is often ambiguous on what is meant by function. Biochemical function refers to the precise chemical reaction catalysed by an enzyme or the precise set of contacts between interacting proteins. Developmental role, on the other hand, refers to the function of the protein in the developing plant. It is easy to imagine that the biochemical function could stay the same, while the developmental role is modified by a change in regulation (Doebley and Lukens, 1998). Alternatively, biochemical function could change and thus modify the developmental role of a protein. Herein a distinction will be made between these two possibilities wherever possible.
|
Phylogenies and alpha taxonomy |
|---|
Identification of important questions for EDG, and interpretation of the results, depends on the availability of good alpha taxonomy and robust phylogenies, and thus EDG will be at its most effective in plant groups for which those resources already exist. For example, studies on the role of CYCLOIDEA-like proteins in the control of zygomorphy (Cubas et al., 2001; Citerne et al., 2003; Hileman and Baum, 2003) are exciting because zygomorphy is known to vary widely among angiosperms, because phylogenetic analyses confirm that there have been multiple origins (and possibly reversals) in the trait (Donoghue et al., 1998), and because the trait appears to be connected with pollination success. Zygomorphy would be of much less interest if it were an occasional population-level polymorphism. Similarly, studies of inflorescence diversification in the grasses are interesting precisely because inflorescence characters differentiate closely related species, genera, and subfamilies (Malcomber et al., 2006). Thus any genes identified that control differences in inflorescence development between two species are prime candidates for investigation across a wider group (Doust and Kellogg, 2002; Doust et al., 2005).
Robust phylogenies are particularly important for laying out precise hypotheses for EDG, and for assessing their generality. An example comes from integument number, which is two in most angiosperms, but reduced to one in some Ericales and also reduced in the remaining asterid taxa, which form the sister clade to the Ericales. Within the Ericales, McAbee et al. (2005) investigated several species of Impatiens in which integument number varies. They hypothesized that the gene INNER NO OUTER (INO), which affects integument development in Arabidopsis, might be deployed differently in species with one integument versus those with two. INO proved to be a very useful marker and led to several interesting hypotheses about the regulation of ovule development, but its expression pattern and hence its presumed developmental role was conserved among all Impatiens species investigated. Thus INO was not responsible for variation in integument number in Impatiens. However, to generalize this result to the non-Ericalean asterids would require a separate study; the experiments on Impatiens are not informative about the major asterid clade because integument is reduced independently in the two lineages.
Within the last decade, phylogenies of sufficient size and with sufficient support have become available for the angiosperms (Qiu et al., 1999), seed plants (Burleigh and Mathews, 2004), and plants as a whole (reviewed in Friedman et al., 2004). Within the angiosperms, comprehensive phylogenies of the monocots are in press (Graham et al., 2006), and studies of asterids (Bremer et al., 2002) and rosids (Jansen et al., 2006) have appeared recently. A multigene phylogeny of the grasses was published in 2001 (Grass Phylogeny Working Group, 2001) and a phylogeny of the Brassicaceae appeared recently (Beilstein et al., 2006).
The Brassicaceae phylogeny illustrates clearly the extent of generalization possible from the model system, Arabidopsis thaliana. For example, the controls of trichome development in Arabidopsis must be different in other members of the family. The stereotypical three-branched trichomes of Arabidopsis are derived in the clade most closely related to Arabidopsis, whereas other clades are characterized by quite different trichome types. Similarly, fruit morphology is highly variable in the family, hinting that the genes underlying fruit development must have different roles in different species.
Phylogenies of species and genera are becoming routine. Any recent issue of Systematic Botany, Molecular Phylogenetics and Evolution, or other systematic publications contains information on relationships among closely related taxa, and fortunately phylogeny and alpha taxonomy now often proceed simultaneously. New molecular markers are being developed continually (Small et al., 2004; Mason-Gamer, 2005; Choi et al., 2006; Doust et al., 2006; Hughes et al., 2006; Whittall et al., 2006), and phylogeneticists have a broad array of tools at their disposal for phylogeny reconstruction. Challenges remain, particularly at the population level or among closely related species in which hybridization, incomplete lineage sorting, and polyploidy all create a phylogenetic history that is more reticulate than bifurcating. Nonetheless, the methods for investigating plant relationships have become relatively straightforward.
In summary, availability of good phylogenies was initially limiting for EDG, but is becoming less and less of a problem. There remain large families, particularly in the tropics, in which the alpha taxonomy and phylogeny critically require attention, but it is easy to foresee a time in the not too distant future in which the major outlines of phylogeny, at least for families, subfamilies, and tribes, will be available for most angiosperms.
|
Description of the phenotype |
|---|
To understand the evolution of morphology, morphological variation needs to be described precisely. Good alpha taxonomy is an important first step in this direction, but is often (necessarily) limited to characters that can be studied in adult plants, generally in the herbarium. At the same time, developmental morphology is a natural extension of alpha taxonomy. The deep knowledge of adult morphology acquired by a monographer could help inform and direct developmental studies. Collection of fixed material in the field could provide multiple developmental stages, and even a developmental description of a single species would move the field forward.
Much excellent developmental morphology took place in the 19th century in Europe (Payer, 1857), but then the field became less central to many of the questions that were being addressed by what came to be called biology (Stevens, 1994). In recent years, developmental morphology has begun something of a renaissance (Endress et al., 2000; Doust and Kellogg, 2002; Endress, 2003; von Balthazar et al., 2004; Matthews and Endress, 2005; McMahon and Hufford, 2005; Rudall et al., 2005; Tucker and Hodges, 2005; Xu and Rudall, 2006), and has provided new insights into the structure and development of angiosperms. Plants that look similar in early development often diverge later to look very different [e.g. ovary position in Saxifragales (Kuzoff et al., 2001; Soltis and Hufford, 2002) or inflorescence form in grasses (Kellogg et al., 2004)].
Friedman and Williams (2004) have noted that a number of generalizations about angiosperm development are in fact based on only a very few data points. The familiar examples of triploid endosperm and a Polygonum-type embryo sac may not to be ancestral, despite being widespread. Except for the phylogenetically critical genus Amborella, species that are sister to all other angiosperms instead have diploid endosperm and four-celled embryo sacs (Friedman and Williams, 2004).
A similar example appears in the literature on inflorescence development in grasses, for which only a few descriptions of development had been published until quite recently. One of the few papers to study early development of the inflorescence of multiple species was that of Evans (1940). He hypothesized that the phyllotaxis of the inflorescence of all grasses was distichous, continuing the phyllotactic pattern of the leaves, whereas in fact the ancestral and common condition for the family is to have spiral phyllotaxis (Moncur, 1981; Malcomber et al., 2006). Evans' error was to restrict his sampling largely to species in what is now known as the subfamily Pooideae, a subfamily for which distichous inflorescence phyllotaxis appears to be derived and universal. It has been found that virtually all observations have had to be re-done and verified. More importantly, developmental descriptions have allowed the characters that are varying in evolutionary time to be revised.
In summary, studying the evolution of development requires a deep understanding of morphology throughout both developmental and evolutionary time. This generally requires a focus on meristems, primordia, tissues, and cells in which genetic changes have their first effects.
|
Gene and genome evolution |
|---|
The various genome projects have produced and are producing vast amounts of sequence data. The amount of data will continue to increase for the foreseeable future. Currently the NIH is supporting development of techniques to lead to a complete genome sequence for US$1000—the ‘$1000 genome’ (Service, 2006). Current prices for sequencing a genome approximately the size of the human genome are still of the order of tens of millions of dollars, but this price is three orders of magnitude less than the US$3 billion estimated at the start of the human genome project (http://www.genome.gov/). Technology already available (e.g. 454 sequencing; Margulies et al., 2005) promises to reduce the cost by at least one or two orders of magnitude. It seems likely that in a few years evolutionary biologists may be liberated from the robotic work of polymerase chain reaction (PCR) and sequencing, or at least the efforts in the laboratory will become more targeted, aimed at filling gaps in the publicly available data.
The availability of massive amounts of sequence data is leading to development of increasingly automated methods for generating phylogenies, for analysing genome organization, and for exploring the analysis of gene families. Examples include Resampled Inference of Orthologs (RIO; Zmasek and Eddy, 2002), OrthologID (Chiu et al., 2006; http://nypg.bio.nyu.edu/orthologid/), and PhiGS (Dehal and Boore, 2006; http://phigs.jgi-psf.org/). A particularly promising approach is the mor project (http://mor.clarku.edu/; Hibbett et al., 2005), developed for phylogenetic placement of new fungal sequences. GenBank is searched automatically at periodic intervals for new rDNA sequences; mor then checks whether they are homobasidiomycetes, downloads them, enters them into an existing alignment, and adds them to an existing phylogenetic tree. This could easily be extended to other genes and other taxa.
Already the availability of whole-genome sequences is making the tedious process of primer design easier. To identify the sequence of a developmentally important gene from a non-model organism, the standard approach is to retrieve from GenBank sequences of orthologous genes from several model species, align the sequences, and then design primers from conserved, but gene-specific regions. For example, in work on genes that control inflorescence form in various grasses, Malcomber et al. (2006) relied on alignments of maize and rice. Many of the genes studied were members of multigene families. Primers were designed such that one or both fell into exons that were locus specific, thus avoiding PCR amplification of the entire gene family. The tedious process of primer design from a multiple nucleotide alignment, which previously took a skilled researcher 4–8 h, has been automated by the program Primaclade (Gadberry et al., 2005).
Genome data are already leading to material progress in phylogenetic reconstruction. Analyses of many genes from multiple whole-genome sequences have been particularly informative in yeast, for which complete genome sequences are available for multiple species. These analyses have shown that thousands of aligned base pairs can lead to robust phylogenetic trees, and that addition of more genes improves phylogenetic accuracy more than does the addition of more taxa (Rokas et al., 2003; Rokas and Carroll, 2005). Similar studies are beginning to appear for plants. For example, de la Torre et al. (2006) have used sequences of expressed sequence tags (ESTs) to assess support for the monophyly of the gymnosperms. Because ESTs are single-pass sequences, they raise real questions about sequence accuracy, although it is possible to compensate for this by requiring contigs to be comprised of multiple reads. In addition to sequence data, information on genome structure could also conceivably be used for phylogenetic reconstruction (Boore, 2006).
Recent estimates suggest that the rice genome includes 37 544 genes (International Rice Genome Sequencing Project, 2005), about half as big again as Arabidopsis, which is estimated to have 25 498 genes (Arabidopsis Genome Initiative, 2000). Many of these genes of course affect morphological development. One surprise has been the frequency of whole-genome duplication, followed by return to a diploid state. Cycles of polyploidization and diploidization have been postulated before, but only with the availability of whole-genome sequences is it possible to see how often this occurs (reviewed in Kellogg and Bennetzen, 2004). Whole-genome duplications have been documented at the origin of Brassicaceae and at the origin of Poaceae, with probable additional duplications near the origin of eudicots and one even more ancient near the base of the angiosperm tree (Arabidopsis Genome Initiative, 2000; Blanc et al., 2000; Paterson et al., 2000).
For studies of EDG, information is needed on the molecular evolution of each gene—patterns of duplications and the speed of evolution. At the moment, this is proceeding gene by gene, or at least gene family by gene family, although the automated programs described above promise to speed this process enormously. Currently most studies of molecular evolution either use whole-genome sequences and analyse an entire gene family [e.g. bHLH (Buck and Atchley, 2003), MADS box (Malcomber and Kellogg, 2005; Theissen et al., 2000), R2R3 Myb (Rabinowicz et al., 1999), and ERF (Nakano et al., 2006)], or use a more laborious PCR approach to determine molecular evolution in the context of a detailed organismal phylogeny (Litt and Irish, 2003; Malcomber and Kellogg, 2006).
Molecular evolution studies are forcing a revision of the terms orthology and paralogy. These terms were first defined by Fitch (1970), and distinguish genes that are connected by speciation (orthologues) from those that are related by duplication (paralogues). Because of the Brassicaceae and Poaceae duplications, no orthologues exist between the two families, by definition. This is true even if one of each gene pair has been lost. More often, one family or the other has genes that the other one does not have. Thus Poaceae has a set of grass-specific genes, and Brassicaceae has a set of crucifer-specific genes. In some cases, it is known that these correlate with distinct morphologies. For example, the grasses have a set of genes that control inflorescence architecture that are not present in Brassicaceae; the grasses also have a number of novel inflorescence structures (Malcomber et al., 2006).
Critical though sequence data are, they cannot by themselves identify which genes have been modified in biochemical function or developmental role at particular times in evolution. Comparative sequence data tend to focus on coding sequences of proteins, which are generally conserved. Although increasingly powerful tests for selection are being developed, most protein sequences appear to be under strong purifying selection (Barrier et al., 2003; Nordborg et al., 2005). For example, Yamasaki et al. (2005) sequenced 1095 genes in 14 inbred lines of maize; of these, 35 genes were invariant among the inbreds and so were sequenced from an additional set of landraces and teosintes. Of these, only eight, or 0.7% of the original set, showed evidence of selection.
Tests for selection generally exclude sequences that vary in length because there is no null model for insertion/deletion mutations as there is for point substitutions. In analyses of genes and gene families, it is common to find regions of the gene in which alignment is difficult because of short repeat sequences (EA Kellogg, personal observation). The number of repeats varies considerably among closely related species, and it is tempting to speculate that they may subtly alter the biochemical function of the protein. It would be of considerable interest to know whether these sequences affect protein function and whether they show any evidence of selection, although the differences are likely to be quantitative rather than qualitative and thus not easily tested. Currently, protein structure modelling is generally not precise enough to determine in silico whether these length differences have an effect on the protein. Techniques for experiments in vitro are so laborious that they cannot be done for large numbers of comparisons, and techniques for analysis in vivo are also slow (see below on transgenics). As protein modelling improves, it may be possible to predict the effect of differences in primary sequence.
Identification of conserved non-coding sequences (CNS) via phylogenetic footprinting also holds some promise for analysis of gene regulation. This requires extensive genome sequence from multiple species, which is only available as genomic sequence data come on-line. EST data, because they are coding sequences, are not useful for such comparative analyses. Development of computational phylogenetic footprinting tools has provided investigators with a rapid means of identifying CNS, thus providing a shortcut to laborious experimental approaches for identifying potential regulatory sequences (Kaplinsky et al., 2002; Guo and Moose, 2003; Inada et al., 2003; Lockton and Gaut, 2005). These methods rely on comparison of taxa sufficiently divergent that regions of high similarity result from active conservation and not chance carryover due to insufficient accumulation of randomizing mutations. Despite the interest in these methods, available analytical methods often fail to detect known transcription factor-binding sites (Hong et al., 2003), suggesting that conservation of primary sequence per se is insufficient for characterizing CNS. Other pattern-matching programs may be helpful in this regard (Yan et al., 2005).
To summarize, sequence data are increasingly available to permit—even demand—phylogenetic analyses. However, identifying the genes that underlie diversification of plant form is proving to be difficult from sequence analysis alone. Differences among both plant and animal species are likely to be due to regulatory evolution (Doebley and Lukens, 1998; Lee et al., 2005; Prud'homme et al., 2006), but most tools for sequence analysis focus on conserved protein-coding sequences. While analyses of coding sequences are necessary for EDG, they are insufficient.
|
Gene expression |
|---|
One way to determine whether a gene could have a different developmental role in different species is to examine its expression pattern. If the expression pattern varies, then it suggests that perhaps the gene is being used in a different way to produce different morphologies. Here the distinction between biochemical function and developmental role becomes particularly critical. Expression studies cannot illuminate biochemical function, nor can they determine whether it is conserved or modified. However, for EDG, a change in developmental role is as interesting as a change in biochemical function.
Comparative gene expression data can help test developmental hypotheses. For example, Vollbrecht et al. (2005) recently cloned the maize gene ramosa1, which they hypothesize is required to suppress outgrowth of short branches (spikelet pairs) in the inflorescence of maize. They then examined the expression of orthologous genes in closely related species (Miscanthus and Sorghum), which developed short inflorescence branches at different times during development. In all cases, onset of ra1 expression correlated with formation of short branches, consistent with the developmental hypothesis. Similarly, Malcomber and Kellogg (2004) tested the hypothesis that LEAFY HULL STERILE1 (LHS1), a SEPALLATA-like MADS box protein, was required to specify the apical flower within the grass spikelet. This hypothesis was first put forward by Cacharrón et al. (1999) to explain the observation that the gene was expressed only in the upper flower of the maize spikelet. However, when multiple disparate species of grasses were investigated, gene expression was restricted to the apical flower only in species with two- or three-flowered spikelets in which floral maturation proceeded from the top down. In species with multiple flowers per spikelet and maturation from the bottom up, LHS1 was expressed in multiple flowers and not restricted solely to the apical one. Thus the protein appears to have changed its developmental role during evolution.
Mapping changes in gene expression pattern on to a phylogeny can help pinpoint the internode along which the gene changed its developmental role and/or biochemical function. For example, Lee et al. (2005) investigated the evolution and expression of CRABS CLAW (CRC) among a variety of eudicots. In Arabidopsis, CRC is required for development of nectaries. Lee et al. (2005) found that orthologous genes are expressed in floral nectaries in several species of Brassicales and in two species of Solanaceae. CRC orthologues were also expressed in extrafloral nectaries in Capparis flexuosa (Capparaceae) and in Gossypium hirsutum (Malvaceae), consistent with the hypothesis that the genes are indeed necessary for nectary development. (This hypothesis was then tested and supported by viral-induced gene silencing, and by transformation experiments in Arabidopsis.) In contrast, CRC was not expressed in nectaries in Aquilegia, a basal eudicot. This suggests that the deployment of CRC for specification of nectary tissue originated at the origin of the core eudicots.
The most precise data on gene expression comes from in situ hybridization data rather than from reverse transcription–PCR (RT–PCR), by providing information on the cellular location of gene expression. Despite the power of in situ hybridization, relatively few laboratories are using it in a comparative context. There are several reasons for this. First, appreciable background information is necessary to interpret the results. The molecular evolution of the gene and gene family must be well enough worked out to develop a gene-specific probe, generally from the 3' untranslated region, which needs to be cloned from each species investigated. Also, the developmental morphology of the plants needs to be understood in enough detail to interpret thin sections from all developmental stages.
The second reason for the paucity of in situ data has to do with challenges of the technique combined with the sociology of science. It requires someone who is familiar enough with anatomy and morphology to carry out careful dissections, fixation, and sectioning—all the skills of a classical morphologist. It also requires someone who is able to work with sequence data and molecular evolution—the skills of a molecular evolutionist. Finally, it requires the patience and ability to work with multistep laboratory procedures, including working with RNA—the skills of a molecular biologist. Few students and post-docs have the time or interest in developing this full range of skills.
Immunolocalization, which assesses the cellular location of proteins rather than RNA, is much less technically challenging than in situ hybridization, and is considerably faster. Polyclonal antibodies are often quite tolerant of modest differences in protein sequence, permitting the same antibodies to be used on multiple species, often as part of the same experiment. (Monoclonal antibodies, because they are based on a single epitope, are likely to be too species specific, making them less suitable for cross-species comparisons.) Immunolocalization was used productively to examine the localization of photosynthetic enzymes in independently derived C4 grasses (Sinha and Kellogg, 1996). These authors found that the two carboxylases, Rubisco and phosphoenolpyruvate carboxylase, were expressed in bundle sheath and mesophyll, respectively, in all C4 lineages, but that pryvuate orthophosphate dikinase (PPDK), light-harvesting chlorophyll a,b-binding protein (LHCP), and the two malic enzymes (NAD-ME and NADP-ME) were expressed in different tissues in representatives of different C4 lineages. More recently, Bharathan et al. (2002) used immunolocalization to study the distribution of KNOTTED1 (KN1)-like proteins in a broad set of land plants, and Kim et al. (2003) applied the technique to the study of PHANTASTICA (PHAN)-like proteins. These two studies together showed that the enormous diversity of leaf morphology seen in seed plants can be explained in part by variations in the localization of KN1-like and PHAN-like proteins.
Despite the appeal of immunolocalization, production of antibodies to plant proteins is difficult, time-consuming, and often unsuccessful. The field of EDG would certainly be advanced by the production of antibodies to as many plant proteins as possible, but this would require a community-wide effort and dedicated funding. Nonetheless, as antibodies become available for developmentally interesting proteins, they could be used much more widely than they are now.
A final limitation to gene expression studies is availability of funding. In the USA at least, funding is difficult to obtain for comparative gene expression as a project of its own. Funding can be garnered for alpha taxonomy and molecular phylogenetics, but simply exploring the expression of a developmentally interesting gene in multiple species is seen as something outside the purview of systematics. At the same time, comparative gene expression data can determine when in evolutionary time gene function (biochemical or developmental) has changed but not how it has changed. The data are thus incomplete from the point of view of a developmental geneticist and not fundable by agencies focused on the identification of gene function.
|
Developmental genetics and quantitative genetics |
|---|
EDG requires information on which genes might affect a given phenotype. The more precisely a set of candidate genes can be specified, the more likely are comparative studies to be able to identify critical changes in those genes. Until the late 1970s and early 1980s when genes controlling morphology began to be cloned from Arabidopsis, there was no gene list—no one knew how many genes might be involved in controlling morphological development, and gene function and developmental role were known for only a tiny handful of genes. The cloning and functional description of the MIKC-type MADS box genes in Arabidopsis, and generation of the ABC model, provided some early hypotheses about the possible control of floral morphology and hence floral diversity (Coen and Meyerowitz, 1991). This provided a small but plausible list of candidates and spawned a minor growth industry in exploring the evolution, expression, and function of MADS box genes. This has led to a number of striking results and emerging hypotheses (Theissen et al., 2000; Münster et al., 2002). Nonetheless, this relatively small number of developmental regulators cannot explain all of angiosperm diversity.
To identify additional genes, reasonable candidates could be identified based on their mutant phenotype in a model organism, and/or the list could be narrowed using quantitative genetics (quantitative trait locus, or QTL, mapping). A striking example of the former approach involves the production of floral pigments in species of Antirrhinum (Schwinn et al., 2006). In this study, three R2R3 Myb genes were cloned and characterized in Antirrhinum majus, which together are responsible for the distribution and intensity of floral colour and patterning. Various accessions of A. majus were then crossed with each of five other species of Antirrhinum. Segregation of the F2 progeny in each of these crosses showed that different alleles of the three Myb genes could account for many of the differences in floral colour among the species of the genus.
In QTL mapping, two plants that differ in one or more interesting traits are crossed, their F1 progeny self-pollinated or backcrossed to one of the parents, and the resulting segregating population of plants used to construct a genetic map using DNA-based markers. Members of the population (backcross or F2) are also scored for the phenotype of interest. If variation in the phenotype correlates with allelic variation at a particular marker, it provides evidence for a gene near that marker controlling the particular phenotype. This approach has been used, for example, to determine the number of genes underlying morphological differences between two species of Mimulus (Bradshaw et al., 1995, 1998), between maize and its wild ancestor teosinte (Doebley and Stec, 1991, 1993), and between species of Antirrhinum (Langlade et al., 2005).
If the DNA-based markers can also be placed on a map or genome sequence of a model organism, it becomes possible to clone the gene underlying the QTL. To date, only a few genes have been cloned in this way (Liu et al., 2002; Ashikari et al., 2005; Li et al., 2006; see also Salvi and Tuberosa, 2005), but this number is likely to increase as more genomic sequence becomes available for more species. Thus, if the gene underlying the QTL can be identified, a function is likely to be ascribed.
QTL studies in non-model species are likely to uncover genes that do not vary in the model. For example, in studies of foxtail millet and its wild relative, green millet, a major QTL was identified on foxtail chromosome VI controlling axillary branching (Doust et al., 2004). The foxtail genome is apparently rearranged in this region compared with maize and rice, but to the extent that comparisons can be made, neither maize nor rice appears to have known mutants mapping to this region with phenotypes that affect vegetative branching.
QTL analyses are statistically powerful, but have low resolution, and it is often difficult to localize the phenotype precisely enough to target a single gene (Whitt and Buckler, 2003; Flint-Garcia et al., 2005). Additional precision can be obtained by generating a set of recombinant inbred lines (RILs) or near isogenic lines (NILs), although this requires a major investment of resources (Lynch and Walsh, 1998).
Another approach that is used in some models is association analysis, which in contrast to QTL analysis can have very high resolution, albeit low power (Whitt and Buckler, 2003; Flint-Garcia et al., 2005). In association analysis, many closely related plants are genotyped, using any of several molecular markers. The resulting genotypes are then searched for mutations that correlate with phenotypic variation. This requires a large amount of genotyping for every correlation detected. The advantage, however, is the high precision with which different alleles can be characterized.
|
Transformation |
|---|
A final challenge, not really solved for any system other than a handful of models, is how to prove that the different forms of the gene of interest really did cause the morphological transition observed. If, for example, it was possible to identify a plausible candidate that affected cotyledon number, and it could be demonstrated that different protein structures and different regulatory patterns correlated with the presence of one versus two cotyledons, how could it be proved that modifications in this gene had led to the origin of the monocots? For many evolutionary biologists a strong correlation would be sufficient evidence, but most geneticists would like additional proof, including information on biochemical function. One way to prove the connection of a genotype with a phenotype is by one or more experiments in which all or part of the locus of interest is placed in a model plant and the resulting phenotype observed.
Most commonly, transformation experiments rely on moving genes and their promoters into a single tractable model (e.g. Arabidopsis), in a genetic analogue of a common garden experiment. In the recent paper on CRC cited above (Lee et al., 2005), the promoter from the CRC gene from Gossypium was fused to a reporter gene and the resulting construct introduced into Arabidopsis. The reporter was expressed in the extrafloral pattern characteristic of Gossypium nectaries, indicating that it could be responsible for formation of nectaries in Gossypium. Similarly, Whipple et al. (2004) expressed B-class MADS box genes from maize in appropriate Arabidopsis mutants, to show that the maize genes had a similar function to the well-characterized B-class genes in Arabidopsis. Maizel et al. (2005) investigated LEAFY (LFY) proteins from a fern (Ceratopteris richardsii), and a moss (Physcomitrella patens). Arabidopsis LFY binds to the promoter of APETALA1, thus regulating the transition to flowering. When the genes were introduced into lfy mutants of Arabidopsis, the Ceratopteris gene partially rescued the mutant, indicating some ability to bind to the appropriate receptors; the moss gene had no effect, indicating no ability to bind. Both the fern and moss genes were also tested in an assay in yeast, and were found to bind weakly (fern) or not at all (moss) to the AP1 promoter. When a critical aspartate residue in the moss protein was replaced with a histidine, making the moss protein more like that of Arabidopsis, DNA binding was partially restored. The authors were thus able to show that the binding specificity of LFY had differentiated since the last common ancestor of mosses and angiosperms.
Moving DNA from several different species into a single heterologous species risks generating phenotypes that are hard to interpret. For example, Yoon and Baum (2004) hypothesized that the LFY protein might be involved in the origin of rosette flowering in several unrelated species of Brassicaceae. In most Brassicaceae, in the transition to flowering the apical meristem shifts from producing leaves with axillary buds to producing suppressed bracts with axillary flowers. In the genera Idahoa, Ionopsidium, and Leavenworthia, on the other hand, flowers are produced in leaf axils. Yoon and Baum cloned the LFY genes and their cis-regulatory sequences from each of the rosette flowering species, and used them to transform lfy mutants of Arabidopsis. Each of the three sorts of transformants was different morphologically. LFY from Ionopsidium largely rescued the Arabidopsis lfy mutation, indicating that modifications of LFY and its promoter were not involved in rosette flowering in Ionopsidium. However, LFY from Idahoa produced some flowering rosettes in Arabidopsis, and LFY from Leavenworthia produced shoots that terminated in flowers. The experiments were thus highly successful in demonstrating that rosette flowering involved different genetic mechanisms in each of the three independent evolutionary origins. Nonetheless, the effects of the Idahoa and Leavenworthia alleles in Arabidopsis are hard to interpret functionally.
This problem would be even more acute if the comparison were between alleles differing in subtle quantitative effects. Moving each allele into, for example, Arabidopsis may simply show that neither allele works very well in the Arabidopsis background. The phenotypes of the transformants could easily be completely uninformative about their actual function in the study species.
A more compelling test would be to perform reciprocal transformations, in which the promoter and coding sequence of a gene from species A was placed into species B, and vice versa (Baum, 2002), in an analogue of a reciprocal transplant experiment. Such experiments could connect the promoter of A with the coding sequence of B and place both into both species to determine if each species could be transformed into something looking like the other. Unfortunately, few systems exist for which reciprocal transformation is possible.
Viral-induced gene silencing (VIGS) has been suggested as a possible ‘universal’ transformation system, and it has been applied successfully in Solanaceae (Ruiz et al., 1998; Burch-Smith et al., 2004), Arabidopsis (Turnage et al., 2002), poppy (Hileman et al., 2005), barley (Holzberg et al., 2002), and wheat (Scofield et al., 2005). The technique has yet to be used to address any questions in EDG, however. VIGS has the advantages of being rapid, working for multiple species, and being able to knock-down the expression of multiple members of a gene family simultaneously (Burch-Smith et al., 2004). On the other hand, it does not lead to a complete gene knock-out, nor does it produce stable transformants (Wang et al., 2006). Reduction of gene expression will certainly be more informative than no manipulation at all, but is often harder to interpret than a complete loss-of-function mutation. The lack of stable transformation likewise is limited because it precludes follow-up genetic studies. Thus, while VIGS holds considerable promise, it will not be a panacea, and will probably create problems in interpretation of phenotypes.
In summary, although transformation is often seen as the critical experiment in studies of developmental genetics, it remains to be seen whether it can produce consistently clear enough answers for EDG.
|
Hypotheses addressed by EDG |
|---|
Multiple hypotheses have been presented regarding the process of evolution at the level of the gene, far too many to present or evaluate here. Some examples include: (i) most genes are under purifying selection (a central tenet of the neutral theory of molecular evolution; Kimura, 1983); (ii) the raw material for genetic diversification comes from gene or genome duplication (Ohno, 1970); (iii) duplicated genes tend to diverge in function (Ohno, 1970; Force et al., 1999); (iv) functional divergence is generally the result of cis-regulatory changes rather than changes in the sequence of the protein itself (Doebley and Lukens, 1998); (v) major morphological changes (‘hopeful monsters’) may be caused by mutations in single genes (Goldschmidt, 1940; Gottlieb, 1984); (vi) similar morphological variation occurs in closely related species by repeated mutations in similar genes (extrapolation from Vavilov's law; Vavilov, 1922); (vii) divergence between closely related species is due to a few genes of large effect (reviewed by Orr and Barton, 2001); (viii) some components of the genetic machinery are more easily modified than others (e.g. Myb domain transcription factors; Schwinn et al., 2006); (ix) major morphological change occurs by changing the relative timing of developmental events (heterochrony; Gould, 1977); and (x) major morphological change occurs by expressing a genetic programme in a novel place (heterotopy; Kellogg, 2000).
Most of these hypotheses are supported by data. However, each is actually a generalization, rather than a single mechanistic hypothesis to be supported or refuted by one critical experiment. Multiple studies will be required on multiple plants to establish how broadly each hypothesis applies. As in most of evolutionary biology, few universal rules are expected, but rather a set of observations that apply to particular groups of plants. Whatever happened at a particular point in evolutionary time was constrained by the prior history of that plant but affected by the particular point in time and space at which change occurred.
Testable hypotheses in EDG have the form ‘Group A has a phenotype X, which is derived from phenotype Y, the latter being present in A's sister groups, B and C. It is hypothesized that the difference between phenotypes X and Y is caused by differences in the biochemical role or expression pattern of gene K'. Therefore, it is necessary to determine whether there are functional (however the term is defined) differences in orthologues of gene K in plants representing A, B, and C. The result of any experiments applies only to groups A, B, and C. If the transition between phenotypes X and Y occurs in other groups of plants, separate experiments are required to determine if the genetics changes are the same or different. There is a complex interplay between phylogeny, hypotheses in EDG, and the ability to generalize results.
|
Desiderata |
|---|
To understand evolutionary history at the level of the gene requires a handful of technical advances plus a great deal more data on gene expression patterns and biochemical function. The technical advances include higher throughput methods for investigating tissue-specific gene expression, which might include production of antibodies to large numbers of plant proteins. Easier and more easily generalized methods of transformation will also permit more exploratory studies of gene function in different species. Most important, however, is the accumulation of more data on more species. For example, a broad survey of the expression of each of the well-characterized MADS-box genes in several representatives of each angiosperm order could quickly establish what the common components were of the MADS-box transcription factor complex. At the same time, it would pinpoint taxa that would repay more in-depth sampling. Such a survey project would be completely feasible with the plant resources of a major botanical garden within 3–5 years, even using current techniques of in situ hybridization. If antibodies were available for even one important protein (e.g. KN1, PHAN), such a survey could be accomplished even more rapidly and with considerable taxonomic depth. Broad surveys of gene expression can then identify genes that would repay the slow process of transformation and evaluating transgenic phenotypes.
An obvious way for EDG to test some of the generalizations that abound in the literature is to explore gene expression the way that sequence variation is currently explored—with many genes and many taxa. Certainly exploration of new techniques would help, but is not absolutely necessary. The major need is simply identification of a source of funding for research that is truly interdisciplinary. The short-term research programme will be too molecular and mechanistic for systematists, and not mechanistic enough for geneticists. However, in the long run, the data produced will illuminate the thousands of ways that gene function can be modified, and at the same time will provide a rigorous genetic basis for the characters used by taxonomists.
|
Acknowledgements |
|---|
I would like to thank N Battey and P Rudall for the invitation to contribute this paper. Work in my laboratory is supported by grants from the National Science Foundation.
|
Abbreviations |
|---|
CNS, conserved non-coding sequence; EDG, evolutionary developmental genetics; EST , expressed sequence tag; QTL, quantitative trait locus; VIGS, viral-induced gene silencing.
|
References |
|---|
Arabidopsis Genome Initiative. (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408:796–815.[CrossRef][Medline]
Ashikari M, Sakakibara H, Lin S, Yamamoto T, Takashi T, Nishimura A, Angeles ER, Qian Q, Kitano H, Matsuoka M. (2005) Cytokinin oxidase regulates rice grain production. Science 309:741–745.
Barrier M, Bustamante CD, Yu J, Purugganan MD. (2003) Selection on rapidly evolving proteins in the Arabidopsis genome. Genetics 163:723–733.
Baum DA. (2002) Identifying the genetic causes of phenotypic evolution: a review of experimental strategies. In Cronk QCB, Bateman RM, Hawkins JA (Eds.). Developmental genetics and evolution(Taylor & Francis, London) pp. 493–507.
Beilstein MA, Al-Shehbaz IA, Kellogg EA. (2006) Brassicaceae phylogeny and trichome evolution. American Journal of Botany 93:607–619.
Bharathan G, Goliber TE, Moore C, Kessler S, Pham T, Sinha NR. (2002) Homologies in leaf form inferred from KNOX1 gene expression during development. Science 296:1858–1860.
Blanc G, Barakat A, Guyot R, Cooke R, Delseny M. (2000) Extensive duplication and reshuffling in the Arabidopsis genome. The Plant Cell 12:1093–1101.
Boore JL. (2006) The use of genome-level characters for phylogenetic reconstruction. Trends in Ecology and Evolution 7:201.
Bradshaw HD, Otto KG, Frewen BE, McKay JK, Schemske DW. (1998) Quantitative trait loci affecting differences in floral morphology between two species of monkeyflower (Mimulus). Genetics 149:367–382.
Bradshaw HD, Wilbert SM, Otto KG, Schemske DW. (1995) Genetic mapping of floral traits associated with reproductive isolation in monkey flowers (Mimulus). Nature 376:762–765.[CrossRef]
Bremer B, Bremer K, Heidari N, Erixon P, Anderberg AA, Olmstead RG, Källersjö M, Barkhordarian E. (2002) Phylogenetics of asterids based on 3 coding and 3 non-coding chloroplast DNA markers and the utility of non-coding DNA at higher taxonomic levels. Molecular Phylogenetics and Evolution 23:274–301.
Buck M and Atchley W. (2003) Phylogenetic analysis of plant basic helix–loop–helix proteins. Journal of Molecular Evolution 56:742–750.[CrossRef][Web of Science][Medline]
Burch-Smith TM, Anderson JC, Martin GB, Dinesh-Kumar SP. (2004) Applications and advantages of virus-induced gene silencing for gene function studies in plants. The Plant Journal 39:734–746.[CrossRef][Web of Science][Medline]
Burleigh JG and Mathews S. (2004) Phylogenetic signals in nucleotide data from seed plants: implications for resolving the seed plant tree of life. American Journal of Botany 91:1599–1613.
Cacharrón J, Saedler H, Theissen G. (1999) Expression of MADS box genes ZMM8 and ZMM14 during inflorescence development of Zea mays discriminates between the upper and the lower floret of each spikelet. Development, Genes and Evolution 209:411–420.[CrossRef][Web of Science][Medline]
Chiu JC, Lee EK, Egan MG, Sarkar IN, Coruzzi GM, DeSalle R. (2006) OrthologID: automation of genome-scale ortholog identification within a parsimony framework. Bioinformatics 22:699–707.
Choi H-K, Luckow MA, Doyle J, Cook DR. (2006) Development of nuclear gene-derived molecular markers linked to legume genetic maps. Molecular Genetics and Genomics 276:56–70.[CrossRef][Web of Science][Medline]
Citerne HL, Luo D, Pennington RT, Coen E, Cronk QCB. (2003) A phylogenomic investigation of CYCLOIDEA-like TCP genes in the Leguminosae. Plant Physiology 131:1042–1053.
Coen ES and Meyerowitz EM. (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353:31–37.[CrossRef][Medline]
Cubas P, Coen E, Zapater JM. (2001) Ancient asymmetries in the evolution of flowers. Current Biology 11:1050–1052.[CrossRef][Web of Science][Medline]
Dehal PS and Boore JL. (2006) A phylogenomic gene cluster resource: the Phylogenetically Inferred Groups (PhIGs) database. BMC Bioinformatics 7:201.[CrossRef][Medline]
De la Torre JE, Egan MG, Katari M, Brenner ED, Stevenson DW, Coruzzi GM, DeSalle R. (2006) ESTimating plant phylogeny: lessons from partitioning. BMC Evolutionary Biology 6:48.
Doebley J and Lukens L. (1998) Transcriptional regulators and the evolution of plant form. The Plant Cell 10:1075–1082.
Doebley J and Stec A. (1991) Genetic analysis of the morphological differences between maize and teosinte. Genetics 129:285–295.[Abstract]
Doebley J and Stec A. (1993) Inheritance of the morphological differences between maize and teosinte: comparison of results for two F2 populations. Genetics 134:559–570.[Abstract]
Donoghue MJ, Ree RH, Baum DA. (1998) Phylogeny and the evolution of flower symmetry in the Asteridae. Trends in Plant Science 3:311–317.[CrossRef][Web of Science]
Doust AN, Devos KM, Gadberry MD, Gale MD, Kellogg EA. (2004) Genetic control of branching in foxtail millet. Proceedings of the National Academy of Sciences, USA 101:9045–9050.
Doust AN, Devos KM, Gadberry MD, Gale MD, Kellogg EA. (2005) The genetic basis for inflorescence variation between foxtail and green millet (Poaceae). Genetics 169:1659–1672.
Doust AN and Kellogg EA. (2002) Inflorescence diversification in the panicoid ‘bristle grass’ clade (Paniceae, Poaceae): evidence from molecular phylogenies and developmental morphology. American Journal of Botany 89:1203–1222.
Doust AN, Penly AM, Jacobs SWL, Kellogg EA. (2006) Congruence, conflict and polyploidization shown by nuclear and chloroplast markers in the monophyletic ‘bristle clade’ (Paniceae, Panicoideae, Poaceae). Systematic Botany (in press).
Endress PK. (2003) Morphology and angiosperm systematics in the molecular era. Botanical Review 68:545–570.[CrossRef]
Endress PK, Baas P, Gregory M. (2000) Systematic plant morphology and anatomy—50 years of progress. Taxon 49:401–434.[CrossRef][Web of Science]
Evans MW. (1940) Developmental morphology of the growing point of the shoot and the inflorescence in grasses. Journal of Agricultural Research 61:481–520.
Fitch WM. (1970) Distinguishing homologous and analogous proteins. Systematic Zoology 19:99–113.
Flint-Garcia SA, Thuillet A-C, Yu J, Pressoir G, Romero SM, Mitchell SE, Doebley J, Kresovich S, Goodman MM, Buckler ES. (2005) Maize association population: a high-resolution platform for quantitative trait locus dissection. The Plant Journal 44:1054–1064.[CrossRef][Web of Science][Medline]
Force A, Lynch M, Pickett FB, Amores A, Yan Y, Postlethwait J. (1999) Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151:1531–1545.
Friedman WE, Moore RC, Purugganan MD. (2004) The evolution of plant development. American Journal of Botany 91:1726–1741.
Friedman WE and Williams JH. (2004) Developmental evolution of the sexual process in ancient flowering plant lineages. The Plant Cell 16:Supplement, S119–S132.
Gadberry MD, Malcomber ST, Doust AN, Kellogg EA. (2005) Primaclade—a flexible tool to find conserved PCR primers across multiple speices. Bioinformatics 21:1263–1264.
Goldschmidt RB. (1940) The material basis of evolution(Yale University Press, New Haven, CT).
Gottlieb LD. (1984) Genetics and morphological evolution in plants. American Naturalist 123:681–709.[CrossRef][Web of Science]
Gould SJ. (1977) Ontogeny and phylogeny(Harvard University Press, Cambridge, MA).
Graham SW, Zgurski JM, McPherson MA, et al. (2006) Robust inference of monocot deep phylogeny using an expanded multigene plastid data set. In Columbus JT, Friar EA, Hamilton CW, Porter JM, Prince LM, Simpson MG (Eds.). Monocots: comparative biology and evolution(Rancho Santa Ana Botanical Garden, Claremont, CA) (in press).
Grass Phylogeny Working Group. (2001) Phylogeny and subfamilial classification of the Poaceae. Annals of the Missouri Botanical Garden 88:373–457.[CrossRef][Web of Science]
Guo H and Moose SP. (2003) Conserved non-coding sequences among cultivated cereal genomes identify candidate regulatory sequence elements and patterns of promoter evolution. The Plant Cell 15:1143–1158.
Hibbett DS, Nilsson RH, Snyder M, Fonseca M, Costanzo J, Shonfeld M. (2005) Automated phylogenetic taxonomy: an example in the homobasidiomycetes (mushroom-forming fungi). Systematic Biology 54:660–668.
Hileman LC and Baum DA. (2003) Why do paralogs persist? Molecular evolution of CYCLOIDEA and related floral symmetry genes in Antirrhineae (Veronicaceae). Molecular Biology and Evolution 20:591–600.
Hileman LC, Drea S, Martino G, Litt A, Irish VF. (2005) Virus-induced gene silencing is an effective tool for assaying gene expression in the basal eudicot species Papaver somniferum (opium poppy). The Plant Journal 44:334–341.[CrossRef][Web of Science][Medline]
Holzberg S, Brosio P, Gross C, Pogue GP. (2002) Barley stripe mosaic virus-induced gene silencing in a monocot plant. The Plant Journal 30:315–327.[CrossRef][Web of Science][Medline]
Hong RL, Hamaguchi L, Busch MA, Weigel D. (2003) Regulatory elements of the floral homeotic gene AGAMOUS identified by phylogenetic footprinting and shadowing. The Plant Cell 15:1296–1309.
Hughes CE, Eastwood RJ, Bailey CD. (2006) From famine to feast? Selecting nuclear DNA sequence loci for plant species-level phylogeny reconstruction. Philosophical Transactions of the Royal Society B: Biological Sciences 361:211–225.
Inada DC, Bashir A, Lee C, Thomas BC, Ko C, Goff SA, Freeling M. (2003) Conserved noncoding sequences in the grasses. Genome Research 13:2030–2041.
International Rice Genome Sequencing Project. (2005) The map-based sequence of the rice genome. Nature 436:793–800.[CrossRef][Medline]
Jansen RK, Kaittanis C, Lee S-B, Saski C, Tomkins J, Alverson AJ, Daniell H. (2006) Phylogenetic analyses of Vitis (Vitaceae) based on complete chloroplast genome sequences: effects of taxon sampling and phylogenetic methods on resolving relationships among rosids. BMC Evolutionary Biology 6:32.
Kaplinsky NJ, Braun DM, Penterman J, Goff SA, Freeling M. (2002) Utility and distribution of conserved noncoding sequences in the grasses. Proceedings of the National Academy of Sciences, USA 99:6147–6151.
Kellogg EA. (1996) Integrating genetics, phylogenetics and developmental biology. In Sobral BWS (Ed.). Impact of plant molecular genetics(Birkhäuser, Boston) pp. 159–172.
Kellogg EA. (2000) The grasses: a case study in macroevolution. Annual Review of Ecology and Systematics 31:217–238.[CrossRef][Web of Science]
Kellogg EA. (2004) Evolution of developmental traits. Current Opinion in Plant Biology 7:92–98.[CrossRef][Web of Science][Medline]
Kellogg EA and Bennetzen JL. (2004) The evolution of nuclear genome structure in seed plants. American Journal of Botany 91:1709–1725.
Kellogg EA, Hiser KM, Doust AN. (2004) Taxonomy, phylogeny, and inflorescence development of the genus Ixophorus (Panicoideae: Poaceae). International Journal of Plant Sciences 165:1089–1105.[CrossRef]
Kellogg EA and Shaffer HB. (1993) Model organisms in evolutionary studies. Systematic Biology 42:409–414.[CrossRef]
Kim M, McCormick S, Timmermans M, Sinha N. (2003) The expression domain of PHANTASTICA determines leaflet placement in compound leaves. Nature 424:438–443.[CrossRef][Medline]
Kimura M. (1983) The neutral theory of molecular evolution(Cambridge University Press, Cambridge).
Kuzoff RK, Hufford L, Soltis DE. (2001) Structural homology and developmental transformations associated with ovary diversification in Lithophragma (Saxifragaceae). American Journal of Botany 88:196–205.
Langlade NB, Feng X, Dransfield T, Copsey L, Hanna AI, Thébaud C, Bingham A, Hudson A, Coen E. (2005) Evolution through genetically controlled allometry space. Proceedings of the National Academy of Sciences, USA 102:10221–10226.
Lee J-Y, Baum SF, Oh S-H, Jiang C-Z, Chen J-C, Bowman JL. (2005) Recruitment of CRABS CLAW to promote nectary development within the eudicot clade. Development 132:5021–5032.
Li C, Zhou A, Sang T. (2006) Rice domestication by reducing shattering. Science 311:1936–1939.
Litt A and Irish VF. (2003) Duplication and diversification in the APETALA1/FRUITFULL floral homeotic gene lineage: implications for the evolution of floral development. Genetics 165:821–833.
Liu J, VanEck J, Cong B, Tanksley SD. (2002) A new class of regulatory genes underlying the cause of pear-shaped tomato fruit. Proceedings of the National Academy of Sciences, USA 99:13302–13306.
Lockton S and Gaut BS. (2005) Plant conserved non-coding sequences and paralogue evolution. Trends in Genetics 21:60–65.[CrossRef][Web of Science][Medline]
Lynch M and Walsh B. (1998) Genetics and analysis of quantitative traits(Sinauer, Sunderland, MA).
Maizel A, Busch MA, Tanahashi T, Perkovic J, Kato M, Hasebe M, Weigel D. (2005) The floral regulator LEAFY evolves by substitutions in the DNA binding domain. Science 308:260–263.
Malcomber ST and Kellogg EA. (2004) Heterogeneous expression patterns of the SEPALLATA-like gene LEAFY HULL STERILE1 (LHS1) in grasses (Poaceae) suggests separate roles in meristem determinacy, palea/lemma identity, and flower sexuality. The Plant Cell 16:1692–1706.
Malcomber ST and Kellogg EA. (2005) SEPALLATA gene diversification: brave new whorls. Trends in Plant Science 10:427–435.[CrossRef][Web of Science][Medline]
Malcomber ST and Kellogg EA. (2006) Evolution of unisexual flowers in grasses (Poaceae) and the putative sex-determination gene, TASSELSEED2 (TS2). New Phytologist 170:885–899.[CrossRef][Web of Science][Medline]
Malcomber ST, Preston JC, Reinheimer R, Kossuth J, Kellogg EA. (2006) Developmental gene evolution and the origin of grass inflorescence diversity. Advances in Botanical Research (in press).
McAbee JM, Kuzoff RK, Gasser CS. (2005) Mechanisms of derived unitegmy among Impatiens species. The Plant Cell 17:1674–1684.
Margulies M, Egholm M, Altman WE, et al. (2005) Genome sequencing in microfabricated high-density picolitre reactors. Nature 437:376–380.[Medline]
Mason-Gamer RJ. (2005) The ß-amylase genes of grasses and a phylogenetic analysis of the Triticeae. American Journal of Botany 92:1045–1058.
Matthews ML and Endress PK. (2005) Comparative floral structure and systematics in Celastrales (Celastraceae, Parnassiaceae, Lepidobotryaceae). Botanical Journal of the Linnean Society 149:129–194.[CrossRef]
McMahon M and Hufford L. (2005) Evolution and development in the Amorphoid clade (Amorpheae: Papilionoideae: Leguminosae): petal loss and dedifferentiation. International Journal of Plant Sciences 166:383–396.[CrossRef]
Moncur MW. (1981) Floral initiation in field crops(CSIRO, Canberra, ACT, Australia).
Münster T, Deleu W, Wingen LU, Ouzunova M, Cacharrón J, Faigl W, Werth S, Kim JTT, Saedler H, Theissen G. (2002) Maize MADS-box genes galore. Maydica 47:287–301.[Web of Science]
Nakano T, Suzuki K, Fujimara T, Shinshi H. (2006) Genome-wide analysis of the ERF gene family in Arabidopsis and rice. Plant Physiology 140:411–432.
Nordborg M, Hu TT, Ishino Y, et al. (2005) The pattern of polymorphism in Arabidopsis thaliana. Public Library of Science Biology 3:e196.
Ohno S. (1970) Evolution by gene duplication(Springer-Verlag, Berlin).
Orr HA and Barton NH. (2001) The genetics of species differences. Trends in Ecology and Evolution 16:343–350.
Paterson AH, Bowers JE, Burow MD, Draye X, Elsik CG, Jiang CX, Katsar CS, Lan TH, Lin YR, Ming R. (2000) Comparative genomics of plant chromosomes. The Plant Cell 12:1523–1540.
Payer J-B. (1857) Traité de organogénie comparée de la fleur(Masson, Paris).
Prud'homme B, Gompel N, Rokas A, Kassneer VA, Williams TM, Yeh S-D, True JR, Carroll SB. (2006) Repeated morphological evolution through cis-regulatory changes in a pleiotropic gene. Nature 440:1050–1053.[CrossRef][Web of Science][Medline]
Qiu Y-L, Lee J, Bernasconi-Quadroni F, Soltis DE, Soltis PS, Zanis MJ, Zimmer EA, Chen Z, Savolainen V, Chase MW. (1999) The earliest angiosperms: evidence from mitochondrial, plastid and nuclear genes. Nature 402:404–407.[CrossRef]
Rabinowicz PD, Braun EL, Wolfe AD, Bowen B, Grotewold E. (1999) Maize R2R3 Myb genes: sequence analysis reveals amplification in higher plants. Genetics 153:427–444.
Rokas A and Carroll SB. (2005) More genes or more taxa? The relative contribution of gene number and taxon number to phylogenetic accuracy. Molecular Biology and Evolution 22:1337–1344.
Rokas A, Williams BL, King N, Carroll SB. (2003) Genome-scale approaches to resolving incongruence in molecular phylogenies. Nature 425:798–804.[CrossRef][Medline]
Rudall PJ, Stuppy W, Cunniff J, Kellogg EA, Briggs BG. (2005) Evolution of reproductive structures in grasses (Poaceae) inferred by sister-group comparison with their putative closest living relatives, Ecdeiocoleaceae. American Journal of Botany 92:1432–1443.
Ruiz MT, Voinnet O, Baulcombe DC. (1998) Initiation and maintenance of virus-induced gene silencing. The Plant Cell 10:937–946.
Salvi S and Tuberosa R. (2005) To clone or not to clone plant QTLs: present and future challenges. Trends in Plant Science 10:297–304.[CrossRef][Web of Science][Medline]
Schwinn K, Venail J, Shang Y, Mackay S, Alm V, Butelli E, Oyama R, Bailey P, Davies K, Martin C. (2006) A small family of MYB-regulatory genes controls floral pigmentation intensity and patterning in the genus Antirrhinum. The Plant Cell 18:831–851.
Scofield SR, Huang L, Brandt AS, Gill BS. (2005) Development of a virus-induced gene-silencing system for hexaploid wheat and its use in functional analysis of the Lr21-mediated leaf rust resistance pathway. Plant Physiology 138:2165–2173.
Service RF. (2006) The race for the $1000 genome. Science 311:1544–1546.
Sinha NR and Kellogg EA. (1996) Parallelism and diversity in multiple origins of C4 photosynthesis in the grass family. American Journal of Botany 83:1458–1470.[CrossRef][Web of Science]
Small RL, Cronn RC, Wendel JF. (2004) Use of nuclear genes for phylogeny reconstruction in plants. Australian Systematic Botany 17:145–170.[CrossRef][Web of Science]
Soltis DE and Hufford L. (2002) Ovary position diversity in Saxifragaceae: clarifying the homology of epigyny. International Journal of Plant Sciences 163:277–293.[CrossRef]
Stevens PF. (1994) The development of biological systematics(Columbia University Press, New York).
Theissen G, Becker A, Rosa AD, Kanno A, Kim JT, Munster T, Winter K-U, Saedler H. (2000) A short history of MADS-box genes in plants. Plant Molecular Biology 42:115–149.[CrossRef][Web of Science][Medline]
Tucker SC and Hodges SA. (2005) Floral ontogeny of Aquilegia, Semiaquilegia, and Enemion (Ranunculaceae). International Journal of Plant Sciences 166:557–574.[CrossRef]
Turnage MA, Muangsan N, Peele CG, Robertson D. (2002) Geminivirus-based vectors for gene silencing in Arabidopsis. The Plant Journal 30:107–117.[CrossRef][Web of Science][Medline]
Vavilov N. (1922) The law of homologous series in variation. Journal of Genetics 12:47–89.[Web of Science]
Vollbrecht E, Springer PS, Goh L, Buckler ES IV, Martienssen R. (2005) Architecture of floral branch systems in maize and related grasses. Nature 436:1119–1126.[CrossRef][Medline]
von Balthazar M, Alverson WS, Schonenberger J, Baum DA. (2004) Comparative floral development and androecium structure in Malvoideae (Malvaceae s.l.). International Journal of Plant Sciences 165:445–473.[CrossRef]
Wang CC, Cai XZ, Wang XM, Zheng Z. (2006) Optimisation of tobacco rattle virus-induced gene silencing in Arabidopsis. Functional Plant Biology 33:347–355.[CrossRef]
Whipple CJ, Ciceri P, Padilla CM, Ambrose BA, Bandong SL, Schmidt RJ. (2004) Conservation of B-class floral homeotic gene function between maize and Arabidopsis. Development 131:6083–6091.
Whitt SR and Buckler ES. (2003) Using natural allelic diversity to evaluate gene function. In Grotewold E (Ed.). Plant functional genomics: methods and protocols(Humana Press, Totowa, NJ) Vol. 236: pp. 123–139.[CrossRef]
Whittall JB, Medina-Marino A, Zimmer EA, Hodges SA. (2006) Generating single-copy nuclear gene data for a recent adaptive radiation. Molecular Phylogenetics and Evolution 39:124–134.[CrossRef][Web of Science][Medline]
Xu F and Rudall PJ. (2006) Comparative floral anatomy and ontogeny in Magnoliaceae. Plant Systematics and Evolution 258:1–15.[CrossRef]
Yamasaki M, Tenaillon MI, Bi IV, Schroeder SG, Sanchez-Villeda H, Doebley JF, Gaut BS, McMullen MD. (2005) A large-scale screen for artificial selection in maize identifies candidate agronomic loci for domestication and crop improvement. The Plant Cell 17:2859–2872.
Yan T, Yoo D, Berardini TZ, Mueller LA, Weems DC, Weng S, Cherry JM, Rhee SY. (2005) PatMatch: a program for finding patterns in peptide and nucleotide sequences. Nucleic Acids Research 33:W262–W266.
Yoon H-S and Baum DA. (2004) Transgenic study of parallelism in plant morphological evolution. Proceedings of the National Academy of Sciences, USA 101:6524–6529.
Zmasek CM and Eddy SR. (2002) RIO: analyzing proteomes by automated phylogenomics using resampled inference of orthologs. BMC Bioinformatics 3:14.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Doust Architectural Evolution and its Implications for Domestication in Grasses Ann. Bot., May 3, 2007; (2007) mcm040v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Flowering Newsletter bibliography for 2006 J. Exp. Bot., April 20, 2007; (2007) erm028v2. [Full Text] [PDF]
|
|
|
|
|
|
R. M. Bateman, J. Hilton, and P. J. Rudall Morphological and molecular phylogenetic context of the angiosperms: contrasting the 'top-down' and 'bottom-up' approaches used to infer the likely characteristics of the first flowers J. Exp. Bot., October 1, 2006; 57(13): 3471 - 3503. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||