| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
American Society of Plant Physiologists
Functional Evolutionary Genetics and Plant AdaptationLinking Phenotype and Genotypeneckardt{at}aspp.org
Conway Morris (2000) LINKAGE DISEQUILIBRIUM AND SINGLE NUCLEOTIDE POLYMORPHISMS
One of the most important concepts in studies of both functional and evolutionary genetics is linkage disequilibrium (LD). The extent and ramifications of LD in Arabidopsis thaliana was a topic of great interest at the workshop and was discussed in detail in talks given by Magnus Nordborg (University of Southern California, Los Angeles) and Karl Schmid (Max Planck Institute for Chemical Ecology). In a population with several polymorphic loci, LD occurs when alleles at two loci occur together more often than expected (Figure 1)
. LD is a statistical measure that quantifies the nonindependence of genotypes at several loci. When genotypes are correlated between loci, the information from a marker may predict genotypic function at another locus—hence, the importance of LD in functional genomics and studies of hu-man disease. In some instances, the occurrence of LD suggests that selection favors chromosomes carrying particular multilocus genotypes. However, Nordborg pointed out that the extent of LD is a result of a complex historical process (M. Nordborg and S. Tavaré provide a comprehensive discussion of this topic in an unpublished review available at http://www-hto.usc.edu/papers/abstracts/tig.html). LD is expected to vary greatly because of the randomness of history, but the average rate of decay of LD (i.e., the genetic or physical distance over which LD can be measured) depends on the demographic history of the population and a number of other factors. In particular, in plant populations, the extent of selfing versus outcrossing can have a strong effect. Nordborg showed that LD is more extensive in selfing species, which is expected because populations of self-pollinating individuals tend to be largely homozygous (Nordborg, 2000
What was of great interest to those at the meeting was that Nordborg also showed that although LD is extensive in A. thaliana, it is far from genome wide (as had been considered by many to be the case). He presented evidence that LD in A. thaliana decays over a distance of 100 to 200 kb. This has significance for those interested in using single nucleotide polymorphisms (SNPs) in association studies (i.e., linking phenotype with a SNP marker). A genome-wide "SNP chip" for conducting association studies with A. thaliana could be constructed using  2000 SNPs spaced at 50- to 100-kb intervals, a prospect that is possible with currently available resources and technologies. Nordborg favored the idea of sequencing 500- to 1000-bp segments every 50 to 100 kb (1% of the genome) in 100 individuals, which could be used to produce a SNP chip with 10 or so tightly linked SNPs every 50 to 100 kb. Schmid discussed efforts of the Max Planck SNP Consortium to develop a database of 1300 SNPs, evenly spaced at 100 kb, by using a selection of 12 A. thaliana accessions that are among the most highly diverged (based on previous amplified fragment length polymorphism studies) plus data from the closely related species Arabidopsis lyrata and Arabis drummondii. GENETIC VARIATION AND THE MATING GAME
One of the major topics of discussion was the idea that the type of mating system (i.e., selfing versus outcrossing) can have a significant effect on the evolution of plant genes. A high degree of selfing produces highly homozygous populations, which effectively reduces recombination (as discussed above) and complicates efforts to determine the adaptive significance of a particular trait or gene. Because A. thaliana is a highly selfing species, a number of evolutionary geneticists have turned to the closely related outcrossing species A. lyrata as a model for investigating adaptive variation (North American and European subspecies formerly known as Arabis lyrata, Arabis petraea, and Cardaminopsis petraea are now considered subspecies of Arabidopsis lyrata [O'Kane and Al-Shehbaz, 1997 The effect of mating system on DNA variation was discussed in a number of presentations. Because selfing produces populations of homozygotes and reduces the effective recombination rate, it is commonly thought that selfing reduces DNA variation. However, selfing actually has little effect on the overall amount of DNA variation within a species. Rather, it affects how variation is structured. DNA variation in a selfing species tends to be structured into distinct haplotypes coupled with extensive LD (a haplotype is a specific multilocus combination of alleles that occurs in an individual, such as A1B1). Conversely, recombination among heterozygotes in an outcrossing species reduces LD and breaks up haplotypes.
Outi Savolainen (University of Oulu, Finland) compared sequence variation in A. lyrata and A. thaliana for alcohol dehydrogenase (Adh), the model gene of sequence variation in plants, and the candidate genes CONSTANS (CO) and FRIGIDA (FRI), which have been associated with flowering time in a number of species. The overall (within-species) level of variation was about the same in both species; however, a higher degree of within-population polymorphism was found in A. lyrata compared with that of A. thaliana populations. Selfing is expected to decrease the effective population size, and thus within-population variation, by twofold relative to outcrossing species, but within-population variation was decreased more than twofold in A. thaliana relative to A. lyrata. In addition to the effect of selfing, the recent and rapid expansion of A. thaliana from Asia to other parts of the world, leading to founder effects or so-called bottlenecks, is seen as an important factor leading to low within-population variation in A. thaliana (Savolainen et al., 2000
The difficulty of inferring the action of natural selection in selfing species was raised in several presentations. For example, Aguadé presented data on sequence variation in genes encoding enzymes of the phenylpropanoid pathway in A. thaliana. Two patterns of variation emerged: in some genes (typified by FAH1), variation was structured into two highly divergent haplotypes that exhibited little within-haplotype diversity, whereas in other genes (typified by CHI), there was no clear evidence for two haplotypes. In outcrossing species, the presence of two highly divergent haplotypes in regions with normal levels of recombination can be a strong indication of balancing selection in that region. However, in a selfing species, this pattern could equally be attributable to genetic drift, given that recombination is effectively reduced (Aguadé, 2001 Information about the potential adaptive significance of genes can come from examining patterns of DNA variation in the context of gene function and phenotype. Michael Purugganan (North Carolina State University, Raleigh) presented data on DNA variation among genes involved in flower development, which again suggested that two basic patterns of variation exist in A. thaliana: genes that are somehow organized into distinct haplotypes versus those that are not. Thus, one group of genes, including TFL, showed a low amount of variation in the coding sequence, but high levels of variation organized into distinct haplotypes in the promoter region; other genes showed either low (LFY) or normal (AP3, PI) levels of variation throughout coding sequence and promoter regions. The low level of variation in the LFY floral meristem identity gene is consistent with the idea that natural selection has driven an advantageous LFY mutation to fixation in all A. thaliana populations (directional selection). But selection on the LFY coding region also could be attributable to genetic "hitchhiking" (e.g., directional selection on a nearby locus, resulting in reduced nucleotide variation for a chromosomal region surrounding the selected locus). The size of the affected region depends on LD; thus, it will be larger in a selfing population. TFL1, which appears to function by repressing LFY, is perhaps even more complex. Although the low amount of variation in the coding sequence may suggest directional (purifying) selection (as for LFY), two distinct promoter haplotypes suggest balancing (diversifying) selection acting on the promoter region. Association studies with 21 A. thaliana accessions suggest that the TFL promoter types are associated with different numbers of inflorescence meristems. Moreover, Purugganan presented results from field experiments conducted by his collaborators, the Arabidopsis ecologists Johanna Schmitt and Cynthia Weinig (Brown University, Providence, RI), which suggest that the number of inflorescence branches has an impact on overall plant fitness.
Wolfgang Stephan (Ludwig-Maxi-milians-Universität, Munich, Germany) presented data on the effects of recombination and mating system on DNA variation in the tomato genus Lycopersicon. His group has studied patterns of nucleotide diversity in the selfing species L. chmielewski and L. pimpinellifolium and the outcrossers L. peruvianum, L. chilense, and L. hirsutum. Although the amount of within-population polymorphism was much lower in the selfing relative to the outcrossing species, the level of nucleotide diversity was not highly correlated with recombination rate, as it is in Drosophila. Thus, there is a strong effect of mating system on the structure of nucleotide diversity, but it is by no means the only factor. Stephan suggested that demographic factors, such as population subdivision and population size, also have dramatic effects on the structure of nucleotide diversity (Stephan and Langley, 1998 DOES EVOLUTION PLAY FAVORITES?
A number of speakers presented evidence that certain loci have been "recruited" for a particular function inde-pendently multiple times, either in different species or within a single species. This pattern has been observed before. For example, Paterson et al. (1995) Morning glory (Ipomoea spp) has a tremendous amount of variation in flower color. Rausher's group examined the phylogeny of 40 Ipomoea spp and found that white flower color has evolved independently many times. Furthermore, transitions in flower color were found to be highly asymmetric; for example, there are many cases of purple to white transitions but not of the opposite transition of white to purple. The data collected so far suggest that this pattern may be the result of "knockout" mutations. For example, I. purpurea (normally purple flower) produces some variants at the W locus that produce white flowers in homozygous individuals and white-purple flowers in heterozygotes. W was cloned and found to be homologous with the myb transcription factor An2 in petunia, which is known to be involved in flower color. The white-flowered I. purpurea variants seem to have frameshift deletions and an introduced stop codon in this gene. Rausher's group will test the knockout hypothesis by introducing a wild-type copy of W, presumably the functional gene, into white-flowered plants. There are at least eight genetic routes that can produce white-flowered plants in Ipomoea. Rausher's group is collecting data to determine the genetic changes associated with flower color in 15 independently evolved white-flowered species. They also will test various hypotheses that the pattern of knockout mutations is not random. For example, there may be fewer deleterious pleiotropic effects associated with regulatory genes that are highly specifically expressed only in certain tissues (such as An2, which is expressed specifically in flower petals in petunia) relative to structural genes or other reg-ulatory genes that are more widely expressed throughout the plant. Another hypothesis is that there are fewer deleterious pleiotropic effects associated with mutations in downstream compared with upstream structural genes. The data collected so far suggest that downstream structural genes may be more subject to mutations that persist than are upstream structural genes.
Caroline Dean (John Innes Centre, Norwich, UK) gave another example of nonrandom occurrence of mutations at the FRI locus in A. thaliana, which is a major determinant of flowering time (Johanson et al., 2000 THE HOLY GRAIL: LINKING PHENOTYPE AND GENOTYPE Significant progress has been made toward linking phenotype and genotype for a number of plant traits and genes. In addition to the research on genes involved in flower development and flowering time discussed above, numerous presentations were given linking genes to a wide variety of phenotypic characteristics, such as trichome development, hypocotyl length and seedling development, seed storage lipids, insect and pathogen resistance, and salt tolerance. A few of these presentations are summarized here.
Flower Shape Daisies and other members of the Asteraceae (composites) have what appear to be radially symmetrical flowers. However, the composite flower consists of inner disc florets and outer ray florets, and although the disc florets are symmetrical, each ray floret is asymmetrical. Coen reported on research on Senecio spp conducted in collaboration with Richard Abbott and Amanda Gillies (University of St. Andrews, UK) and Pilar Cubas (Universidad Autónoma de Madrid, Spain). Senecio vulgaris, a composite native to the United Kingdom, has yellow disc florets but no ray florets. The Italian species S. squalidus, which has yellow disc and ray florets, escaped from the Oxford botanical gardens 300 years ago and is now widespread throughout the United Kingdom. The two species interbred, and a rayed variant of S. vulgaris appeared. Restriction fragment length polymorphism analysis has shown polymorphism cosegregating with the rayed and nonrayed forms of Senecio that is tightly linked to a CYCLOIDIA-like gene, which is expressed only in floral meristem tissue that forms into ray florets. Future work will focus on providing definitive evidence that this gene is responsible for the rayed versus the nonrayed phenotype. In another example, Coen showed a striking mutant of Cosmos, another composite with ray and disc florets, in which the ray florets formed tubes instead of flattened petals. Floral symmetry is thought to have evolved independently many times. It will be very interesting to determine if the Cosmos mutation is associated with a CYCLOIDIA-like gene and if a similar mutant can be created in Senecio.
Insect Resistance
Pathogen Resistance
Starch Biosynthesis In summary, the renewed interest in collaborations among functional and evolutionary geneticists is highly promising for plant genetics research. Functional geneticists can provide information and ideas about genes and mechanisms that may be ecologically significant targets of natural selection. Evolutionary geneticists can provide information on the degree and structure of DNA variation, which can form the basis for the development of new tools, such as mapped recombinant inbred lines and marker densities for LD mapping, and the isolation of new genes involved in development and physiology. A number of plant geneticists began studying the outcrossing species A. lyrata with the idea that it might provide a plant equivalent of Drosophila, which has large random mating populations in which levels of within-population DNA variation are correlated with recombination rate in local chromosomal regions. The structure of variation in A. lyrata may resemble that of Drosophila more than that of A. thaliana does. However, Nordborg pointed out that the main reason Drosophila is a "good model" for population genetics is that it can fly, so there is random mating on a global scale, and that population structure, perhaps more than the type of mating system, has a critical influence on patterns of DNA variation. Thus, it should not be surprising to find that the pattern of DNA variation in A. lyrata does not resemble that of Drosophila very closely. Nonetheless, evolutionary genetics should benefit greatly from more extensive analyses and comparisons of A. lyrata and A. thaliana. Perhaps the joint efforts of plant functional and evolutionary geneticists will redefine what it means to be a "good model" for population genetics. Acknowledgments I thank Monserrat Aguadé, Magnus Nordborg, Thomas Mitchell-Olds, and Michael Purugganan for helpful discussions and comments. References
Aguadé, M. (2001). Nucleotide sequence variation at two genes of the phenylpropanoid pathway, the FAH1 and F3H genes, in Arabidopsis thaliana. Mol. Biol. Evol. 18, 1–9. Conway Morris, S. (2000). Evolution: Bringing molecules into the fold. Cell 100, 1–11.[ISI][Medline]
Johanson, U., West, J., Lister, C., Michaels, S., Amasino, R., and Dean, C. (2000). Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290, 344–347. Kliebenstein, D.J., and Mitchell-Olds, T. (2001). Comparative quantitative trait locus mapping of aliphatic, indolic and benzylic glucosinolate production in Arabidopsis thaliana leaves and seeds. Genetics, in press.
Kliebenstein, D.J., Lambrix, V.M., Reichelt, M., Gershenzon, J., and Mitchell-Olds, T. (2001a). Gene duplication in the diversification of secondary metabolism: Tandem 2-oxoglutarate–dependent dioxygenases control glucosinolate biosynthesis in Arabidopsis. Plant Cell 13, 681–693. Kliebenstein, D.J., Kroymann, J., Brown, P., Figuth, A., Pedersen, D., Gershenzon, J., and Mitchell-Olds, T. (2001b). Genetic control of natural variation in Arabidopsis thaliana glucosinolate accumulation. Plant Physiol, in press.
Nordborg, M. (2000). Linkage disequilibrium, gene trees and selfing: An ancestral recombination graph with partial self-fertilization. Genetics 154, 923–929. O'Kane, S.L., and Al-Shehbaz, I.A. (1997). A synopsis of Arabidopsis (Brassicaceae). Novon 7, 323–327.[CrossRef][ISI]
Olsen, K.M., and Schaal, B.A. (2001). Microsatellite variation in cassava (Manihot esculenta, Euphorbiaceae) and its wild relatives: Further evidence for a southern Amazonian origin of domestication. Am. J. Bot. 88, 131–142. Parker, J.E., Feys, B.J., Van Der Biezen, E.A., Noël, L., Aarts, N., Austin, M.J., Botella, M.A., Frost, L.N., Daniels, M.J., and Jones, J.J. (2000). Unravelling R gene–mediated disease resistance pathways in Arabidopsis. Mol. Plant Pathol. 1, 17–24.[CrossRef]
Paterson, A.H., Lin, Y.-R., Li, Z., Schertz, K.F., Doebley, J.F., Pinson, S.R.M., Liu, S.-C., Stansel, J.W., and Irvine, J.E. (1995). Convergent domestication of cereal crops by independent mutations at corresponding genetic loci. Science 269, 1714–1718.
Savolainen, O., Langley, C.H., Lazzaro, B.P., and Fréville, H. (2000). Contrasting patterns of nucleotide polymorphism at the alcohol dehydrogenase locus in the outcrossing Arabidopsis lyrata and the selfing Arabidopsis thaliana. Mol. Biol. Evol. 17, 645–655.
Stephan, W., and Langley, C.H. (1998). DNA polymorphism in Lycopersicon and crossing-over per physical length. Genetics 150, 1585–1593.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ASPB Publications | THE PLANT CELL | PLANT PHYSIOLOGY | |
|---|---|---|---|
