Green Genes—Comparative Genomics of the Green Branch of Life

Home , Green algae, Lycophyte, Red algae

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Leading Edge Essay

John L. Bowman,1,* Sandra K. Floyd,1 and Keiko Sakakibara1 1School of Biological Sciences, Monash University, Clayton Campus, Melbourne, Victoria 3800, Australia *Correspondence: [email protected] DOI 10.1016/j.cell.2007.04.004

As more plant genome sequences become available, researchers are increasingly using comparative genomics to address some of the major questions in plant biology. Such questions include the evolution of photosynthesis and multicellularity, the developmental genetic changes responsible for alterations in body plan, and the origin of important plant innovations such as roots, leaves, and vascular tissue.

All plants are descended from a sin- scapes and whose evolution allowed genome. Plastid genome sequences gle eukaryotic ancestor that acquired the subsequent colonization of land are available from all major plant lin- a photosynthetic cyanobacterium as by the metazoans. Plastid genome eages, and perhaps surprisingly, the an endosymbiont (the ancestral plas- sequences are available for species plastid genomes from all lineages are tid). The acquisition of a cyanobacte- in all major lineages of plants, and similar in size and gene content. The rial endosymbiont was a momentous nuclear genome sequences have plastids of the glaucophytes (called event in the evolution of life on Earth been determined for a red alga, two cyanelles) still retain a peptidoglycan leading to a shift of most primary chlorophytes, and three distinct line- cell wall characteristic of the ances- production from prokaryotic cyano- ages of land plants. Here, we high- tral cyanobacterial endosymbiont and bacteria to photosynthetic eukaryo- light some of the major evolutionary yet have genomes similar to those of tes. Although the endosymbiosis of a transitions in the evolution of land other plants. The continuing nuclear cyanobacterium was a singular event plants and some key questions that bombardment of plastid-derived in the history of life, plastids have are beginning to be addressed by DNA is thought to have contributed also been transmitted horizontally comparing genome sequences from significantly to the genome content to other eukaryotic lineages via sec- a diverse range of plant species. of plants, with as much as 18% of ondary endosymbiotic events where the genome of the model flowering unrelated eukaryotes acquired endo- The Algal Origins of the plant (angiosperm) Arabidopsis thal- symbiotic plants. There are five or so Photosynthetic Eukaryotes iana thought to have been derived eukaryotic lineages, one of which is Following the capture of a cyanobac- from the cyanobacterial endosym- plants (Keeling et al., 2005). Within terium by the plant ancestor, the evo- biont (Martin et al., 2002). The con- the plants, three distinct groups have lution of the endosymbiont genome tribution of cyanobacterial genes to been identified (Figure 1): the glau- was characterized by wholesale algal nuclear genomes has not been cophytes (little-known freshwater transfer of genetic material to the host analyzed in detail, but substantial dif- algae), rhodophytes (red algae), and nuclear genome, resulting in a reduc- ferences might have contributed to the green plants (which include green tion in the endosymbiont genome differing genomic trajectories in the algae and land plants). The rhodo- and an enrichment of the host nuclear major plant lineages. phytes are primarily marine algae and include reef-building coralline algae; Table 1. Plant Genomes for Which Sequence Is Available (circa March 2007) they provide a source of agar and Size (Mb) # of Genes form the basis of the billion-dollar Cyanidioschyzon merolae (unicellular red alga) 16.5 5331 nori industry in Japan. The green plants, by far the most diverse of the Ostreococcus tauri (unicellular green alga) 12.56 8166 three groups, comprise two major Chlamydomomas reinhardtii (unicellular green alga) 136 >15,000 clades: the chlorophytes (freshwater Physcomitrella patens (moss) 487 >20,000 and marine algae) and the strepto- phytes (including the paraphyletic Selaginella moellendorfii( lycophyte) 85? ? charophycean freshwater algae and Oryza sativa (rice) 389 41,000 the land plants). It was the land plants Populus trichocarpa (popular tree) 485 45,000 (embryophytes) that colonized and eventually dominated terrestrial land- Arabidopsis thaliana (flowering plant) 140 27,500

Cell 129, April 20, 2007 ©2007 Elsevier Inc. 229 Figure 1. Phylogenetic Relationships among Plants Depicted are relationships among the three lineages of plants: glaucophytes (freshwater algae; blue), rhodophytes (red algae; red), and the green plants (chlorophytes, charophytes, and land plants; green). Estimated dates for some nodes are listed in millions of years be- fore present. The primary endosymbiotic event is estimated to have occurred at least 1.6 billion years ago. A deep split within the green lineage created the chlorophyte clade and the charophyte plus land plant clade. Note that both the charophytes and the bryophytes are grades and are not monophyletic. Major events in the evolution of land plants are de- marcated with arrows. Species for which complete nuclear genome sequences are available are listed in color (photographs at right; the three angiosperm species are pictured upper left). Species positioned in large phylogenetic gaps where genome sequences would be informative (black) include the following: the basal lineage of land plants, the liverworts, charophycean algal lineages (Chara, Coleo- chaete) that are sisters to land plants, and the gymnosperms, which are the sister group to flowering plants (angiosperms). Also included is a multicellular chlorophytic green alga. Sec- ondary endosymbiotic events have occurred within both the red algae (e.g., diatoms, pictured) and green plants. Pie chart shows the relative species richness of the major clades. The vast majority of species within the Plantae are angiosperms (250,000 species), with other groups having substantially fewer described species (numbers approximated): glaucophytes 13; rhodophytes 5,920; chlorophytes 3,720, charophytes 3,400; bryophytes 17,000 (liverworts 7,000, mosses 10,000, hornworts 100); lycophytes 1,225; ferns 12,000; gymnosperms 800. Photos from top: Opuntia basilaris, Ginkgo biloba, Selaginella kraussiana, Physcomitrella patens, Marchantia polymorpha, Chara sp., Coleochaete sp., Chlamydomomas reinhardtii, Hydrodictyon sp., Ostreococcus tauri, Cyanidioschyzon merolae. Photos courtesy of Gayle Dupper, Institute of Forest Genetics, Placerville, CA, USA (poplar), Charles Delwiche, University of Maryland (Chara), James Umen, Salk Institute (Chlamydomonas), Hervé Moreau, Université Pierre et Marie Curie-Paris (Ostreococcus), and Tsuneyoshi Kuroiwa, Rikkyo University (Cyanidioschyzon).

There are three plant species for between genes (Derelle et al., 2006). mitochondrial) and for both forward which almost complete genome One remarkable feature of the O. and reverse genetics (reviewed in information for nucleus, chloroplast, tauri genome is its extreme heteroge- Grossman et al., 2007). Additional red and mitochondrion is available: the neity with 2 chromosomes differing and green algal genome sequences, red alga Cyanidioshyzon merolae, the from the other 18 in GC content and such as the sequences of two other marine prasinophycean green alga transposable element distribution, Ostreococcus genomes (US Depart- Ostreococcus tauri, and the chloro- suggesting horizontal acquisition of ment of Energy Joint Genome Insti- phycean green alga Chlamydomonas at least one of its chromosomes. In tute, www.jgi.doe.gov), are required reinhardtii (Table 1). These unicel- contrast, the C. reinhardtii genome to assess whether these character- lular algae are ideal models for cell is larger and contains more genes. A istics are unique or more general for biology because the cells are mono- comparative genomics study using these taxa. plastidic, with C. merolae and O. tauri the Chlamydomonas, Arabidopsis, By producing oxygen as a waste cells also containing only a single and human genomes facilitated the product, the evolution of photosyn- mitochondrion and Golgi body, the identification of genes involved in thetic cyanobacteria 3.5 billion years division of which can be synchro- flagellar development and function in ago dramatically altered the Earth’s nized. C. merolae lives in acidic hot both Chlamydomonas and humans, ecosystem. Following the primary springs but can be grown in culture. including genes involved in human endosymbiotic event that defines Its genome is compact with most disease (angiosperms lack the flag- plants, this eukaryotic lineage evolved genes lacking introns (Matsuzaki et ellated sperm found in many other to become the dominant primary pro- al., 2004). The phytoplankton O. tauri organisms) (Li et al., 2004). C. rein- ducer in both aquatic and terrestrial is a picoeukaryote comprising cells hardtii is a sophisticated model for habitats. Comparisons among algal that are about the size of prokaryotes investigating photosynthesis and fun- genome sequences can provide (about 1 µm in diameter). Its genome damental cell biology with tools avail- information to elucidate characteris- is similarly compact with an aver- able for transformation of all three tics of the ancestral photosynthetic age spacing of only 197 basepairs genomes (nuclear, chloroplast, and eukaryotes. For example, the O. tauri

230 Cell 129, April 20, 2007 ©2007 Elsevier Inc. genome includes genes potentially algal ancestors. Comparison of mito- ways of gametophyte development involved in C4 photosynthesis, which chondrial genomes of the charophyc- or was it due to the origin of de novo enhances photosynthetic capabili- ean alga Chara, the liverwort March- developmental genes and networks? ties under low CO2 conditions and antia polymorpha, and other land The earliest land plants most likely has evolved repeatedly in several plants provides some of the strong- had a haploid-dominant life cycle, angiosperm lineages. C4 photo- est evidence for the sister relation- with an ephemeral-dependent spo- synthesis in O. tauri would confer a ship of Chara to land plants and of rophyte, and this has been retained in significant advantage under specific liverworts to other land plants (Turmel the extant bryophytes (mosses, horn- environmental conditions, suggest- et al., 2003). Unlike the development worts, and liverworts). Flowering plant ing that this capability may have been of their closely related multicellular models such as Arabidopsis, rice, and present at an early stage of green charophycean algal relatives, land poplar all represent diploid-dominant algal evolution (Derelle et al., 2006). plants exhibit growth from an api- plants in which the sporophyte is long Additionally, genomic comparisons cal meristem that produces a three- lived and complex and the gameto- between C. merolae and the green dimensional body that becomes pat- phyte is diminutive and ephemeral. algae will provide insight into both terned to produce distinct tissues. One of the major questions in plant shared and specific genetic charac- One of the key questions is how pro- evolution concerns the evolution of the ters in the two algal lineages. grams for development and growth sporophyte developmental program, were changed to allow the produc- which was modified through time so Becoming Multicellular tion and patterning of tissues. The that sporophytes became larger and The emergence of multicellular ability to compare the ancestral land acquired the ability to branch, develop organisms from unicellular ances- plant genome with that of algal rela- conducting tissues, and produce tors occurred repeatedly in the evo- tives would facilitate the identifica- roots, leaves, seeds, and flowers. lution of eukaryotes, most notably tion of the genetic bases for the key Another key question concerns in the metazoan and land plant line- innovations that allowed green plants the relationship between radial and ages. The origin of multicellularity is to evolve from aquatic ancestors and bilateral or dorsiventral develop- thus one of the key questions in the adapt to life on land. Such key innova- ment in land plants (Friedman et al., evolution of life on Earth. Compara- tions include the perception of envi- 2004). The most familiar instance tive genomics suggests that a com- ronmental cues (light and gravity), the of dorsiventral development is that bination of co-opting existing genes origin of extracellular matrices (spo- of leaves. There has been a great for new functions and the evolution ropollenin, lignin, and pectic acid), deal of research interest in under- of new proteins from novel combina- establishment of intercellular com- standing the genetics of polarity and tions of pre-existing protein domains munication networks (plasmodes- growth of leaves in flowering plants. contributed to the emergence of mul- mata, plant hormones, receptors, Organs referred to as leaves occur ticellularity in metazoans (Ruiz-Trillo et and their ligands), and diversification in all extant vascular plants, but in al., 2007). Multicellularity has evolved of gene regulatory networks promot- at least three cases these leaves numerous times within the red and ing cell differentiation. Because of the evolved independently, in lyco- green algae. Are similar or distinct enormous evolutionary divergence phytes, ferns, and seed plants. The genetic programs recruited to pattern between chlorophytes and strepto- earliest vascular plants (known only multicellular algal taxa? Is multicellu- phytes, sequencing of the genome of from fossils) lacked laminar, lateral, larity in land plants fundamentally dif- a charophycean alga (such as Chara) vascularized appendages. Thus, in ferent or similar to that of their algal will be required to assess the ori- vascular plants, organs with dorsiv- relatives? Comparative genomics of gins of genetic mechanisms in land entral polarity evolved in the spo- multicellular organisms from distinct plants. rophyte generation that had radial lineages should shed light on these The closest relatives of land plants, patterning mechanisms. Variability questions. However, given that only the charophycean algae, have a hap- in growth form also exists in game- unicellular algae have thus far had lontic life cycle in which the zygote is tophytes: some have radial organi- their genomes sequenced, sequenc- the only diploid cell. All land plants have zation (mosses, whisk ferns) and ing of the genomes of multicellular a life cycle that includes an alterna- others have dorsiventral or thalloid red and green algae will be required tion of generations involving a haploid organization (liverworts, hornworts, to address this issue. phase (gametophyte) in which gam- ferns). Evidence from the earliest etes are produced and a diploid phase land plant fossils suggests that the Conquering the Land (sporophyte) that produces spores. earliest land plants may have been The origin of land plants from aquatic Thus, multicellularization of the zygote liverworts or liverwort-like plants ancestors marks a major evolution- evolved early during land plant evolu- with a thalloid gametophyte. If this is ary transition in the history of green tion. Was the initial elaboration of the true, then a transition from dorsiven- plants. Land plants inherited many zygote to produce a multicellular dip- tral to radial gametophyte develop- biochemical, ultrastructural, and loid sporophyte due to a co-option of ment must have occurred within land physiological characters from their already existing developmental path- plants, and more than once.

Cell 129, April 20, 2007 ©2007 Elsevier Inc. 231 Liverworts represent the sister programs, and whether body plans in developmental gene families that group to all other extant land plants. the different generations require dif- are shared with flowering plants and The best hope of assessing the nature ferent developmental programs. some that are not (reviewed in Floyd of the land plant ancestral genome and Bowman, 2007). Two gene fami- for comparison with algal genomes Becoming Large, the Evolution of lies important for leaf development will require comparison of a liverwort Vasculature in flowering plants, Class III HD-Zip genome with that of other land plant Another plant genome that has been and KANADI, are both present in the genomes. For example, the genome sequenced and awaits assembly and genome of S. moellendorffii. How- of the thalloid liverwort M. polymor- annotation is that of the lycophyte ever, the subclade of Class III HD-Zip pha will provide the basis for com- Selaginella moellendorffii. There are genes involved in leaf polarity in flow- paring the developmental genetics of two major lineages of extant vascular ering plants has no ortholog in S. moe- plants with dorsiventral development plants, the lycophytes (spike mosses, llendorffii. Likewise, the YABBY gene and those with radial development. club mosses, quillworts) and the family, important for abaxial identity Efforts toward obtaining the nuclear euphyllophytes (ferns, horsetails, seed and laminar outgrowth in flowering genome sequence of M. polymor- plants), representing an ancient diver- plants, has not been found in the S. pha include the construction of BAC gence of a vascular plant ancestor. moellendorffii genomic sequence. libraries (Green Plant BAC project), These lineages separated prior to the With the completion of the assembly end-sequences of both BAC and EST evolution of many features we com- and annotation of the S. moellendorffii libraries (T. Kohchi, personal commu- monly associate with plants. Leaves, genome the full assessment of many nication), and submission of a pilot roots, and complex vascular archi- gene families will be possible and we proposal for whole-genome shotgun tectures have evolved independently can begin to address long-standing sequencing (JGI). within both lineages from a morpho- questions in vascular plant evolution Although genome sequences from logically simpler common ancestor. with a new set of genetic tools. Tech- liverworts and charophycean algae Despite millions of years of evolution, niques for genetic transformation are not yet available, the nuclear lycophytes have also retained many in S. moellendorffii enabling trans- genome of the moss, Physcomitrella developmental features thought to genic approaches for studying gene patens, has recently been sequenced be ancestral or primitive for vascular expression may also be possible. along with a large number of cDNA plants. These include an apical mer- clones derived from various devel- istem with one or a few apical initial Genome Duplications and opmental stages, including leafy cells, apical dichotomous branch- Morphological Innovations in shoots of gametophytes and sporo- ing, and a protostelic vasculature Flowering Plants phytes (M. Hasebe, personal com- with xylem surrounded by phloem. One surprising discovery from the munication). Initial analyses of cDNA The differences between lycophytes genome sequences of the model sequences suggest that mosses and euphyllophytes highlight some plants A. thaliana and rice (Oryza and angiosperms have largely the of the major questions in vascular sativa) is evidence for repeated same types of gene families, includ- plant evolution. How were the com- whole-genome duplications, despite ing most of the gene families impli- plex shoot apical meristems of seed the diploid nature of the two species. cated in developmental patterning plants derived from simpler ancestral Flowering plants offer an attractive in angiosperms. This suggests a co- meristems? Are both simple and com- system to study the consequences option of existing genes rather than plex meristems regulated by the same of whole-genome duplications due to the evolution of new genes in the tran- gene regulatory networks? How might their propensity for polyploidization. sition from a gametophyte-dominant these networks have changed as the Flowering plants have likely under- life cycle to a sporophyte-dominant simpler ancestral meristems evolved? gone multiple rounds of polyploiz- one (Nishiyama et al., 2003; Floyd Are the independently acquired leaves idation in the past 150–200 million and Bowman, 2007). However, the and roots of these organisms pat- years. In contrast, in mammals poly- gene families have markedly diver- terned by the same or different genetic ploidization has been suppressed sified in the angiosperms relative to programs? Are the vascular tissues over the same timeframe due to the mosses (Floyd and Bowman, 2007). analogous or homologous to the con- presence of the X-Y sex chromosome Functional analyses using homolo- ducting tissues in mosses? system. There are three key ques- gous recombination knockout tech- In the case of leaves some insight tions: (1) How does the process of nology in P. patens will be required has already been gained from diploidization occur? (2) Do whole- to clarify the questions of whether the genomic data in addition to using genome duplications correlate with same genetic networks function in candidate gene approaches (Floyd speciation events? and (3) Do whole- both haploid and diploid generations and Bowman, 2006, Harrison et al., genome duplications correspond to of land plants, whether the radial 2005). Although the S. moellendorf- an explosive evolution of morphol- shoots of the moss gametophyte and fii genome has not been assembled ogy by providing the raw material of the vascular plant sporophyte are yet, searches of the unassembled entire genetic pathways for selection, regulated by similar developmental sequences have identified many as suggested by Ohno (1970)? The

232 Cell 129, April 20, 2007 ©2007 Elsevier Inc. availability of multiple whole-genome and Wolfe, 2004). Such preferential have contributed to one gene being duplications of varying antiquity in retention could also be evidence of specialized for carpel development flowering plants facilitates the formu- a requirement to maintain an appro- (an angiosperm-specific structure) lation of some hypotheses to address priate stoichiometry in protein com- and the other gene specialized for these questions. plexes and a selection for increased ovule and integument development Several groups have tried to date diversity of secondary metabolites (shared by both angiosperms and the three whole-genome duplica- involved in defense. The preferential gymnosperms). In contrast, second- tions (called 1R, 2R, and 3R) that retention of transcription factors is ary growth (the production of wood) have left their imprint in the organi- also consistent with the higher per- has secondarily evolved multiple zation of the Arabidopsis genome centage of transcription factors in times within the angiosperms from (reviewed in De Bodt et al., 2005). multicellular plants (12%–15%) rela- herbacious ancestors, suggest- Although there is much uncertainty tive to unicellular plants (2%–4%). ing that most if not all angiosperm in the precise timings, the earliest The increase in genes encoding sig- species still possess the ancestral event could correspond with the naling molecules and transcription genetic programs. This implies that origin of extant flowering plants, factors is also consistent with the secondary growth may involve co- and the second event with the ori- idea of neofunctionalization follow- option of pre-existing genetic pro- gin of the eudicots (the most spe- ing gene duplication contributing to grams via changes in gene regu- cies-rich clade of flowering plants). the evolution of morphological com- lation mediated, for example, by Likewise, a whole-genome duplica- plexity. The evolutionary process of modifications to chromatin. Thus, tion event dating to the base of the diploidization, whereby a polyploid it is likely that both the evolution monocots, the second large clade decays to become a diploid, is enig- of new genes, via gene duplication of angiosperms, is evident in the matic. However, the study of recent events, and the co-option of existing genome of rice. That the whole- polyploids suggests that massive genetic programs contributed to the genome duplications correlate in gene loss accompanied by struc- evolution of morphological diversity time with origins or major radiations tural evolution of chromosomes within the angiosperms, and that of flowering plants is consistent and epigenetic reprogramming of the ample genetic material provided with Ohno’s idea that such events retained genes may influence chro- by whole-genome duplications has facilitate major leaps in morphologi- mosome pairing and thus contribute played a major role in the rise of the cal evolution. More precise dating to diploidization. angiosperms as the dominant land of the whole-genome duplication The evolution of the flower and plant vegetation on the planet today. events relative to the adaptive radia- the carpel were key innovations that tions of angiosperms will require the allowed angiosperms to engage in Conclusion analysis of other phylogenetically specialized animal pollination sys- Genomes from plants representing informative angiosperm genomes tems and seed dispersal mecha- different phylogenetic lineages, lev- such as that of the basal eudicot nisms. Did the evolution of these els of organization, and body plan Aquilegia (columbine; in progress innovations require the evolution of will soon be available for comparative at JGI). new genes, or were already exist- genomic analyses and for functional Based on comparison of the ing genetic programs co-opted to analysis of development using reverse genome sequences of Arabidopsis new roles? Similar to the situation genetics and transgenic techniques. and of the poplar tree (Populus), described earlier in the evolution of As researchers begin to mine the rich it has been estimated that their vascular plants from a bryophyte- source of data from Cyanidioshyzon, common ancestor possessed only like ancestor, orthologs of flower Ostreococcus, Chlamydomonas, 12,000–14,000 genes. This sug- patterning genes are present in Physcomitrella, and Selaginella to gests differential retention of genes gymnosperms (that is, nonflowering compare with Arabidopsis, Oryza, in the different lineages, possibly seed plants such as conifers). For and Populus, we look ahead to the related to their different life his- example, B and C class MADS box addition of still more plant genomes tories, that is, ephemeral annual genes that pattern the reproductive such as those of Marchantia and versus long-lived and sometimes organs of angiosperm flowers are Chara to bridge some of the vast evo- vegetatively propagated perennial also expressed in the reproductive lutionary gaps that remain. We are on (Tuskan et al., 2006; Maere et al., organs of gymnosperms. However, the verge of a new and exciting era of 2005). Analysis of the Arabidop- whereas gymnosperms appear to comparative genomics for the major sis genome has also revealed the have single copies of a C class gene, lineage of photosynthetic organisms, preferential retention of specific angiosperms harbor multiple copies and the future looks very green. classes of genes—such as those derived from gene/genome dupli- encoding transcription factors, sig- cations within the angiosperm line- Acknowledgments naling molecules, and secondary age. At least two C class genes are Authors are supported by the US National metabolism enzymes—following present in all angiosperms, suggest- Science Foundation and the Australian whole-genome duplications (Blanc ing that neofunctionalization could Research Council.

Cell 129, April 20, 2007 ©2007 Elsevier Inc. 233 References and Spalding, M.H. (2007). Curr. Opin. Plant Natl. Acad. Sci. USA 99, 12246–12251. Biol. 10, 190–198. Matsuzaki, M., Misumi, O., Shin-I, T., Blanc, G., and Wolfe, K.H. (2004). Plant Cell Harrison, C.J., Corley, S.B., Moylan, E.C., Al- Maruyama, S., Takahara, M., Miyagishima, S., 16, 1679–1691. exander, D.L., Scotland, R.W., and Langdale, Mori, T., Nishida, K., Yagisawa, F., Nishida, K., et al. (2004). Nature 428, 653–657. De Bodt, S., Maere, S., and Van de Peer, Y. J.A. (2005). Nature 434, 509–514. (2005). Trends Ecol. Evol. 20, 591–597. Keeling, P.J., Burger, G., Durnford, D.G., Lang, Nishiyama, T., Fujita, T., Shin-I, T., Seki, M., Nishide, H., Uchiyama, I., Kamiya, A., Carninci, Derelle, E., Ferraz, C., Rombauts, S., Rouze, B.F., Lee, R.W., Pearlman, R.E., Roger, A.J., and Gray, M.W. (2005). Trends Ecol. Evol. 20, P., Hayashizaki, Y., Shinozaki, K., et al. (2003). P., Worden, A.Z., Robbens, S., Partensky, F., Proc. Natl. Acad. Sci. USA 100, 8007–8012. Degroeve, S., Echeynié, S., Cooke, R., et al. 670–676. (2006). Proc. Natl. Acad. Sci. USA 103, 11647– Li, J.B., Gerdes, J.M., Haycraft, C.J., Fan, Ohno, S. (1970). Evolution by Gene Duplica- 11652. tion (George Allen and Unwin, London). Y., Teslovich, T.M., May-Simera, H., Li, H., Floyd, S.K., and Bowman, J.L. (2006). Curr. Blacque, O.E., Li, L., Leitch, C.C., et al. (2004). Ruiz-Trillo, I., Burger, G., Holland, P.W.H., King, Biol. 16, 1911–1917. Cell 117, 541–552. N., Lang, B.F., Roger, A.J., and Gray, M.W. (2007). Trends Genet. 23, 113–118. Floyd, S.K., and Bowman, J.L. (2007). Int. J. Maere, S., De Bodt, S., Raes, J., Casneuf, Plant Sci. 168, 1–35. T., Van Montagu, M., Kuiper, M., and Van de Turmel, M., Otis, C., and Lemieux, C. (2003). Peer, Y. (2005). Proc. Natl. Acad. Sci. USA 102, Plant Cell 15, 1888–1903. Friedman, W.E., Moore, R.C., and Purugga- 5454–5459. nan, M.D. (2004). Am. J. Bot. 91, 1726–1741. Tuskan, G.A., Difazio, S., Jansson, S., Bohl- Martin, W., Rujan, T., Richly, E., Hansen, A., mann, J., Grigoriev, I., Hellsten, U., Putnam, Grossman, A.R., Croft, M., Gladyshev, V.N., Cornelsen, S., Lins, T., Leister, D., Stoebe, B., N., Ralph, S., Rombauts, S., Salamov, A., et al. Merchant, S.S., Posewitz, M.C., Prochnik, S., Hasegawa, M., and Penny, D. (2002). Proc. (2006). Science 313, 1596–1604.