The Plant Cell, Vol. 16, S61–S71, Supplement 2004, www.plantcell.org ª 2004 American Society of Plant BiologistsSex-Determining Mechanisms in Land PlantsMilos Tanurdzic and Jo Ann Banks 1Purdue Genetics Program and Department of Botany and Plant Pathology, Purdue University, West Lafayette, Indiana 47906INTRODUCTIONSex determination is a process that leads to the physicalseparation of male and female gamete-producing structures todifferent individuals of a species. Even though sexually reproducingspecies have only three possible options—to relegate thetwo sexes to separate individuals, to keep them together onthe same individual, or to have a combination of both—plantsin particular display a great variety of sexual phenotypes. Inangiosperms, a sex-determining process is manifest in speciesthat are monoecious, in which at least some flowers areunisexual but the individual is not, or dioecious, in whichunisexual plants produce flowers of one sex type. In plants thatproduce no flowers and are homosporous, sex determination ismanifest in the gametophyte generation with the production ofegg- and sperm-forming gametangia on separate individualgametophytes. The determinants of sexual phenotype in plantsare diverse, ranging from sex chromosomes in Marchantiapolymorpha and Silene latifolia to hormonal regulation in Zeamays and Cucumis sativa to pheromonal cross-talk betweenindividuals in Ceratopteris richardii. Here, we highlight recentefforts aimed at understanding the genetic and molecularmechanisms responsible for sex determination in several plantspecies that separate their sexes into two individuals or flowers.Representatives of all major land plant lineages are included togive an evolutionary perspective, which is important in understandinghow different sex-determining mechanisms evolvedand their consequences in plant development and evolution.Although great progress has been made in genetically identifyingthe genes that regulate sex expression in these species, few ofthem have been cloned. Because a ‘‘one-size-fits-all’’ mechanismof sex determination will not account for the variety ofsexual systems in plants, future efforts at cloning these genesin several well-chosen model systems will be necessary tounderstand these processes at the molecular level. There areseveral excellent recent reviews of sex determination thatdescribe species that have not been included to which thereader is directed (Ainsworth, 1999, 2000; Geber et al., 1999;Matsunaga and Kawano, 2001; Negrutiu et al., 2001; Barrett,2002; Charlesworth, 2002). We begin with the most basal lineageof the land plants.1To whom correspondence should be addressed. E-mail email@example.com; fax 765-494-5896.Article, publication date, and citation information can be found atwww.plantcell.org/cgi/doi/10.1105/tpc.016667.THE BRYOPHYTESThe bryophytes are a group of plants that includes the modernliverworts, hornworts, and mosses. In this group, the haploidgametophyte is the dominant phase of the life cycle, which isillustrated in Figure 1. Their diploid spore-producing sporophytesare very small, short-lived, and parasitic upon the independentgamete-producing gametophyte generation. All bryophytes arehomosporous, producing only one type of spore, yet all threegroups of extant bryophytes are represented by species that arehomothallic, with one individual gametophyte producing bothmale and female sex organs (gametangia), and species that areheterothallic, with one individual producing only male or femalegametangia (Smith, 1955). Sexual dimorphism in heterothallicspecies can be extreme, as exemplified by members of thegenus Micromitrium, in which the dwarf male gametophytegrows on the leaves of the markedly larger female plant (Smith,1955).In many species of bryophytes, heterothallism (unisexuality)has been correlated with the presence of sex chromosomes(Smith, 1955). Although the extent of heterothallism and sexchromosomes in the bryophytes has not been assessedsystematically, this is the only known group of homosporousplants that uses sex chromosomes in sex determination. Todate, studies of bryophyte sex determination have focused onthe heterothallic liverwort Marchantia polymorpha. In thisspecies, the male and female thalli (vegetative gametophytes)look alike, although males and females can be distinguishedeasily by differences in the morphology of the sexual structureeach produces. A gametophyte bears gametangia on stalkedbranches called antheridiophores (if male) or archegoniophores(if female) that arise from the upper surface of thethallus (Figure 1). Antheridiophores produce sperm-formingantheridia, and archegoniophores produce egg-forming archegonia.The sex of each haploid gametophyte is determined bycytologically distinct sex chromosomes, with males having onevery small Y chromosome and no X chromosome and femaleshaving one X chromosome and no Y chromosome (Lorbeer,1934). In addition to its rapid growth (it is often an invasiveweed in greenhouses), its ability to be propagated vegetativelyby gemma cups (Figure 1), and its ability to be transformed(Takenaka et al., 2000), Marchantia has a relatively smallgenome size of 280 Mbp distributed among eight autosomesplus one sex chromosome (Okada et al., 2000), making ita worthy model organism amenable to genomics-style investigations.Working on the assumption that sex-determining factors existon the Marchantia sex chromosomes, Okama and colleagues
S62The Plant CellFigure 1. The Life Cycle of the Liverwort Marchantia polymorpha.Haploid gametophytes develop gametangia in antheridiophores and archegoniophores that produce the sperm and egg, respectively. Uponfertilization, which is facilitated by raindrops, the diploid sporophyte remains attached to the archegoniophore and produces yellow sporangia, inwhichhaploid spores are formed after meiosis. The spores are liberated and germinate to form a new gametophyte thallus. The sex of the thallus depends onwhich sex chromosome it inherits. Photographs courtesy of Katsuyuki Yamato, Kyoto University.(2000) set out to identify these factors by constructing separatemale and female P1-derived artificial chromosome (PAC)libraries and identifying clones specific to either the male or thefemale genome. Their screen resulted in 70 male-specific PACclones that hybridized only the Y chromosome by fluorescence insitu hybridization. No female-specific clones were found, indicatingthat the X chromosome does not harbor long stretchesof unique sequences, as does the Y chromosome. To date, twomale-specific PAC clones with insert sizes totaling 126 kb havebeen sequenced (Okada et al., 2001; Ishizaki et al., 2002). Thisand other analyses have revealed that approximately one-fourthto one-third of the 10-Mb Y chromosome of Marchantia consistsof an estimated 600 to 15,000 copies of an element of variablelength (0.7 to 5.2 kb) that contains other smaller repetitiveelements (Okada et al., 2001; Ishizaki et al., 2002). Of the sixputative protein-encoding genes found embedded within therepeats, all are present in multiple copies on the Y chromosomebased on DNA gel blot hybridization. Two of these genes, namedORF162 (Okada et al., 2001) and M2D3.5 (Ishizaki et al., 2002),are unique to the Y chromosome; the remaining four genes arepresent in low copy number on the X chromosome or theautosomes. ORF162 encodes a putative protein with a RINGfinger domain; M2D3.5 is a member of the same gene family.ORF162 transcripts are detectable only in the male sexualorgans, indicating that the gene family represented by ORF162and M2D3.5 may be important in the development of theantheridiophore. Of the four genes also present on the Xchromosome or the autosomes, only one (M2D3.4) is restrictedin its expression to the male gametophyte, indicating that theM2D3.4 X or automosmal homolog might be a pseudogene.M2D3.4 encodes a putative protein similar to a Lilium longiflorumgene that is expressed exclusively in the male gametic cells. Theremaining three genes are not sex specific in their expression.The functions of the Y chromosome–encoded genes are as yetunknown.Although only a small portion of the Marchantia Y chromosomehas been sequenced, it is sufficient to make meaningfulcomparisons with the euchromatic male-specific region (MSY)of the human Y chromosome, the sequence of which waspublished recently (Skaletsky et al., 2003; see also Hawley,2003). The sequence of the human MSY region and limitedcomparative sequencing of the MSY regions in great apes(Rozen et al., 2003) have added new insights to our understandingof how the mammalian testis gene families on the Ychromosome have been maintained over the course of evolution.As will be shown, the similarities between the human andliverwort Y chromosomes are striking and may reflect a commonmechanism underlying the evolution of the Y chromosome inthese two disparate organisms. The MSY region of the human Ychromosome is made up of three classes of sequences: Xtransposed, X degenerate, and ampliconic, the latter representing30% of the MSY euchromatin. Because the Marchantia Xchromosome has not been sequenced, it is only possible tomake comparisons between the human ampliconic sequences,which are Y specific, and the Marchantia Y chromosomesequences. Like the Marchantia Y chromosome sequences,the ampliconic regions of the MSY consist of highly repetitivesequences unique to the Y chromosome, although the sizes,sequences, and stoichiometries of the repetitive elements veryconsiderably between the two species. Protein-encoding genes
Sex DeterminationS63or gene families (six genes in Marchantia and nine gene familiesin human) occur within repetitive elements, and all are presentin multiple copies on the Y chromosome. In both organisms,homologs also may be present on the X or autosomal chromosomesin low copy number. Although all protein-encodinggenes found within the human ampliconic sequences are expressedonly in the testis, at least some of the protein-encodinggenes identified in Marchantia are male organ specific in theirexpression.According to the prevailing theory of mammalian sex chromosomeevolution (Graves and Schmidt, 1992; Jegalian and Page,1998; Lahn and Page, 1999), the X and Y chromosomes arederived from an ancient autosomal pair of chromosomes. The Ychromosome acquired genes, especially those that enhancemale fertility, by a series of autosomal transpositions that werethen amplified on the Y chromosome, whereas the X chromosomemaintained its ancestral genes. Other genes present on theY chromosome (i.e., X homologs) were lost, probably aided bya lack of X-Y recombination, leading to a mostly degenerate Ychromosome. The recent sequencing results suggest that theuniformity of the ampliconic repetitive sequences of the human Ychromosome is maintained from generation to generation byintrachromosomal Y-Y gene conversion. This occurs at relativelyhigh rates: 600 nucleotides per newborn human male areestimated to have undergone Y-Y gene conversion (Rozen et al.,2003). Although it is not known if there are higher orderpalindromic repeated sequences in Marchantia as there are inhumans (these palindromic sequences may be necessary forgene conversion), the homogeneity of repetitive sequences in theMarchantia Y chromosome and the relatively high frequency ofmale-specific genes within these repetitive elements suggestgene conversion playing a role in maintaining the repetitiveelements of the Marchantia Y chromosome.One important distinction between Marchantia and humans isthat X-X recombination cannot occur in Marchantia because alldiploid sporophytes are X-Y. This suggests that forces other thanhomologous interchromosomal recombination are responsiblefor preventing the degeneration of the X chromosome, althoughsize alone may not be an indicator of chromosome degeneracy.Future efforts to sequence the entire Y and X chromosomes inMarchantia will be very important in understanding how sexchromosomes specify sexual phenotype in this species and howboth the X and Y chromosomes evolved in this ancient lineage ofplants.THE LYCOPHYTESAnother group of land plants deserving attention from anevolutionary perspective is the lycophyte lineage, which includesthe modern Lycopodiales genera Selaginella and Isoetes. Thislineage is most closely related to the earliest vascular plants thatfirst appeared on land 250 to 400 million years ago (Kenrickand Crane, 1997). Although the earliest lycophytes and extantmembers of the Lycopodiales are homosporous and produceonly one type of spore, Selaginella and Isoetes are heterosporous,with their sporophytes producing free-living megasporesand microspores that give rise to the female and malegametophytes, respectively. Of the lycophytes, Selaginella hasgreat potential as a useful comparative system for the study ofsex determination in plants, in part because many species havevery small genome sizes (J.A. Banks, unpublished observations).Selaginella produces microsporangia and megasporangia thatare born on the same strobilus or in different strobili of the sameplant, depending on the species; strictly unisexual species havenot been reported. In Selaginella moellendorfii, for example, eachstrobilis bears both kinds of sporangia, with microsporangia atthe bottom and macrosporangia toward the top of each strobilis.In such plants, sex determination would be viewed as theprocess that regulates the sexual identity of the sporangia in thestrobilis, a mechanism that has clear parallels with floral organidentity in angiosperms that produce perfect flowers. Althoughwe know virtually nothing about this group of plants beyonddescriptive biology, this lineage holds important clues forunderstanding the evolution of heterospory from homospory,a switch that occurred many times during land plant evolution(Stewart and Rothwell, 1993) and that has had a major impact onthe timing of sex determination from the gametophyte tosporophyte generation in plants (Sussex, 1966). The questionof how heterospory evolved from homospory is difficult to studyin the heterosporous angiosperm lineage because their homosporousprogenitors are probably extinct.THE HOMOSPOROUS FERNSRecent phylogenetic analyses of vascular seed–free plantsgroup the leptosporangiate and eusporangiate ferns andmembers of Equisetum and Psilotum into a monophyletic cladethat is sister to the seed plants (Pryer et al., 2001). With fewexceptions, they are homosporous plants. The one plant forwhich a sex-determining pathway has been genetically welldefined is the leptosporangiate fern Ceratopteris richardii. LikeMarchantia, Ceratopteris is homosporous and produces onlyone type of haploid spore. Although the sex of the Marchantiagametophyte is determined genetically by sex chromosomes,the sex of the Ceratopteris gametophyte (male or hermaphroditic)is determined epigenetically by the pheromone antheridiogen.Since their discovery by Dopp (1950) in the fern Pteridiumaquilinum, antheridiogens have been identified and characterizedfrom many species of leptosporangiate ferns (reviewed byNaf, 1979; Yamane, 1998), suggesting that it is a common modeof regulating sexual phenotypes in this group of plants. Althoughthe structure of the Ceratopteris antheridiogen is unknown, allother fern antheridiogens characterized to date are mostly novelgibberellins (Yamane, 1998).The Ceratopteris male and hermaphroditic gametophytes areeasy to distinguish not only by the type of gamete produced butalso by the presence or absence of a multicellular meristem. Asillustrated in Figure 2, the hermaphrodite forms a single meristemand meristem notch that gives the hermaphrodite itsheart-shaped appearance. Cells of this meristem differentiate asegg-forming archegonia, sperm-forming antheridia, or simplyenlarge, adding to the growing sheet of photosynthetic parenchymacells that make up most of the hermaphrodite prothallus.A Ceratopteris spore grown in isolation always develops asa hermaphrodite. A male gametophyte develops from a sporeonly if the spore is placed in medium that had previously
S64The Plant CellFigure 2. The Sex-Determining Mutants of Ceratopteris richardii.The her1 (hermaphroditic) mutant and the wild-type hermaphrodite are indistinguishable, as are the tra1 (transformer) mutant and the wild-type males,except that the her1 and tra1 mutants are insensitive to the absence or presence of A CE .TheA CE -insensitive fem1 (feminization) gametophyte producesno antheridia. The man1 (many antheridia) mutant produces 10 times more antheridia than hermaphrodites, whereas the not1 (notchless) mutantrarely produces antheridia. The meristem notch normally present on the hermaphrodite often is missing in the not1 mutant, giving it a cup-shapedappearance. The novel phenotypes of the fem1 tra1 and fem1 not1 tra1 mutants are shown. an, antheridia; ar, archegonia; mn, meristem notch.supported the growth of a hermaphrodite. Lacking a multicellularmeristem, almost all cells of the male gametophyte terminallydifferentiate as antheridia. The male-inducing pheromone thatis secreted by the Ceratopteris hermaphrodite is called A CE forantheridiogen Ceratopteris. Based on physiological studies inCeratopteris (Banks et al., 1993), A CE is not secreted by thehermaphrodite until after it loses the competence to respond toits male-inducing effects, which corresponds to the initiation ofthe meristem. A gametophyte will develop as a male only if it isexposed continuously to A CE from a very young age, between 2to 4 days after spore inoculation. Thus, in a population of spores,those that germinate first become A CE -secreting meristichermaphrodites, whereas those that germinate later becomeameristic males under the influence of A CE secreted by itsneighboring hermaphrodites.To understand how A CE represses the development of femaletraits (i.e., archegonia and meristem) and promotes thedevelopment of male traits (i.e., antheridia) in Ceratopteris,a genetics approach has been used to identify the genesinvolved in this response (Banks, 1998; Strain et al., 2001). Todate, five phenotypic classes of mutants have been identified;representatives of each class are illustrated in Figure 2. Inaddition to those that are always hermaphroditic (the hermaphroditicmutants), always male (the transformer [tra] mutants), oralways female (the feminization [fem] mutants) regardless of theabsence or presence of A CE , there are mutants that produceexcessive antheridia (the many antheridia mutants) as well asthe feminizing mutants that often lack a meristem notch (thenotchless mutants). By comparing the phenotypes of doublemutant gametophytes to each single mutant gametophyteparent, the epistatic interactions among these genes have beenassessed. One particularly informative phenotype is the fem1tra1 double mutant, also illustrated in Figure 2. Unlike otherdouble mutant combinations, this one has a novel phenotypeunlike that of either parent. This finding suggests that these twogenes (FEM1 and TRA1) define two separate pathways, onespecifying male development and the other female development.The sex-determining mutants have been ordered into a geneticsex-determining pathway, illustrated in Figure 3, that is mostconsistent with the genetic data. In this pathway, the sex of thegametophyte ultimately depends on the activities of two genes,one specifying the development of male traits (FEM1) and theother specifying the development of female traits (TRA). Thesegenes also repress each other, so that when TRA is active, FEM1is not and visa versa. What determines which of these two genesis expressed in the gametophyte (and thus its sex) is A CE , whichultimately represses the TRA genes, as described in the legendto Figure 3.In comparing mechanisms of gametophytic sex determinationin homosporous bryophytes and ferns, one obvious questionthat arises is what drove Marchantia to an X-Y chromosomalmechanism of sex determination and Ceratopteris to an epigeneticallyregulated mechanism dependent on pheromonalcross-talk between individuals? The answer to this questionprobably lies in the different ratios of males and females orhermaphrodites that occur in the populations of each species. InMarchantia, the segregation of X and Y sex chromosomes duringmeiosis in the sporophyte ensures that each gametophyteprogeny has an equal probability of being either male or female,barring selection. In Ceratopteris, the A CE response allowsthe ratio of males to hermaphrodites to vary depending onthe density of the population, such that as the population densityincreases, the proportion of males also increases. Althoughthe underlying sex-determining mechanism is inflexible in
Sex DeterminationS65Figure 3. The Genetic Sex-Determining Pathways in the Fern Ceratopteris,the Fly Drosophila melanogaster, and the Nematode Caenorhabditiselegans.The genetic model of sex determination in Ceratopteris (Strain et al.,2001) is dependent on two genes, FEM1 and TRA (there are at least twoTRA genes), which promote the differentiation of male (antheridia) andfemale (meristem and archegonia) traits, respectively. FEM1 and TRAalso antagonize each other such that if FEM1 is active, TRA is not, andvice versa. What determines which of these two genes prevails in thegametophyte and thus its sex is the pheromone A CE , which activates theHER genes, of which there are at least five, and sets into motion a seriesof switches that ultimately result in male development (i.e., FEM1 on andTRA off). These switches are thrown in the opposite direction whenspores germinate in the absence of A CE . Although FEM1 represses TRAand TRA represses FEM1, they do not do so directly. TRA activatesMAN1, which represses FEM1, and FEM1 activates NOT1, whichrepresses TRA. Because TRA and FEM1 are the primary regulators ofsex, NOT1 and MAN1 are considered regulators of the regulators. Thesex determination pathway in C. elegans (Hodgkin, 1987; Villeneuve andMeyer, 1990) is linear and consists of a series of negative controlswitches. The state of the initial switch gene (xol-1) is dependent on theratio of X to autosomal (A) chromosomes. If the ultimate downstreamgene in this pathway (TRA-1) is high, the nematode develops asa hermaphrodite. If TRA-1 is low, it develops as a male. The linear sexdetermination pathway shown for Drosophila is from the mid-1980s(Baker and Ridge, 1980; Cline, 1983). There are actually other sexdeterminingpathways that account for most aspects of sexualphenotype (reviewed by Oliver, 2002); the pathway shown is the somaticpathway. In the soma, Sxl is the key regulator of sex, and its state ofactivity is determined by the X:A ratio. The dsx gene is the downstreamregulatory gene that ultimately determines whether male or female genesare expressed in the soma.Marchantia, it is flexible enough in Ceratopteris to allow eachindividual to determine its sex according to the size of thepopulation in which it resides and the speed at which it germinatesrelative to its neighbors. The flexibility of the Ceratopterissex-determining mechanism is reflected in its sex-determiningpathway, and this becomes especially apparent compared withthe sex-determining pathways known from other organisms,including Drosophila melanogaster and Caenorhabditis elegans,which are illustrated in Figure 3. In both of these animals, anindividual’s sex (male or female in D. melanogaster and male orhermaphrodite in C. elegans) is determined genetically by theratio of X to autosomal chromosomes. This ratio is read andeither activates or represses the activities of downstream genesin each pathway. In both animals, the sex ultimately depends onthe state of the terminal gene in each linear pathway, TRA1 in thecase of C. elegans. The Ceratopteris sex-determining pathway isdistinctly different from those of D. melanogaster and C. elegansin that it is not linear and there are two sex-determining genes,one for male and one for female development. Their ability torepress each other endows each Ceratopteris spore with theflexibility to determine its sex upon germination based onenvironmental cues.So why would a flexible mechanism of sex determination thatallows sex ratios to vary be adaptive in ferns but not inbryophytes? The answer to this question may lie in the ephemeralnature of the fern gametophyte. Although the gametophytes ofbryophytes are persistent, the gametophytes of ferns are not. InCeratopteris, for example, gametophytes reach sexual maturityonly 14 days after spore inoculation and die once they arefertilized. The limited time that a fern gametophyte is able to becrossed by another might favor a sex-determining system thatwould promote outcrossing by increasing the proportion ofmales when population densities are high and ensuring a highproportion of hermaphrodites capable of self-fertilization whenpopulation densities are low. Because there are a variety of sexdeterminingmutants available in Ceratopteris, the hypothesisrelated to the consequences of variable versus fixed sex ratioscan be tested easily, at least under defined laboratory conditions.Future studies to clone the sex-determining genes in Ceratopteriswill be necessary to understand their biochemicalfunctions and to test their interactions predicted by the geneticmodel. Although the size of the Ceratopteris genome is probablyvery large (n ¼ 37), the ability to inactivate genes in theCeratopteris gametophyte by RNA interference (Stout et al.,2003; G. Rutherford, M. Tanurdzic, and J.A. Banks, unpublishedobservations) provides an important means to study the effectsof inactivating potential sex-determining genes in the Ceratopterisgametophyte.THE FLOWERING PLANTSAlthough unisexuality is very common in animals, hermaphroditismis the rule in angiosperms. Approximately 90% of allangiosperm species have perfect flowers with specializedorgans producing microspores or megaspores from which themale or female gametophytes develop. Of the remaining species,approximately half are monoecious, producing unisexual flowersof both sexes on the same individual, and the other half aredioecious, with unisexual male and female flowers arising onseparate individuals (Yampolsky and Yampolsky, 1922). Thedistribution of dioecy and monoecy within the angiospermphylogenetic tree strongly favors the evolutionary scenario inwhich unisexual flowers evolved from perfect flowers multipletimes in the angiosperm lineage (Lebel-Hardenack and Grant,1997; Charlesworth, 2002). Not surprisingly, there are a variety ofsex determination mechanisms in the angiosperms. For organizationalpurposes only, sex determination in monoecious anddioecious species are treated separately in this review.
S66The Plant CellThe Monoecious AngiospermsThere are numerous terms to describe the variety of sexualphenotypes observed in monoecious plant species (defined andcompiled by Sakai and Weller, 1999). Although this rich nomenclatureis appropriate, it tends to confound the problemof sex determination in this group of plants. For simplicity,monoecious species here are grouped into two categories: thosethat produce only unisexual male and female flowers on the sameplant, and those that produce both unisexual and perfect flowerson the same plant.Zea mays (maize) is an example of a monoecious species thatproduces only unisexual flowers in separate male and femaleinflorescences, referred to as the tassel and ear, respectively.Unisexuality in maize occurs through the selective elimination ofstamens in ear florets (flowers) and by the elimination of pistils intassel florets (reviewed by Irish, 1999). Two general classes ofsex-determining mutants have been identified in maize, includingthose that masculinize ears and those that feminize tassels. Theanther ear (an1) and dwarf (d1, d2, d3, and d5) mutants of maizeare recessive and masculinize ears by preventing stamenabortion in the female florets (Wu and Cheung, 2000). Thedominant dwarf mutation, d8, has a similar phenotype. The D1,D3, and AN1 genes encode enzymes involved in the biosynthesisof the plant hormone gibberellin (GA) (Bensen et al., 1995;Winkler and Helentjaris, 1995; Spray et al., 1996). The D8 geneencodes the maize homolog of GIBBERELLIN INSENSITIVE/REPRESSOR OF gal-3 (Peng et al., 1999), a family oftranscription factors that negatively regulate GA responses inplants (Richards et al., 2001; reviewed by Olszewski et al., 2002).GA signaling is thought to be derepressed in the d8 mutant,resulting in a dominant phenotype. The molecular identity ofthese genes provides direct evidence that endogenous GAshave a feminizing role in sex determination in maize.The tasselseed1 (ts1) and ts2 mutants of maize feminize maleflorets in the tassel, as illustrated in Figure 4. In addition to thetassel phenotype, the second floret of the ear spikelet, whichnormally fails to develop, develops normally in the ts mutants,leading to a double-kerneled spikelet in the ear (Dellaporta andCalderon-Urrea, 1994; Irish, 1999). The TS2 gene has beencloned and shown to encode a putative short-chain alcoholdehydrogenase with signature motifs of steroid dehydrogenases(DeLong et al., 1993). TS2 mRNA is expressed in the subepidermallayer of the gynoecium before its abortion, whichcorrelates well with the timing and location of cell deathresponsible for the pistil abortion in wild-type male florets(Calderon-Urrea and Dellaporta, 1999). The expression of ts2requires the wild-type TS1 gene. Although the pistils of ear floretsalso express TS1 and TS2, they are protected from a deathly fateby the action of another gene, SILKLESS1, that interacts directlyor indirectly with TS2 (Calderon-Urrea and Dellaporta, 1999).Although the cloning of the sex-determining genes in maizedemonstrates that GAs and possibly other steroid-like hormonesplay a pivotal role in stamen abortion and feminization of flowers,the spatial distribution of these molecules could have an effect onthe sex determination process, as exemplified by a steepgradient in GA abundance along the maize shoot (Rood et al.,1980), which correlates well with the male-suppressing andFigure 4. The ts4 Mutant of Maize.The wild-type male tassel (left) produces only male florets, whereasplants homozygous for the ts4 allele (right) form hermaphroditic flowerswith functional pistils, allowing seeds to form in the tassel. Photographscourtesy of Erin Irish, University of Iowa.female-promoting phenotypic effects of GA. How the synthesisor transport of these molecules is regulated, and the identityof the downstream targets of these hormones, remain to bediscovered.Another monoecious plant, Cucumis sativus (cucumber), hasserved as a model system for sex determination studies since the1950s and early 1960s, driven by breeding programs for hybridseed production. Cucumber plants are mostly monoecious butcan be dioecious or hermaphroditic, depending on the genotype.Regardless of their sex, all floral buds are initially hermaphroditic,and it is the arrest of stamen or pistil development that leads tounisexual flowers. In monoecious cucumbers, male flowers format the bottom and female flowers form at the top of each shoot.There are three major genes that affect the arrangement ofunisexual flowers or their sex; they are designated the F, A, and Mgenes (following the nomenclature of Pierce and Werner, 1990).The F gene is semidominant and affects the expression offemaleness along the plant, causing it to extend the gradientof femaleness toward the bottom of the plant. The A geneis epistatic to F, and it too is required for the expression offemaleness. The M gene is required for maleness in that it is notrequired for the establishment of the gender gradient along theshoot but rather for the selective abortion of pistils and stamensin female and male flowers, respectively (Perl-Treves, 1999).With respect to the F and M genes, M-ff plants are monoecious,M-F- plants are female, mmF- plants are hermaphroditic, andmmff plants are andromonoecious (with male and hermaphroditicflowers).In addition to the sex-determining genes, plant hormones havelong been implicated in the sex-determining process in cucumber.GA and ethylene application and the use of GA and ethyleneinhibitors can subvert the genotypic constitution of the plant, withGA acting mainly as a masculinizing agent and ethylene actingas a feminizing agent (Perl-Treves, 1999). By treating monoeciousand andromonoecious cucumber plants with various
Sex DeterminationS67combinations of GA and ethrel or GA and ethylene inhibitors, Yinand Quinn (1995) demonstrated that ethylene is the main regulatorof sex determination, with GA functioning upstream ofethylene, possibly as a negative regulator of endogenousethylene production. These findings led them to propose a modelfor how sex determination might occur (Yin and Quinn, 1995),with ethylene serving both as a promoter of the female sex and aninhibitor of the male sex. The basic tenets of the model are thatthe F gene should encode a molecule that would determine therange and gradient of ethylene production along the shoot,thereby acting to promote femaleness, whereas the M geneshould encode a molecule that perceives the ethylene signal andinhibits stamen development above threshold ethylene levels.This model is consistent with how unisexual flowers might arisevery early and very late during shoot development; however, themodel also predicts an entire range of intermediate types rarelyor never seen in cucumber. As suggested by Perl-Treves (1999),variations in the model of Yin and Quinn and the incorporation ofadditional factors in the sex-determining process in cucumbercould account for the observed lack of intermediate types.Recent results from several laboratories have providedmolecular evidence in favor of the ethylene theory of sexdetermination in cucumber. Two 1-aminocyclopropane-1-carboxylicacid synthase genes, CS-ACS1 and CS-ACS2, havebeen identified in cucumber, and one of them (CS-ACS1) mapsto the F locus (Trebitsh et al., 1997). The monoecious cucumbergenome has only one copy (Cs-ACS1), whereas the gynoeciousgenome has both copies. The expression of both genescorrelates with sexual phenotype, with gynoecious plantsaccumulating more transcript than monoecious or andromonoeciousplants (Kamachi et al., 1997; Yamasaki et al., 2001).Although these studies are consistent with the female-promotingeffects of ethylene, they do not address the question ofhow ethylene inhibits stamen abortion in gynoecious and notandromonoecious plants. Yamasaki et al. (2001) providedevidence suggesting that the product of the M locus mediatesthe inhibition of stamen development by ethylene (i.e., M affectssensitivity to ethylene). This finding indicates that ethyleneconcentration, which is likely to be dependent on the F locus,and the differential sensitivity of males and females to ethylene,which is likely to be dependent on the M locus, are bothimportant in regulating sexual phenotype in cucumber. Definitivecloning of the M and F genes will allow these hypotheses to betested directly.Because sex determination in cucumber and most otherangiosperm species occurs via selective abortion of flowerorgans, Kater et al. (2001) set out to establish whether thisabortion is based on organ identity or positional informationwithin the flower. The availability of cucumber homologs of theMADS box ABC homeotic genes and the ability to express themectopically in cucumber allowed these authors to show that thesex determination machinery in cucumber selectively aborts sexorgans based on their position rather than their identity (i.e., inmale flowers, carpels are aborted only in the fourth whorl, and infemale flowers, stamens abort only in the third whorl). In addition,because nonreproductive organs that develop in the inner whorlsof a C-class homeotic mutant are not aborted, Kater et al. (2001)speculated that C-class gene products might be targets of thesex-determining process. Even though this is a temptingspeculation, it leaves the question of abortion timing unresolved.Several studies have addressed the role of the MADS box floralhomeotic genes in the sex determination process in many monoeciousand dioecious plants, including Asparagus officinalis(Park et al., 2003), Betula pendula (Elo et al., 2001), Gerberahybrida (Yu et al., 1999), Populus deltoides (Sheppard et al.,2000), Rumex acetosa (Ainsworth et al., 1995), Silene latifolia(Hardenack et al., 1994), and maize (Heuer et al., 2001). In allcases, the expression of B- or C-function floral homeotic genesin carpels or stamens was shown to decline in organs targetedfor abortion in the unisexual flower. However, it is not clear ifthe changes in expression of these genes are a cause or a consequenceof organ abortion. Furthermore, evidence that sexdeterminingmutants cosegregate with MADS box genes islacking. That plants may not use MADS box genes for sexdetermination would not be surprising given that the unisexualflowers of most monoecious and dioecious plants are derivedfrom bisexual flowers with all sex organ primordia present.Another plant within the monoecious group of angiospermsthat has been studied is Carica papaya (papaya). Papaya isa polygamous species with three sexes—females, males, andhermaphrodites. In this species, sexual phenotype is controlledby a single gene with three alleles: the dominant M (for male), thedominant M h (for hermaphrodite), and the recessive m (forfemale) alleles (Storey, 1938). The progeny of self-fertilizedhermaphrodites (genotypically M h m) segregate hermaphroditesand females in a 2:1 ratio. The lack of male progeny indicate thatthe MM genotype is lethal, perhaps because of a lethal geneclosely linked to the M locus. This and other crosses led earlyworkers to the conclusion that all males are genotypically M h M.Thus, females are the homogametic and males the heterogameticsex. Papaya flowers of different sexes also displaysecondary sexual characteristics that cosegregate with the Mallele. The apparent lack of recombination between the lociresponsible for primary and secondary sex traits indicates thatboth loci are tightly linked and inherited en bloc, much like a sexchromosome (Storey, 1953), although heteromorphic chromosomesdo not exist (Kumar et al., 1945).The economical importance of papaya has driven sex determinationresearch in this species, because the pyriform fruitsfrom hermaphroditic trees are preferred by consumers more thanthe spherical fruits produced by the female trees. Because it isonly upon flowering that the sex of the individual papaya tree canbe determined, molecular markers that cosegregate with the Malleles have been sought intensively. To date, two groups havereported randomly amplified polymorphic DNA markers that arehighly specific for males and hermaphrodites but absent infemales (Deputy et al., 2002; Urasaki et al., 2002). One randomlyamplified polymorphic DNA marker was also shown by DNA gelblot hybridization to be completely absent from the female genome(Urasaki et al., 2002), providing the first molecular evidencefor genomic differentiation between the sexes. Although papayais a tropical tree and not considered a model plant system, it hasa small genome of 371 Mbp (Arumuganathan and Earle, 1991)and may be transformable (Fitch et al., 1992). These facts,coupled with its economic importance in tropical and subtropicalregions of the world, make it is a species worthy of further study.
S68The Plant CellThe Dioecious AngiospermsSilene latifolia is a dioecious species with individual plantsproducing either all female or all male flowers. As it is by far thebest characterized dioecious species to date, our review of sexdeterminingmechanisms in dioecious plants will focus on thisspecies. In male and female S. latifolia flowers, the gynoeciumand androecium initiate but arrest development prematurely,leading to functionally unisexual flowers (Grant et al., 1994). Thesexual phenotype of individuals is determined by sex chromosomes;males are heterogametic (XY) and females are homogametic(XX). Early cytogenetic studies of sex-determining mutantsin S. latifolia led Westergaard (1946, 1958) to conclude that theY chromosome is divided into three regions relevant to sexexpression: one required for the suppression of female developmentand two required for the promotion of male development.None of these regions would be necessary for thedevelopment of female reproductive organs, because thesefunctions would reside on the X or autosomal chromosomes.Additional sex-determining mutants have been generated recentlyby x-ray mutagenesis of pollen and selecting bothhermaphrodites and asexual F1 progeny (Farbos et al., 1999;Lardon et al., 1999; Lebel-Hardenack et al., 2002). Thesemutants verify the earlier work of Westergaard (1946, 1958; hislines were apparently lost) and have resulted in the identificationof two additional classes of mutants, those that are not Y linkedand hermaphroditic and those that are Y linked and asexual. Thehermaphroditic deletion mutants are likely to contain a gene(s)necessary for the female-suppressing function, whereas theasexual deletion mutants likely contain the male-promotinggene(s). Genetic screens to identify mutant XX hermaphroditesor asexuals have not been reported. Although these sexdetermininggenes have not been cloned, the construction ofan amplified fragment length polymorphism map of the Ychromosome using lines deleted for overlapping regions of theY chromosome will be useful for genetic and physical mapping ofthe sex-determining mutants (Lebel-Hardenack et al., 2002) andmay ultimately lead to their cloning.Another approach to identify sex-determining genes in S.latifolia has been to clone genes that are expressed specifically inthe male flowers and determine their linkage to the Y chromosome(reviewed by Charlesworth, 2002). Of the >50 isolatedgenes that have been correlated with sex expression in S. latifoliato date, only the four listed in Table 1 have been shown to belinked to the Y chromosome. Genes linked exclusively to the Ychromosome have not been found, because all of the Y-linkedgenes have X or autosomal homologs. These data indicatethat the S. latifolia Y chromosome most closely resembles theX-degenerate class of sequences of the human Y chromosome.Genes within this class have an X homolog, the X andY homologs encode similar but nonidentical isoforms, many ofthem are ubiquitously expressed, and all are present in low copynumber in the genome (Skaletsky et al., 2003). As shown in Table1, many of these features are shared with the Y-linked genes ofS. latifolia.Although molecular approaches have not yet succeeded inidentifying the major regulatory sex-determining genes in S.latifolia, this work has and will continue to test theories of how Ychromosomes evolved from an ancestral pair of autosomes inplants (Charlesworth, 1996, 2002; Negrutiu et al., 2001). Thesetheories state that once mutations that result in geneticallydetermined males (where female genes are repressed) andfemales (where male genes are repressed) occur, recombinationbetween the sex-determining genes must be suppressed toavoid recombinant asexual or hermaphroditic offspring. Anotherconsequence of nonrecombination is the certainty that unisexualmale and female offspring will be produced in equal ratios.Recombination between homologous chromosomes often issuppressed by the accumulation of chromosomal inversions onone homolog (in this case, the Y). The lack of X-Y recombinationwould eventually lead to degeneracy and loss of gene function onthe Y chromosome, with the exception of genes required for malefertility and those necessary to suppress female fertility. The S.latifolia Y chromosome appears not to fit this paradigm forseveral reasons. First, the Y chromosome is 1.4 times larger thanthe X chromosome and is largely euchromatic, indicating that itmay not be degenerate (Ciupercescu et al., 1990; Grabowska-Joachimiak and Joachimiak, 2002). Second, measurements ofDNA polymorphism in sex-linked gene pairs have revealed thatalthough the DNA polymorphism of S4Y-1 is greater than that ofS4X-1, the DNA polymorphism of SlY-1 is 20-fold lower than thatof SlX-1 (Filatov et al., 2000, 2001; Filatov and Charlesworth2002). Given that the S. latifolia sex chromosomes diverged lessthan 20 million years ago (Desfeux et al., 1996; Charlesworth,2002), which is much less than the 240- to 320-million-yeartimeline for human sex chromosome evolution (Lahn and Page,1999), it is likely that the S. latifolia Y chromosome is at a relativelyyoung stage of evolution (Charlesworth, 2002). Comparativesequencing of the S. latifolia sex chromosomes will be importantin understanding the evolution of the Y chromosome in plants,especially compared with the Y chromosome of Marchantia andthe sex-determining chromosome region of papaya.Table 1. Sex Chromosome–Linked Genes in S. latifoliaY-Linked Gene X-Linked Gene Male-Specific Expression Function ReferenceSlY-1 SlX-1 SlY-1 yes, SlX-1 no WD-repeat protein Delichere et al., 1999SlY-4 SlX-4 No Fructose-2,6-bisphosphatase Atanassov et al., 2001MRO3-Y a MRO3-X a Yes Unknown Matsunaga et al., 1996; Guttman andCharlesworth 1998DD44Y DD44X No Oligomycin sensitivity–conferringproteinMoore et al., 2003a The Y homolog is inactive and the X homolog is active.
Sex DeterminationS69FUTURE DIRECTIONSSex determination in plants is a fundamental developmentalprocess that is particularly important for economic reasons,because the sexual phenotypes of commercially important cropsdictate how they are bred and cultivated. Although most cropplants are not considered model systems—and sex determinationis not a problem that can be addressed in the modelangiosperm Arabidopsis—the economic value in manipulatingthe sexual phenotypes of crop plants should continue todrive interest in this area of research. Recent studies of sexdeterminingmechanisms have demonstrated clearly that angiosperms,including crop plants, have evolved a variety ofsex-determining mechanisms that involve a number of differentgenetic and epigenetic factors, from sex chromosomes to planthormones. Although the determinants of sexual phenotype arediverse, determining whether the downstream master sexregulatorygenes that specify male or female development areheld in common or not will require cloning the sex-determininggenes from a variety of plant species.Choosing to study sex determination in plants representingother major land plant lineages will allow several broaderdevelopmental and evolutionary questions to be addressed.One unresolved question is how heterospory evolved fromhomospory. By identifying the sex-determining genes in homosporousplants such as Ceratopteris and examining theexpression of possible homologous genes in closely relatedheterosporous species, one can test the hypothesis that theswitch from homospory to heterospory involved a heterochronicshift in the timing of expression of these genes from thegametophyte to the sporophyte generation. Other questions tobe resolved are how sex chromosomes evolved in plants andwhether similar processes led to distinct sex chromosomes inplants and animals. Comparing the Y chromosome sequences ofMarchantia and S. latifolia, for example, will be invaluable inunderstanding how male-promoting genes and female-suppressinggenes became localized to a Y chromosome and howrecombination between the Y and its homolog was and continuesto be suppressed.ACKNOWLEDGMENTSSupport was provided by the National Science Foundation (MCB-9723154). This is journal paper 17271 of the Purdue UniversityAgricultural Experiment Station.Received August 25, 2003; accepted January 19, 2004.REFERENCESAinsworth, C., ed (1999). Sex Determination in Plants. (Oxford, UK:BIOS Scientific Publishers).Ainsworth, C. (2000). Boys and girls come out to play: The molecularbiology of dioecious plants. Ann. Bot. 86, 211–221.Ainsworth, C., Crossley, S., Buchanan-Wollaston, V., Thangavelu,M., and Parker, J. (1995). Male and female flowers of the dioeciousplant sorrel show different patterns of MADS box gene expression.Plant Cell 7, 1583–1598.Arumuganathan, K., and Earle, E.D. (1991). Nuclear DNA content ofsome important plant species. Plant Mol. Biol. Rep. 9, 208–218.Atanassov, I., Delichere, C., Filatov, D., Charlesworth, D., Negrutiu,I., and Moneger, F. (2001). Analysis and evolution of two functionalY-linked loci in a plant sex chromosome system. Mol. Biol. Evol.18, 2126–2168.Baker, B.S., and Ridge, K.A. (1980). Sex and the single cell. I. On theaction of major loci affecting sex determination in Drosophilamelanogaster. Genetics 94, 383–423.Banks, J. (1998). Sex determination in the fern Ceratopteris. TrendsPlant Sci. 2, 175–180.Banks, J., Hickok, L., and Webb, M.A. (1993). The programming ofsexual phenotype in the homosporous fern, Ceratopteris richardii. Int.J. Plant Sci. 154, 522–534.Barrett, S. (2002). The evolution of plant sexual diversity. Nat. Rev. 3,274–284.Bensen, R., Johal, J., Crane, V.C., Tossberg, J.T., Schnable, P.S.,Meeley, R.B., and Briggs, S.P. (1995). Cloning and characterizationof the maize An1 gene. Plant Cell 7, 75–84.Calderon-Urrea, A., and Dellaporta, S.L. (1999). Cell death and cellprotection genes determine the fate of pistils in maize. Development126, 435–441.Charlesworth, D. (1996). The evolution of chromosomal sex determinationand dosage compensation. Curr. Biol. 6, 149–162.Charlesworth, D. (2002). Plant sex determination and sex chromosomes.Heredity 88, 94–101.Ciupercescu, D., Veuskens, J., Mouras, A., Ye, D., Briquet, M., andNegrutiu, I. (1990). Karyotyping Melandrium album, a dioecious plantwith heteromorphic sex chromosomes. Genome 33, 556–562.Cline, T.W. (1983). The interaction between daughterless and sex-lethalin triploids: A lethal sex-transforming maternal effect linking sexdetermination and dosage compensation in Drosophila melanogaster.Dev. Biol. 95, 260–274.Delichere, C., Veuskens, J., Hernould, M., Barbacar, N., Mouras, A.,Negrutiu, I., and Moneger, F. (1999). SlY1, the first active genecloned from a plant Y chromosome, encodes a WD-repeat protein.EMBO J. 11, 4169–4179.Dellaporta, S.L., and Calderon-Urrea, A. (1994). The sex determinationprocess in maize. Science 266, 1501–1505.DeLong, A., Calderon-Urrea, A., and Dellaporta, S.L. (1993). Sexdetermination gene TASSELSEED2 of maize encodes a short-chainalcohol dehydrogenase required for stage-specific floral organabortion. Cell 27, 757–768.Deputy, J.C., Ming, R., Ma, H., Liu, Z., Fitch, M.M., Wang, M.,Manshardt, R., and Stiles, J.I. (2002). Molecular markers for sexdetermination in papaya (Carica papaya L.). Theor. Appl. Genet. 106,107–111.Desfeux, C., Maruice, S., Henry, J.P., Lejeune, B., and Gouyon, P.H.(1996). Evolution of reproductive systems in the genus Silene. Proc.R. Soc. Lond. Ser. B 263, 409–414.Dopp, W. (1950). Eine die Antheridienbildung bei farnen forderndeSubstanz in den Prothallien von Pteridium aquilinum L. Kuhn. Ber.Dtsch. Bot. Ges. 63, 139–147.Elo, A., Lemmetyinen, J., Turunen, M., Tikka, L., and Sopanen, T.(2001). Three MADS-box genes similar to APETALA1 and FRUITFULLfrom silver birch (Betula pendula). Physiol. Plant. 112, 95–103.Farbos, I., Veuskens, J., Vyskot, B., Oliveira, M., Hinnisdaels, S.,Aghmir, A., Mouras, A., and Negrutiu, I. (1999). Sexual dimorphismin white campion: Deletion of the Y chromosome results in a floralasexual phenotype. Genetics 151, 1187–1196.Filatov, D., and Charlesworth, D. (2002). Substitution rates in theX- and Y-linked genes of the plants Silene latifolia and S. dioica. Mol.Biol. Evol. 19, 898–907.
S70The Plant CellFilatov, D., Laporte, V., Vitte, C., and Charlesworth, D. (2001). DNAdiversity in sex-linked and autosomal genes of the plant speciesSilene latifolia and Silene dioica. Mol. Biol. Evol. 18, 1442–1454.Filatov, D., Moneger, F., Negrutiu, I., and Charlesworth, D. (2000).Low variability in a Y-linked plant gene and its implications forY-chromosome evolution. Nature 404, 388–390.Fitch, M.M., Manshardt, R., Gonsalves, D., and Slightom, J.L. (1992).Virus resistant papaya derived from tissue bombarded with the coatprotein gene of papaya ringspot virus. Bio/Technology 10, 1466–1472.Geber, M.A., Dawson, T.E., and Delph, L.F., eds (1999). Gender andSexual Dimorphism in Flowering Plants. (Berlin: Springer).Grabowska-Joachimiak, A., and Joachimiak, A. (2002). C-bandedkaryotypes of two Silene species with heteromorphic sex chromosomes.Genome 45, 243–252.Grant, S., Hunkirchen, B., and Saedler, H. (1994). Developmentaldifferences between male and female flowers in the dioecious plantSilene latifolia. Plant J. 6, 471–480.Graves, J., and Schmidt, M. (1992). Mammalian sex chromosomes:Design or accident? Curr. Opin. Genet. Dev. 2, 890–901.Guttman, D., and Charlesworth, D. (1998). An X-linked gene witha degenerate Y-linked homologue in a dioecious plant. Nature 393,263–266.Hardenack, S., Ye, D., Saedler, H., and Grant, S. (1994). Comparisonof MADS box gene expression in developing male and female flowersof the dioecious plant white campion. Plant Cell 6, 1775–1787.Hawley, R.S. (2003). The human Y chromosome: Rumors of its deathhave been greatly exaggerated. Cell 113, 825–828.Heuer, S., Hansen, S., Bantin, J., Brettschneider, R., Kranz, E., Lorz,H., and Dresselhaus, T. (2001). The maize MADS box geneZmMADS3 affects node number and spikelet development and isco-expressed with ZmMADS1 during flower development, in eggcells, and early embryogenesis. Plant Physiol. 127, 33–45.Hodgkin, J. (1987). Sex determination and dosage compensation inCaenorhabditis elegans: Variations on a theme. Annu. Rev. Genet. 21,133–154.Irish, E.E. (1999). Maize sex determination. In Sex Determination inPlants, C. Ainsworth, ed (Oxford, UK: BIOS Scientific Publishers), pp.183–188.Ishizaki, K., Shimizu-Ueda, Y., Okada, S., Yamamoto, M., Fujisawa,M., Yamato, K.T., Fukuzawa, H., and Ohyama, K. (2002). Multicopygenes uniquely amplified in the Y chromosome-specific repeats of theliverwort Marchantia polymorpha. Nucleic Acids Res. 30, 4675–4681.Jegalian, K., and Page, D. (1998). A proposed path by which genescommon to mammalian X and Y chromosomes evolve to become Xinactivated. Nature 394, 776–780.Kamachi, S., Sekimoto, H., Kondo, H., and Sakai, S. (1997). Cloningof a cDNA for a 1-aminocyclopropane-1-carboxylate synthase that isexpressed during development of female flowers at the apices ofCucumis sativus L. Plant Cell Physiol. 38, 1197–1206.Kater, M.M., Franken, J., Carney, K., Colombo, L., and Angenent,G.C. (2001). Sex determination in the monoecious species cucumberis confined to specific floral whorls. Plant Cell 13, 481–493.Kenrick, P., and Crane, P.R. (1997). The origin and early evolution ofplants on land. Nature 389, 33–39.Kumar, L., Abraham, A., and Srinivasan, V. (1945). The cytology ofCarica papaya Linn. Indian J. Agric. Sci. 15, 242–253.Lahn, B., and Page, D. (1999). Four evolutionary strata on the human Xchromosome. Science 286, 964–967.Lardon, A., Seorgiev, S., Aghmir, A., Le Merrer, G., and Negruitiu, I.(1999). Sexual dimorphism in white campion: Complex control ofcarpel number is revealed by Y chromosome deletions. Genetics 151,1173–1185.Lebel-Hardenack, S., and Grant, S.R. (1997). Genetics of sexdetermination in flowering plants. Trends Plant Sci. 2, 130–136.Lebel-Hardenack, S., Hauser, E., Law, T., Schmidt, J., and Grant, S.(2002). Mapping of sex determination loci on the white campion(Silene latifolia) Y chromosome using amplified fragment lengthpolymorphism. Genetics 160, 717–725.Lorbeer, G. (1934). Die Zytologie der Lebermoose mit besondererBerucksichtingung allgemeiner Chromosomenfragen. Jahrb. Wiss.Bot. 80, 567–817.Matsunaga, S., and Kawano, S. (2001). Sex determination by sexchromosomes in dioecious plants. Plant Biol. 3, 481–488.Matsunaga, S., Kawano, S., Takano, H., Uchida, H., Sakai, A.,and Kuroiwa, T. (1996). Isolation and developmental expression ofmale reproductive organ-specific genes in a dioecious campion,Melandrium album (Silene latifolia). Plant J. 10, 679–698.Moore, R.C., Kozyreva, O., Lebel-Hardenack, S., Siroky, J., Hobza,R., Vyskot, B., and Grant, S.R. (2003). Genetic and functionalanalysis of DD44, a sex-linked gene from the dioecious plant Silenelatifolia, provides clues to early events in sex chromosome evolution.Genetics 163, 321–334.Naf, U. (1979). Antheridiogens and antheridial development. In TheExperimental Biology of Ferns, A.F. Dyer, ed (New York: AcademicPress), pp. 436–470.Negrutiu, I., Vyskot, B., Barbacar, N., Georgiev, S., and Moneger, F.(2001). Dioecious plants: A key to the early events of sex chromosomeevolution. Plant Physiol. 127, 1418–1424.Okada, S., et al. (2000). Construction of male and female PAC genomiclibraries suitable for identification of Y-chromosome-specific clonesfrom the liverwort, Marchantia polymorpha. Plant J. 24, 421–428.Okada, S., et al. (2001). The Y chromosome in the liverwort Marchantiapolymorpha has accumulated unique repeat sequences harboringa male-specific gene. Proc. Natl. Acad. Sci. USA 98, 9454–9459.Oliver, B. (2002). Genetic control of germline sexual dimorphism inDrosophila. Int. Rev. Cytol. 219, 1–60.Olszewski, N., Sun, T.-p., and Gubler, F. (2002). Gibberellin signaling:Biosynthesis, catabolism, and response pathways. Plant Cell 14(suppl.), S61–S80.Park, H.H., Ishikawa, Y., Yoshida, R., Kanno, A., and Kameya, T.(2003). Expression of AODEF, a B-functional MADS-box gene, instamens and inner sepals of the dioecious species Asparagusofficinalis L. Plant Mol. Biol. 51, 867–875.Peng, J., et al. (1999). ‘‘Green revolution’’ genes encode mutantgibberellin response modulators. Nature 400, 256–261.Perl-Treves, R. (1999). Male to female conversion along the cucumbershoot: Approaches to studying sex genes and floral development inCucumis sativus. In Sex Determination in Plants, C. Ainsworth, ed(Oxford, UK: Bios Scientific Publishers), pp. 189–216.Pierce, L.K., and Werner, T.C. (1990). Review of genes and linkagegroups in cucumber. HortScience 25, 605–615.Pryer, K.M., Schneider, H., Smith, A.R., Cranfill, R., Wolf, P.G.,Hunt, J.S., and Sipes, S.D. (2001). Horsetails and ferns are a monophyleticgroup and the closest living relatives to seed plants. Nature409, 618–622.Richards, D.E., King, K.E., Ait-ali, T., and Harberd, N.P. (2001). Howgibberellin regulates plant growth and development: A moleculargenetic analysis of gibberellin signaling. Annu. Rev. Plant Physiol.Plant Mol. Biol. 52, 67–88.Rood, S.B., Pharis, R.P., and Major, D.J. (1980). Changes ofendogenous gibberellin-like substances with sex reversal of theapical inflorescence of corn. Plant Physiol. 66, 793–796.Rozen, S., Skaletsky, H., Marszalek, J.D., Minx, P.J., Cordum, H.S.,Waterston R.H., Wilson R.K., and Page D.C. (2003). Abundant gene
Sex DeterminationS71conversion between arms of palindromes in human and ape Ychromosomes. Nature 423, 873–876.Sakai, A.K., and Weller, S.G. (1999). Gender and sexual dimorphism inflowering plants: A review of terminology, biogeographic patterns,ecological correlates and phylogenetic approaches. In Gender andSexual Dimorphism in Flowering Plants, M.S. Geber, T.E. Dawson,and L.F. Delph, eds (Berlin: Springer), pp. 1–31.Sheppard, L.A., Brunner, A., Krutovskii, K., Rottmann, W., Skinner,J., Vollmer, S., and Strauss, S.H. (2000). A DEFICIENS homolog fromthe dioecious tree black cottonwood is expressed in female and malefloral meristems of the two-whorled, unisexual flowers. Plant Physiol.124, 627–640.Skaletsky, H., et al. (2003). The male-specific region of the human Ychromosome is a mosaic of discrete sequence clusters. Nature 423,825–837.Smith, G.M. (1955). Cryptogamic Botany. Vol. II. Bryophytes andPteridophytes. (New York: McGraw-Hill).Spray, C.R., Kobayashi, M., Suzuki, Y., Phinney, B.O., Gaskin, P.,and MacMillan, J. (1996). The dwarf-1 (d1) mutant of Zea maysblocks steps in the gibberellin-biosynthetic pathway. Proc. Natl. Acad.Sci. USA 93, 10515–10518.Stewart, W., and Rothwell, G. (1993). Paleobotany and the Evolution ofPlants. (New York: Cambridge University Press).Storey, W.B. (1938). Segregation of sex types in solo papayaand their application to the selection of seed. Am. Soc. Hortic. Sci. 35,83–85.Storey, W.B. (1953). Genetics of papaya. J. Hered. 44, 70–78.Stout, S.C., Clark, G.B., Archer-Evans, S., and Roux, S.J. (2003).Rapid and efficient suppression of gene expression in a single-cellmodel system, Ceratopteris richardii. Plant Physiol. 131, 1165–1168.Strain, E., Hass, B., and Banks, J. (2001). Characterization ofmutations that feminize gametophytes of the fern Ceratopteris.Genetics 159, 1271–1281.Sussex, I. (1966). The origin and development of heterospory invascular plants. In Trends in Plant Morphogenesis, E. Cutter, ed(New York: John Wiley & Sons), pp. 140–152.Takenaka, M., Yamaoka, S., Hanajiri, T., Shimizu-Ueda, Y., Yamato,K.T., Fukuzawa, H., and Ohyama, K. (2000). Direct transformationand plant regeneration of the haploid liverwort Marchantia polymorphaL. Transgenic Res. 9, 179–185.Trebitsh, T., Rudich, J., and Riov, J. (1997). Identification of a1-aminocyclopropane-1-carboxylic acid synthase gene linked tothe Female (F) locus that enhances female sex expression in cucumber.Plant Physiol. 113, 987–995.Urasaki, N., Tokumoto, M., Tarora, K., Ban, Y., Kayano, T., Tanaka,H., Oku, H., Chinen, I., and Terauchi, R. (2002). A male andhermaphrodite specific RAPD marker for papaya (Carica papaya L.).Theor. Appl. Genet. 104, 281–285.Villeneuve, A.M., and Meyer, B.J. (1990). The regulatory hierarchycontrolling sex determination and dosage compensation in Caenorhabditiselegans. Adv. Genet. 27, 117–188.Westergaard, M. (1946). Aberrant Y chromosomes and sex expressionin Melandrium album. Hereditas 32, 419–443.Westergaard, M. (1958). The mechanism of sex determination indioecious flowering plants. Adv. Genet. 9, 217–281.Winkler, R.G., and Helentjaris, T. (1995). The maize Dwarf3 geneencodes a cytochrome P450-mediated early step in gibberellinbiosynthesis. Plant Cell 7, 1307–1317.Wu, H.-M., and Cheung, A.Y. (2000). Programmed cell death in plantreproduction. Plant Mol. Biol. 44, 267–281.Yamane, H. (1998). Fern antheridiogens. Int. Rev. Cytol. 184, 1–32.Yamasaki, S., Fujii, N., Matsuura, S., Mizusawa, H., and Takahashi,H. (2001). The M locus and ethylene-controlled sex determination inandromonoecious cucumber plants. Plant Cell Physiol. 42, 608–619.Yampolsky, C., and Yampolsky, H. (1922). Distribution of the sexforms in the phanerogamic flora. Bibl. Genet. 3, 1–62.Yin, T., and Quinn, J.A. (1995). Tests of a mechanistic model of onehormone regulating both sexes in Cucumis sativus (Cucurbitaceae).Am. J. Bot. 82, 1537–1546.Yu, D., Kotilainen, M., Pollanen, E., Mehto, M., Elomaa, P.,Helariutta, Y., Albert, V., and Teeri, T. (1999). Organ identity genesand modified patterns of flower development in Gerbera hybrida(Asteraceae). Plant J. 17, 51–62.