Portal de Periódicos da CAPES

The Evolutionary Consequences of Polyploidy

2007; Cell Press; Volume: 131; Issue: 3 Linguagem: Inglês

10.1016/j.cell.2007.10.022

1097-4172

Sarah P. Otto,

CRISPR and Genetic Engineering

Polyploidization, the addition of a complete set of chromosomes to the genome, represents one of the most dramatic mutations known to occur. Nevertheless, polyploidy is well tolerated in many groups of eukaryotes. Indeed, the majority of flowering plants and vertebrates have descended from polyploid ancestors. This Review examines the short-term effects of polyploidization on cell size, body size, genomic stability, and gene expression and the long-term effects on rates of evolution. Polyploidization, the addition of a complete set of chromosomes to the genome, represents one of the most dramatic mutations known to occur. Nevertheless, polyploidy is well tolerated in many groups of eukaryotes. Indeed, the majority of flowering plants and vertebrates have descended from polyploid ancestors. This Review examines the short-term effects of polyploidization on cell size, body size, genomic stability, and gene expression and the long-term effects on rates of evolution. One of the most striking features of genome structure is its lability. From small-scale rearrangements to large-scale changes in size, genome comparisons among species reveal that variation is commonplace. Even over the short time course of laboratory experiments, chromosomal rearrangements, duplications/deletions of chromosome segments, and shifts in ploidy have been observed and have contributed to adaptation (Dunham et al., 2002Dunham M.J. Badrane H. Ferea T. Adams J. Brown P.O. Rosenzweig F. Botstein D. Characteristic genome rearrangements in experimental evolution of Saccharomyces cerevisiae.Proc. Natl. Acad. Sci. USA. 2002; 99: 16144-16149Crossref PubMed Scopus (417) Google Scholar, Gerstein et al., 2006Gerstein A.C. Chun H.J. Grant A. Otto S.P. Genomic convergence toward diploidy in Saccharomyces cerevisiae.PLoS Genetics. 2006; 2: e145https://doi.org/10.1371/journal.pgen.0020145Crossref PubMed Scopus (142) Google Scholar, Riehle et al., 2001Riehle M.M. Bennett A.F. Long A.D. Genetic architecture of thermal adaptation in Escherichia coli.Proc. Natl. Acad. Sci. USA. 2001; 98: 525-530Crossref PubMed Scopus (192) Google Scholar). Changes in genome structure typically have immediate effects on the phenotype and fitness of an individual. Beyond these immediate effects, changes in genome structure might allow evolutionary transitions that were previously impossible. For example, by introducing an additional complement of chromosomes, polyploidization might release gene duplicates from the constraints of having to perform all of the functions of a gene (pleiotropy), providing extra “degrees of freedom” upon which selection can act to favor new functions. Polyploidization can also stimulate further structural changes in the genome, providing polyploid lineages with genomic variation not available to diploid organisms. Indeed, it has been proposed that tetraploidy may be an intermediate stage in some cancers, facilitating a cascade of structural changes that disrupt normal controls to cell growth (Storchova and Pellman, 2004Storchova Z. Pellman D. From polyploidy to aneuploidy, genome instability and cancer.Nat. Rev. Mol. Cell Biol. 2004; 5: 45-54Crossref PubMed Scopus (580) Google Scholar). Here, I discuss the evolutionary impact of polyploidization, beginning with the prevalence of polyploidy, and then I explore the longer-term consequences of polyploidy on the rate and nature of evolutionary transitions. Polyploidization is the increase in genome size caused by the inheritance of an additional set (or sets) of chromosomes (Figure 1). The duplicated sets of chromosomes may originate from the same or a closely related individual (“autopolyploid”) or from the hybridization of two different species (“allopolyploidy”). When polyploidization involves duplicated sets of chromosomes that share homology but are sufficiently distinct due to their separate origins, these pairs of chromosomes are referred to as homeologs (see Figure 1). Polyploidy is especially prevalent among hybrid taxa, an association thought to be driven by problems with meiotic pairing in diploid hybrids, which are solved if each homeologous chromosome has its own pairing partner. Additionally, diploid hybrids form unreduced gametes (which have the same number of chromosomes as somatic cells) at unusually high rates (Ramsey and Schemske, 2002Ramsey J. Schemske D.W. Neopolyploidy in flowering plants.Annu. Rev. Ecol. Syst. 2002; 33: 589-639Crossref Scopus (692) Google Scholar), increasing the rate of formation of polyploids from hybrid lineages. By combining traits from two parental species and ensuring fair segregation of these traits, allopolyploids potentially benefit from “hybrid vigor” (where hybrids have characteristics that make them superior to both parental species) and an altered ecological niche without the problems associated with segregation and breakdown at the F2 generation that occurs among diploid hybrids. Chromosomes that have previously diverged and been brought together by hybridization typically segregate as bivalents (Figure 1C; Ramsey and Schemske, 2002Ramsey J. Schemske D.W. Neopolyploidy in flowering plants.Annu. Rev. Ecol. Syst. 2002; 33: 589-639Crossref Scopus (692) Google Scholar). Genomic analyses have begun to unravel the genes responsible for bivalent pairing, ensuring that homologs rather than homeologs pair during meiosis (e.g., at the Ph1 locus in polyploid wheat; Griffiths et al., 2006Griffiths S. Sharp R. Foote T.N. Bertin I. Wanous M. Reader S. Colas I. Moore G. Molecular characterization of Ph1 as a major chromosome pairing locus in polyploid wheat.Nature. 2006; 439: 749-752Crossref PubMed Scopus (353) Google Scholar). In contrast, autopolyploids more often exhibit multivalent pairing than allopolyploids (Figure 1B), ∼3.5 times more so in plants (Ramsey and Schemske, 2002Ramsey J. Schemske D.W. Neopolyploidy in flowering plants.Annu. Rev. Ecol. Syst. 2002; 33: 589-639Crossref Scopus (692) Google Scholar). In most cases, descendants of a polyploidization event in the distant past (“paleopolyploids”) exhibit bivalent pairing of chromosomes and disomic inheritance (as if diploid). This observation has traditionally led to the conclusion that autopolyploids are ephemeral whereas allopolyploids give rise to the majority of long-lasting lineages (Grant, 1971Grant V. Plant Speciation. Columbia University Press, New York1971Google Scholar, Stebbins, 1950Stebbins G.L. Variation and Evolution in Plants. Columbia University Press, New York1950Crossref Google Scholar). Bivalent pairing, however, can occur and is even more prevalent (63.7%) than multivalent pairing (28.8%) among newly formed autotetraploid plants (Ramsey and Schemske, 2002Ramsey J. Schemske D.W. Neopolyploidy in flowering plants.Annu. Rev. Ecol. Syst. 2002; 33: 589-639Crossref Scopus (692) Google Scholar). Although bivalent pairing is initially nonpreferential in autopolyploids, leading to tetrasomic inheritance (Figure 1), segregating polymorphisms, especially rearrangements and indels (insertions and deletions), may increase pairing fidelity over time ultimately yielding disomic inheritance. Increased pairing fidelity may also stem from genetic changes, such as at the Ph1 locus in wheat (Griffiths et al., 2006Griffiths S. Sharp R. Foote T.N. Bertin I. Wanous M. Reader S. Colas I. Moore G. Molecular characterization of Ph1 as a major chromosome pairing locus in polyploid wheat.Nature. 2006; 439: 749-752Crossref PubMed Scopus (353) Google Scholar). Thus, one cannot assume that a paleopolyploid is necessarily allopolyploid solely because it exhibits disomic inheritance; additional evidence is needed, such as phylogenetic evidence that different genomic regions are more closely related to different parental species. Autopolyploids are also less frequently recognized as distinct species than allopolyploids, even when they are reproductively isolated and morphologically differentiated from their diploid parents (Soltis et al., 2007Soltis D.E. Soltis P.S. Schemske D.W. Hancock J.F. Thompson J.N. Husband B.C. Judd W.S. Autopolyploidy in angiosperms: have we grossly underestimated the number of species?.Taxon. 2007; 56: 13-30Google Scholar). Thus, recent papers have argued that autopolyploidy may contribute more to evolution and species diversification than traditionally thought (Soltis et al., 2007Soltis D.E. Soltis P.S. Schemske D.W. Hancock J.F. Thompson J.N. Husband B.C. Judd W.S. Autopolyploidy in angiosperms: have we grossly underestimated the number of species?.Taxon. 2007; 56: 13-30Google Scholar). Mutations affecting ploidy occur relatively frequently in both plants and animals. Plants produce unreduced gametes, a common route to polyploidization, at an average rate of ∼0.5% per gamete (Ramsey and Schemske, 1998Ramsey J. Schemske D.W. Pathways, mechanisms, and rates of polyploid formation in flowering plants.Annu. Rev. Ecol. Syst. 1998; 29: 467-501Crossref Scopus (1329) Google Scholar). Both unreduced gametes and polyspermy contribute to the production of polyploid animals. Among chicken embryos, 0.9% are triploid or tetraploid (Bloom, 1972Bloom S.E. Chromosome abnormalities in chicken (Gallus domesticus) embryos: types, frequencies and phenotypic effects.Chromosoma. 1972; 37: 309-326Crossref PubMed Scopus (45) Google Scholar), and among spontaneous human abortions, 5.3% are triploid or tetraploid (Creasy et al., 1976Creasy M.R. Crolla J.A. Alberman E.D. A cytogenetic study of human spontaneous abortions using banding techniques.Hum. Genet. 1976; 31: 177-196Crossref PubMed Scopus (179) Google Scholar). Placing these rates in context, gene duplication events are much rarer, with a roughly 10−8 chance of occurring per gene copy per generation (Lynch, 2007Lynch M. The Origins of Genome Architecture. Sinauer Associates, Inc., Sunderland, MA2007Google Scholar). Conversely, aneuploidy, the gain or loss of a single copy of a chromosome, can be much more frequent; for example, aneuploids were four times more common than polyploids among the aborted fetuses examined by Creasy et al., 1976Creasy M.R. Crolla J.A. Alberman E.D. A cytogenetic study of human spontaneous abortions using banding techniques.Hum. Genet. 1976; 31: 177-196Crossref PubMed Scopus (179) Google Scholar. The evolutionary contribution of structural alterations to the genome depends on their ability to persist. Although the probability of a gene duplication is low, the half-life of gene duplicates is very long (over a million generations; Lynch, 2007Lynch M. The Origins of Genome Architecture. Sinauer Associates, Inc., Sunderland, MA2007Google Scholar). Conversely, aneuploids often have low fitness and, in mammals, rarely survive to reproduce. Indeed, it is exceedingly rare for a homologous chromosome pair to be lost or duplicated among all members of a population. In mammals and birds, ploidy changes are also typically fatal, with polyploids dying early during development. Interestingly, polyploidy is lethal regardless of the sexual phenotype of the embryo (e.g., triploid XXX humans, which develop as females, die, as do triploid ZZZ chickens, which develop as males), and polyploidy causes much more severe defects than trisomy involving the sex chromosomes (diploids with an extra X or Y chromosome). Thus, in mammals and birds, evidence suggests that a general disruption of development—not problems restricted to sex determination—is the root cause of the failure of polyploids to persist. Specific developmental problems in human polyploid aborted fetuses have been attributed to abnormal imprinting and placental development (see references in Otto and Whitton, 2000Otto S.P. Whitton J. Polyploid incidence and evolution.Annu. Rev. Genet. 2000; 34: 401-437Crossref PubMed Scopus (1525) Google Scholar). In many taxa besides mammals and birds, however, polyploidy is surprisingly well tolerated. Polyploid lineages often persist in plants, which is immediately apparent from the excess of even over odd chromosome numbers (Figure 2). Because doubling a number always generates an even number, this excess of even chromosome numbers is a predictable consequence of polyploidization. By contrast, the fusion or fission of chromosomes simply switches whether a cell has an even or odd number of chromosomes. The excess of even chromosome numbers can be used to infer how often changes in chromosome number are due to polyploidization (42% of the time in ferns, 32% in monocots, and 18% in dicots; Otto and Whitton, 2000Otto S.P. Whitton J. Polyploid incidence and evolution.Annu. Rev. Genet. 2000; 34: 401-437Crossref PubMed Scopus (1525) Google Scholar). In various plant genera, the rate at which polyploids arise and persist is on the order of 0.01 per lineage per million years, roughly 1/10th the rate of speciation (Meyers and Levin, 2006Meyers L.A. Levin D.A. On the abundance of polyploids in flowering plants.Evolution Int. J. Org. Evolution. 2006; 60: 1198-1206PubMed Google Scholar). With such a high rate of polyploidization per speciation, we would expect a large fraction of plant species to have undergone polyploidization at some point in their evolutionary past. Previous studies had suggested that polyploidy occurred sometime in the past of 57% (Grant, 1963Grant V. The Origin of Adaptations. Columbia University Press, New York1963Google Scholar) to 70% (Goldblatt, 1980Goldblatt P. Polyploidy in angiosperms: monocotyledons.in: Lewis W.H. Polyploidy: Biological Relevance. Plenum Press, New York1980: 219-239Crossref Google Scholar, Masterson, 1994Masterson J. Stomatal size in fossil plants: Evidence for polyploidy in majority of angiosperms.Science. 1994; 264: 421-423Crossref PubMed Scopus (865) Google Scholar) of flowering plants, based solely on chromosome numbers among extant species. Recent genomic analyses indicate that an early polyploidization event may predate the radiation of flowering plants (Bowers et al., 2003Bowers J.E. Chapman B.A. Rong J.K. Paterson A.H. Unravelling angiosperm genome evolution by phylogenetic analysis of chromosomal duplication events.Nature. 2003; 422: 433-438Crossref PubMed Scopus (1105) Google Scholar), suggesting that 100% of angiosperms are paleopolyploid. Unfortunately, evidence for such ancient polyploidization events is almost always tentative because of the loss of sequence homology and synteny over evolutionary time. Although polyploidization is less prevalent in animals, nearly 200 independent examples of polyploidy have been reported in insects and vertebrates (Table 1), with many more cases known among other invertebrate groups (Gregory and Mable, 2005Gregory T.R. Mable B.K. Polyploidy in animals.in: Gregory T.R. The Evolution of the Genome. Elsevier, San Diego2005: 427-517Crossref Scopus (115) Google Scholar).Table 1Polyploidization in Insects and VertebratesReproductionInsectsFishAmphibiaReptilesBirdsMammalsTotalParthenogenesis89931500106Sexual22326101aThe only reported case of polyploidization in mammals involves the related red and golden viscacha rats (Gallardo et al., 2004).54?018100019A summary of data on the number of polyploidzation events. (Data derived from Otto and Whitton, 2000Otto S.P. Whitton J. Polyploid incidence and evolution.Annu. Rev. Genet. 2000; 34: 401-437Crossref PubMed Scopus (1525) Google Scholar, online Table 1; and Gregory and Mable, 2005Gregory T.R. Mable B.K. Polyploidy in animals.in: Gregory T.R. The Evolution of the Genome. Elsevier, San Diego2005: 427-517Crossref Scopus (115) Google Scholar, Table 1). Mode of reproduction of the polyploid is specified, where known.a The only reported case of polyploidization in mammals involves the related red and golden viscacha rats (Gallardo et al., 2004Gallardo M.H. Kausel G. Jiménez A. Bacquet C. González G. Figueroa J. Köhler N. Ojeda R. Whole-genome duplications in South American desert rodents (Octodontidae).Biol. J. Linn. Soc. 2004; 82: 443-451Crossref Scopus (46) Google Scholar). Open table in a new tab A summary of data on the number of polyploidzation events. (Data derived from Otto and Whitton, 2000Otto S.P. Whitton J. Polyploid incidence and evolution.Annu. Rev. Genet. 2000; 34: 401-437Crossref PubMed Scopus (1525) Google Scholar, online Table 1; and Gregory and Mable, 2005Gregory T.R. Mable B.K. Polyploidy in animals.in: Gregory T.R. The Evolution of the Genome. Elsevier, San Diego2005: 427-517Crossref Scopus (115) Google Scholar, Table 1). Mode of reproduction of the polyploid is specified, where known. If polyploidy were an evolutionary dead end, we would expect polyploid taxa to be near the tips of the tree of life and to be relatively species poor. Instead, several polyploidization events are ancient, and several of these events gave rise to species-rich groups. For example, multiple independent polyploidization events occurred early in the evolution of plants (Freeling and Thomas, 2006Freeling M. Thomas B.C. Gene-balanced duplications, like tetraploidy, provide predictable drive to increase morphological complexity.Genome Res. 2006; 16: 805-814Crossref PubMed Scopus (342) Google Scholar), fish (the diverse group of ray-finned fishes, Catostomidae, Salmonidae, and different groups of Cyprinidae), and amphibia (Syrinidae), with evidence for two additional rounds of genome duplication at the base of the vertebrate tree of life (Dehal and Boore, 2005Dehal P. Boore J.L. Two rounds of whole genome duplication in the ancestral vertebrate.PLoS Biol. 2005; 3: e314https://doi.org/10.1371/journal.pbio.0030314Crossref PubMed Scopus (960) Google Scholar), as proposed by Ohno, 1970Ohno S. Evolution by Gene Duplication. George Allen and Unwin, London1970Crossref Google Scholar; see Otto and Whitton, 2000Otto S.P. Whitton J. Polyploid incidence and evolution.Annu. Rev. Genet. 2000; 34: 401-437Crossref PubMed Scopus (1525) Google Scholar for additional examples). This data is sufficient to conclude that polyploidy is not an evolutionary dead end, but it does not prove that polyploidization contributes to evolutionary success. Whether polyploidization increases the longevity and species richness of a group relative to taxa that have not undergone polyploidization is a question that has yet to be answered. Why is it that the duplication of a whole genome often gives rise to lineages that persist over evolutionary time whereas the duplication of a single chromosome virtually never does? One plausible explanation is that polyploidization preserves the balance of gene products (Guo et al., 1996Guo M. Davis D. Birchler J.A. Dosage effects on gene expression in a maize ploidy series.Genetics. 1996; 142: 1349-1355PubMed Google Scholar, Papp et al., 2003Papp B. Pal C. Hurst L.D. Dosage sensitivity and the evolution of gene families in yeast.Nature. 2003; 424: 194-197Crossref PubMed Scopus (607) Google Scholar). This “balance hypothesis” is an old idea that traces back to the pre-genomics era; for instance, Haldane (1932) pointed out that morphological changes were more marked in trisomic than in triploid plants, arguing that “In the latter case the number of genes of all sorts is increased equally, in the former the balance is upset” (p. 29 in Haldane, 1990Haldane J.B.S. The Causes of Evolution. Princeton University Press, Princeton, NJ1990Google Scholar). Consistent with the balance hypothesis, genes duplicated by polyploidization persist longer, on average, than genes duplicated individually (Lynch, 2007Lynch M. The Origins of Genome Architecture. Sinauer Associates, Inc., Sunderland, MA2007Google Scholar). Another explanation is that organisms have evolved mechanisms to cope with changes in ploidy because of the natural variation in genome copy number associated with mitotic and meiotic cell cycles (as DNA replicates and cells divide). According to this “evolved-robustness hypothesis,” organisms with a regular alternation of generations, with mitoses in both haploid and diploid phases, are predicted to be especially tolerant of shifts in ploidy. In addition, somatic variation in ploidy (“endopolyploidy”) is a normal part of development in many animals as well as plants (Gregory, 2005Gregory T.R. The evolution of the genome. Elsevier Academic, Burlington, MA2005Google Scholar). Most famously, the chromosomes of the salivary gland are highly replicated in flies, leading to visible polytene chromosomes. In mammals, multinucleate cells are found in hepatocytes and osteoclasts, whereas megakaryocytes, trophoblasts, and hepatocytes display endopolyploidy (that is, have a nucleus with multiple copies of the normal complement of DNA). In summary, the existence of regular mitotic cell cycles, an alternation of generations, as well as endopolyploidy ensures that organisms have experienced, and survived, an evolutionary history at different ploidy levels. The balance hypothesis and the evolved-robustness hypothesis are not opposing explanations. Instead, they may serve as proximate and ultimate explanations for the same phenomenon—present-day organisms function better with a balanced set of chromosomes because their evolutionary past involved changes in ploidy that preserved the balance, but not the absolute number, of chromosomes. At a phenotypic level, the effects of polyploidization are often mild and idiosyncratic. Cell volume generally rises with increasing genome size (Cavalier-Smith, 1978Cavalier-Smith T. Nuclear volume control by nucleoskeletal DNA, selection for cell volume and cell growth rate, and the solution of the DNA C-value paradox.J. Cell Sci. 1978; 34: 247-278PubMed Google Scholar, Gregory, 2001Gregory T.R. Coincidence, coevolution, or causation? DNA content, cell size, and the C-value enigma.Biol. Rev. Camb. Philos. Soc. 2001; 76: 65-101Crossref PubMed Scopus (386) Google Scholar), although the exact relationship between ploidy and cell volume varies among environments and taxa. In yeast, for example, the volume of diploid cells in rich media is 2.4 times that of haploid cells at 30°C but only 1.1 times at 37°C (Mable, 2001Mable B.K. Ploidy evolution in the yeast Saccharomyces cerevisiae: A test of the nutrient limitation hypothesis.J. Evol. Biol. 2001; 14: 157-170Crossref Scopus (35) Google Scholar). Larger cells tend to have smaller surface area to volume ratios, a phenomenon thought to lower the growth rate of polyploid cells. Whether or not cell geometry affects growth rate depends on the environment (Adams and Hansche, 1974Adams J. Hansche P.E. Population studies in microorganisms I. Evolution of diploidy in Saccharomyces cerevisiae.Genetics. 1974; 76: 327-338PubMed Google Scholar, Mable, 2001Mable B.K. Ploidy evolution in the yeast Saccharomyces cerevisiae: A test of the nutrient limitation hypothesis.J. Evol. Biol. 2001; 14: 157-170Crossref Scopus (35) Google Scholar), as expected given that transport across the membrane limits growth only under certain circumstances (Weiss et al., 1975Weiss R.L. Kukora J.R. Adams J. The relationship between enzyme activity, cell geometry, and fitness in Saccharomyces cerevisiae.Proc. Natl. Acad. Sci. USA. 1975; 72: 794-798Crossref PubMed Scopus (59) Google Scholar). Perhaps as a consequence of a slower metabolism, polyploids tend to exhibit slower development (Levin, 1983Levin D.A. Polyploidy and novelty in flowering plants.Am. Nat. 1983; 122: 1-25Crossref Scopus (639) Google Scholar), but this pattern is not universally true and can be reversed (e.g., in groups where polyploid eggs or seeds are larger). Interestingly, not all features of a cell scale with ploidy level, which can have important side consequences. For example, Storchova et al., 2006Storchova Z. Breneman A. Cande J. Dunn J. Burbank K. O'Toole E. Pellman D. Genome-wide genetic analysis of polyploidy in yeast.Nature. 2006; 443: 541-547Crossref PubMed Scopus (242) Google Scholar showed that kinetochore size and length of the pre-anaphase spindle do not scale with ploidy level, whereas the spindle pole body does. These authors argue that changes in ploidy thus upset geometric relationships among key components of the machinery used to segregate chromosomes during meiosis, potentially explaining the higher rate of chromosome nondisjunction in tetraploids. To prevent the detrimental effects of genomic instability, animals may have evolved mechanisms limiting the proliferation of tetraploid cells (see Essay by N.J. Ganem and D. Pellman on page 437 of this issue), which would present another barrier to the establishment of polyploid lineages. Although cell size typically is larger in polyploids, adult size may or may not be altered; as a rough generalization, polyploidization is more likely to increase adult body size in plants and invertebrates than in vertebrates (Gregory and Mable, 2005Gregory T.R. Mable B.K. Polyploidy in animals.in: Gregory T.R. The Evolution of the Genome. Elsevier, San Diego2005: 427-517Crossref Scopus (115) Google Scholar, Otto and Whitton, 2000Otto S.P. Whitton J. Polyploid incidence and evolution.Annu. Rev. Genet. 2000; 34: 401-437Crossref PubMed Scopus (1525) Google Scholar). The poor correlation between cell size and organismal size was even remarked upon by Albert Einstein, who wrote “Most peculiar for me is the fact that in spite of the enlarged single cell the size of the animal is not correspondingly increased” (Fankhauser, 1972Fankhauser G. Memories of great embryologists. Reminiscences of F. Baltzer, H. Spemann, F. R. Lillie, R. G. Harrison, and E. G. Conklin.Am. Sci. 1972; 60: 46-55PubMed Google Scholar). The key to accurately predicting the effects of ploidy on body size must come from developmental biology. In cases where morphogen gradients guide development, ploidy need not affect adult body size (Day and Lawrence, 2000Day S.J. Lawrence P.A. Measuring dimensions: the regulation of size and shape.Development. 2000; 127: 2977-2987PubMed Google Scholar) because ploidy need not alter the overall density of cellular material, only how it is packaged (i.e., into cells that are twice as large and carry twice as much DNA). By contrast, where growth is determined by cell-cell interactions or where there is a fixed number of cells in the adult, ploidy, by altering cell size, should directly influence adult size (Gregory et al., 2000Gregory T.R. Hebert P.D. Kolasa J. Evolutionary implications of the relationship between genome size and body size in flatworms and copepods.Heredity. 2000; 84: 201-208Crossref PubMed Scopus (96) Google Scholar). A surprising feature of many newly formed polyploids is that their genomes are unstable and undergo rapid repatterning (Wendel, 2000Wendel J.F. Genome evolution in polyploids.Plant Mol. Biol. 2000; 42: 225-249Crossref PubMed Scopus (1292) Google Scholar). For example, Song et al., 1995Song K. Lu P. Tang K. Osborn C.T. Rapid genome change in synthetic polyploids of Brassica and its implications for polyploid evolution.Proc. Natl. Acad. Sci. USA. 1995; 92: 7719-7723Crossref PubMed Scopus (800) Google Scholar observed extensive genomic rearrangements and fragment loss within five generations in newly created polyploid Brassica hybrids, and more recent studies have documented genomic changes soon after formation of wheat and Arabidopsis allopolyploids (but not in cotton or Spartina; Chen and Ni, 2006Chen Z.J. Ni Z. Mechanisms of genomic rearrangements and gene expression changes in plant polyploids.Bioessays. 2006; 28: 240-252Crossref PubMed Scopus (315) Google Scholar). In most examples studied to date, rapid genomic repatterning has been observed in allopolyploids, and there are many reasons to expect that hybridization may be causally responsible. Transposable elements that are repressed within each parent lineage but activated in hybrids can facilitate the movement of genes and promote unequal crossing over. For example, Josefsson et al., 2006Josefsson C. Dilkes B. Comai L. Parent-dependent loss of gene silencing during interspecies hybridization.Curr. Biol. 2006; 16: 1322-1328Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar found that maternally derived siRNAs are not sufficient to repress retrotransposons in the paternal genome of Arabidopsis thaliana × A. arenosa hybrids. Divergence of centromeres and centromeric histones can lead to segregation distortion and nondisjunction in hybrids (Malik and Bayes, 2006Malik H.S. Bayes J.J. Genetic conflicts during meiosis and the evolutionary origins of centromere complexity.Biochem. Soc. Trans. 2006; 34: 569-573Crossref PubMed Scopus (35) Google Scholar). In addition, nonhomologous recombination and nonreciprocal exchanges are particularly likely among homeologous chromosomes that bear structural rearrangements. Nevertheless, genomic repatterning in polyploids is not entirely driven by hybridization. In autotetraploids of both Candida albicans (Bennett et al., 2003Bennett R.J. Uhl M.A. Miller M.G. Johnson A.D. Identification and characterization of a Candida albicans mating pheromone.Mol. Cell. Biol. 2003; 23: 8189-8201Crossref PubMed Scopus (129) Google Scholar) and S. cerevisiae (Gerstein et al., 2006Gerstein A.C. Chun H.J. Grant A. Otto S.P. Genomic convergence toward diploidy in Saccharomyces cerevisiae.PLoS Genetics. 2006; 2: e145https://doi.org/10.1371/journal.pgen.0020145Crossref PubMed Scopus (142) Google Scholar), reduction in genome size through chromosome loss has been observed, largely restoring the diploid complement. The exact mechanism by which this reduction occurs is unknown, but similar reductive divisions (termed “neosis”) have been observed in human cell lines that have undergone endopolyploidization in response to carcinogens (Rajaraman et al., 2005Rajaraman R. Rajaraman M.M. Rajaraman S.R. Guernsey D.L. Neosis–a paradigm of self-renewal in cancer.Cell Biol. Int. 2005; 29: 1084-1097Crossref PubMed Scopus (71) Google Scholar). By altering the ge