Genome duplication and the origin
of angiosperms
Stefanie De Bodt*
, Steven Maere*
and Yves Van de Peer
Department of Plant Systems Biology, Flanders Interuniversity Institute for Biotechnology (VIB), Ghent University,
Technologiepark 927, B-9052 Ghent, Belgium
Despite intensive research, little is known about the
origin of the angiosperms and their rise to ecological
dominance during the Early Cretaceous. Based on
whole-genome analyses of Arabidopsis thaliana, there
is compelling evidence that angiosperms underwent
two whole-genome duplication events early during their
evolutionary history. Recent studies have shown that
these events were crucial for the creation of many
important developmental and regulatory genes found in
extant angiosperm genomes. Here, we argue that these
ancient polyploidy events might have also had an
important role in the origin and diversification of the
angiosperms.
The rise of the angiosperms
`An abominable mystery' is how Charles Darwin referred
to the rise and early diversification of the angiosperms
(flowering plants), one of the greatest terrestrial radi-
ations that has resulted in O250 000 species (Figure 1).
Since then, plant biologists have been studying the
evolution of the angiosperms to understand their origin
and their rise to ecological dominance. However, in spite of
much research and analyses of different sources of data
(e.g. fossil record and phylogenetic analyses using
molecular and morphological characters), the origin of
the angiosperms remains unclear.
Angiosperms appear rather suddenly in the fossil
record during the Jurassic [208­145 million years ago
(Mya)], with no obvious ancestors for a period of 80­90
million years before their appearance. Nevertheless, the
existence during the Jurassic of all known sister taxa to
the angiosperms implies that the angiosperm lineage
must have been established by that time [1]. However, this
ancestral lineage, coined `angiophytes', is unlikely to be
equivalent to angiosperms as known from the Cretaceous
(145 Mya) through to recent forms because it might have
lacked many of the characteristic angiosperm features [2].
It is presumed that angiophytes went through a period of
little diversification during the Late Triassic (220 Mya)
and Jurassic, either because the diversity-enhancing
features, such as flowers, of the crown-group angiosperms
had not yet evolved in stem angiophytes or because the
diversity among angiophytes was inhibited during the
Jurassic by environmental conditions or biotic inter-
actions [2].
Evidence from the history of other major clades of land
plants, such as seed ferns, suggests that the characteristic
features of angiosperms were acquired sequentially
through time [3]. The recent transitional­combinational
theory of the angiosperm origin suggests an evolution
from Jurassic seed ferns through three fundamental
transitions: (i) evolution of the carpel; (ii) emergence of
double fertilization; and (iii) origin of the flower [4]. The
extant (or modern) angiosperms did not appear until the
Early Cretaceous (145­125 Mya), when the final combin-
ation of these three angiosperm features occurred, as
supported by evidence from micro- and macrofossils [4].
This combination of features might have provided the
opportunity for explosive evolutionary diversification,
especially in response to selection from insect pollinators
and herbivores, accompanied by modifications in compati-
bility and breeding systems.
The fossil record provides excellent evidence for this
rapid diversificationin floralformduringtheearliestphases
of recorded flowering plant history [5]. Only 10­12 million
years elapsed between the first fossil records (w130 Mya)
and clear documentation of all of the major lines offlowering
plants [1,6]. This diversification of angiosperms occurred
during a period (the Aptian, 125­112 Mya; Figure 1) when
their pollen and megafossils were rare components of
terrestrial floras and species diversity was low [1,6].
Angiosperm fossils show a dramatic increase in diversity
between the Albian (112­99.6 Mya) and the Cenomanian
(99.6­93.5 Mya) at a global scale [7­10] (Figure 1).
The angiosperm radiation yielded species with new
growth architectures and new ecological roles. Early
angiosperms had small flowers with a limited number of
parts that were probably pollinated by a variety of insect
taxa but specialized for none. Accordingly, Cenomanian
flowers do not yet provide strong evidence for specializ-
ation of pollination syndromes. However, by the Turonian
(93.5­89.3 Mya), flowering plants had a wide variety of
features that are, in extant species, closely associated with
several types of specialized insect pollination and with
high species diversity within angiosperm subclades. The
evolution of larger seed size in many angiosperm lineages
during the early Cenozoic (from 65 Mya) indicates that
animal-mediated dispersal and shade-tolerant life-history
strategies had become common among angiosperms by
this time [2].
Corresponding author: Van de Peer, Y. (Yves.VandePeer@psb.ugent.be).
* Contributed equally.
Available online 9 August 2005
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005
www.sciencedirect.com 0169-5347/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2005.07.008
In summary, fossils with affinities to diverse angio-
sperm lineages, including monocots, are all found in Early
Cretaceous floras [5,8,11] and w42 of the 94 extant orders
(44%) and w63 of the 439 extant families (14%) of
flowering plants occurred for the first time during the
Cretaceous [2]. However, the question remains why this
was such a decisive time in the evolution of plants. Here,
we discuss why whole-genome duplication events might
have had a key role in the origin of angiosperms and their
morphological and ecological diversification.
Of polyploids and paleopolyploids
One of the most striking features of angiosperms is that
many have experienced one or more episodes of polyploidy
(see Glossary) in their ancestry [12]. Apart from species
that are currently polyploid, such as many crops, others
are considered to have paleopolyploid genomes. In 1996,
when the sequencing of the flowering plant Arabidopsis
thaliana (Brassicaceae) genome began, this model plant,
with its small genome, was not expected to be an ancient
polyploid. However, five years after the release of its
genome sequence [13], there is compelling evidence that
the Arabidopsis genome, or rather that of its ancestors,
has been duplicated three times (events referred to here as
1R, 2R and 3R) during the past 250 million years [14,15].
Although duplicated genes and genomes can provide
the raw material for evolutionary diversification and the
functional divergence of duplicated genes might offer a
selective advantage to polyploids over a long time period, a
beneficial effect of these duplications is assumed shortly
after the duplication event. In other words, if a new
genome doubling is to survive long enough to exert its
long-term evolutionary effects, it must provide an immedi-
ate selective advantage that enables its establishment
(Box 1; reviewed in [16]).
The fate of duplicated genes
Ancient polyploidy events might have directly influenced
the increase in the number of plant species and plant
complexity observed since the Early Cretaceous (Box 1;
Figure 1). However, other factors, such as expansion and
functional diversification of specific gene families follow-
ing a polyploidy event, are likely to have been more
influential and could explain, at least in part, the origin
and fast diversification of angiosperm lineages. Several
authors have provided evidence that gene retention after
duplication is biased according not only to the function of
the genes, but also to the timing and mode of the
duplication events [17­19]. Blanc and Wolfe [18] studied
the relationship between gene function and duplicate loss
after the most recent polyploidy event (3R). Similarly,
Seoighe and Gehring [19] analyzed the survivability of
duplicates of various functions following 3R. Recently,
Maere et al. [17] developed an evolutionary model that
simulates the population dynamics of duplicate genes
belonging to different functional categories, based on the
KS distribution of the Arabidopsis paranome. They took
into account the three major genome-wide duplication
events (1R, 2R and 3R) and a continuous mode of small-
scale gene duplications (referred to as 0R). These studies
all concluded that genes involved in transcriptional
regulation and signal transduction have been preferen-
tially retained following genome duplications. Similarly,
developmental genes have been observed to be retained
following genome duplications [17,18], particularly
following the two oldest events (1R and 2R) [17] (Box 2).
However, few regulatory and developmental gene
duplicates appear to have survived small-scale duplication
events [17]. Their rapid loss can be explained by the fact
that transcription factors and genes involved in signal
transduction tend to show a high dosage effect in multi-
cellular eukaryotes [20]. The expression of a wide range of
genes regulated by these proteins show major pertur-
bations when only one regulatory component is dupli-
cated, rather than all components that govern a certain
pathway [21,22]. Furthermore, that transcription factors
and kinases are often active as protein complexes and
must be present in stoichiometric quantities for their
correct functioning is congruent with their high retention
rate following whole-genome instead of small-scale dupli-
cation events [22,23].
Regulatory and developmental genes are thought to
have been of primordial importance for the evolution of
morphological complexity in plants and animals [24­26].
Overall, the three polyploidy events in the ancestors of
Arabidopsis might have been responsible for O90% of the
transcription factors, signal transducers and develop-
mental genes created during the past w250 million
years [17]. Although this is only an approximate assess-
ment from a single study, it is tempting to suggest that if
only small-scale gene duplications had occurred in the
evolutionary past of Arabidopsis and of angiosperms in
Glossary
Allopolyploidy: an allo(tetra)polyploid originates by the fusion of the genomes
of two different, but closely related, species.
Autopolyploidy: a polyploid in which all the chromosomes come from the
same species; i.e., a tetraploid is formed by the doubling of its own genome.
Eudicots: a monophyletic clade that is strongly supported by molecular data
and by a single morphological synapomorphy, namely triaperturate pollen
(pollen with three grooves). This pollen type is distinct from the uniaperturate
pollen of basal dicots (other dicots that are not monophyletic with the eudicots)
and monocots. The eudicot clade contains most of the flowering plants,
outnumbering all other plant groups put together.
Hybrid vigour: heterosis, the phenomenon whereby hybrid individuals display
characteristics that exceed even the better of the two parents. Heterosis is
produced by increased heterozygosity.
KS: the number of synonymous substitutions per synonymous site. Because
synonymous substitutions do not result in amino-acid replacements, the rate of
fixation of these substitutions is expected to be relatively constant in different
protein-encoding genes, and to reflect the overall mutation rate. The time of
duplication (T) between two sequences can then be calculated as TZKS/2l,
where KS is the fraction of synonymous substitutions per synonymous site, and
l is the mean rate of synonymous substitution.
Monocots: the number of cotyledons found in the embryo is the basis for
distinguishing the two classes of flowering plants, and is the source of the
names monocots and dicots. The cotyledons are the seed leaves produced by
the embryo and serve to absorb nutrients packaged in the seed until the
seedling is able to produce its first true leaves and begins photosynthesizing.
The monocots comprise approximately one-quarter of all flowering plant
species and include some of the largest and most familiar groups of plants,
such as lilies, orchids, agaves, palms and grasses. Some monocots, such as
corn, rice, wheat, and barley, are among our most important food crops. Sugar
cane, pineapples, dates, bananas, and many of our familiar tropical fruits also
come from monocots.
Paranome: collection of all duplicated genes in a genome.
Polyploid: a polyploid organism has more than two sets of chromosomes.
Stoichiometric quantities (of protein complex partners): quantities of protein
complex partners that reflect their relative abundance in the complex.
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005592
www.sciencedirect.com
general, the expansion of regulatory and developmental
genes would have been severely hampered.
However, according to Maere et al. [17], genes involved
in secondary metabolism or responses to biotic stimuli,
such as pathogen attack, tend to be preserved regardless
of the mode of duplication. Because plants are sessile
organisms, secondary metabolite pathways, as well as
genes governing responses to biotic stimuli, are crucial to
the development of survival strategies against herbivores,
insects, snails and plant pathogens [27]. Additionally, in
angiosperms, anthocyanins and other secondary metab-
olites give rise to colourful and scented flowers that attract
pollen- and nectar-collecting animals [27]. Thus, second-
ary metabolite diversification might have led to more
efficient seed dispersal (compared with wind pollination,
which is widespread in most seed plants) and might have
provided new possibilities for reproductive isolation and
the elevation of speciation rates [6]. Moreover, coevolu-
tionary interactions between early angiosperms and, in
particular, insect pollinators are thought to have been of
major importance for the angiosperm radiation [28,29]
(Box 3). The finding that genes involved in secondary
metabolism and responses to biotic stimuli are also
strongly retained following (continuously occurring)
Aptian
Late
Jurassic
Middle
Jurassic
Early
Jurassic
Albian
Cenomanian
Turonian
89.3
93.5
99.6
112
125
65 Mya
145 Mya
208 Mya
(Core)
EudicotsMonocots
Gymnosperms Angiosperms
Ferns
Mosses
3R
2R
1R
245 Mya
10 000 species
Basal
dicots
TRENDS in Ecology & Evolution
Paleozoic
Triassic
Jurassic
Cretaceous
Cenozoic
Figure 1. Simplified view of the rise and diversification of the mosses, ferns, gymnosperms and angiosperms, as inferred from fossil data [10,74]. `Basal dicots' refers to the
primitive dicotyledons, such as the magnoliids, whereas (core) eudicots refers to the monophyletic grouping of the other dicotyledonous plant families. The three whole-
genome duplication events (1R, 2R and 3R), for which evidence can be found in the Arabidopsis genome, are also indicated.
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005 593
www.sciencedirect.com
small-scale gene duplications might reflect the continuous
interaction between plants and animals, fungi or plant
pathogens imposing a constant need for adaptation. By
contrast, genes involved in responses to abiotic stress,
such as drought, cold and salinity, appear to have been
only moderately retained after small-scale gene dupli-
cation events [17], indicating that they might have been
required at more specific times in evolution, such as
during major environmental changes or adaptation to new
niches. Interestingly, 1R and 2R might have occurred
during a period of increased tectonic activity linked to
highly elevated atmospheric CO2 levels [29] (Boxes 2,3).
The hypotheses outlined here should be approached
with caution, as they are based on a few computational
studies on a single flowering plant, Arabidopsis thaliana
[17­19]. Similar analyses on other flowering plants are
needed to accept or reject these hypotheses.
Darwin's abominable mystery revisited
Here, we propose that the ancient polyploidy events that
occurred during early angiosperm evolution (Box 2) have
created much of the genetic material found in extant angio-
sperms. In particular, genes involved in specific processes,
such as development, transcriptional regulation and signal-
ling, appear to have been retained in this manner [17­19]. It
is therefore tempting to speculate that much of the
evolutionary success of polyploidy in angiosperms, which
entails a saltatory doubling of regulatory factors, is tied
directly to regulatory gene evolution, both at the protein
level and through the differential regulation of the
regulatory genes themselves [30].
Proliferation of transcription factor genes, followed by
the recruitment of these genes as upstream or down-
stream regulators in developmental pathways, is believed
to have contributed substantially to the evolution of
morphological diversity in animals and plants [24­26].
Furthermore, transcriptional regulators, activated by
signal transducers, can act as key switches in plant
development [24,31,32]. Subject to diverse selective
pressures, these extra genes might have subsequently
evolved novel functions (e.g. [33]) that resulted in major
changes in biological complexity [34,35]. The first genome
duplication event appears to have occurred during the
Late Jurassic or Early Cretaceous, a period during which
the angiosperms originated and rose to ecological dom-
inance [36­38] (Figure 1; Box 2). The high retention of
duplicated genes underlying transcriptional regulatory
networks and development after 1R suggests that this
event provided the additional genetic material to produce
the observed increase in plant complexity. For instance,
co-sexual flowers originated, and major developmental
changes in leaf morphology occurred [6,39]. The occur-
rence of 1R before the monocot­eudicot split appears to
correlate well with this decisive moment in plant evolution
[15] (Box 2).
According to the dating described in Box 2, 2 R occurred
w66­109 Mya, probably after the divergence of monocots
and eudicots [15], in a time period when flowering plants
developed a wide variety of specialized insect pollination
strategies and where species diversity within the angio-
sperm subclades increased rapidly [28]. The magnitude of
Box 1. Polyploidy, adaptation and speciation
Genome duplication events double the amount of genetic material
on which evolution can work and the functional divergence of
duplicated genes is considered important for biological evolution
and increases in biological complexity [33­35]. However, apart from
duplicated genes adopting novel functions, polyploidy events can
also contribute to evolution in other ways.
Polyploidy and hybrid vigour
Differences in phenotype between polyploids and their diploid
progenitors can be caused by increased variation in dosage-
regulated gene expression [45]. By having a different number of
alleles at a locus, polyploid plants can differ from their diploid
ancestors in overall gene expression levels. For genes having allele-
dosage effects, polyploidy increases the potential variation in
expression levels and can provide selective advantages [46]. For
instance, changes in enzyme levels can affect many aspects of plant
physiology, morphology and life history [47]. Because the combin-
ation of different genomes (or the increase in heterozygosity) can
lead to hybrid vigour, the newly formed polyploid can have a
selective advantage compared with closely related diploid organ-
isms. In accordance with Spring [48], Rieseberg et al. [49] suggest
that hybridization provides a mechanism for large and rapid
adaptive transitions, made possible by the genetic variation of
hundreds or thousands of genes in a single generation.
Polyploidy and sympatric speciation
Hybridization is a significant evolutionary force that creates
opportunities for adaptive evolution and speciation [50­52]. It is
estimated that 2­4% of the speciation events in flowering plants can
be attributed to ploidy changes that have potentially broad-scale
effects on gene regulation and developmental processes, effects that
can produce immediate shifts in morphology, breeding system and
ecological tolerances [44]. Increased cell volume is one of the
most common and universal phenotypic effects of polyploidization
(e.g. [53]). Changes in cell volume result in changes in surface:
volume ratios, which can alter the rate of metabolic processes.
Consequently, growth and developmental rates are often affected in
polyploid plants.
Whereas, in general, developmental rates are lower for polyploid
plants than for their diploid progenitors, larger seed sizes can have
an opposite effect on the rate of (early) development and might affect
the likelihood of establishing seedlings in resource-limited environ-
ments. They might also result in niche differentiation as a byproduct
of polyploidization [44]. Differences in vegetative traits, such as
ecological tolerance (e.g. drought), and susceptibility to arthropod
infestation and fungal diseases, have been documented between
polyploid and diploid plants [44,52]. In addition, several reproductive
traits differ significantly, such as the initiation and duration of
flowering time, fertility, self-compatibility, germination and the
relative sizes and spatial relations of floral organs. Again, polyploids
that persist are likely to be able to inhabit niches that are different
from those of their diploid progenitors, which can then result in
speciation [44].
Polyploidy and allopatric speciation
As well as its role in sympatric speciation, polyploidy can also have a
role in allopatric speciation. Based on isozyme studies, Werth and
Windham [54] developed a model in which the reciprocal silencing of
genes in geographically separated populations promotes speciation,
an idea that was revived in a model called `divergent resolution' [55]. In
this model, the loss or silencing of gene duplicates was postulated to
be more important for the evolution of species diversity than was the
acquisition of newfunctions by duplicated genes. Divergent resolution
occurs in allopatric populations when different copies of duplicated
genes are lost from different chromosomes, thereby creating genetic
barriers to reproduction [55,56].
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005594
www.sciencedirect.com
these events was phrased aptly by Crepet [28]: `Had
Darwin witnessed such a pattern, he might have been
even more astonished by the rapid ascension of the
flowering plants.' Would he have considered it a second
abominable mystery? 2R might have provided, at least in
eudicots, the raw material on which evolution could act to
produce the observed specialized floral phenotypes with
insects providing the necessary selective forces [6]. In
support of these theories, Zahn et al. [40] recently
provided evidence that, on two separate occasions, several
duplication events have occurred simultaneously in
certain subfamilies of MADS-box transcription factors,
known to be important in floral development. Both of these
duplication bursts, one supposedly close to the origin of
angiosperms, and the other in the ancestor of the core
eudicots, might be linked to genome duplication events
[40]. Given the role of MADS-box genes in flower
development, these gene duplications, possibly linked to
1R and 2R, might have had an important role in the
morphological invention of the flower and subsequent
diversification of floral forms [40].
Developmentally related genes were found to be less
strongly, although significantly, retained after the
youngest genome duplication event [17,18], dated before
the divergence of Arabidopsis and Brassica and after the
divergence of Brassicales and Malvales [15,41], possibly
close to the emergence of the crucifers [41]. Compared to
1R and 2R, gene retention following 3R is lower and less
biased towards functional class [17]. This does not imply
that the youngest genome duplication did not contribute to
plant evolution, but it might have done so in a different
and less dramatic way. For instance, the youngest genome
duplication in Arabidopsis has created many duplicates
[14,15] that could be divergently resolved (Box 1), and
such genes might have had a prominent role in the
radiation of the crucifer family, which comprises O3500
species [42].
Conclusion
It has been suggested that large-scale gene duplication
or whole-genome duplication events can be associated
with important evolutionary transitions, major leaps
in development, and/or adaptive radiations of species
[34,35,43,44]. So far, evidence linking major duplication
events with evolution or biological innovation has been
scarce. However, because Arabidopsis has undergone
Box 2. Inferring the age of genome duplications
The timing of 1R, 2R, and 3R in the evolutionary past of Arabidopsis
thaliana is still debated. Based on phylogenetic analysis, Bowers et al.
[15] concluded that 3R occurred 14.5­86 Mya, after the divergence of
Brassicales and Malvales, but before the divergence of Arabidopsis
and Brassica. Blanc et al. [41] narrowed this to 24­40 Mya. However,
the age of the older genome duplications is less clear. Arguably, 2R
occurred after the divergence of monocots and eudicots, whereas 1R
appears to predate the monocot­eudicot divergence but could not be
positioned relative to the angiosperm­gymnosperm divergence owing
to the sparseness of appropriate sequence data [15,37]. Based on
simulation of the KS distribution of the Arabidopsis paranome, Maere
et al. [17] recently proposed KS values for 3R (KSZ0.7­0.8), 2R (KSZ2.0­
2.1) and 1R (KSZ3.1­3.2).
One way of relating these KS estimates to time is to use molecular
clocks, or synonymous substitution rate estimates, such as that of
Koch et al. [57] (1.5!10K8
ss siteK1
yearK1
) or Lynch and Conery [58]
(6.1!10K9
ss siteK1
yearK1
) (grey lines in Figure I). However, there are
theoretical and empirical concerns about the accuracy of molecular
clocks [59­62], exemplified by the large difference in the clock rate
estimates of Koch et al. [57] and Lynch and Conery [58]. Some of the
major issues are rate heterogeneity in and between lineages caused by
evolutionary factors (e.g. generation time), difficulties in interpreting
the fossil data used to calibrate the clock, and rate variation among
genes, even at synonymous sites. Because rates of nuclear gene
evolution vary widely within an organism [63,64], sampling only a few
genes in molecular clock calibrations can be risky.
To overcome at least some of these limitations, we could use the KS
estimate for 3R (KSw0.75) to calibrate the clock. By avoiding inter-
lineage comparison, and by the many duplicates involved, the KS
estimates for large-scale duplications should be more representative of
the average rate of synonymous substitution in Arabidopsis and its
ancestors. Because 3R is the most recent genome duplication event, it
is the most obvious choice. There is still a large population of retained
duplicates from 3R compared with 1R and 2R, and 3R can be more
accurately positioned phylogenetically. Unfortunately, no accurate
fossil calibration point for 3R is currently available and the 3R age
intervals proposed by Blanc et al. [41] and Bowers et al. [15] are
themselves dependent on a molecular clock [65]. A more-accurate
dating of the 3R event would require a better phylogenetic positioning
of the 3R event on the tree of life (i.e. a better sampling of the
Brassicales) and an adequate fossil calibration point. Here, we used the
Blanc et al. [41] interval (3R occurred 24­40 Mya) as a best guess and
extrapolated the resulting relationship to obtain estimated age
intervals for 1R and 2R. The 1R interval (101­168 Mya) overlaps with
the occurrence of the first bona fide angiosperm fossils, approximately
130 Mya [4], whereas the 2R interval (66­109 Mya) is centred around
the Turonian (89.3­93.5 Mya). Interestingly, the fossil record shows a
dramatic modification of angiosperm flowers at about this time [28,66].
Additionally, the suggested time interval for 2R roughly corresponds
to the time period during which the core eudicots emerged and
radiated [9,67].
Lynch
and
Conery
3R 2R 1R
Age(Myr)
0 1 2 3 4 5
0
100
200
300
400
Bowers et al.
Blanc et al.
Koch et al.
Angiosperm origin
Angiosperm radiation
TRENDS in Ecology & Evolution
KS
Figure I. Estimated ages for 1R, 2R and 3R, based on KS data for Arabidopsis
thaliana [17]. The grey lines depict the relationship between time and KS, based
on the molecular clock estimates of Koch et al. [57] and Lynch and Conery [58].
The beige and green areas represent the KS versus time window extrapolated
from the 3R age interval proposed by Bowers et al. [15] and Blanc et al. [41]
respectively, using KSZ0.75 as the average KS value for 3R duplicates [17].
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005 595
www.sciencedirect.com
several well-documented rounds of genome duplication, it
is an ideal model system with which to study the effect of
such events. As has been shown in recent studies, the
increase in the number of genes that are important in
development, transcriptional regulation and signalling
has been mainly the result of large-scale gene dupli-
cations; these genes might not have been retained in
small-scale duplication events [17]. Given that such genes
are considered important for introducing phenotypic
variation and increase in biological complexity, linking
ancient polyploidy events with decisive moments in
evolution becomes less speculative and the origin and
evolution of angiosperms perhaps less of a mystery.
Acknowledgements
S.D.B. thanks the Institute for the Promotion of Innovation by Science
and Technology in Flanders (IWT) for a predoctoral fellowship. S.M. is a
Research Fellow of the Fund for Scientific Research (FWO). The authors
also acknowledge the constructive comments of three anonymous
referees.
References
1 Doyle, J.A. and Donoghue, M.J. (1993) Phylogenies and angiosperm
diversification. Paleobiology 19, 141­167
2 Wing, S.L. and Boucher, L.D. (1998) Ecological aspects of the
cretaceous flowering plant radiation. Annu. Rev. Earth Planet Sci.
26, 379­421
3 Friis, E.M. et al. (2005) When Earth started blooming: insights from
the fossil record. Curr. Opin. Plant Biol. 8, 5­12
4 Stuessy, T.F. (2004) A transitional­combinatorial theory for the origin
of angiosperms. Taxon 53, 3­16
5 Friedman, W.E. et al. (2004) The evolution of plant development. Am.
J. Bot. 91, 1726­1741
6 Crane, P.R. et al. (1995) The origin and early diversification of
angiosperms. Nature 374, 27­33
7 Lidgard, S. and Crane, P.R. (1990) Angiosperm diversification and
Cretaceous floristic trends; a comparison of palynofloras and leaf
macrofloras. Paleobiology 16, 77­93
8 Crane, P.R. and Lidgard, S. (1989) Angiosperm diversification and
paleolatitudinal gradients in Cretaceous floristic diversity. Science
246, 675­678
9 Magallón-Puebla, S. et al. (1999) Phylogenetic pattern, diversity and
diversification of eudicots. Ann. Mo. Bot. Gard. 86, 297­372
10 Crane, P.R. et al. (2004) Fossils and plant phylogeny. Am. J. Bot. 91,
1683­1699
11 Friis, E.M. et al. (2000) Reproductive structure and organization of
basal angiosperms from the Early Cretaceous (Barremian or Aptian)
of Western Portugal. Int. J. Plant Sci. 161, S169­S182
12 Tate, J.A. et al. (2005) Polyploidy in plants. In The Evolution of the
Genome (Gregory, T., ed.), pp. 371­426, Elsevier, Academic Press
13 The Arabidopsis Genome Initiative. (2000) Analysis of the genome
sequence of the flowering plant Arabidopsis thaliana. Nature 408,
796­815
14 Simillion, C. et al. (2002) The hidden duplication past of Arabidopsis
thaliana. Proc. Natl. Acad. Sci. U. S. A. 99, 13627­13632
15 Bowers, J.E. et al. (2003) Unravelling angiosperm genome evolution
by phylogenetic analysis of chromosomal duplication events. Nature
422, 433­438
16 Van de Peer, Y. and Meyer, A. (2005) Large-scale gene and ancient
genome duplications. In The Evolution of the Genome (Gregory, T.,
ed.), pp. 330­368, Elsevier, Academic Press
17 Maere, S. et al. (2005) Modeling gene and genome duplications in
eukaryotes. Proc. Natl. Acad. Sci. U. S. A. 102, 5454­5459
18 Blanc, G. and Wolfe, K.H. (2004) Functional divergence of duplicated
genes formed by polyploidy during Arabidopsis evolution. Plant Cell
16, 1679­1691
19 Seoighe, C. and Gehring, C. (2004) Genome duplication led to highly
selective expansion of the Arabidopsis thaliana proteome. Trends
Genet. 20, 461­464
20 Birchler, J.A. et al. (2001) Dosage-dependent gene regulation in
multicellular eukaryotes: implications for dosage compensation,
aneuploid syndromes, and quantitative traits. Dev. Biol. 234, 275­288
21 Amoutzias, G.D. et al. (2004) Convergent networks by single-gene
duplications in higher eukaryotes. EMBO Rep. 5, 274­279
22 Papp, B. et al. (2003) Dosage sensitivity and the evolution of gene
families in yeast. Nature 424, 194­197
23 Krylov, D.M. et al. (2003) Gene loss, protein sequence divergence, gene
dispensability, expression level, and interactivity are correlated in
eukaryotic evolution. Genome Res. 13, 2229­2235
24 Kellogg, E.A. (2004) Evolution of developmental traits. Curr. Opin.
Plant Biol. 7, 92­98
25 Carroll, S.B. (2000) Endless forms: the evolution of gene regulation
and morphological diversity. Cell 101, 577­580
26 Gibson, G. and Honeycutt, E. (2002) The evolution of developmental
regulatory pathways. Curr. Opin. Genet. Dev. 12, 695­700
27 Chen, F. et al. (2003) Biosynthesis and emission of terpenoid volatiles
from Arabidopsis flowers. Plant Cell 15, 481­494
28 Crepet, W.L. (2000) Progress in understanding angiosperm history,
success, and relationships: Darwin's abominably `perplexing phenom-
enon'. Proc. Natl. Acad. Sci. U. S. A. 97, 12939­12941
29 Barrett, P.M. and Willis, K.J. (2001) Did dinosaurs invent flowers?
Dinosaur­angiosperm coevolution revisited. Biol. Rev. Camb. Philos.
Soc. 76, 411­447
Box 3. Dinosaurs, bugs and CO2
Alternative scenarios have been putforward to explain theorigin of the
angiosperms and their subsequent tempo and pattern of diversifica-
tion, most of which could be compatible with the genome duplication
theory proposed here. For instance, some authors (e.g. [68]) have
speculated that the origin and diversification of angiosperms was
mediated by changes in the browsing behavior of dinosaurs (reviewed
in [29]). Moreover, the ecological radiation of angiosperms has been
associated with the evolution of complex jaw mechanisms, such as
pleurokinesis (sideway chewing movement) among ornithischian
dinosaurs such as ceratopsians and stegosaurs. However, others
have refuted the idea of dinosaurs being causative agents in the origin
of angiosperms [29].
Insect pollinators are thought to have had a much larger role in
angiosperm diversification. During the Cretaceous, the development
of specialized floral features, indicating a high level of pollinator
specificity [66], seems to coincide well with the origin and
diversification of many insect clades, such as lepidopterans,
hymenopterans and dipterans. It is now widely accepted that
coevolution between early angiosperms and anthophilous insects
has been of major importance for the evolution of angiosperms [69].
Furthermore, insect herbivores such as aphids and beetles might
have also had an important role in many of the adaptive radiations
that account for the present diversity of flowering plants (and also
beetles) [70,71].
Barrett and Willis [29] suggest that the diversification of
angiosperms might have been the direct result of a major geological
event that led to increased levels of atmospheric CO2. During the
past 500 million years, there have been five periods of increased
tectonic activity, the most recent of which occurred between
120 Mya and 80 Mya, at the time of the final break-up of Pangaea.
Such periods of increased activity are associated with significant
changes in atmospheric CO2 levels owing to increased volcanic
activity [29,72]. In turn, elevated CO2 levels could increase the ability
of plants to colonize drier and/or more nutrient-poor sites, thereby
increasing the range of niches available for plant growth and, thus,
the opportunities for the emergence of novel plant forms [29]. Apart
from the colonization of new habitats, increased levels of CO2 could
also have an effect on the rate of turnover of plant generations [73]. If
generation turnover rates increase, the rate of speciation is also
expected to increase. Therefore, as stated by Barrett and Willis [29],
higher levels of CO2 might have increased the evolutionary and
selective pressures on plants owing to the induction of (abiotic)
stress from which plants cannot escape; evolving to cope with the
changing environment might have been the only option for plants.
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005596
www.sciencedirect.com
30 Grotewold, E. (2005) Plant metabolic diversity: a regulatory perspec-
tive. Trends Plant Sci. 10, 57­62
31 Doebley, J. and Lukens, L. (1998) Transcriptional regulators and the
evolution of plant form. Plant Cell 10, 1075­1082
32 Tautz, D. (2000) Evolution of transcriptional regulation. Curr. Opin.
Genet. Dev. 10, 575­579
33 Taylor, J.S. and Raes, J. (2004) Duplication and divergence: the
evolution of new genes and old ideas. Annu. Rev. Genet. 9, 615­643
34 Aburomia, R. et al. (2003) Functional evolution in the ancestral
lineage of vertebrates or when genomic complexity was wagging its
morphological tail. J. Struct. Funct. Genom. 3, 45­52
35 Ohno, S. (1970) Evolution by Gene Duplication, Springer-Verlag
36 Chaw, S.M. et al. (2004) Dating the monocot­dicot divergence and the
origin of core eudicots using whole chloroplast genomes. J. Mol. Evol.
58, 424­441
37 Chapman, B.A. et al. (2004) A comparative phylogenetic approach for
dating whole-genome duplication events. Bioinformatics 20, 180­185
38 Davies, T.J. et al. (2004) Darwin's abominable mystery: insights from a
supertree of the angiosperms. Proc. Natl. Acad. Sci. U. S. A. 101,
1904­1909
39 Cronk, Q.C. (2001) Plant evolution and development in a post-genomic
context. Nat. Rev. Genet. 2, 607­619
40 Zahn, L.M. et al. (2005) The evolution of the SEPALLATA subfamily of
MADS-box genes: a preangiosperm origin with multiple duplications
throughout angiosperm history. Genetics 169, 2209­2223
41 Blanc, G. et al. (2003) A recent polyploidy superimposed on older
large-scale duplications in the Arabidopsis genome. Genome Res. 13,
137­144
42 Cronquist, A. (1981) An Integrated System of Classification of
Flowering Plants, Columbia University Press
43 Wendel, J.F. (2000) Genome evolution in polyploids. Plant Mol. Biol.
42, 225­249
44 Otto, S.P. and Whitton, J. (2000) Polyploid incidence and evolution.
Annu. Rev. Genet 34, 401­437
45 Guo, M. et al. (1996) Dosage effects on gene expression in a maize
ploidy series. Genetics 142, 1349­1355
46 Osborn, T.C. et al. (2003) Understanding mechanisms of novel gene
expression in polyploids. Trends Genet. 19, 141­147
47 Thompson, J.N. et al. (1997) Plant polyploidy and insect/plant
interactions. Am. Nat. 150, 730­743
48 Spring, J. (2003) Major transitions in evolution by genome fusions:
from prokaryotes to eukaryotes, metazoans, bilaterians and verte-
brates. J. Struct. Funct. Genom. 3, 19­25
49 Rieseberg, L.H. et al. (2003) Major ecological transitions in wild
sunflowers facilitated by hybridization. Science 301, 1211­1216
50 Ehrendorfer, F. (1980) Polyploidy and distribution. In Polyploidy ­
Biological relevance (Lewis, W.H., ed.), pp. 45­60, Plenum Press
51 Arnold, M.L. (1997) Natural Hybridization and Evolution, Oxford
University Press
52 Ramsey, J. and Schemske, D.W. (1998) Pathways, mechanisms, and
rates of polyploid formation in flowering plants. Annu. Rev. Ecol. Syst.
29, 467­501
53 Masterston, J. (1994) Stomatal size in fossil plants: evidence for
polyploidy in majority of angiosperms. Science 264, 421­423
54 Werth, C.R. and Windham, M.D. (1991) A model for divergent,
allopatric speciation of polyploid pteridophytes resulting from
silencing of duplicate-gene expression. Am. Nat. 137, 515­526
55 Lynch, M. and Force, A. (2000) The probability of duplicate gene
preservation by subfunctionalization. Genetics 154, 459­473
56 Taylor, J.S. et al. (2001) Genome duplication, divergent resolution and
speciation. Trends Genet. 17, 299­301
57 Koch, M.A. et al. (2000) Comparative evolutionary analysis of chal-
cone synthase and alcohol dehydrogenase loci in Arabidopsis, Arabis,
and related genera (Brassicaceae). Mol. Biol. Evol. 17, 1483­1498
58 Lynch, M. and Conery, J.S. (2000) The evolutionary fate and
consequences of duplicate genes. Science 290, 1151­1155
59 Soltis, P.S. et al. (2002) Rate heterogeneity among lineages of tracheo-
phytes: integration of molecular and fossil data and evidence for
molecular living fossils. Proc. Natl. Acad. Sci. U. S. A. 99, 4430­4435
60 Strauss, E. (1999) Can mitochondrial clocks keep time? Science 283,
1435, 1437­1438
61 Senchina, D.S. et al. (2003) Rate variation among nuclear genes and
the age of polyploidy in Gossypium. Mol. Biol. Evol. 20, 633­643
62 Sanderson, M.J. et al. (2004) Molecular evidence on plant divergence
times. Am. J. Bot. 91, 1656­1665
63 Gaut, B. (1998) Molecular clocks and nucleotide substitution rates in
higher plants. In Evolutionary Biology (Hecht, M.K., ed.), pp. 93­120,
Plenum Press
64 Zhang, L. et al. (2002) Patterns of nucleotide substitution among
simultaneously duplicated gene pairs in Arabidopsis thaliana. Mol.
Biol. Evol. 19, 1464­1473
65 Yang, Y.W. et al. (1999) Rates of nucleotide substitution in angiosperm
mitochondrial DNA sequences and dates of divergence between
Brassica and other angiosperm lineages. J. Mol. Evol. 48, 597­604
66 Crepet, W. (1996) Timing in the evolution of derived floral characters:
Upper Cretaceous (Turonian) taxa with tricolpate and tricolpate-
derived pollen. Rev. Palaeobot. Palynol. 90, 339­359
67 Wikstro¨m, N. et al. (2001) Evolution of the angiosperms: calibrating
the family tree. Proc. R. Soc. London B Biol. Sci. 268, 2211­2220
68 Bakker, R.T. (1978) Dinosaur feeding behaviour and the origin of
flowering plants. Nature 274, 661­663
69 Grimaldi, D.A. (1999) The co-radiations of pollinating insects and
angiosperms in the Cretaceous. Ann. Mo. Bot. Gard. 86, 373­406
70 Farrell, B.D. (1998) `Inordinate fondness' explained: why are there so
many beetles? Science 281, 555­559
71 Wilf, P. et al. (2000) Timing the radiations of leaf beetles: hispines on
gingers from the latest Cretaceous to Recent. Science 289, 291­294
72 Sheridan, R.E. (1997) Pulsation tectonics as a control on the dispersal
and assembly of supercontinents. J. Geodynam. 23, 173­196
73 Phillips, O.L. and Gentry, A.H. (1994) Increasing turnover through
time in tropical forests. Science 263, 954­958
74 Friis, E.M. et al. (2004) Araceae from the Early Cretaceous of
Portugal: evidence on the emergence of monocotyledons. Proc. Natl.
Acad. Sci. U. S. A. 101, 16565­16570
Opinion TRENDS in Ecology and Evolution Vol.20 No.11 November 2005 597
www.sciencedirect.com