Zimmermann’s telome theory of megaphyll leaf evolution:
a molecular and cellular critique
David J Beerling and Andrew J Fleming
Megaphyll leaf evolution was a critical event in Earth history that
had major consequences for the biotic regulation of the global
environment. Zimmermann’s telome theory has been widely
accepted for over seventy years as the leading explanation for
this evolutionary innovation. According to the telome theory,
megaphylls evolved from the three-dimensional lateral
branches of early vascular land plants in a hypothetical series of
three transformations; first, the formation of determinate lateral
branches (overtopping); second, the development of ‘flattened’
branch systems (planation); and third, the fusion of planated
branches with lateral outgrowths of photosynthetic mesophyll
tissue to form the leaf blade (webbing). A critical review of the
molecular and cellular evidence identifies plausible genetic,
cellular and physiological mechanisms in extant higher plants
for overtopping and planation but more limited evidence for the
process of webbing (lateral outgrowth fusion). We highlight key
outstanding questions concerning the telome theory that are
likely to be resolved when gene identification and functional
analysis techniques are applied to photosynthetic organisms
that have different evolutionary histories.
Addresses
Department of Animal and Plant Sciences, University of Sheffield,
Sheffield, S10 2TN, UK
Corresponding author: Beerling, David J ([email protected])
Current Opinion in Plant Biology 2007, 10:4–12
This review comes from a themed issue on
Growth and development
Edited by Cris Kuhlemeier and Neelima Sinha
Available online 1st December 2006
1369-5266/$ – see front matter
# 2006 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.pbi.2006.11.006
Introduction
Megaphyll leaves, with their flat blades and complex
vasculature, are fundamental light-capture organs formed
as determinate lateral outgrowths from the flanks of the
shoot apical meristem (SAM). Today, the vast majority of
euphyllophytes (i.e. ferns, gymnosperms and angiosperms) utilize the megaphyll leaf form to harvest solar
energy, and this underpins the net annual primary production of the contemporary terrestrial biosphere of
around 56 billion tonnes of carbon (C) (56.4 1015 g
C year 1) [1]. This situation contrasts dramatically with
that in the late-Silurian (420 Myr ago) when the earliest
Current Opinion in Plant Biology 2007, 10:4–12
tracheophytes, such as Cooksonia, had neither lateral
branches nor leaves but consisted of naked dichotomized
branching stem systems [2]. The subsequent origination
and spread of megaphyll leaves throughout terrestrial
floras did not take place until some 40 Myr later, from
the close of the Devonian (Famennian; 359 Myr ago)
onwards. It profoundly altered the evolutionary trajectory
of plant and animal life, as well as the exchange of
energy and materials between the land surface and the
atmosphere [1]. In particular, by accelerating the chemical weathering of Ca-Mg silicate rocks, the long-term
sink for CO2, the evolutionary diversification of land
plants appeared to create a network of feedback
cycles that regulate the long-term geochemical carbon
cycle and climate on a multi-million-year timescale [3].
The telome theory was first proposed seventy years ago
by the German palaeobotanist Walter Zimmermann in his
1930 book entitled ‘Die Phylogenie de Pflanzen’. In this
book, Zimmermann describes the evolutionary sequence
of modifications leading up to the appearance of megaphyll leaves on ancestral early axial vascular land plants
[4]. Zimmermann’s ‘theory’ was really a description of
three fundamental steps — overtopping, planation, and
webbing — for transforming a stem (a telome) into a
laminated leaf blade by the fusion of lateral branch
systems or telomes (Figure 1). Each step was presumably
underpinned by modification of the regulation of the
genetic networks in those descendent organisms that
acquired it, and (presumably) also conferred ecological
or physiological advantages on them. It also seems
likely, given the independent evolution of megaphyll
leaves in all major clades of the Euphyllophytina [5],
that each of the three transformations has been recruited
multiple times during the evolutionary history of land
plants.
Here, we critically review and discuss insights gained
from recent advances in molecular and cell biology that
contribute towards a fundamental understanding of the
telome theory. Our aim is twofold. First, to determine
what molecular biology can say about how each step
might have arisen. This information addresses one of
the long-standing criticisms levelled at the telome theory,
that it lacks a developmental mechanism. It might also
shed light on other issues, including why certain morphological transformations occur as opposed to others,
and why the particular sequence of processes foreshadowed the appearance of leaves [6,7]. Our second aim is
to highlight future areas of investigation that might
address some of the outstanding issues.
www.sciencedirect.com
Megaphyll leaf evolution Beerling and Fleming 5
Figure 1
The three major transformations proposed by Zimmermann in his telome theory of megaphyll leaf evolution. Reproduced from [1] with permission
of Oxford University Press.
Zimmermann’s telome theory
In its general form, the telome theory proposes that, by
combining and modifying simple multicellular telomes
through one or more developmental processes, it is possible
to explain all extant and extinct plant morphologies [7]. In
the case of leaves, Zimmermann essentially envisaged the
transformation of a stem into a leaf by modification of
existing organs rather than through a major change in body
plan. He postulated that the evolution of a megaphyll leaf
proceeded through a series of hypothetical transformations
that were based on modifications to a lateral branch of a
dichotomized three-dimensional branching stem system,
as typified by early Psilophyton- and Rhynia-like plants. The
telome module is the distal portion of the axis, the proximal
portion is known as a mesome. The first step, ‘overtopping’, occurs when one branch of a dichotomous pair
overgrows another to allow the main axis to become a stem
and the sub-ordinate stem (i.e. the overtopped branch) to
branch out in three dimensions. In the next step, ‘planation’, neighbouring lateral branch systems of terete (cylindrical, tapering) stems become orientated into a single plane,
i.e. the branch system is ‘flattened’ not the stem. Finally,
lateral outgrowths of photosynthetic tissue (a ‘webbing’)
evolved to join the segments of the planated lateral
branches to form a laminate leaf blade. It is interesting
to note that no single plant lineage has been found to
exhibit all three crucial steps in its fossil history. Instead,
each one is recognized independently in several different
fossil lineages, with putative intermediate stages documented in many early fossil taxa [5].
www.sciencedirect.com
In the succeeding sections, we summarize the fossil data
supporting each hypothesized transformation of the telome
theory, and attempt to relate each phase to the molecular
developmental and physiological mechanisms identified to
date in present day plants (mainly angiosperms).
Transformation one: overtopping
Overtopping, the formation of determinate lateral branch
systems, is the limited elongation of branch tips. According to the telome theory, overtopped branches are the
necessary precursors to megaphyll leaves [4]. In overtopped branches, vascularisation is primarily channelled
into a single or a few vein branches. Determinate threedimensional lateral branching systems became widespread following the rapid evolutionary radiation of plants
during the Early Devonian (Pragian-Emsian) [8]. The
dichotomizing axial forms remained common during this
period but were confined to the rhyniophytes such as
Horneophyton lignieri (Kidston and Lang) Barghoorn and
Darrah [9]. Caution is required in interpreting the ‘protoleaves’ of Early Devonian fossils as lateral appendages
because the natural orientation of the axis can be difficult
to ascertain and because compression during fossilization
could artificially flatten branch systems [10]. Nevertheless, the morphology of Early Devonian branched axial
systems is thought to have been diverse, with some axes
terminating at sterile tips and others with sporgangia.
The ‘overtopping’ of branched systems, and the appearance of determinate laterals, is recognized as a genuine
evolutionary innovation within the Early Devonian
Current Opinion in Plant Biology 2007, 10:4–12
6 Growth and development
trimerophytes. In this group, branching was nearly
always consistently three-dimensional, as seen in Pertica
quadrifaria [11] and Psilophyton Dawson [12]. There is
also anatomical evidence in permineralised proto-leaves
of Psilophyton coniculum that the vascular traces were
markedly differentiated from those of the main axis,
indicating that they developed laterally [13].
In developmental terms, overtopping requires, first, the
initiation of branches, and second, the control of the
relative growth rate of different branches. Extant plants
exhibit two types of overtopping: sympodial branching
where overtopping is combined with reduction, which is
typical of (but not exclusive to) lycopods and some ferns,
and monopodial branching, which is typical of the main
branching system of the majority of higher plants today.
Branching in angiosperms requires the formation of lateral secondary meristems, termed axillary meristems.
These are distributed along the stem proximal to the
primary SAM. Although there is some debate as to
whether axillary meristems represent pockets of cells
that are left behind by the primary SAM or are formed
de novo, molecular markers for axillary meristems have
been identified and mutation in such genes (e.g. LATERAL SUPPRESSOR [LAS]) leads to loss of axillary
branching [14]. Thus, at least some of the transcription
factors that are involved in defining potential branch
points in angiosperms have been identified and their
evolutionary history can be investigated.
Once formed, an axillary meristem might grow out or
remain dormant. A host of environmental and endogenous signals have been identified that lead to speciesspecific branching patterns. Nevertheless, a central role
for the growth factor auxin in controlling axillary meristem outgrowth (i.e. apical dominance) is a long-standing
tenet in plant physiology (reviewed in [15]). More
recently, a novel carotenoid-associated signalling system
that is involved in axillary branch outgrowth has been
identified [16–19]. An interesting idea arising from this
research is that many of the factors that control axillary
meristem activation converge on auxin transport, and that
the relative growth of a branch might be dependent on its
relative ability to transport auxin [20].
The evolution of axillary meristems is not explicitly
invoked in the telome theory, yet they clearly represent
the major mechanism for branching in angiosperms.
Thus, whether the mechanisms identified to date in
the control of angiosperm axillary meristem formation
support or refute the telome theory is a moot point. As will
be seen later, the involvement of auxin flux is a recurring
theme in the mechanisms relating to leaf development.
The investigation of auxin flux in non-angiosperm species
with respect to branch growth would shed light on its
potential role in leaf evolution.
Current Opinion in Plant Biology 2007, 10:4–12
The formation of an axillary meristem is intimately linked
with leaf initiation at the primary SAM. The SAM is
distinguished by being developmentally indeterminate,
that is, it has the potential to form a large number of leaves
and associated axillary meristems. The angiosperm leaf,
on the other hand, is distinguished by being determinate,
i.e. the extent of leaf growth is limited. This difference in
determinacy between angiosperm meristem and leaf is
set by the expression of a developmental module of
transcription factors, the KNOX (KNOTTED-like homeobox)/ARP (AS1-RS2-PHAN) module (reviewed in [21]).
Thus, most angiosperm SAMs are characterised by the
expression of KNOX genes and by the lack of ARP gene
expression, whereas leaves are characterised by the
expression of ARP genes and the consequential suppression of KNOX gene expression. Within the eudicots, the
KNOX/ARP regulatory module has been conserved, but
the timing and positioning of the switch within the leaf
from a KNOX ‘ON’ to KNOX ‘OFF’ state (mediated by
ARP gene expression) appears to have shifted in evolution. A delay in switching to a KNOX ‘OFF’ is associated
with the acquisition of more complex leaf forms (e.g.
compound rather than simple and palmate rather than
pinnate leaves) [22,23]. Moreover, the KNOX/ARP module appears to have been recruited in the acquisition of
lateral (leaf) organs on more than one occasion over
evolutionary time. For example, the pattern of KNOX/
ARP gene expression in microphyll leaf formation (e.g. by
Selaginella kraussiana) is highly reminiscent of that
observed in angiosperms, i.e. KNOX is ‘OFF’, ARP is
‘ON’ [24]. In Selaginella, both ARP and KNOX genes are
expressed within the SAM itself, and it has been suggested that this expression pattern is linked with the
ability of these meristems to bifurcate, although the
mechanism by which this might occur is unclear [24].
Nevertheless, it is possible that the KNOX/ARP module
might have played an important role in the initial acquisition of branching upon which overtopping is theoretically
based (Figure 2).
The evolutionary importance of the KNOX/ARP module
is also suggested by comparison of its functions in Arabidopsis thaliana and the closely related Cardomine hirsute
[22]. The switch from the simple leaf morphology of
Arabidopsis to the compound leaf morphology of Cardomine can be linked to an altered timing of KNOX gene
expression, probably through alterations in promoter
sequences that control the timing and/or position of
expression. This is in contrast to the situation for the
evolution of the LEAFY (LFY) gene (which encodes a
transcription factor that is important in the control of
flowering). Homologues of LFY are present in non-flowering plants, and it seems that the incorporation of this
transcription factor into flowering gene networks has been
associated with mutation of LFY’s DNA-binding domain
rather than with elements that control gene expression
[25].
www.sciencedirect.com
Megaphyll leaf evolution Beerling and Fleming 7
Figure 2
The potential relationship of the KNOX/ARP transcription module with determinate branching. (a) A theoretical meristem (depicted as a light blue
circle) is maintained in an indeterminate state by co-expression of the KNOX and ARP transcription factors. (b,c) The growth of such a meristem
leads to the eventual splitting (branching) of the tissue to generate two equivalent indeterminate meristems, both with KNOX and ARP in the ‘ON’
state. (d) At some time in evolution, the KNOX/ARP module became recruited to distinguish tissue that had an indeterminate fate (KNOX ‘ON’
and ARP ‘OFF’, light blue) from that with a determinate fate (KNOX ‘OFF’, ARP ’ON’, dark blue). (e) The indeterminate part of the meristem
(light blue) would form a branch (stem) that would ‘overtop’ the determinate part, which might form a leaf or a branch (dark blue).
These emerging molecular genetic data support the concept of bricolage: that ‘evolutionary diversification
involves a constant tinkering with a highly conserved
set of molecules to produce the most diverse organismal
forms’ [26]. KNOX/ARP would seem to represent such a
conserved module in leaf evolution.
might function to maximize spore dispersal, light interception and possibly also to offer greater mechanical
stability [6]. As with the fossil evidence regarding overtopping, caution is required in the interpretation of this
innovation from fossils because the compression of initially terete structures during preservation can confer the
appearance of planation and lamination [10].
Transformation two: planation
Planation is the flattening of the three-dimensional terete
stem segments into a single plane. When overtopped
morphologies are combined with planated lateral
branches, computer model simulations indicate that they
www.sciencedirect.com
Non-laminate proto-leaves are more commonly planated
in Mid than in Early Devonian fossil plant assemblages,
although three-dimensional branching structures are
retained by many species, such as the progymnosperms
Current Opinion in Plant Biology 2007, 10:4–12
8 Growth and development
Figure 3
Tetraxylopteris schmidtii [27] and Actinoxylon banksii [28],
and the cladoxylalean Pseudosporochnus nodosus Leclercq
and Banks [29]. Typical examples of planated branched
systems include those of the cladoxylalean Cladoxylon
scoparium Kräusel and Weyland [30] and the putative
sphenophyte Ibyka amphikoma [31]. In the Late Devonian, the finely branching non-laminate ‘fronds’ of the
putative primitive fern Rhacophyton ceratangium are a
particularly noteworthy example of planation, reaching
lengths in excess of 30 cm [32].
No modern-day representatives of planated branching
stems are known, although the sub-tropical gymnosperm
genus Phyllocladus has greatly reduced true leaves and
phylloclades might represent planated shoot systems
[33]. In the Araucariceae, large multi-veined leaves are
planated into large compound shoots, and the Cupressaceae and Podocarpaceae leaves with a single vein are reorientated to produce a two-dimensional flattening of
shoots [34].
In developmental terms, planation requires the juxtaposition of branches and hence would involve the positioning
of branch initiation points and the co-ordination of relative
branch growth. In modern angiosperms, branch points (i.e.
axillary meristems) are intimately associated with leaf
position. Significant strides have been made in our understanding of the molecular control of leaf initiation and its
patterning within the SAM in the past few years (reviewed
in [35]). As in axillary meristem outgrowth, auxin plays a
central role in the control of leaf initiation and its patterning, with auxin flux through plant tissue directed by the
pattern of expression of a series of auxin efflux (PINFORMED [PIN]-encoded) and influx (AUXIN1 [AUX])
carriers. With respect to leaf (and, thus, associated branch)
positioning, analysis of PIN and AUX expression patterns
in the primary SAM indicates that local accumulation of
auxin at discrete sites within the meristem dictates the
presumptive position of leaf formation [36,37]. On the
basis of this system, different patterns of auxin flux can be
modelled [38–40], leading to different patterns of leaf
initiation. Depending on the extent of internode expansion that occurs between these leaf initiation events
(which might be influenced by KNOX gene expression
The potential role of auxin in patterning branching. (a) The site of
determinate branch formation (dark blue) is dictated by local
accumulation of auxin (AUX) within the SAM (depicted as a light blue
circle). (b) The dynamic integration of auxin flux within the SAM leads to
the shifting of the local auxin maximum with time so that adjacent
portions of the SAM become successively determined to form branches.
Recent data based on PIN distribution provide models as to how this
might occur. (c) If internode growth between successive branches is
limited, a series of adjacent co-planar branches could be formed (i.e.
planation could occur). The relative outgrowth of each branch itself
might depend on auxin transport within the branch, and the
determinancy or indeterminancy of each branch could vary depending
on its KNOX/ARP status.
Current Opinion in Plant Biology 2007, 10:4–12
www.sciencedirect.com
Megaphyll leaf evolution Beerling and Fleming 9
[24]), various constellations of leaves (and associated
axillary meristems) could form in three-dimensional space,
a pre-requisite for planation (Figure 3).
Transformation three: webbing
The fossil record shows that as photosynthetic branching
systems were successively overtopped and planated, closely aligned branches became webbed or infilled by thin
lateral outgrowths of photosynthetic mesophyll tissue to
produce the broad lamina of a leaf blade. True laminate
leaves are atypical of Mid Devonian floras. They appear
more frequently only in Late Devonian forest floras
dominated by the progymnosperm Archaeopteris Dawson
[41], although these structures were by no means a
ubiquitous adaptation among terrestrial plants at this
time. Fossil megaphylls vary enormously in shape,
size and morphology: from the large laminate leaves of
Archaeopteris obtusa to the finely dissected branches
of Archaeopteris fissilis [42,43]. By the Mississippian
(Tournasian-Viséan) plants that had megaphyll leaves
were confined to the ferns, pteridosperms and sphenophytes, as arborescent lycophytes with long microphyll
leaves dominated floras. The leaves of the early sphenophytes, such as Archaeocalamites radiatus (Brongniartt)
Stur and Sphenophyllum tenerrimum Ettingshausen,
remained as lateral branches without laminae, with laminate forms only appearing in later groups [44]. In the
pteridosperms Diplopteridium teilianum (Kidston) Walton
and Rhacopteris circularis Walton, and possibly in other
members of the same group, leaves were exclusively
laminate, closely resembling those of modern ferns. In
other pteridosperms, taxa with broad lamina, such as
Charbeckia macrophylla [45], coexisted with taxa such as
Diplopteridium holdenii that possessed highly dissected
terete forms [42]. Only with the rise of the gymnosperms
and the decline of the arborescent lycophytes (from the
Late Carboniferous onwards) does the laminate leaf-form
become firmly established in floras worldwide.
In developmental terms, the webbing of telomes to produce a laminate leaf blade requires, first, the production of
lateral outgrowths and, second, either congenital or postgenital fusion of adjacent branches. The lateral growth of
angiosperm leaves to form a lamina is under the control of a
complex transcriptional and signalling network that
requires the juxtaposition of fields of cells that express
either adaxial or abaxial identity genes [46–49]. The interaction of these two domains leads to the gradual restriction
of growth to a plane that is centred around the primary
vascular element. Interestingly, the adaxial leaf domain in
angiosperms is characterised by the expression of ARP
genes, suggesting a close mechanistic interplay between
determinate lateral organ formation, proximal–distal
growth and lamina formation [21,22,23,50,51].
In modern angiosperms, fusion events between adjacent
primordia can be observed in certain genetic contexts.
www.sciencedirect.com
Thus, the CUP SHAPED COTYLEDON (CUC) genes
(encoding NAC [NAM-ATF1-CUC2] transcription factors) in Arabidopsis prevent the fusion of cotyledons [52],
and mutation of the CUPULLIFORMIS gene in Antirrhinum leads to the dramatic formation of fused organs
throughout the plant [53]. These genes must be expressed
between forming primordia to prevent fusion, and fusion
occurs in their absence. In addition to these transcription
factors, auxin has also been implicated in the control of
lateral organ fusion. Thus, disruption of auxin flux often
leads to fused organs, and it is possible to induce collartype leaves by manipulating auxin patterns around the
SAM [54,55]. With respect to postgenital fusion, the
cuticle composition of the cell wall is a potential target.
Thus, manipulations that lead to loss or disruption of the
cuticle lead to postgenital fusion events [56].
The lateral organs of angiosperms can clearly be persuaded to fuse. During angiosperm leaf initiation and
development, however, there are no obvious fusion
events. Most notably, at the time of initiation, there is
no overt vascular differentiation in the angiosperm leaf.
Recent data using PIN proteins as markers of auxin flux
and thus, by implication, of vascular differentiation, show
that the influx of auxin at the site of presumptive leaf
formation on the SAM predicts the position at which the
primary vascular tissue (i.e. the midrib) will form
[37,57]. Subsequently, the pattern of vascular formation
appears to be prescribed by the pattern of PIN protein,
and thus, auxin flux. This patterning process is thought to
depend on a complex and not yet fully resolved feedback
loop of auxin flux and PIN protein distribution [58]. In
this system, there is no evidence of the postgenital fusion
of plant tissue to form a leaf. Rather, a tissue volume is
formed and, at the same time, patterning processes define
a simple vascular network. A feedback system then
enables a co-ordination of leaf lamina growth and more
complex vascular patterning [57].
High carbon dioxide: an environmental
barrier to leaf evolution?
Large megaphyll leaves became widespread some 40–50
Myr after the origination of vascular land plants. Fossil
evidence indicates, however, that at least one group of
plants was capable of producing megaphyll leaves millions
of years earlier. Early Devonian (Pragian) proto-leaves of
rare Chinese fossils of Eophyllophyton bellum [59] are tiny,
planated and webbed by thin lamina (thickness 40–
200 mm) [59]. They occur on lateral branches and are
not confined to terminal fertile regions, like similar lamina
structures in Adoketophyton subverticillatum Li and Edwards
and Calatheca beckii Hao and Gensel from the same flora,
and so are not thought to have a role protecting the
sporangia. Consequently, these proto-leaves have been
interpreted as true photosynthetic megaphyll leaves with a
distinctive morphology of deeply incised lobes reaching
2.0–4.5 mm in length and 1.4–4.0 mm in width [60]. The
Current Opinion in Plant Biology 2007, 10:4–12
10 Growth and development
rare occurrence of E. bellum megaphylls suggests that the
basic molecular ‘toolkit’ for producing and assembling
leaves was in place tens of millions of years before this
evolutionary innovation became widespread [1], and raises
the question of what prevented plants from releasing this
morphogenetic potential ?
One hypothesis links the inordinately long delay to the
high CO2 concentration in the Silurian atmosphere; large
megaphyll leaves only appear and become widespread
after a 90% drop in CO2 concentration during the Devonian and early Carboniferous, which corresponds to a 100fold rise the stomatal density of fossil leaves [61]. This
inverse response of stomatal numbers to CO2 concentration on an evolutionary timescale is consistent with the
effects of CO2 seen in extant plants over centuries [62]
and might also have a genetic basis [63]. Experimental
studies using the C24 Arabidopsis accession identified
HIGH CARBON DIOXIDE (HIC) as a gene that is
expressed exclusively in guard cells and is putatively
linked to long-chain fatty acid synthesis [63]. When
the expression of HIC was reduced, stomatal density
increased by up to 40% when grown at elevated CO2
concentrations, whereas normal (wildtype) plants showed
either a small reduction in stomatal density or no
change in response to elevated CO2 [63]. Gray et al.
[63] hypothesized that HIC negatively regulates stomatal
development in normal plants through the production of
long-chain fatty acids that are deposited in the guard cells.
These fatty acids block the diffusion of an unknown
signalling compound that represses stomatal development. Under conditions of elevated CO2 concentration,
fatty acid synthesis is disrupted and production of the
repressing signalling compound inhibited, allowing
stomatal density to increase.
The functional significance that this response holds for
leaf evolution has been revealed by theoretical calculations of the energy budgets of various hypothetical and
actual photosynthetic structures [61]. For an early axial
land plant in the Silurian high CO2 atmosphere, whose
simple structure intercepts minimal quantities of solar
energy, stem temperatures remain safely below the lethal
temperatures despite having a very restricted transpirational cooling stream dictated (in part) by few stomatal
pores. By contrast, such a low transpiration rate is insufficient to prevent the temperature of large megaphyll
leaves exceeding the highly conserved thermal limit
within the plant kingdom, because of their much greater
interception of solar energy. In fact, according to theory,
large megaphylls only become a viable option after stomatal density and transpirational cooling capacity
increase with the well-documented 90% drop in CO2
concentration in the late Palaeozoic [61].
The ‘high CO2’ hypothesis is independently supported
by quantitative analyses from the fossil record. These
Current Opinion in Plant Biology 2007, 10:4–12
show, first, a 25-fold enlargement in the maximum size of
leaf blades that tracks the late Palaeozoic drop in atmospheric CO2 [43]; second, increases in leaf size in two
phylogenetically independent clades (progymnosperms
and pteridosperms) [42,43]; and third, the first enlargement of Archaeopteris leaves in association with an eightfold rise in stomatal density [43]. It is possible, therefore,
that high CO2 concentrations acted as an environmental
‘barrier’ that prevented the origination and spread of large
megaphylls. Once that environmental barrier was
removed, ecological processes, especially competition
between species for light, soil water and nutrients, would
have driven both the co-evolution of the root and shoot
and the well-documented increase in plant height [3].
Future directions and conclusions
The first years of this century have seen significant steps
forward in the identification of genetic mechanisms that
provide a basis for interpreting leaf evolution. Much of
this work has concentrated on model angiosperms, in
which rapid systems for both gene identification and
functional analysis have been established. The next
few years will see the application and development of
such technologies in an ever wider spectrum of plant
species, covering the breadth of distinct extant plant
clades. Indeed, for some representatives of mosses and
ferns, genomic approaches are well underway [64–66] and
have provided insightful data linking KNOX gene expression, auxin and organogenesis [67,68]. These approaches
will provide the tools with which at least some of the ideas
described in this article can be investigated. In particular,
hypotheses on the role of environmental factors as driving
forces for the selection or adaptation leaf form will
become testable.
With respect to the telome theory, several questions
remain unanswered (listed in Box 1). At present, overtopping and planation seem, at least conceptually, to be
highly plausible mechanisms for leaf evolution. Geneticbased mechanisms have been identified in present day
plants that might have been transposed from early plants.
With respect to webbing, a mechanism that involves only
Box 1 Outstanding questions.
1. Are the genetic loci that control branching in angiosperms of
evolutionary significance?
2. Is the system of auxin flux identified in angiosperms conserved in
evolution?
3. To what extent does the KNOX/ARP transcription factor module
represent a basic mechanism in lateral organ formation? How is
the expression of this module integrated with auxin signalling?
4. Is the abaxial–adaxial patterning process and lamina formation an
innovation in angiosperms?
5. How widespread are genes such as HIC in controlling stomatal
responses to CO2 concentration in plants with different evolutionary origins?
6. What are the physiological or ecological advantages of each
transformation?
www.sciencedirect.com
Megaphyll leaf evolution Beerling and Fleming 11
lateral outgrowth from individual branches seems more
likely than a step requiring branch fusion. As a consequence, if the fusion element of webbing is not a requirement for leaf evolution, then the requirement for
planation is not obvious. At a more general level, there
is also a need to investigate the ecological advantages
gained by plants that acquired each transformation, particularly in respect of the interception of solar energy and
the efficiency of CO2 uptake in relation to water loss, in
the high CO2 of the Silurian atmosphere.
Acknowledgements
We thank Colin Osborne and Jane Langdale for comments on the
manuscript. We acknowledge inspiration for this article from the writings
of the distinguished palaeobotanist Thomas Harris (1903–1983) who
commented in 1940, ‘to me telomes and phyllomes are so to speak the
atoms of comparative morphology into which all organs are analysable’
(Ann Bot 1940, 4:713-734).
References and recommended reading
Papers of particular interest, published within the period of review,
have been highlighted as:
of special interest
of outstanding interest
1.
Beerling DJ: Leaf evolution: gases, genes and geochemistry.
Ann Bot 2005, 96:345-352.
2.
Edwards D, Feehan J, Smith DG: A late Wenlock flora from Co.
Tipperary, Ireland. Bot J Linn Soc 1983, 86:19-36.
3.
Beerling DJ, Berner RA: Feedbacks and the coevolution of
plants and atmospheric CO2. Proc Natl Acad Sci USA 2004,
102:1302-1305.
4.
Zimmermann W: Main results of the telome theory.
Palaeobotanist 1952, 1:456-470.
5.
Kenrick P, Crane PR: The Origin and Early Diversification of
Land Plants. A Cladistic Study. Smithsonian Institute; 1997.
6.
Niklas KJ: Computer models of early land plant evolution.
Annu Rev Earth Planet Sci 2004, 32:47-66.
7.
Kenrick P: The telome theory. In Developmental Genetics and
Plant Evolution. Edited by Cronk QCB, Bateman RM, Hawkins JA.
Taylor and Francis; 2002:365-387.
16. Sorefan K, Booker J, Haurogne K, Goussot M, Bainbridge K,
Foo E, Chatfield S, Ward S, Beveridge C, Rameau C, Leyser O:
MAX4 and RMS1 are orthologous dioxygenase-like genes that
regulate shoot branching in Arabidopsis and pea. Genes Dev
2003, 17:1469-1474.
17. Booker J, Auldridge M, Wills S, McCarty D, Klee H, Leyser O:
MAX3/CCD7 is a carotenoid cleavage dioxygenase required
for the synthesis of a novel plant signaling molecule. Curr Biol
2004, 14:1232-1238.
18. Booker J, Sieberer T, Wright W, Williamson L, Willett B,
Stirnberg P, Turnbull C, Srinivasan M, Goddard P, Leyser O: MAX1
encodes a cytochrome P450 family member that acts
downstream of MAX3/4 to produce a carotenoid-derived
branch-inhibiting hormone. Dev Cell 2005, 8:443-449.
19. Lazar G, Goodman HM: MAX1, a regulator of the flavonoid
pathway, controls vegetative axillary bud outgrowth in
Arabidopsis. Proc Natl Acad Sci USA 2006, 103:472-476.
20. Bennett T, Sieberer T, Willett B, Booker J, Luschnig C,
Leyser O: The Arabidopsis MAX pathway controls shoot
branching by regulating auxin transport. Curr Biol 2006,
16:553-563.
The stems of plants that have mutations in MAX genes show an unexpected increased capacity for auxin transport, most probably via a PINbased mechanism. The authors propose that relative auxin transport
capacity might play a key role in axillary meristem growth.
21. Tsiantis M, Hay A: Comparative plant development: the time of
the leaf? Nat Rev Genet 2003, 4:169-180.
22. Hay A, Tsiantis M: The genetic basis for differences in leaf form
between Arabidopsis thaliana and its wild relative Cardamine
hirsute. Nat Genet 2006, 38:942-949.
An elegant series of experiments indicates that evolutionary tinkering with
the KNOX/ARP transcription module underlies the modulation of leaf form
observed in closely related Arabidopsis and Cardamine species.
23. Kim M, McCormick S, Timmermans M, Sinha N: The expression
domain of PHANTASTICA determines leaflet placement in
compound leaves. Nature 2003, 424:438-443.
24. Harrison CJ, Corley SB, Moylan EC, Alexander DL, Scotland RW,
Langdale JA: Independent recruitment of a conserved
developmental mechanism during leaf formation. Nature 2005,
434:509-514.
Analysis of the expression of the KNOX/ARP transcription module in
Selaginalla indicates a pattern that is conserved in angiosperms. Transgenic experiments confirm conservation of gene function, consistent with
the idea that modulation of the timing and spatial expression of the KNOX/
ARP module is basic to lateral organ formation in different evolutionary
contexts.
8.
Gensel PG, Kotyk ME, Basinger JF: Morphology of above- and
below-ground structures in Early Devonian (Pragian-Emsian)
plants. In Plant Invade the Land. Edited by Gensel PG, Edwards D.
Columbia; 2001: 83–102.
25. Maizel A, Busch MA, Tanahashi T, Perkovic J, Kato M,
Hasebe M, Weigel D: The floral regulator LEAFY evolves by
substitutions in the DNA binding domain. Science 2005,
308:260-263.
This paper shows how changes to the protein sequence of a transcription
factor over evolutionary time can lead to its incorporation into a novel
developmental function, in this case flowering.
9.
Eggert DA: The sporangium of Horneophyton lignieri
(Rhyniophytina). Am J Bot 1974, 61:405-413.
26. Duboule D, Wilkins AS: The evolution of ‘bricolage’. Trends
Genet 1998, 14:54-59.
10. Chaloner WG: Plant and spore compression in sediments. In
Fossil Plants and Spores – Modern Techniques. Edited by Jones
TP, Rowe NP. Geological Society; 1999:36-40.
27. Beck CB: Tetraxylopteris schmidtii gen. et sp. nov., a probable
pteridosperm precursor from the Devonian of New York.
Am J Bot 1957, 44:350-367.
11. Kasper AE, Andrews HN: Pertica, a new genus of Devonian
plants from northern Maine. Am J Bot 1972, 59:897-911.
28. Matten JC: Actinoxylon banksii gen. et sp. nov.: a
progymnosperm from the Middle Devonian of New York.
Am J Bot 1968, 55:773-782.
12. Gensel PG, Andrews HN: Plant Life in the Devonian Period.
Praeger Scientific; 1984.
13. Trant CA, Gensel PG: Branching in Psilophyton: a new species
from the Lower Devonian of New Brunswick, Canada. Am J Bot
1985, 72:1256-1273.
14. Greb T, Clarenz O, Schäfer E, Müller D, Herrero R, Schmitz G,
Theres K: Molecular analysis of the LATERAL SUPPRESSOR
gene in Arabidopsis reveals a conserved control mechanism
for axillary meristem formation. Genes Dev 2003, 17:1175-1187.
15. Leyser O: The fall and rise of apical dominance. Curr Opin Genet
Dev 2005, 15:468-471.
www.sciencedirect.com
29. Berry CM, Farion-Demaret M: A reinvestigation of the
cladoxylopsid Pseudosporochnus nodosus Leclercq et Banks
from the middle Devonian of Goé, Belgium. Int J Plant Sci 1997,
158:350-372.
30. Leclercq S: Classe des Cladoxylopsida Pichi-Sermoli 1959.
In Traite de Paleobotanique, Vol. IV. Edited by Boureau E.
Masson et Cie; 1970:119-165.
31. Skog JE, Banks HE: Ibyka amphikoma gen. et sp. n., a new
protoarcticulate precursor from the late Middle Devonian of
New York state. Am J Bot 1973, 60:366-380.
Current Opinion in Plant Biology 2007, 10:4–12
12 Growth and development
32. Andrews HN, Phillips TL, Radforth NW: Palaeobotanical studies
in Arctic Canada. I. Archaeopteris from Ellesmere Island.
Can J Bot 1968, 43:545-556.
52. Aida M, Ishida T, Fukaki H, Fujisawa H, Tasaka M: Genes involved
in organ separation in Arabidopsis: an analysis of the
cup-shaped cotyledon mutant. Plant Cell 1997, 9:841-857.
33. Tomlinson PB, Takasp T, Rattenbury JA: Developmental shoot
morphology in Phyllocladus (Podocarpaceae). Bot J Linn Soc
1989, 99:223-248.
53. Weir I, Lu J, Cook H, Causier B, Schwarz-Sommer Z, Davies B:
CUPULIFORMIS establishes lateral organ boundaries in
Antirrhinum. Development 2004, 131:915-922.
34. Hill RS, Broadribb TJ: Turner Review No. 2. Southern conifers in
space and time. Aust J Bot 1999, 47:639-696.
54. Liu C, Xu Z, Chua NH: Auxin polar transport is essential for the
establishment of bilateral symmetry during early plant
embryogenesis. Plant Cell 1993, 5:621-630.
35. Fleming AJ: Formation of primordia and phyllotaxy. Curr Opin
Plant Biol 2005, 8:53-58.
36. Reinhardt D, Pesce E-R, Stieger P, Mandel T, Baltensperger K,
Bennett M, Traas J, Friml J, Kuhlemeier C: Regulation of
phyllotaxis by polar auxin transport. Nature 2003, 426:255-260.
37. Benková E, Michniewicz M, Sauer M, Teichmann T, Seifertová D,
Jürgens G, Friml J: Local, efflux-dependent auxin gradients
as a common module for plant organ formation. Cell 2003,
115:591-602.
55. Reinhardt D, Mandel T, Kuhlemeier C: Auxin regulates the
initiation and radial position of plant lateral organs.
Plant Cell 2001, 12:507-518.
56. Sieber P, Schorderet M, Ryser U, Buchala A, Kolattukudy P,
Métraux JP, Nawrath C: Transgenic Arabidopsis plants
expressing a fungal cutinase show alterations in the structure
and properties of the cuticle and postgenital organ fusions.
Plant Cell 2000, 12:721-738.
39. Smith RS, Guyomarc’h S, Mandel T, Reinhardt D, Kuhlemeier C,
Prusinkiewicz P: A plausible model of phyllotaxis. Proc Natl
Acad Sci USA 2006, 103:1301-1306.
57. Scarpella E, Marcos D, Friml J, Berleth T: Control of leaf vascular
patterning by polar auxin transport. Genes Dev 2006,
20:1015-1027.
Analysis of PIN protein pattern within developing leaves indicates that the
predicted flux of auxin prescribes the future pattern of vascular differentiation within the leaf, leading to a sequential increase in vascular
complexity coupled with leaf growth.
40. de Reuille Barbier PB, Bohn-Courseau I, Ljung K, Morin H,
Carraro N, Godin C, Traas J: Computer simulations reveal
properties of the cell–cell signaling network at the shoot apex
in Arabidopsis. Proc Natl Acad Sci USA 2006, 103:1627-1632.
58. Paciorek T, Zazimalová E, Ruthardt N, Petrásek J, Stierhof Y-D,
Kleine-Vehn J, Morris DA, Emans N, Jürgens G, Geldner N, Friml J:
Auxin inhibits endocytosis and promotes its own efflux from
cells. Nature 2005, 435:1251-1256.
41. Beck CB: Predominance of Archaeopteris in an upper
Devonian flora of western Catskills and adjacent
Pennsylvania. Bot Gaz 1964, 125:125-128.
59. Hao SG, Beck CB: Further observations on Eophyllophyton
bellum from the Lower Devonian (Siegenian) of Yunnan, China.
Palaeontographica 1993, B230:27-41.
42. Boyce CK, Knoll AH: Evolution of developmental potential and
the multiple independent origins of leaves in Paleozoic
vascular plants. Paleobiology 2002, 28:70-100.
60. Hao SG, Beck BC, Deming W: Structure of the earliest leaves:
adaptations to high concentrations of CO2. Int J Plant Sci 2003,
164:71-75.
43. Osborne CP, Beerling DJ, Lomax BH, Chaloner WG: Biophysical
constraints on the origin of leaves inferred from the fossil
record. Proc Natl Acad Sci USA 2004, 101:10360-10362.
61. Beerling DJ, Osborne CP, Chaloner WG: Evolution of leaf-form in
land plants linked to atmospheric CO2 decline in the Late
Palaeozoic. Nature 2001, 410:352-354.
44. Steward WN, Rothwell GW: Paleobotany and the Evolution of
Plants. 2nd Edn. Cambridge University Press; 1993.
62. Woodward FI: Stomatal numbers are sensitive to CO2
increases from pre- industrial levels. Nature 1987,
327:617-618.
38. Jönsson H, Heisler MG, Shapiro BE, Meyerowitz EM, Mjolsness E:
An auxin-driven polarized transport model for phyllotaxis.
Proc Natl Acad Sci USA 2006, 103:1633-1638.
45. Knaus MJ, Upchurch GR, Gillespie WH: Charbeckia macrophylla
gen. et sp. nov. from the lower Mississippian price (Pocono)
formation in southeastern Virginia. Rev Palaeobot Palynol 2000,
111:71-92.
46. Waites R, Selvadurai HRN, Oliver IR, Hudson A: The
PHANTASTICA gene encodes a MYB transcription factor
involved in growth and dorsoventrality of lateral organs in
Antirrhinum. Cell 1998, 93:779-789.
47. Siegfried KR, Eshed Y, Baum SF, Otsuga D, Drews GN,
Bowman JL: Members of the YABBY gene family specify
abaxial cell fate in Arabidopsis. Development 1999,
126:4117-4128.
48. Kerstetter RA, Bollman K, Taylor RA, Bomblies K, Poethig RS:
KANADI regulates organ polarity in Arabidopsis. Nature 2001,
411:706-709.
49. McConnell JR, Emery J, Eshed Y, Bao N, Bowman J, Barton MK:
Role of PHABULOSA and PHAVOLUTA in determining radial
patterning in shoots. Nature 2000, 411:709-713.
50. Bharathan G, Goliber TE, Moore C, Kessler S, Pham T, Sinha NR:
Homologies in leaf form inferred from KNOXI gene expression
during development. Science 2002, 296:1858-1860.
51. McHale NA, Koning RE: PHANTASTICA regulates development
of the adaxial mesophyll in Nicotiana leaves. Plant Cell 2004,
16:1251-1262.
Current Opinion in Plant Biology 2007, 10:4–12
63. Gray JE, Holroyd GH, van der Lee FM, Bahrami AR, Sijmons PC,
Woodward FI, Schuch W, Hetherington AM: The HIC signalling
pathway links CO2 perception to stomatal development.
Nature 2000, 408:713-716.
64. Bezanilla M, Pan A, Quatrano RS: RNA interference in the moss
Physcomitrella patens. Plant Physiol 2003, 133:470-474.
65. Hiwatashi Y, Nishiyama T, Fujita T, Hasebe M: Establishment of
gene-trap and enhancer-trap systems in the moss
Physcomitrella patens. Plant J 2001, 28:105-116.
66. Nishiyama T, Fujita T, Shin-I T, Seki M, Nishide H, Uchiyama I,
Kamiya A, Carninci P, Hayashizaki Y, Shinozaki K et al.:
Comparative genomics of Physcomitrella patens
gametophytic transcriptome and Arabidopsis thaliana:
implications for land plant evolution. Proc Natl Acad Sci USA
2003, 100:8007-8012.
67. Sakakibara K, Nishiyama T, Sumikawa N, Kofuji R, Murata T,
Hasebe M: Involvement of auxin and a homeodomain-leucine
zipper I gene in rhizoid development of the moss
Physcomitrella patens. Development 2003, 130:4835-4846.
68. Sano R, Juarez CM, Hass B, Sakakibara K, Ito M, Banks JA,
Hasebe M: KNOX homeobox genes potentially have similar
function in both diploid unicellular and multicellular
meristems, but not in haploid meristems. Evol Dev 2005,
7:69-78.
www.sciencedirect.com