Directory UMM :Data Elmu:jurnal:P:PlantScience:Plant Science_BioMedNet:201-220:
trends in plant science
perspectives
16 Hunter, J.P. (1998) Key innovations and the ecology of macroevolution,
Trends Ecol. Evol. 13, 31–36
17 Tomlinson, P.B. (1994) Functional morphology of saccate pollen in
conifers with special reference to Podocarpaceae, Int. J. Plant Sci. 155,
699–715
18 Wilson, V. and Owens, J.N. The reproductive cycle in Podocarpus totara,
Am. J. Bot. (in press)
19 Tomlinson, P.B., Braggins, J.E. and Rattenbury, J.A. (1991) Pollination drop
in relation to cone morphology in Podocarpaceae: a novel reproductive
mechanism, Am. J. Bot. 78, 1289–1303
20 Takaso, T. and Owens, J.N. (1995) Pollination drop and microdrop secretions
in Cedrus, Int. J. Plant Sci. 156, 640–649
21 Singh, H. and Owens, J.N. (1982) Sexual reproduction in grand fir (Abies
grandis), Can. J. Bot. 60, 2197–2214
22 Owens, J.N., Simpson, S.J. and Molder, M. (1981) The pollination mechanism
and the optimal time of pollination in Douglas-fir (Pseudotsuga menziesii),
Can. J. For. Res. 11, 36–50
23 Owens, J.N., Morris, S. and Catalano, G. (1994) How the pollination
mechanism and prezygotic and postzygotic events affect seed production in
Larix occidentalis, Can. J. For. Res. 24, 917–927
24 Takaso, T. and Owens, J.N. (1996) Postpollination-prezygotic ovular
secretions into the micropylar canal in Pseudotsuga menziesii (Pinaceae),
J. Plant Res. 109, 147–160
25 Haines, R.J., Prakash, N. and Nikles, D.G. (1984) Pollination in Araucaria
Juss., Aust. J. Bot. 32, 583–594
26 Eames, A.J. (1913) The morphology of Agathis australis, Ann. Bot. 27, 1–36
27 Owens, J.N. et al. (1995) The reproductive biology of Kauri (Agathis australis).
I. Pollination and prefertilization development, Int. J. Plant Sci. 156, 257–269
28 Singh, H. (1978) Embryology of Gymnosperms, Gebrüder Borntraeger
29 Colangeli, A.M. and Owens, J.N. (1989) Postdormancy seed-cone
development and the pollination mechanism in western hemlock (Tsuga
heterophylla), Can. J. For. Res. 19, 44–53
30 Owens, J.N. and Blake, M.D. (1983) Pollen morphology and development of
the pollination mechanisms in Tsuga heterophylla and T. mertensiana, Can. J.
Bot. 61, 3041–3048
31 McWilliam, J.R. (19958) The role of the micropyle in the pollination of Pinus,
Bot. Gaz. (Chicago) 120, 109–117
32 Ziegler, H. (1959) Uber die Zusammensetzung des Ëbestaubungstropfensí und
den Mechanismus seiner Sekretion, Planta 52, 587–599
33 Serdi-Benkaddour, R. and Chesnoy, L. (1985) Secretion and composition of
the pollination drop in the Cephalotaxus drupacea (Gymnosperm,
Cephalotaxeae), in Sexual Reproduction in Higher Plants (Cristi, M., Gori, P.
and Pacini, E., eds), pp. 345–350, Springer-Verlag
34 Runions, C.J. and Owens, J.N. Evidence of prezygotic self-incompatibility in a
gymnosperm, in Proceedings: Reproductive Biology ’96 in Systematics,
Conservation and Economic Botany (1–5 Sept. 1996), Royal Botanical
Gardens, Kew, UK (in press)
John N. Owens* is at the Centre for Forest Biology,
PO Box 3020 STN CSC, Victoria, BC, Canada V8W 3N5;
Tokushiro Takaso is at the Iromote Station, Tropical Biosphere
Research Centre, University of the Ryukyus, 870 Uehara,
Taketomi-cho, Okinawa 907-1541, Japan;
C. John Runions is in the Section of Ecology and Systematics,
Corson Hall, Cornell University, Ithaca, NY 14853-2701, USA.
*Author for correspondence (tel 11 250 721 7113;
fax 11 250 721 6611; e-mail [email protected]).
Ecological and evolutionary
genetics of Arabidopsis
Massimo Pigliucci
The crucifer Arabidopsis thaliana has been the subject of intense research into
molecular and developmental genetics. One of the consequences of having
this wealth of physiological and molecular data available, is that ecologists
and evolutionary biologists have begun to incorporate this model system into
their studies. Current research on A. thaliana and its close relatives ably
illustrates the potential for synergy between mechanistic and organismal
biology. On the one hand, mechanistically oriented research can be placed in
an historical context, which takes into account the particular phylogenetic
history and ecology of these species. This helps us to make sense of
redundancies, anomalies and sub-optimalities that would otherwise be difficult
to interpret. On the other hand, ecologists and evolutionary biologists now
have the opportunity to investigate the physiological and molecular basis for
the phenotypic changes they observe. This provides new insight into the
mechanisms that influence evolutionary change.
B
iology is experiencing the age of
model systems1. Our present understanding of genetics would have been
very different if laboratories throughout the
world had not agreed to concentrate their efforts on the fruit fly Drosophila melanogaster
at the beginning of the century. Similarly,
different branches of biology have adopted
distinct organisms as being particularly convenient for the type of study at hand. As a consequence, we have considerable knowledge of
the physiology of mice, the developmental
biology of sea urchins, the molecular biology
of Escherichia coli, and an understanding of
disease resistance in tobacco. There are, of
course, limits to this strategy of focusing on a
1360 - 1385/98/$ – see front matter © 1998 Elsevier Science. All rights reserved. PII: S1360-1385(98)01343-0
reduced number of organisms. Although it has
been possible to understand their biology in
depth, it is also clear that we are forfeiting anything more than a superficial knowledge of the
overwhelming majority of living organisms.
Fortunately, research in evolutionary biology can help to broaden the scope of our
investigations. All organisms were derived
from a single common ancestor, which is why
they share the same genetic/molecular machinery. Thus, we can apply what we learn
about a small number of organisms to the majority – at least as long as we do not extrapolate too far from our starting point in either
ecological or phylogenetic space. The real
question is how many model systems we need,
and how far these generalizations can reasonably be extended.
Arabidopsis as a model system
Arabidopsis thaliana (L.) Heynh. is a small
annual, white-flowered member of the Brassicaceae family, and is allied to other crucifers
such as mustard, Brassica napus and broccoli.
Arabidopsis thaliana was first adopted as a
model system in plant genetics in the 1950s,
largely as a target for mutagenesis studies2.
More recently, A. thaliana has been the focus
of physiological, developmental and genetic
research that has made it the reference point
for plant molecular biology3.
December 1998, Vol. 3, No. 12
485
trends in plant science
perspectives
From the standpoint of reductionist biology,
the main advantages of A. thaliana reside in its
small genome and reduced number of chromosomes; the ease with which it can be transformed; the fact that it is easy to grow at high
density; and the fast life cycle of some of its
ecotypes. The word ecotype is used loosely in
the case of A. thaliana. The original use of the
term4 was to indicate a genetically adapted ecological specialist, such as an alpine form, or a
heavy metal tolerant form. Arabidopsis thaliana’s ecotypes are usually distinguished into
‘early’ and ‘late’ flowering, depending on the
length of their vegetative phase and their type
of phenology – spring ephemerals or winter
annuals.
Recently, however, this species and its
close relatives have also been the subject of
ecological and evolutionary research5–7. The
underlying idea is to exploit the wealth of information on the molecular and physiological
machinery of A. thaliana to shed some light
on ecological and evolutionary questions. The
following review briefly summarizes what is
known about the ecology and evolution of A.
thaliana and its close phylogenetic neighbors.
The ultimate aim of this is to facilitate the integration of many biological disciplines and to
encourage a multifaceted approach to understanding a single group of organisms.
A simple plant?
Arabidopsis thaliana is a relatively simple
flowering plant, in that it is a typical annual
weed, equipped with a minimum phenotype
and characterized by a straightforward life
cycle. The phenotype of A. thaliana features
a filamentous root system, a basal rosette of
leaves, a main flowering stem (which usually
develops lateral branches), and optional additional flowering branches stemming from the
base of the plant. Its life cycle is roughly divided
into four stages. First, the germinating seedling elongates and etiolates; the cotyledons
then open and the first true leaves are formed.
Second, the rosette grows to a number of leaves
over a period of time that varies with the population and the environmental conditions7 –
plants can produce anywhere between five to
six and hundreds of leaves. Third, the main
stem begins to elongate (bolting), and the first
flowers open within a few days. Under favorable environmental conditions, additional basal
meristems also develop into inflorescences, so
that the total fruit/seed output can vary dramatically. Finally, leaves and fruits begin to
senesce, the fruits split open, and the seeds are
released. Seeds can germinate immediately or
go through a period of dormancy, again depending on the population and the environment.
This apparently simple scenario is complicated by a variety of responses to environmental conditions, which have been the focus of
much ecological research. In particular, the time
486
December 1998, Vol. 3, No. 12
of flowering can vary considerably, thereby
affecting the entire life cycle and reproductive fitness of the plant. At least four groups
of populations have been distinguished8. A
very early flowering kind (presumably spring
ephemerals) bolts within 40 days of sowing
(this includes virtually all the ecotypes used in
molecular genetic research), and does not respond to vernalization unless grown under short
days. A second type (probably early summer
annuals) flowers somewhat later and responds
only slightly to vernalization. In the third class
(which is thought to correspond to the ecological type of late summer annuals) bolting is delayed even further, and this effect is heightened
by a lack of vernalization. The fourth group behave as winter annuals and flower in four to six
months, although this can be reduced to a single month by vernalization (e.g. 70 days at 28C).
It has to be pointed out that this classification
is based on physiological data alone, and that
we still lack convincing field evidence that the
assumed ecology of these types corresponds to
their habit under natural conditions. No examples of biennial A. thaliana have been described
to date, although some of its relatives are biennial, and some genotypes do not flower within
one season if exposed to short photoperiod
(M. Pigliucci and H. Callahan, unpublished).
From an ecological perspective it is interesting that there is no clear association between
the flowering habit of A. thaliana and its geographic distribution. Populations from both
Scandinavia and Tennessee are winter-annual,
as are the genotypes from warm and dry Moravia and wet and cold Rhineland in Europe.
Furthermore, Cologne (Germany) and Grenoble
(France) are characterized by similar temperatures and precipitation, but they host summer annuals and winter annuals, respectively. Several
genes have been found to affect both flowering
time and its response to photoperiod and vernalization (the effect is also mediated by gibberellic acid)9. Thus, this deceptively simple
aspect of the plant’s ecological genetics is likely
to keep researchers busy for some time to come.
Another supposedly established fact about
A. thaliana is that it is predominantly a selfing species. Interestingly, this characteristic is
usually invoked as an objection to its use in
evolutionary studies: selfed organisms are considered to be evolutionary dead-ends, because
their populations do not harbor significant
genetic variation. Although an estimate of the
outcrossing rate of this species gave a value of
0.3% based on allozyme markers10, several
lines of evidence point to higher values (around
1.2–2.2%)11, implying that it is a more interesting subject for studies on population genetics.
Even the allozyme studies revealed 45 different multilocus genotypes based on seven polymorphic loci when 470 individuals were
sampled from 16 populations (with up to eight
genotypes per population); these figures are
not indicative of low genetic variation. Furthermore, quantitative genetic studies of several
characters in A. thaliana have shown significant heritability and genetic variation both
within and among populations12. Also, molecular studies using DNA sequencing of the
inter-spacer (ITS) region of the chloroplast
genome have revealed remarkable variation
both among and within populations (M. Cruzan
and M. Pigliucci, unpublished). Finally, protogyny may significantly contribute to outcrossing in A. thaliana (I. Al-Shehbaz, pers.
commun.). Clearly, there is the need to further
assess the mating system and population
genetics of this species.
The complexities of weed ecology
Arabidopsis thaliana is an excellent example
of what a weed can do from an ecological standpoint. Although it is rarely locally abundant
(its populations seldom count more than a few
scattered individuals or clumps), and it does
not withstand competition with grasses or other
tall vegetation, its geographical range is nonetheless impressive. The northernmost limit of
its distribution is northern Scandinavia, and
the southernmost tip of its range is in northern
Africa and the Middle East. To date, and in
spite of all we know about the physiology and
molecular biology of A. thaliana, no systematic investigation of what makes this astounding geographical range possible with such a
‘simple’ phenotype has been undertaken.
What is known, is that A. thaliana responds
dramatically to environmental variation over
a wide range of spatial scales. When used for
bioassays in a series of explant trials within an
area of 50 m2 of undisturbed forest, it was
found that the environmental variance (measured by the response of the plant) between
sites, increased with distance5. However, the
correlation between sites decreased with separation in the same fashion at all spatial scales
examined. In other words, environments – as
perceived by A. thaliana – display the same
degree of complexity at all spatial scales.
An elegant experimental design was used
to investigate the response of A. thaliana to
selection over very short spatial scales13. Random combinations of seeds of three phenotypically distinguishable lines were sown
over and across a series of 5 3 1 m transects.
If no selection occurred, one would expect to
find random mixtures of mature plants at the
end of the season. Instead, the results showed
a very significant tendency for genotypes of the
same kind to cluster together, and eliminate
the other two kinds from each microsite. The
success of a given genotype depended on the
microsite, and the range of this effect was less
than 50 cm. This suggests that very localized
environmental conditions may favor one or another genotype, yielding an apparently random
patchwork of genotypes across the landscape.
trends in plant science
perspectives
(a)
(b)
250
17
Number of fruits
Number of leaves
15
13
11
9
7
200
150
100
50
5
3
0
Neutral shade
Low R:FR
Neutral shade
Treatment
Low R:FR
Treatment
Fig. 1. Reaction norms for number of rosette leaves (a) and fruit production (b) in response to altered light quality [low Red (R):Far Red (FR),
on the right side of the abscissa, simulating canopy shade]. Landsberg erecta (blue line) is an inbred line used under standard laboratory conditions. The mutants functionally lack all phytochromes, and therefore cannot perceive red light; they are the classic hy1 (red line) and hy2
(green line) chromophore mutants, which correspond to two different loci with similar biochemical and phenotypic effects. Notice the similarity between the reaction norms of both wild type and mutants for the two traits.
Because the distance over which the homogenizing effect of selection occurred turned out
to be very small (with random distributions
above that scale), one can reasonably infer that
most of A. thaliana’s ecology is local.
However, these experiments did not address
the factors responsible for this extremely localized response to selection. To address this
point, an investigation into the spatiotemporal
effects of nitrogen and litter availability was
carried out14. It was found that some English
populations undergo two life cycles within a
given year. Some plants flower in the spring
(May–June), and their offspring do not go
through dormancy, but germinate immediately
and flower in late summer (September–October). The seeds of the summer generation then
overwinter, to germinate the following spring.
These investigations revealed that litter from
the spring individuals fosters reproductive output in the summer generation (mostly very small
individuals responding to high density – a result of phenotypic plasticity). By contrast, the
overwintering seeds cannot use the litter of the
previous generation because it is washed away.
However, because they are at a much lower
population density, they produce plants with
large rosettes, which accumulate photosynthate
and yield a significant number of siliques.
These experiments demonstrate how little
we know about the ecology of A. thaliana. For
example, we do not know how widespread the
two-generation cycle is, or which ecological
factors may determine its occurrence. Furthermore, because there have been few experimental studies of A. thaliana in the field5,14,15,
we are in no position to interrelate the microand macro-ecology of this plant.
Phenotypic plasticity
Several laboratories have focused on the use
of A. thaliana in studies on phenotypic plasticity – the way in which genotypes respond
to heterogeneous environmental conditions7,16.
The phytochrome genes in A. thaliana (which
control response to light availability, light
quality and photoperiod) are presently some of
the best known examples of plasticity genes.
A plasticity gene is a receptor of environmental change that effects a cascade of specific
and fitness-enhancing developmental/phenotypic changes17. An investigation into the plasticity of a wild-type inbred line and of several
single-gene mutants representing potential
plasticity genes is summarized in Fig. 1. This
study demonstrates that the elimination of all
of the phytochromes (which control the response to canopy shade), transforms the plastic wild type into a genotype that is totally
unresponsive to this specific environmental
change, although there is no effect on other
kinds of plasticity. This lack of plasticity directly translates into a proportional reduction
in reproductive fitness of the mutants (which
are otherwise physiologically and developmentally normal). This investigation demonstrates the adaptive implications of phenotypic
plasticity18.
Contrary to expectations based on its putative selfing habit, A. thaliana shows a tremendous degree of variation for phenotypic
plasticity to an array of environmental circumstances. A study of four populations (two early
flowering and two late flowering ecotypes), revealed significant responses to water, nutrient
and light gradients, with significant genetic
variation among populations for several traits in
reaction to nutrient and light availability19. A
larger survey of 26 populations found significant population by environment interaction
(i.e. genetic variation for plasticity) for five
out of nine traits20. Size-fecundity relationships also varied significantly in response to
light, nutrient and pot volume gradients21. This
latter study identified a minimum size threshold that the plant has to reach to be able to
flower. However, the value of the threshold is
itself environmentally dependent (i.e. plastic).
Other researchers have used A. thaliana as
a model system to explore the effects of herbivory15, and the interaction between herbivory and abiotic factors22. A poorly explored
aspect of plasticity in A. thaliana (and in other
plants and animals as well) is the interaction
between development and environment. Ecological studies have demonstrated developmental stage-specific variation for plasticity12,
and research carried out on mutants has pinpointed the complex and antagonistic molecular basis of developmental windows of
plasticity23. Although a general framework for
this kind of research has not yet emerged, it is
a growing area and demonstrates the trend for
a direct interaction between ecological and
molecular genetics.
Quantitative genetics and constraints
Arabidopsis thaliana is surprisingly genetically variable, both at the DNA sequence and
at the quantitative genetic level. This has afforded some interesting evolutionary studies
of this plant. One old problem in evolutionary
genetics is represented by dominance24,25. To
understand the evolution of dominance we
need studies addressing both its proximate
causes (the genetic and physiological bases),
and its ultimate causes (i.e. under what circumstances natural selection favors a certain
degree of dominance). It has been shown that
dominance of two mutations in A. thaliana
that affect the production of abscisic acid is
modulated by maternal effects26, which provides the proximate cause of dominance in this
instance. This kind of work can be integrated
into studies on the selective advantage of different degrees of dominance and of the evolutionary dynamics of maternal effects, to yield
insights into the broader question of why
dominance evolves.
December 1998, Vol. 3, No. 12
487
trends in plant science
perspectives
Leaf number
(a)
12
11
10
9
8
7
6
5
4
3
0
500
Fitness
D
W
0
L
5
10 15 20 25
Bolting time (days)
30
Leaf number
(b)
12
11
10
9
8
7
6
5
4
3
0
500
Fitness
D
L
W
0
5
10 15 20 25
Bolting time (days)
30
Leaf number
(c)
12
11
10
9
8
7
D
6
5 W
4
3
0
5
0
500
Fitness
L
10 15 20 25
Bolting time (days)
December 1998, Vol. 3, No. 12
The big picture: evolution and
systematics of Arabidopsis
No matter how interesting a model system is
from a molecular or even an ecological standpoint, the crucial test of its relevance from an
evolutionary perspective is provided by the
exploration of its phylogenetic neighborhood.
Until recently, the systematics of Arabidopsis
was confusing34, but several authors have concentrated on revising the situation, applying
modern molecular and cladistic analyses to the
group34–36.
Many of the species that used to be classified as being Arabidopsis are currently considered as being fairly distant relatives, whereas
taxa once in Arabis or Cardaminopsis are now
thought of as being Arabidopsis in the strictest
sense. The emerging phylogenetic picture of
this genus immediately creates the opportunity
to examine interesting questions from an evolutionary perspective. For example, A. thaliana is an annual, but several of its relatives are
biennials or perennials. How often does a new
flowering habit emerge through history? How
are these evolutionary shifts related to the
ecology, and therefore what are the selective
pressures these plants experience? Furthermore, not all members of Arabidopsis are
diploid. In fact, one of A. thaliana’s closest
relatives is A. arenosa, which is a tetraploid.
Recent research has confirmed an old speculation that A. thaliana and A. arenosa are
respectively the female and male parent of a
third taxon, A. suecica37,38. The wealth of molecular and physiological information available
for A. thaliana opens up previously unforeseen possibilities for investigating the ecology
and evolution of hybridization and polyploidy
(as well as the evolution of mating systems),
because A. thaliana is predominantly selfcrossing, while the other two are outcrossing.
30
Fig. 2. Relationship between bolting time
and leaf number in Arabidopsis thaliana.
The solid red circles represent the three
wild-type [Dijon (D), Landsberg (L) and
Wassilewskija (W)] populations. All
other circles (diameter proportional to reproductive fitness, measured as the number of fruit produced) are EMS mutants
from the (a) Dijon, (b) Landsberg and (c)
Wassilewskija backgrounds. Note how
genetic variation generated by mutation
can dramatically alter the relationship between the two traits. Also, it is interesting to note that – contrary to expectation
– many mutants display a higher reproductive fitness than the wild types. This
effect might be because the three wild
types have been selected for a very short
life cycle, which had the pleiotropic effect
of reducing the reproductive output (M.
Pigliucci and N. Turner, unpublished).
488
Recently, quantitative genetics has acquired a new tool with which to merge evolutionary and molecular genetics. It is now
possible to use molecular markers to identify
and map the positions of genetic regions
affecting quantitative characters in natural
populations. These elements are known as
Quantitative Trait Loci (QTLs)27. QTL mapping has been used to study phenotypic plasticity, flowering time and life history28,29.
However, a word of caution is necessary. If
the stated goal is to investigate loci affecting
life-history evolution in natural populations, it
is inappropriate to compare artificially derived
lines grown under continuous light30, as this is
an unnatural condition. Care has to be taken to
prevent our enthusiasm for interdisciplinary
research and new techniques from getting
in the way of our understanding of the underlying biology.
Another long-standing evolutionary problem is the question of constraints31,32. A very
well known genetic constraint in A. thaliana
is represented by a strong phenotypic and
genetic correlation between flowering time
and leaf production (i.e. between size and age
at maturity)30,33. It is interesting that the intensity of this constraint depends on the
genetic background, as well as on the environment in which the plant grows. It is also
possible to break the constraint to some extent, and obtain plants with a very different
combination of the two traits compared to
the wild type. This can be achieved by inducing mutations in different backgrounds,
and then comparing the position in phenotypic space of the mutants and the respective
wild-type (Fig. 2). Such an approach represents one of the first experimental attempts
to address the evolutionary problem of
constraints.
A. thaliana – Columbia Derivative
A. thaliana – Columbia
A. thaliana – Landsberg erecta
A. thaliana – Landsberg
A. thaliana – Nossen
A. thaliana – Moscow
Hylandra
Cardaminopsis petraea
A. griffithiana
A. pumila
0
20
40
60
Fig. 3. Evolution of phenotypic plasticity of leaf production in response to light quality in
Arabidopsis and allied species. The length of the bars to the right of each taxon indicates
the intensity of the plastic response: the color of the bar indicates distribution in Central
Europe (yellow), Eastern Europe (red) and Northern Europe (blue). Hylandra (also known as
A. suecica) indicates a hybrid between A. thaliana and (presumably) A. arenosa (belonging
to the Cardaminopsis clade). The phylogenetic position of A. thaliana’s ecotypes is inferred
from their genetic distances. Note how strong or weak plasticity has apparently evolved more
than once in the whole group.
trends in plant science
perspectives
The availability of a phylogeny of the Arabidopsis genus and its more or less distant relatives
has also afforded some first-time comparative
analyses addressing evolutionary ecological
questions. One such question concerns the evolution of phenotypic plasticity. For example, in
the case of plasticity of leaf production, a high
degree of plasticity occurs in some distant relatives outside A. thaliana’s clade, but it is lost in
early flowering taxa, only to re-evolve in late
flowering ones (Fig. 3). It would be interesting
to map the evolution of the known genes affecting flowering time upon the same cladogram.
Can we put everything together?
Attempts to unify biology have characterized
the history of this discipline throughout the
20th century. One of the most important results
of this trend has been the neo-darwinian synthesis that led to modern evolutionary theory.
Recently, many people have invoked a longawaited merger between organismal and molecular/developmental biology. This goal now
appears attainable thanks to the reliability, sophistication, and low cost of modern molecular
techniques. However, given the limited funding
for the basic sciences, and the relatively small
number of scientists that study either molecular
or evolutionary biology, a true synthesis seems
possible – at least at present – only by concentrating on a few well chosen organisms. Together with other classical model systems, such
as E. coli and Drosophila, Arabidopsis and its
relatives are clearly gaining permanent status
as members of this intensively studied group.
Acknowledgements
I wish to thank Jacqui Shykoff and PierreHenri Guyon, of the University of Paris-Sud,
for their hospitality and discussions while
preparing this review. Hilary Callahan, Mark
Camara and Carolyn Wells have been instrumental in developing some of the ideas discussed in this paper. Melissa Brenneman,
Jacqui Shykoff and Ihsan Al-Shehbaz have
also made helpful comments.
References
1 Kellogg, E.A. and Shaffer, H.B. (1993) Model
organisms in evolutionary studies, System. Biol.
42, 409–414
2 Redei, G.P. (1992) A heuristic glance at the past of
Arabidopsis genetics, in Methods in Arabidopsis
Research (Koncz, N-H.C.C. and Schell, J., eds),
pp. 1–15, World Scientific, Singapore
3 Goodman, H.M., Eckers, J.R. and Dean, C.
(1995) The genome of Arabidopsis thaliana,
Proc. Natl Acad. Sci. U. S. A. 92, 10831–10835
4 Turesson, G. (1922) The genotypical response of
the plant species to the habitat, Hereditas 3, 211–350
5 Bell, G. and Lechowicz, M.J. (1991) The ecology
and genetics of fitness in forest plants. I.
Environmental heterogeneity measured by
explant trials, J. Ecol. 79, 663–685
6 Aarssen, L.W. and Clauss, M.J. (1992) Genotypic
variation in fecundity allocation in Arabidopsis
thaliana, J. Ecol. 80, 109–114
7 Pigliucci, M. and Schlichting, C.D. (1998)
Reaction norms of Arabidopsis. V. Flowering
time controls phenotypic architecture in response
to nutrient stress, J. Evol. Biol. 11, 285–301
8 Napp-Zinn, K. (1985) Arabidopsis thaliana, in
CRC Handbook of Flowering (Halevy, A.H., ed.),
pp. 492–503, CRC, Boca Raton, FL
9 Coupland, G. (1995) Genetic and environmental
control of flowering time in Arabidopsis, Trends
Genet. 11, 393–397
10 Abbott, R.J. and Gomes, M.F. (1989) Population
genetic structure and outcrossing rate of Arabidopsis
thaliana (L.) Heynh, Heredity 62, 411–418
11 Snape, J.W. and Lawrence, M.J. (1971) The
breeding system of Arabidopsis thaliana,
Heredity 27, 299–302
12 Pigliucci, M. (1997) Ontogenetic phenotypic
plasticity during the reproductive phase in
Arabidopsis thaliana (Brassicaceae), Am. J. Bot.
84, 887–895
13 Stratton, D.A. and Bennington, C.C. (1996)
Measuring spatial variation in natural selection
using randomly-sown seeds of Arabidopsis
thaliana, J. Evol. Biol. 9, 215–228
14 Thompson, L. (1994) The spatiotemporal effects
of nitrogen and litter on the population dynamics
of Arabidopsis thaliana, J. Ecol. 82, 63–68
15 Mauricio, R. and Rausher, M.D. (1997)
Experimental manipulation of putative selective
agents provides evidence for the role of natural
enemies in the evolution of plant defense,
Evolution 51, 1435–1444
16 Zhang, J. and Lechowicz, M.J. (1994)
Correlation between time of flowering and
phenotypic plasticity in Arabidopsis thaliana
(Brassicaceae), Am. J. Bot. 81, 1336–1342
17 Pigliucci, M. (1996) How organisms respond to
environmental changes: from phenotypes to
molecules (and vice versa), Trends Ecol. Evol.
11, 168–173
18 Dudley, S.A. and Schmitt, S.A. (1996) Testing
the adaptive plasticity hypothesis: densitydependent selection on manipulated stem length
in Impatiens capensis, Am. Nat. 147, 445–465
19 Pigliucci, M., Whitton, J. and Schlichting, C.D.
(1995) Reaction norms of Arabidopsis. I. Plasticity
of characters and correlations across water, nutrient
and light gradients, J. Evol. Biol. 8, 421–438
20 Pigliucci, M. and Schlichting, C.D. (1995)
Reaction norms of Arabidopsis (Brassicaceae).
III. Response to nutrients in 26 populations from a
worldwide collection, Am. J. Bot. 82, 1117–1125
21 Clauss, M.J. and Aarssen, L.W. (1994)
Phenotypic plasticity of size-fecundity
relationships in Arabidopsis thaliana, J. Ecol. 82,
447–455
22 Grant-Petersson, J. and Renwick, J.A.A. (1996)
Effects of ultraviolet-B exposure of Arabidopsis
thaliana on herbivory by two crucifer-feeding
insects (Lepidoptera), Physiol. Chem. Ecol. 25,
135–142
23 Mozley, D. and Thomas, B (1995) Developmental
and photobiological factors affecting photoperiodic
induction in Arabidopsis thaliana Heynh.
Landsberg erecta, J. Exp. Bot. 46, 173–179
24 Kacser, H. and Burns, J.A. (1981) The molecular
basis of dominance, Genetics 97, 639–666
25 Keightley, P.D. and Kacser, H. (1987)
Dominance, pleiotropy and metabolic structure,
Genetics 117, 319–329
26 Finkelstein, R.R. (1994) Maternal effects govern
variable dominance of two abscisic acid response
mutations in Arabidopsis thaliana, Plant Physiol.
105, 1203–1208
27 Mitchell-Olds, T. (1995) The molecular basis of
quantitative genetic variation in natural
populations, Trends Ecol. Evol. 10, 324–327
28 Kuittinen, H., Sillanpaa, M.J. and Savolainen, O.
(1997) Genetic basis of adaptation: flowering
time in Arabidopsis thaliana, Theor. Appl. Genet.
95, 573–583
29 van der Schaar, W. et al. (1997) QTL analysis of seed
dormancy in Arabidopsis using recombinant inbred
lines and MQM mapping, Heredity 79, 190–200
30 Mitchell-Olds, T. (1996) Genetic constraints on
life-history evolution: quantitative-trait loci
influencing growth and flowering in Arabidopsis
thaliana, Evolution 50, 140–145
31 Maynard Smith, J. et al. (1985) Developmental
constraints and evolution, Q. Rev. Biol. 60, 265–287
32 Schlichting, C.D. and Pigliucci, M. (1998)
Phenotypic Evolution. A Reaction Norm
Perspective, Sinauer
33 Koornneef, M., Hanhart, C.J. and
van den Veen, J.H. (1991) A genetic and physiological analysis of late flowering mutants in
Arabidopsis thaliana, Mol. Gen. Genet. 229, 57–66
34 Price, R.A., Palmer, J.D. and Al-Shehbaz, I.A.
(1994) Systematic relationships of Arabidopsis:
a molecular and morphological perspective, in
Arabidopsis (Meyerowitz, E.M. and
Somerville, C.R., eds), pp. 7–20, Cold Spring
Harbor Laboratory Press
35 Kamm, A. et al. (1995) Analysis of a repetitive
DNA family from Arabidopsis arenosa and
relationships between Arabidopsis species, Plant
Mol. Biol. 27, 853–862
36 O’Kane, S.L.J. and Al-Shehbaz, I.A. (1997)
A synopsis of Arabidopsis (Brassicaceae), Novon
7, 323–327
37 Mummenhoff, K. and Hurka, H. (1995)
Allopolyploid origin of Arabidopsis suecica
(Fries) Norrlin: evidence from chloroplast and
nuclear genome markers, Bot. Acta 108, 449–456
38 O’Kane, S.L.J., Schaal, B.A. and Al-Shehbaz, I.A.
(1996) The origins of Arabidopsis suecica
(Brassicaceae) as indicated by nuclear rDNA
sequences, System. Bot. 21, 559–566
Massimo Pigliucci is at the Depts of Botany,
and Ecology and Evolutionary Biology,
University of Tennessee, Knoxville, TN, USA
(tel 11 423 974 6221; fax 11 423 974 0978;
e-mail [email protected]).
December 1998, Vol. 3, No. 12
489
perspectives
16 Hunter, J.P. (1998) Key innovations and the ecology of macroevolution,
Trends Ecol. Evol. 13, 31–36
17 Tomlinson, P.B. (1994) Functional morphology of saccate pollen in
conifers with special reference to Podocarpaceae, Int. J. Plant Sci. 155,
699–715
18 Wilson, V. and Owens, J.N. The reproductive cycle in Podocarpus totara,
Am. J. Bot. (in press)
19 Tomlinson, P.B., Braggins, J.E. and Rattenbury, J.A. (1991) Pollination drop
in relation to cone morphology in Podocarpaceae: a novel reproductive
mechanism, Am. J. Bot. 78, 1289–1303
20 Takaso, T. and Owens, J.N. (1995) Pollination drop and microdrop secretions
in Cedrus, Int. J. Plant Sci. 156, 640–649
21 Singh, H. and Owens, J.N. (1982) Sexual reproduction in grand fir (Abies
grandis), Can. J. Bot. 60, 2197–2214
22 Owens, J.N., Simpson, S.J. and Molder, M. (1981) The pollination mechanism
and the optimal time of pollination in Douglas-fir (Pseudotsuga menziesii),
Can. J. For. Res. 11, 36–50
23 Owens, J.N., Morris, S. and Catalano, G. (1994) How the pollination
mechanism and prezygotic and postzygotic events affect seed production in
Larix occidentalis, Can. J. For. Res. 24, 917–927
24 Takaso, T. and Owens, J.N. (1996) Postpollination-prezygotic ovular
secretions into the micropylar canal in Pseudotsuga menziesii (Pinaceae),
J. Plant Res. 109, 147–160
25 Haines, R.J., Prakash, N. and Nikles, D.G. (1984) Pollination in Araucaria
Juss., Aust. J. Bot. 32, 583–594
26 Eames, A.J. (1913) The morphology of Agathis australis, Ann. Bot. 27, 1–36
27 Owens, J.N. et al. (1995) The reproductive biology of Kauri (Agathis australis).
I. Pollination and prefertilization development, Int. J. Plant Sci. 156, 257–269
28 Singh, H. (1978) Embryology of Gymnosperms, Gebrüder Borntraeger
29 Colangeli, A.M. and Owens, J.N. (1989) Postdormancy seed-cone
development and the pollination mechanism in western hemlock (Tsuga
heterophylla), Can. J. For. Res. 19, 44–53
30 Owens, J.N. and Blake, M.D. (1983) Pollen morphology and development of
the pollination mechanisms in Tsuga heterophylla and T. mertensiana, Can. J.
Bot. 61, 3041–3048
31 McWilliam, J.R. (19958) The role of the micropyle in the pollination of Pinus,
Bot. Gaz. (Chicago) 120, 109–117
32 Ziegler, H. (1959) Uber die Zusammensetzung des Ëbestaubungstropfensí und
den Mechanismus seiner Sekretion, Planta 52, 587–599
33 Serdi-Benkaddour, R. and Chesnoy, L. (1985) Secretion and composition of
the pollination drop in the Cephalotaxus drupacea (Gymnosperm,
Cephalotaxeae), in Sexual Reproduction in Higher Plants (Cristi, M., Gori, P.
and Pacini, E., eds), pp. 345–350, Springer-Verlag
34 Runions, C.J. and Owens, J.N. Evidence of prezygotic self-incompatibility in a
gymnosperm, in Proceedings: Reproductive Biology ’96 in Systematics,
Conservation and Economic Botany (1–5 Sept. 1996), Royal Botanical
Gardens, Kew, UK (in press)
John N. Owens* is at the Centre for Forest Biology,
PO Box 3020 STN CSC, Victoria, BC, Canada V8W 3N5;
Tokushiro Takaso is at the Iromote Station, Tropical Biosphere
Research Centre, University of the Ryukyus, 870 Uehara,
Taketomi-cho, Okinawa 907-1541, Japan;
C. John Runions is in the Section of Ecology and Systematics,
Corson Hall, Cornell University, Ithaca, NY 14853-2701, USA.
*Author for correspondence (tel 11 250 721 7113;
fax 11 250 721 6611; e-mail [email protected]).
Ecological and evolutionary
genetics of Arabidopsis
Massimo Pigliucci
The crucifer Arabidopsis thaliana has been the subject of intense research into
molecular and developmental genetics. One of the consequences of having
this wealth of physiological and molecular data available, is that ecologists
and evolutionary biologists have begun to incorporate this model system into
their studies. Current research on A. thaliana and its close relatives ably
illustrates the potential for synergy between mechanistic and organismal
biology. On the one hand, mechanistically oriented research can be placed in
an historical context, which takes into account the particular phylogenetic
history and ecology of these species. This helps us to make sense of
redundancies, anomalies and sub-optimalities that would otherwise be difficult
to interpret. On the other hand, ecologists and evolutionary biologists now
have the opportunity to investigate the physiological and molecular basis for
the phenotypic changes they observe. This provides new insight into the
mechanisms that influence evolutionary change.
B
iology is experiencing the age of
model systems1. Our present understanding of genetics would have been
very different if laboratories throughout the
world had not agreed to concentrate their efforts on the fruit fly Drosophila melanogaster
at the beginning of the century. Similarly,
different branches of biology have adopted
distinct organisms as being particularly convenient for the type of study at hand. As a consequence, we have considerable knowledge of
the physiology of mice, the developmental
biology of sea urchins, the molecular biology
of Escherichia coli, and an understanding of
disease resistance in tobacco. There are, of
course, limits to this strategy of focusing on a
1360 - 1385/98/$ – see front matter © 1998 Elsevier Science. All rights reserved. PII: S1360-1385(98)01343-0
reduced number of organisms. Although it has
been possible to understand their biology in
depth, it is also clear that we are forfeiting anything more than a superficial knowledge of the
overwhelming majority of living organisms.
Fortunately, research in evolutionary biology can help to broaden the scope of our
investigations. All organisms were derived
from a single common ancestor, which is why
they share the same genetic/molecular machinery. Thus, we can apply what we learn
about a small number of organisms to the majority – at least as long as we do not extrapolate too far from our starting point in either
ecological or phylogenetic space. The real
question is how many model systems we need,
and how far these generalizations can reasonably be extended.
Arabidopsis as a model system
Arabidopsis thaliana (L.) Heynh. is a small
annual, white-flowered member of the Brassicaceae family, and is allied to other crucifers
such as mustard, Brassica napus and broccoli.
Arabidopsis thaliana was first adopted as a
model system in plant genetics in the 1950s,
largely as a target for mutagenesis studies2.
More recently, A. thaliana has been the focus
of physiological, developmental and genetic
research that has made it the reference point
for plant molecular biology3.
December 1998, Vol. 3, No. 12
485
trends in plant science
perspectives
From the standpoint of reductionist biology,
the main advantages of A. thaliana reside in its
small genome and reduced number of chromosomes; the ease with which it can be transformed; the fact that it is easy to grow at high
density; and the fast life cycle of some of its
ecotypes. The word ecotype is used loosely in
the case of A. thaliana. The original use of the
term4 was to indicate a genetically adapted ecological specialist, such as an alpine form, or a
heavy metal tolerant form. Arabidopsis thaliana’s ecotypes are usually distinguished into
‘early’ and ‘late’ flowering, depending on the
length of their vegetative phase and their type
of phenology – spring ephemerals or winter
annuals.
Recently, however, this species and its
close relatives have also been the subject of
ecological and evolutionary research5–7. The
underlying idea is to exploit the wealth of information on the molecular and physiological
machinery of A. thaliana to shed some light
on ecological and evolutionary questions. The
following review briefly summarizes what is
known about the ecology and evolution of A.
thaliana and its close phylogenetic neighbors.
The ultimate aim of this is to facilitate the integration of many biological disciplines and to
encourage a multifaceted approach to understanding a single group of organisms.
A simple plant?
Arabidopsis thaliana is a relatively simple
flowering plant, in that it is a typical annual
weed, equipped with a minimum phenotype
and characterized by a straightforward life
cycle. The phenotype of A. thaliana features
a filamentous root system, a basal rosette of
leaves, a main flowering stem (which usually
develops lateral branches), and optional additional flowering branches stemming from the
base of the plant. Its life cycle is roughly divided
into four stages. First, the germinating seedling elongates and etiolates; the cotyledons
then open and the first true leaves are formed.
Second, the rosette grows to a number of leaves
over a period of time that varies with the population and the environmental conditions7 –
plants can produce anywhere between five to
six and hundreds of leaves. Third, the main
stem begins to elongate (bolting), and the first
flowers open within a few days. Under favorable environmental conditions, additional basal
meristems also develop into inflorescences, so
that the total fruit/seed output can vary dramatically. Finally, leaves and fruits begin to
senesce, the fruits split open, and the seeds are
released. Seeds can germinate immediately or
go through a period of dormancy, again depending on the population and the environment.
This apparently simple scenario is complicated by a variety of responses to environmental conditions, which have been the focus of
much ecological research. In particular, the time
486
December 1998, Vol. 3, No. 12
of flowering can vary considerably, thereby
affecting the entire life cycle and reproductive fitness of the plant. At least four groups
of populations have been distinguished8. A
very early flowering kind (presumably spring
ephemerals) bolts within 40 days of sowing
(this includes virtually all the ecotypes used in
molecular genetic research), and does not respond to vernalization unless grown under short
days. A second type (probably early summer
annuals) flowers somewhat later and responds
only slightly to vernalization. In the third class
(which is thought to correspond to the ecological type of late summer annuals) bolting is delayed even further, and this effect is heightened
by a lack of vernalization. The fourth group behave as winter annuals and flower in four to six
months, although this can be reduced to a single month by vernalization (e.g. 70 days at 28C).
It has to be pointed out that this classification
is based on physiological data alone, and that
we still lack convincing field evidence that the
assumed ecology of these types corresponds to
their habit under natural conditions. No examples of biennial A. thaliana have been described
to date, although some of its relatives are biennial, and some genotypes do not flower within
one season if exposed to short photoperiod
(M. Pigliucci and H. Callahan, unpublished).
From an ecological perspective it is interesting that there is no clear association between
the flowering habit of A. thaliana and its geographic distribution. Populations from both
Scandinavia and Tennessee are winter-annual,
as are the genotypes from warm and dry Moravia and wet and cold Rhineland in Europe.
Furthermore, Cologne (Germany) and Grenoble
(France) are characterized by similar temperatures and precipitation, but they host summer annuals and winter annuals, respectively. Several
genes have been found to affect both flowering
time and its response to photoperiod and vernalization (the effect is also mediated by gibberellic acid)9. Thus, this deceptively simple
aspect of the plant’s ecological genetics is likely
to keep researchers busy for some time to come.
Another supposedly established fact about
A. thaliana is that it is predominantly a selfing species. Interestingly, this characteristic is
usually invoked as an objection to its use in
evolutionary studies: selfed organisms are considered to be evolutionary dead-ends, because
their populations do not harbor significant
genetic variation. Although an estimate of the
outcrossing rate of this species gave a value of
0.3% based on allozyme markers10, several
lines of evidence point to higher values (around
1.2–2.2%)11, implying that it is a more interesting subject for studies on population genetics.
Even the allozyme studies revealed 45 different multilocus genotypes based on seven polymorphic loci when 470 individuals were
sampled from 16 populations (with up to eight
genotypes per population); these figures are
not indicative of low genetic variation. Furthermore, quantitative genetic studies of several
characters in A. thaliana have shown significant heritability and genetic variation both
within and among populations12. Also, molecular studies using DNA sequencing of the
inter-spacer (ITS) region of the chloroplast
genome have revealed remarkable variation
both among and within populations (M. Cruzan
and M. Pigliucci, unpublished). Finally, protogyny may significantly contribute to outcrossing in A. thaliana (I. Al-Shehbaz, pers.
commun.). Clearly, there is the need to further
assess the mating system and population
genetics of this species.
The complexities of weed ecology
Arabidopsis thaliana is an excellent example
of what a weed can do from an ecological standpoint. Although it is rarely locally abundant
(its populations seldom count more than a few
scattered individuals or clumps), and it does
not withstand competition with grasses or other
tall vegetation, its geographical range is nonetheless impressive. The northernmost limit of
its distribution is northern Scandinavia, and
the southernmost tip of its range is in northern
Africa and the Middle East. To date, and in
spite of all we know about the physiology and
molecular biology of A. thaliana, no systematic investigation of what makes this astounding geographical range possible with such a
‘simple’ phenotype has been undertaken.
What is known, is that A. thaliana responds
dramatically to environmental variation over
a wide range of spatial scales. When used for
bioassays in a series of explant trials within an
area of 50 m2 of undisturbed forest, it was
found that the environmental variance (measured by the response of the plant) between
sites, increased with distance5. However, the
correlation between sites decreased with separation in the same fashion at all spatial scales
examined. In other words, environments – as
perceived by A. thaliana – display the same
degree of complexity at all spatial scales.
An elegant experimental design was used
to investigate the response of A. thaliana to
selection over very short spatial scales13. Random combinations of seeds of three phenotypically distinguishable lines were sown
over and across a series of 5 3 1 m transects.
If no selection occurred, one would expect to
find random mixtures of mature plants at the
end of the season. Instead, the results showed
a very significant tendency for genotypes of the
same kind to cluster together, and eliminate
the other two kinds from each microsite. The
success of a given genotype depended on the
microsite, and the range of this effect was less
than 50 cm. This suggests that very localized
environmental conditions may favor one or another genotype, yielding an apparently random
patchwork of genotypes across the landscape.
trends in plant science
perspectives
(a)
(b)
250
17
Number of fruits
Number of leaves
15
13
11
9
7
200
150
100
50
5
3
0
Neutral shade
Low R:FR
Neutral shade
Treatment
Low R:FR
Treatment
Fig. 1. Reaction norms for number of rosette leaves (a) and fruit production (b) in response to altered light quality [low Red (R):Far Red (FR),
on the right side of the abscissa, simulating canopy shade]. Landsberg erecta (blue line) is an inbred line used under standard laboratory conditions. The mutants functionally lack all phytochromes, and therefore cannot perceive red light; they are the classic hy1 (red line) and hy2
(green line) chromophore mutants, which correspond to two different loci with similar biochemical and phenotypic effects. Notice the similarity between the reaction norms of both wild type and mutants for the two traits.
Because the distance over which the homogenizing effect of selection occurred turned out
to be very small (with random distributions
above that scale), one can reasonably infer that
most of A. thaliana’s ecology is local.
However, these experiments did not address
the factors responsible for this extremely localized response to selection. To address this
point, an investigation into the spatiotemporal
effects of nitrogen and litter availability was
carried out14. It was found that some English
populations undergo two life cycles within a
given year. Some plants flower in the spring
(May–June), and their offspring do not go
through dormancy, but germinate immediately
and flower in late summer (September–October). The seeds of the summer generation then
overwinter, to germinate the following spring.
These investigations revealed that litter from
the spring individuals fosters reproductive output in the summer generation (mostly very small
individuals responding to high density – a result of phenotypic plasticity). By contrast, the
overwintering seeds cannot use the litter of the
previous generation because it is washed away.
However, because they are at a much lower
population density, they produce plants with
large rosettes, which accumulate photosynthate
and yield a significant number of siliques.
These experiments demonstrate how little
we know about the ecology of A. thaliana. For
example, we do not know how widespread the
two-generation cycle is, or which ecological
factors may determine its occurrence. Furthermore, because there have been few experimental studies of A. thaliana in the field5,14,15,
we are in no position to interrelate the microand macro-ecology of this plant.
Phenotypic plasticity
Several laboratories have focused on the use
of A. thaliana in studies on phenotypic plasticity – the way in which genotypes respond
to heterogeneous environmental conditions7,16.
The phytochrome genes in A. thaliana (which
control response to light availability, light
quality and photoperiod) are presently some of
the best known examples of plasticity genes.
A plasticity gene is a receptor of environmental change that effects a cascade of specific
and fitness-enhancing developmental/phenotypic changes17. An investigation into the plasticity of a wild-type inbred line and of several
single-gene mutants representing potential
plasticity genes is summarized in Fig. 1. This
study demonstrates that the elimination of all
of the phytochromes (which control the response to canopy shade), transforms the plastic wild type into a genotype that is totally
unresponsive to this specific environmental
change, although there is no effect on other
kinds of plasticity. This lack of plasticity directly translates into a proportional reduction
in reproductive fitness of the mutants (which
are otherwise physiologically and developmentally normal). This investigation demonstrates the adaptive implications of phenotypic
plasticity18.
Contrary to expectations based on its putative selfing habit, A. thaliana shows a tremendous degree of variation for phenotypic
plasticity to an array of environmental circumstances. A study of four populations (two early
flowering and two late flowering ecotypes), revealed significant responses to water, nutrient
and light gradients, with significant genetic
variation among populations for several traits in
reaction to nutrient and light availability19. A
larger survey of 26 populations found significant population by environment interaction
(i.e. genetic variation for plasticity) for five
out of nine traits20. Size-fecundity relationships also varied significantly in response to
light, nutrient and pot volume gradients21. This
latter study identified a minimum size threshold that the plant has to reach to be able to
flower. However, the value of the threshold is
itself environmentally dependent (i.e. plastic).
Other researchers have used A. thaliana as
a model system to explore the effects of herbivory15, and the interaction between herbivory and abiotic factors22. A poorly explored
aspect of plasticity in A. thaliana (and in other
plants and animals as well) is the interaction
between development and environment. Ecological studies have demonstrated developmental stage-specific variation for plasticity12,
and research carried out on mutants has pinpointed the complex and antagonistic molecular basis of developmental windows of
plasticity23. Although a general framework for
this kind of research has not yet emerged, it is
a growing area and demonstrates the trend for
a direct interaction between ecological and
molecular genetics.
Quantitative genetics and constraints
Arabidopsis thaliana is surprisingly genetically variable, both at the DNA sequence and
at the quantitative genetic level. This has afforded some interesting evolutionary studies
of this plant. One old problem in evolutionary
genetics is represented by dominance24,25. To
understand the evolution of dominance we
need studies addressing both its proximate
causes (the genetic and physiological bases),
and its ultimate causes (i.e. under what circumstances natural selection favors a certain
degree of dominance). It has been shown that
dominance of two mutations in A. thaliana
that affect the production of abscisic acid is
modulated by maternal effects26, which provides the proximate cause of dominance in this
instance. This kind of work can be integrated
into studies on the selective advantage of different degrees of dominance and of the evolutionary dynamics of maternal effects, to yield
insights into the broader question of why
dominance evolves.
December 1998, Vol. 3, No. 12
487
trends in plant science
perspectives
Leaf number
(a)
12
11
10
9
8
7
6
5
4
3
0
500
Fitness
D
W
0
L
5
10 15 20 25
Bolting time (days)
30
Leaf number
(b)
12
11
10
9
8
7
6
5
4
3
0
500
Fitness
D
L
W
0
5
10 15 20 25
Bolting time (days)
30
Leaf number
(c)
12
11
10
9
8
7
D
6
5 W
4
3
0
5
0
500
Fitness
L
10 15 20 25
Bolting time (days)
December 1998, Vol. 3, No. 12
The big picture: evolution and
systematics of Arabidopsis
No matter how interesting a model system is
from a molecular or even an ecological standpoint, the crucial test of its relevance from an
evolutionary perspective is provided by the
exploration of its phylogenetic neighborhood.
Until recently, the systematics of Arabidopsis
was confusing34, but several authors have concentrated on revising the situation, applying
modern molecular and cladistic analyses to the
group34–36.
Many of the species that used to be classified as being Arabidopsis are currently considered as being fairly distant relatives, whereas
taxa once in Arabis or Cardaminopsis are now
thought of as being Arabidopsis in the strictest
sense. The emerging phylogenetic picture of
this genus immediately creates the opportunity
to examine interesting questions from an evolutionary perspective. For example, A. thaliana is an annual, but several of its relatives are
biennials or perennials. How often does a new
flowering habit emerge through history? How
are these evolutionary shifts related to the
ecology, and therefore what are the selective
pressures these plants experience? Furthermore, not all members of Arabidopsis are
diploid. In fact, one of A. thaliana’s closest
relatives is A. arenosa, which is a tetraploid.
Recent research has confirmed an old speculation that A. thaliana and A. arenosa are
respectively the female and male parent of a
third taxon, A. suecica37,38. The wealth of molecular and physiological information available
for A. thaliana opens up previously unforeseen possibilities for investigating the ecology
and evolution of hybridization and polyploidy
(as well as the evolution of mating systems),
because A. thaliana is predominantly selfcrossing, while the other two are outcrossing.
30
Fig. 2. Relationship between bolting time
and leaf number in Arabidopsis thaliana.
The solid red circles represent the three
wild-type [Dijon (D), Landsberg (L) and
Wassilewskija (W)] populations. All
other circles (diameter proportional to reproductive fitness, measured as the number of fruit produced) are EMS mutants
from the (a) Dijon, (b) Landsberg and (c)
Wassilewskija backgrounds. Note how
genetic variation generated by mutation
can dramatically alter the relationship between the two traits. Also, it is interesting to note that – contrary to expectation
– many mutants display a higher reproductive fitness than the wild types. This
effect might be because the three wild
types have been selected for a very short
life cycle, which had the pleiotropic effect
of reducing the reproductive output (M.
Pigliucci and N. Turner, unpublished).
488
Recently, quantitative genetics has acquired a new tool with which to merge evolutionary and molecular genetics. It is now
possible to use molecular markers to identify
and map the positions of genetic regions
affecting quantitative characters in natural
populations. These elements are known as
Quantitative Trait Loci (QTLs)27. QTL mapping has been used to study phenotypic plasticity, flowering time and life history28,29.
However, a word of caution is necessary. If
the stated goal is to investigate loci affecting
life-history evolution in natural populations, it
is inappropriate to compare artificially derived
lines grown under continuous light30, as this is
an unnatural condition. Care has to be taken to
prevent our enthusiasm for interdisciplinary
research and new techniques from getting
in the way of our understanding of the underlying biology.
Another long-standing evolutionary problem is the question of constraints31,32. A very
well known genetic constraint in A. thaliana
is represented by a strong phenotypic and
genetic correlation between flowering time
and leaf production (i.e. between size and age
at maturity)30,33. It is interesting that the intensity of this constraint depends on the
genetic background, as well as on the environment in which the plant grows. It is also
possible to break the constraint to some extent, and obtain plants with a very different
combination of the two traits compared to
the wild type. This can be achieved by inducing mutations in different backgrounds,
and then comparing the position in phenotypic space of the mutants and the respective
wild-type (Fig. 2). Such an approach represents one of the first experimental attempts
to address the evolutionary problem of
constraints.
A. thaliana – Columbia Derivative
A. thaliana – Columbia
A. thaliana – Landsberg erecta
A. thaliana – Landsberg
A. thaliana – Nossen
A. thaliana – Moscow
Hylandra
Cardaminopsis petraea
A. griffithiana
A. pumila
0
20
40
60
Fig. 3. Evolution of phenotypic plasticity of leaf production in response to light quality in
Arabidopsis and allied species. The length of the bars to the right of each taxon indicates
the intensity of the plastic response: the color of the bar indicates distribution in Central
Europe (yellow), Eastern Europe (red) and Northern Europe (blue). Hylandra (also known as
A. suecica) indicates a hybrid between A. thaliana and (presumably) A. arenosa (belonging
to the Cardaminopsis clade). The phylogenetic position of A. thaliana’s ecotypes is inferred
from their genetic distances. Note how strong or weak plasticity has apparently evolved more
than once in the whole group.
trends in plant science
perspectives
The availability of a phylogeny of the Arabidopsis genus and its more or less distant relatives
has also afforded some first-time comparative
analyses addressing evolutionary ecological
questions. One such question concerns the evolution of phenotypic plasticity. For example, in
the case of plasticity of leaf production, a high
degree of plasticity occurs in some distant relatives outside A. thaliana’s clade, but it is lost in
early flowering taxa, only to re-evolve in late
flowering ones (Fig. 3). It would be interesting
to map the evolution of the known genes affecting flowering time upon the same cladogram.
Can we put everything together?
Attempts to unify biology have characterized
the history of this discipline throughout the
20th century. One of the most important results
of this trend has been the neo-darwinian synthesis that led to modern evolutionary theory.
Recently, many people have invoked a longawaited merger between organismal and molecular/developmental biology. This goal now
appears attainable thanks to the reliability, sophistication, and low cost of modern molecular
techniques. However, given the limited funding
for the basic sciences, and the relatively small
number of scientists that study either molecular
or evolutionary biology, a true synthesis seems
possible – at least at present – only by concentrating on a few well chosen organisms. Together with other classical model systems, such
as E. coli and Drosophila, Arabidopsis and its
relatives are clearly gaining permanent status
as members of this intensively studied group.
Acknowledgements
I wish to thank Jacqui Shykoff and PierreHenri Guyon, of the University of Paris-Sud,
for their hospitality and discussions while
preparing this review. Hilary Callahan, Mark
Camara and Carolyn Wells have been instrumental in developing some of the ideas discussed in this paper. Melissa Brenneman,
Jacqui Shykoff and Ihsan Al-Shehbaz have
also made helpful comments.
References
1 Kellogg, E.A. and Shaffer, H.B. (1993) Model
organisms in evolutionary studies, System. Biol.
42, 409–414
2 Redei, G.P. (1992) A heuristic glance at the past of
Arabidopsis genetics, in Methods in Arabidopsis
Research (Koncz, N-H.C.C. and Schell, J., eds),
pp. 1–15, World Scientific, Singapore
3 Goodman, H.M., Eckers, J.R. and Dean, C.
(1995) The genome of Arabidopsis thaliana,
Proc. Natl Acad. Sci. U. S. A. 92, 10831–10835
4 Turesson, G. (1922) The genotypical response of
the plant species to the habitat, Hereditas 3, 211–350
5 Bell, G. and Lechowicz, M.J. (1991) The ecology
and genetics of fitness in forest plants. I.
Environmental heterogeneity measured by
explant trials, J. Ecol. 79, 663–685
6 Aarssen, L.W. and Clauss, M.J. (1992) Genotypic
variation in fecundity allocation in Arabidopsis
thaliana, J. Ecol. 80, 109–114
7 Pigliucci, M. and Schlichting, C.D. (1998)
Reaction norms of Arabidopsis. V. Flowering
time controls phenotypic architecture in response
to nutrient stress, J. Evol. Biol. 11, 285–301
8 Napp-Zinn, K. (1985) Arabidopsis thaliana, in
CRC Handbook of Flowering (Halevy, A.H., ed.),
pp. 492–503, CRC, Boca Raton, FL
9 Coupland, G. (1995) Genetic and environmental
control of flowering time in Arabidopsis, Trends
Genet. 11, 393–397
10 Abbott, R.J. and Gomes, M.F. (1989) Population
genetic structure and outcrossing rate of Arabidopsis
thaliana (L.) Heynh, Heredity 62, 411–418
11 Snape, J.W. and Lawrence, M.J. (1971) The
breeding system of Arabidopsis thaliana,
Heredity 27, 299–302
12 Pigliucci, M. (1997) Ontogenetic phenotypic
plasticity during the reproductive phase in
Arabidopsis thaliana (Brassicaceae), Am. J. Bot.
84, 887–895
13 Stratton, D.A. and Bennington, C.C. (1996)
Measuring spatial variation in natural selection
using randomly-sown seeds of Arabidopsis
thaliana, J. Evol. Biol. 9, 215–228
14 Thompson, L. (1994) The spatiotemporal effects
of nitrogen and litter on the population dynamics
of Arabidopsis thaliana, J. Ecol. 82, 63–68
15 Mauricio, R. and Rausher, M.D. (1997)
Experimental manipulation of putative selective
agents provides evidence for the role of natural
enemies in the evolution of plant defense,
Evolution 51, 1435–1444
16 Zhang, J. and Lechowicz, M.J. (1994)
Correlation between time of flowering and
phenotypic plasticity in Arabidopsis thaliana
(Brassicaceae), Am. J. Bot. 81, 1336–1342
17 Pigliucci, M. (1996) How organisms respond to
environmental changes: from phenotypes to
molecules (and vice versa), Trends Ecol. Evol.
11, 168–173
18 Dudley, S.A. and Schmitt, S.A. (1996) Testing
the adaptive plasticity hypothesis: densitydependent selection on manipulated stem length
in Impatiens capensis, Am. Nat. 147, 445–465
19 Pigliucci, M., Whitton, J. and Schlichting, C.D.
(1995) Reaction norms of Arabidopsis. I. Plasticity
of characters and correlations across water, nutrient
and light gradients, J. Evol. Biol. 8, 421–438
20 Pigliucci, M. and Schlichting, C.D. (1995)
Reaction norms of Arabidopsis (Brassicaceae).
III. Response to nutrients in 26 populations from a
worldwide collection, Am. J. Bot. 82, 1117–1125
21 Clauss, M.J. and Aarssen, L.W. (1994)
Phenotypic plasticity of size-fecundity
relationships in Arabidopsis thaliana, J. Ecol. 82,
447–455
22 Grant-Petersson, J. and Renwick, J.A.A. (1996)
Effects of ultraviolet-B exposure of Arabidopsis
thaliana on herbivory by two crucifer-feeding
insects (Lepidoptera), Physiol. Chem. Ecol. 25,
135–142
23 Mozley, D. and Thomas, B (1995) Developmental
and photobiological factors affecting photoperiodic
induction in Arabidopsis thaliana Heynh.
Landsberg erecta, J. Exp. Bot. 46, 173–179
24 Kacser, H. and Burns, J.A. (1981) The molecular
basis of dominance, Genetics 97, 639–666
25 Keightley, P.D. and Kacser, H. (1987)
Dominance, pleiotropy and metabolic structure,
Genetics 117, 319–329
26 Finkelstein, R.R. (1994) Maternal effects govern
variable dominance of two abscisic acid response
mutations in Arabidopsis thaliana, Plant Physiol.
105, 1203–1208
27 Mitchell-Olds, T. (1995) The molecular basis of
quantitative genetic variation in natural
populations, Trends Ecol. Evol. 10, 324–327
28 Kuittinen, H., Sillanpaa, M.J. and Savolainen, O.
(1997) Genetic basis of adaptation: flowering
time in Arabidopsis thaliana, Theor. Appl. Genet.
95, 573–583
29 van der Schaar, W. et al. (1997) QTL analysis of seed
dormancy in Arabidopsis using recombinant inbred
lines and MQM mapping, Heredity 79, 190–200
30 Mitchell-Olds, T. (1996) Genetic constraints on
life-history evolution: quantitative-trait loci
influencing growth and flowering in Arabidopsis
thaliana, Evolution 50, 140–145
31 Maynard Smith, J. et al. (1985) Developmental
constraints and evolution, Q. Rev. Biol. 60, 265–287
32 Schlichting, C.D. and Pigliucci, M. (1998)
Phenotypic Evolution. A Reaction Norm
Perspective, Sinauer
33 Koornneef, M., Hanhart, C.J. and
van den Veen, J.H. (1991) A genetic and physiological analysis of late flowering mutants in
Arabidopsis thaliana, Mol. Gen. Genet. 229, 57–66
34 Price, R.A., Palmer, J.D. and Al-Shehbaz, I.A.
(1994) Systematic relationships of Arabidopsis:
a molecular and morphological perspective, in
Arabidopsis (Meyerowitz, E.M. and
Somerville, C.R., eds), pp. 7–20, Cold Spring
Harbor Laboratory Press
35 Kamm, A. et al. (1995) Analysis of a repetitive
DNA family from Arabidopsis arenosa and
relationships between Arabidopsis species, Plant
Mol. Biol. 27, 853–862
36 O’Kane, S.L.J. and Al-Shehbaz, I.A. (1997)
A synopsis of Arabidopsis (Brassicaceae), Novon
7, 323–327
37 Mummenhoff, K. and Hurka, H. (1995)
Allopolyploid origin of Arabidopsis suecica
(Fries) Norrlin: evidence from chloroplast and
nuclear genome markers, Bot. Acta 108, 449–456
38 O’Kane, S.L.J., Schaal, B.A. and Al-Shehbaz, I.A.
(1996) The origins of Arabidopsis suecica
(Brassicaceae) as indicated by nuclear rDNA
sequences, System. Bot. 21, 559–566
Massimo Pigliucci is at the Depts of Botany,
and Ecology and Evolutionary Biology,
University of Tennessee, Knoxville, TN, USA
(tel 11 423 974 6221; fax 11 423 974 0978;
e-mail [email protected]).
December 1998, Vol. 3, No. 12
489