Directory UMM :Data Elmu:jurnal:P:PlantScience:Plant Science_BioMedNet:041-060:
trends in plant science
Perspectives
Coping with multiple enemies:
an integration of molecular
and ecological perspectives
Nigel D. Paul, Paul E. Hatcher and Jane E. Taylor
How plants respond to attack by the range of herbivores and pathogens that
confront them in the field is the subject of considerable research by both
molecular biologists and ecologists. However, in spite of the shared focus of
these two bodies of research, there has been little integration between them. We
consider the scope for such integration, and how greater dialogue between
molecular biologists and ecologists could advance understanding of plant
responses to multiple enemies.
R
ecent research has highlighted how
plants respond to the diversity of organisms that attack them in the field1–4.
Much of this current interest derives from cellular studies of plant defence pathways. These
studies take a reductionist approach and attempt
to identify individual signalling pathways that
link wounding or attack by a pathogen or herbivore to specific gene expression events. For
example, in Arabidopsis this approach has been
successful in identifying key signalling molecules and their relative position in particular
pathways. It has also identified possible points
of cross-talk between pathways and synergistic
or antagonistic events. Recently, there has been
increasing interest in the possible interactions
between plant signalling pathways, including
those involved in responses to disease and
herbivory4.
Induced defence is also an important area of
ecological study5. In contrast with molecular
studies, the ecological view of induced defence
is broad, and is seen within the context of the
overall ‘defence strategy’ of the host plant. This
wider strategy includes constitutive defences,
the ‘quality’ of host tissues as a resource for
consumers, and tolerance (i.e. the capacity of
the plant to adjust its physiology to buffer the
effects of herbivory or disease). With few
exceptions6, ecologists have paid little attention
to plant responses to multiple enemies, and
studies of non-crop hosts have concentrated
largely on defence against herbivores.
Here we consider induced resistances from
the contrasting perspectives of molecular biology and ecology, and whether integration of
these different standpoints offers any novel
insights into plant defence against multiple
Box 1. Rumex –Gastrophysa–Uromyces: an ecological model for
host–herbivore–pathogen interactions
The beetle Gastrophysa viridula and the biotrophic rust fungus Uromyces rumicis (Fig. 1) both
attack the leaves of Rumex species (docks). The physiology and ecology of this system have
been intensively studied. Rust infection of dock reduces the growth, survivorship and fecundity of G. viridula (Fig. 2), which is associated with changes in the nutritional status of host
tissues following infection26. Conversely, G. viridula feeding induces resistances to the rust
fungus, both locally and systemically (Fig. 2)27. These effects are not confined to controlled
environments: rust infection is significantly and substantially reduced in the field by both beetle grazing28 and artificial wounding27. Thus, the combined effect of the beetle and the rust fungus on the host might be expected to be inhibited (i.e. the effect less than that of one of the
agents alone). However, the effect is usually either equivalent (in Rumex obtusifolius, damage
is equivalent to that of the agents; Fig. 2) or additive (in R. crispus, damage is equivalent to the
sum of the effects of each agent alone; Fig. 2)29,30, illustrating the difficulty in extrapolating
from the effects of the agents on each other to their effect on the host. This effect is robust even
when plants are grown with differing levels of nitrogen fertilization29. Tripartite interactions in
dock are not confined to Gastrophysa and Uromyces, because herbivory also significantly
reduced field infection by two other fungal pathogens of dock, the necrotroph Ramularia
rubella, and the hemibiotroph Venturia rumicis28. We are currently applying molecular techniques to the Rumex system to try to identify genes that are up-regulated in response to attack
by either the pathogen or the herbivore, and then to examine their spatial and temporal expression patterns in response to multiple attack. Our aim is to apply these techniques to investigate
the mechanisms of interactions between Rumex, Gastrophysa and Uromyces in the field.
220
May 2000, Vol. 5, No. 5
enemies. We will concentrate on three broad
issues: cross-talk between pathways, specificity of induced resistances and the evolution
of induced resistances in the context of attack
by multiple enemies.
Cellular perspective on systemic
induced resistance
Of the several pathways of systemic induced
defence in plants the best characterized are systemic acquired resistance (SAR), which provides enhanced resistance to pathogen
infection, and the wound response pathway,
which provides resistance to insect feeding.
Both pathways have been shown to induce the
expression of distinct sets of genes. SAR has
been shown to be induced by salicylic acid and,
in at least some systems, salicylic acid appears
to be both necessary and sufficient for inducing SAR. By contrast, the wound response
pathway appears to be induced by jasmonic
acid and ethylene without salicylic acid.
Although it has been widely accepted that the
pathways responsive to pathogen or herbivore
attack are distinct, some recent evidence suggests that jasmonic acid also mediates induced
resistance to some pathogens7,8. Thus, the strict
division of induced responses might be incorrect. In addition, the salicylic acid and jasmonic acid signalling pathways can interact1.
Although the nature of the cross-talk is not yet
fully defined, recent data suggest that induction of the salicylic acid pathway often inhibits
the systemic induction of the jasmonic acid
pathway by later herbivory. Conversely, the
induction of the jasmonic acid pathway can
reduce induction of the salicylic acid pathway
following subsequent pathogen attack9–12.
Ecological perspective on systemic
induced resistance
The molecular biology of induced resistances
has been investigated mostly in herbaceous
species. More than 100 species are known to
express induced resistance following herbivory, but two-thirds of these are trees or
shrubs5. The vast majority of ecological
research into induced resistance deals with herbivores, ranging from mites to large mammals,
and it is clear that induced systemic resistances
are widespread and of sufficient magnitude to
significantly reduce herbivore populations5,13.
In long-lived plants, induced changes can
persist to reduce herbivore populations in the
years following induction14, and in several forest systems the decay of induced resistance can
be over several herbivore generations14.
Where the mechanisms underlying these
ecologically significant induced defences
have been defined, it is clear that different
plant species use a great diversity of
responses. As well as induced chemical and
morphological changes, responses include
elements that are more closely linked to
1360 - 1385/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01603-4
trends in plant science
Perspectives
tolerance (for example, altered protein or
nitrogen content in regrowth leaves) than
resistance per se5. Ecologists have also considered the broader issue of the selective
advantages of induced resistance, in particular the relationship between the benefits and
costs of induction15. This in turn leads to questions concerning the population genetics and
evolution of induced defence. Current theories
are based on the concept that current herbivory
or disease is correlated with the risk of future
attack16. In this sense, current or past attack
can be seen as having the potential to provide
information about the future environment. On
this basis, induced resistances will be selected
for only if (i) current attack is a reliable predictor of future attack, and (ii) if attack reduces
plant fitness.
This body of ecological information on
induced defence against herbivores is in stark
contrast with the limited knowledge of comparable responses to pathogens. Given the
limited understanding of plant pathogens in
natural systems it is not surprising that
responses to multiple enemies are also poorly
understood. One of the few examples of multiple defence to have been studied in a noncrop system – Rumex spp. attacked by the
beetle Gastrophysa viridula and the rust fungus Uromyces rumicis – highlights the potential complexity of such interactions (Box 1;
Figs 1 and 2). Several whole plant studies
using crops also highlight the diversity of herbivore–pathogen interactions: with examples
of attack by one organism inducing resistance
to attack by another, but there are also many
examples of exceptions6.
Defence against multiple enemies
Cross-talk
Molecular biologists have considered
‘defence on multiple fronts’ in terms of interactions between different pathways of induced
resistance1,3,17. For example, whether induction of the salicylic acid pathway influences
subsequent induction of the jasmonic acid
pathway and vice versa. In this respect there
is evidence that resistance induced via one
pathway can suppress induced responses
(b)
Number of eggs laid by
G. viridula
1000
900
800
700
600
500
400
300
200
100
0
Control
plants
Rusted
leaves
mediated via another pathway9–11. However,
this is only part of the story, and there are
examples of induction by one agent resulting
in increased resistance to another12. It is also
increasingly clear that the concept of specific
pathogen and wounding (herbivore)-induced
pathways mediated by salicylic acid and
jasmonic acid, respectively, is simplistic1.
Jasmonic acid also mediates resistances to certain pathogens, and both the salicylic acid and
jasmonic acid pathways induce a spectrum of
(c)
35
120
Plant dry weight (g)
U. rumicis pustules (cm–2)
(a)
Fig. 1. (a) Rumex obtusifolius foliage attacked by the rust fungus Uromyces rumicis.
(b) Rumex obtusifolius foliage attacked by the chrysomelid beetle Gastrophysa viridula,
which attacks Rumex spp. throughout its larval stages (illustrated), and as an adult.
30
25
20
15
10
5
100
80
60
40
20
0
0
Control
plants
Ungrazed Grazed
leaf
leaf
Infected plant
Control Beetle
Rust
RB
Trends in Plant Science
Fig. 2. (a) Fecundity of Gastrophysa viridula reared on Rumex obtusifolius (white) and R. crispus (grey), infected by Uromyces rumicis (rust).
Adults had been fed on the no-choice treatments since hatching in Petri dishes, given excess food, and reared at constant temperature. Data for
control and infected plants are means of 9–15 plants 6 1 SE. Modified from Ref. 26. (b) Infection of Rumex obtusifolius (white) and R. crispus
(grey) by Uromyces rumicis (rust) seven days after feeding by the beetle Gastrophysa viridula. Data for control plants, and grazed and ungrazed
leaves of grazed plants are means of 14–15 plants 6 1 SE. Modified from Ref. 27. (c) The effects of grazing by the beetle Gastrophysa viridula
and infection by Uromyces rumicis, alone and in combination, on the growth of Rumex obtusifolius (white) and R. crispus (grey). Key:
control 5 plants with neither beetle nor rust; beetle 5 plants with beetles only; rust 5 plants with rust only; RB 5 plants infected by U. rumicis that were then grazed by G. viridula. Data are means of 15 plants 6 1 SE. Modified from Ref. 30.
May 2000, Vol. 5, No. 5
221
trends in plant science
Perspectives
Herbivore
Pathogen
b
d
e
d
e
Recognition
Signalling pathways
Gene products (+ or —)
a
c
f
g
h
Effects on attackers
Effects on host plant
Overlap between responses
Trends in Plant Science
Fig. 3. Specificity and overlap in induced resistances: a question of perspective? There is no
question that the host plant has the capacity to recognize attack by different organisms with
a high degree of specificity. At the level of recognition, it is likely that there is no ‘overlap’
or ‘cross-talk’ between responses to different attackers. The signal transduction pathways
involved in induction following attack by herbivores or pathogens also appear to be largely
specific, but recent evidence suggests that there are elements of cross-talk, both positive and
negative. It is clear that different attackers can affect the expression of distinct sets of genes,
and result in different sets of gene products, but such ‘fine-tuning’ might also include individual elements that are less specific17. For example, a herbivore might induce the expression of genes abcde and a pathogen might induce genes defgh. This could be taken as
evidence of specificity or overlap. A more concrete example is that both herbivores and
pathogens might induce specific sets of genes related to the synthesis of phenolics, but
induction of PAL, and/or the resultant general up-regulation of phenolic metabolism is common to a wide range of attackers. The effects of these gene products on attackers can be even
less specific: the major classes of secondary chemicals that are inducible by arthropods have
been shown to have anti-microbial properties and vice versa6. In addition, ‘non-defence’
changes, such as alterations in tissue nitrogen might have wider-ranging effects. Finally,
pathogen and herbivore-induced changes viewed from the perspective of their effects on the
overall functioning and fitness of the host plant, including both defence and ‘civilian’
responses are likely to show considerable overlap.
gene products that although distinct to the
pathway show considerable overlap between
the two (Fig. 3). The relationship between specific gene products and resistance is an area of
research that could benefit from closer liaison
between molecular biologists and ecologists.
We echo the recent suggestion that the value
of plant defence investigations would be
increased if they were to concentrate on how
a putative defence actually contributes to
resistance18.
Likewise, the literature dealing with herbivores and pathogens, rather than altered levels
222
May 2000, Vol. 5, No. 5
of gene products, frequently indicates that the
effects of induced resistances are often not specific to any particular attacker. Thus, although
induced resistances might show some degree
of ‘fine-tuning’ (Fig. 3), it might be expected
that resistance induced by one attacker would
be effective against a range of other potential
attackers. Molecular studies focusing on biochemical mechanisms of ‘cross-talk’ between
pathways would benefit from complementary
studies considering, for example, the extent to
which induced changes are specific in terms of
their effects on a range of attackers.
Specificity
As we have discussed, many of the biochemical responses induced by resistance pathways
are effective against a wide range of organisms. This specificity of effect, or lack of, must
be distinguished from the plant’s specificity of
response (Fig. 3). The plant’s specificity of
response is the ability of the plant to distinguish between different types of biotic challenge and to produce different sets of
chemicals in response to each challenge.
Research suggests that plants do possess
specificity of response10,19.
Reviews of the molecular biology of
induced resistance suggest that plants might
‘fine-tune’ induced resistances to specific
attackers3,17. These arguments seem consistent
with current evolutionary theory. Induced
resistance is selected for when past or current
attack is correlated with the risk of future
attack. Such correlation is likely to be greatest when considering current and future attack
by the same organism, favouring selection of
a ‘fine-tuned’ induced response. However,
selection reflects the interplay of benefits and
costs of induced resistances, which should be
considered when determining the selective
advantage of ‘fine-tuned’ resistances. There
are many questions here, such as:
• Is the cost of a fine-tuned resistance less
than that of a broad-spectrum resistance?
• Is induced resistance finely tuned to a particular attacker likely to confer more or less
benefit than a broad-spectrum induced
resistance?
Although these questions remain unanswered
it is notable that the suggestion of fine-tuning
appears to be at odds with much of the literature that shows that resistance induced by one
organism is often effective against a range of
potential attackers. This might be explained
partly by a wider range of induced changes
than are often considered as resistance mechanisms. Systemic changes in the nutritional
quality of host tissues (e.g. nitrogen chemistry) following pathogen infection might
have rather non-specific effects on herbivores,
and potentially modify responses to induced
allelochemicals. There is less evidence that
herbivore-induced changes in ‘non-defence’
chemistry have systemic effects on pathogens.
However, these broader changes in host chemistry should not be overlooked when considering the biology of plant responses to
multiple enemies.
Evolution of induced resistances in the
context of attack by multiple enemies
There has been considerable debate over
whether a characteristic of the host affecting a
particular attacker has necessarily resulted
from selection pressure imposed by that
attacker5. Of course, this also applies to characteristics acting against multiple enemies.
trends in plant science
Perspectives
Herbivore attack
Pathogen
Civilian
Defence
Systemin
JA-dependent
SA-dependent
Damage
Jasmonic acid
or ethylene
Proteinase
inhibitors
Basic PR
proteins
Salicylic acid
Acidic PR proteins,
defensins,
thionins
Wound-induced SA-independent
SAR
SAR
?
?
Tolerance
SA-induced
SAR
Host growth and development
Host fitness
Trends in Plant Science
Fig. 4. ‘Defence’ and ‘civilian’ elements of induced responses and the possible interactions between them. The range of responses in host
metabolism and physiology that result from attack by herbivores and attack by pathogens can be divided into ‘defence’ and ‘civilian’ reactions.
Induced defence includes responses mediated via the salicylic acid and jasmonic acid pathways, which are well defined. Civilian responses
reflect the interplay of damage and tolerance – physiological and metabolic adjustments made in response to damage. The nature of damage
varies markedly with the attacking organism, but adjustment at the whole-plant level shares a degree of commonality between, for example, the
range of organisms that attack the foliage. The regulation of the balance between damage and tolerance remains poorly defined. Even less
defined are the interactions between the civilian and defence responses. Possible interactions include the effects of salicylic acid and jasmonic
acid on civilian responses independent of defence, and the effects of metabolic shifts necessary for induced resistance on civilian responses.
Equally, the shifts in metabolism and physiology inherent in civilian responses might influence the expression of induced resistances (at the
level of the production of resistance compounds and/or the responses of attackers to those compounds). Host growth, development and, ultimately, fitness cannot be seen as a function of defence or civilian responses, but as the outcome of the interaction between both sets of responses.
Further, because the products of induced
resistance(s) might be active against a wide
range of organisms, it is possible that broadspectrum induced resistance is the result
entirely of selection by a single attacker, rather
than integrated selection by diverse enemies.
However, given that attack by multiple and
diverse enemies is commonplace in the field,
can current models for the selection of induced
resistances be applied to this situation? Such
models are based on ‘information’, defined as
the ability of current attack to predict future
attack. Selection for induced resistance is predicted when past or current attack provides
‘information’ concerning (i.e. is well correlated with) the risk of future attack. Such correlation has been shown experimentally in a
few systems20 and might be expected for many
others. However, is it likely that attack by a
herbivore contains any information about
future attack by pathogens, or vice versa?
Arguably, herbivory, by creating wounds that
might be vulnerable to invasion by microorganisms2,5 predicts an increased risk of future
pathogen attack. However, this is a localized
response, consistent with the induction of
resistance in the vicinity of wounds. Is it likely
that herbivory predicts the risk of pathogen
attack in ungrazed tissues? Equally, on a
whole-plant basis, is past or current pathogen
attack likely to be closely correlated to future
herbivory? If such correlations between current and future attack are unlikely across multiple enemies, this suggests that selection of
broad-spectrum induced resistance is unlikely.
Selection based on ‘information’ is also difficult to relate to negative interactions between
induced responses, such that the response to
one attacker might constrain the plant’s ability to respond to another. Indeed, recent argu-
ments suggest that such interactions reflect the
manipulation of pathways by attackers21,
although this appears to be based on the concept of specific herbivore (jasmonic acid) and
pathogen (salicylic acid) pathways, which is
at odds with some recent evidence.
An alternative view of selection arises if we
develop the suggestion5 that induced resistance should be considered in the context of the
overall response of a plant to attack. This
includes what have been called ‘civilian’
responses in host physiology5. For example,
following foliar attack, civilian responses
include remobilization of reserves from storage tissues, increased partitioning of assimilate to leaves rather than roots and
up-regulation of photosynthesis in undamaged
leaves. Mechanisms of tolerance appear common to attack by both leaf-feeding herbivores
and foliar pathogens. These physiological
May 2000, Vol. 5, No. 5
223
trends in plant science
Perspectives
responses to attack buffer the host from the
effects of an initial attack but can increase
vulnerability to the effects of future attack.
Taking up this theme, a more phytocentric
viewpoint of selection for induced resistance
might be that current attack (regardless of the
organisms doing the attacking) conveys
information about the vulnerability of the
plant to the physiological effects of any
future attack. We hypothesize that attack by
any (foliar) attacker is a reliable predictor of
the physiological effect of future attack by
any (other) foliar attacker(s). This is also
consistent with the suggestion that induced
resistances allow maximum expression of
the plant’s potential to tolerate herbivory or
disease5, because the spatial distribution of
attack over the plant is altered, maintaining
undamaged tissues capable of compensatory
physiological changes5.
Viewing current attack as providing information about the physiological effects of
future attacks requires integration between
induced resistance and changes in the
primary physiology of the plant following
attack. Physiological responses to salicylic
acid and jasmonic acid are known to extend
far beyond induced resistance, but their
effects on ‘civilian’ responses are poorly
understood22–24. However, it is notable that
exogenous applications of both salicylic acid
and jasmonic acid have been shown to inhibit
photosynthesis, reducing levels of Rubisco
and chlorophylls24. Exogenous applications
of salicylic acid or jasmonic acid lack the
temporal and spatial organization of
responses to attack by consumer organisms.
However, these effects of exogenous jasmonic acid and salicylic acid are in marked
contrast with the increase in photosynthesis
in undamaged leaves that is a common
response to herbivory and disease. This paradox highlights the need for a greater understanding of post-induction changes in
‘civilian’ genes that have no direct relationship to resistance.
Salicylic acid and jasmonic acid also interact with other plant growth regulators, including abscisic acid and ethylene, which are
central to the coordination of responses to the
abiotic environment4. The relationships
between plant responses to biotic and abiotic
stress have recently been discussed in terms of
how abiotic stress might influence induced
resistances deployed in agriculture1 or whether
the abiotic environment provides ‘information’
about the risk of future herbivory or disease16.
We argue that responses to biotic and abiotic
stress are linked because optimizing induced
responses to minimize the physiological
effects of attack is highly dependent on the
abiotic environment. This includes civilian
responses such as altered partitioning between
shoot and root, as well as chemical resistance6.
224
May 2000, Vol. 5, No. 5
Conclusions
Acknowledgements
We have approached the issue of defence
against multiple enemies from the contrasting
backgrounds of molecular biology and plant
ecology. Once the initial hurdle of the lack of a
common terminology was overcome, it became
clear that the major limitation to integrating
these viewpoints was the gap between the level
of organization we normally dealt with. A key
We thank NERC for financial support, and
Peter Ayres and John Whittaker for many
valuable discussions.
“…tolerance and induced
resistances can be considered
as complementary elements
of phenotypic plasticity”
constraint on the integration of the details of
signalling pathways and specific gene products
in the few model systems with field data was
the relative scarcity of whole plant investigations of induced responses to herbivory and/or
disease. There is a risk that quantifying the
effects of induced responses on the herbivores
or on the pathogens rather than on the host plant
has falsely emphasized mechanisms that induce
resistance, but fail to benefit plants, while
underestimating the significance of tolerance
and other induced ‘civilian’ responses21.
Consideration of this ‘middle ground’ has
highlighted the need to integrate the available
information from this ‘phytocentric’ viewpoint. We echo the view5 that defence and
civilian aspects of induced responses to attack
should be considered together (Fig. 4). By
analogy with the term used in the context of
abiotic stress, induced resistance and tolerance
can be treated as elements of acclimation to
attack by herbivores and/or pathogens.
Although tolerance and (constitutive) resistance might be alternative adaptive strategies25,
tolerance and induced resistances can be considered as complementary elements of phenotypic plasticity. Thus, we would argue that it
would be fruitful for molecular biologists,
ecologists and plant physiologists to consider
the functional links between tolerance and
induced resistance.
Clearly, progress in this area will benefit
from the use of appropriate model systems. We
agree with the recent suggestion that resolving
uncertainties about signalling conflicts and
synergies in induced resistance will benefit
from the use of systems that are convenient for
genetic, molecular, biochemical and physiological studies1. However, the use of standard
laboratory models can provide little information about the ecology of plant responses to
multiple enemies. We would argue that molecular biologists should not shy away from
applying well-established techniques to native
plants whose responses to multiple attackers
have been defined in the field.
References
1 Bostock, R.M. (1999) Signal conflicts and
synergies in induced resistances to multiple
attackers. Physiol. Mol. Plant Pathol. 55, 99–109
2 Heil, M. (1999) Systemic acquired resistance:
available information and open ecological
questions. J. Ecol. 87, 341–346
3 Maleck, K. and Dietrich, R.A. (1999) Defense on
multiple fronts: how do plants cope with diverse
enemies? Trends Plant Sci. 4, 215–219
4 Genoud, T. and Metraux, J.P. (1999) Crosstalk in
plant cell signalling: structure and function of the
genetic network. Trends Plant Sci. 4, 503–507
5 Karban, R. and Baldwin, I.T. (1997) Induced
Responses to Herbivory, University of Chicago
Press
6 Hatcher, P.E. (1995) Three-way interactions
between plant pathogenic fungi, herbivorous
insects and their host plants. Biol. Rev. 70,
639–694
7 Schweizer, P. et al. (1998) Induced systemic
resistance in wounded rice plants. Plant J. 14,
475–481
8 Terras, F.R.G. et al. (1998) Evidence that the role
of plant defensins in radish defense responses is
independent of salicylic acid. Planta 206,
117–124
9 Felton, G.W. et al. (1999) Inverse relationship
between systemic resistance of plants to
microorganisms and to insect herbivory. Curr.
Biol. 9, 317–320
10 Stout, M.J. et al. (1998) Specificity of induced
resistance in the tomato, Lycopersicon
esculentum. Oecologia 113, 74–81
11 Thaler, J.S. et al. (1999) Trade-offs in plant
defense against pathogens and herbivores: a field
demonstration of chemical elicitors of induced
resistance. J. Chem. Ecol. 25, 1597–1609
12 Stout, M.J. et al. (1999) Plant-mediated
interactions between pathogens and herbivores of
the tomato, Lycopersicon esculentum. Physiol.
Mol. Plant Pathol. 54, 115–130
13 Underwood, N.C. (1999) The influence of plant
and herbivore characteristics on the interactions
between induced resistance and herbivore
population dynamics. Am. Nat. 153, 282–294
14 Haukioja, E. and Neuvonen, S. (1987) Insect
population dynamics and induction of plant
resistance: the testing of hypotheses. In Insect
Outbreaks (Barbosa, P. and Schultz, J.C., eds),
pp. 411–432, Academic Press
15 Simms, E.L. (1992) Costs of plant resistance to
herbivores. In Plant Resistance to Herbivores and
Pathogens, Ecology, Evolution and Genetics
(Fritz, R.S. and Simms, E.L., eds), pp. 392–425,
University of Chicago Press
16 Karban, R. et al. (1999) Induced plant responses
and information content about risk of herbivory.
Trends Ecol. Evol. 14, 443–447
trends in plant science
Perspectives
17 Pieterse, C.M.J. and van Loon, L.C. (1999)
Salicylic acid-independent plant defence
pathways. Trends Plant Sci. 4, 52–58
18 Hammerschmidt, R. and Nicholson, R.L. (1999)
A survey of plant defense responses to pathogens.
In Induced Plant Defenses Against Pathogens
and Herbivores: Biochemistry, Ecology and
Agriculture (Agrawal, A.A. et al., eds),
pp. 55–71, APS Press
19 Felton, G.W. et al. (1994) Oxidative responses in
soybean foliage to herbivory by bean leaf beetle
and three-cornered alfalfa hopper. J. Chem. Ecol.
20, 639–649
20 Karban, R. and Adler, F.R. (1996) Induced
resistance to herbivores and the information
content of early season attack. Oecologia 107,
379–385
21 Karban, R. and Kuc, J. (1999) Induced resistance
against pathogens and herbivores: an overview. In
Induced Plant Defenses Against Pathogens and
Herbivores: Biochemistry, Ecology and Agriculture
(Agrawal, A.A. et al., eds), pp. 1–16, APS Press
22 Hammerschmidt, R. and Smith-Becker, J.A. (1999)
The role of salicylic acid in disease resistance. In
Induced Plant Defenses Against Pathogens and
23
24
25
26
27
28
Herbivores: Biochemistry, Ecology and Agriculture
(Agrawal, A.A. et al., eds), pp. 37–53, APS Press
Sembdner, G. and Parthier, B. (1993) The
biochemistry and the physiological and molecular
actions of jasmonates. Annu. Rev. Plant Physiol.
Plant Mol. Biol. 44, 569–589
Reinbothe, S. et al. (1994) JIPs and RIPs: the
regulation of plant gene expression by jasmonates
in response to environmental cues and pathogens.
Plant Cell 6, 1197–1209
van der Meijden, E. et al. (1988) Defence and
regrowth: alternative plant strategies in the
struggle against herbivores. Oikos 51, 355–363
Hatcher, P.E. et al. (1994) The effect of a foliar
disease (rust) on the development of Gastrophysa
viridula (Coleoptera: Chrysomelidae). Ecol.
Entomol. 19, 349–360
Hatcher, P.E. et al. (1994) Interactions between
Rumex spp., herbivores and a rust fungus:
Gastrophysa viridula grazing reduces subsequent
infection by Uromyces rumicis. Funct. Ecol. 8,
265–272
Hatcher, P.E. and Paul, N.D. Beetle grazing
reduces natural infection of Rumex obtusifolius
by fungal pathogens. New Phytol. (in press)
Does the plant mitochondrion
integrate cellular stress and
regulate programmed cell
death?
Alan Jones
Research on programmed cell death in plants is providing insight into the
primordial mechanism of programmed cell death in all eukaryotes. Much of the
attention in studies on animal programmed cell death has focused on determining
the importance of signal proteases termed caspases. However, it has recently been
shown that cell death can still occur even when the caspase cascade is blocked,
revealing that there is an underlying oncotic default pathway. Many programmed
plant cell deaths also appear to be oncotic. Shared features of plant and animal
programmed cell death can be used to deduce the primordial components of
eukaryotic programmed cell death. From this perspective, we must ask whether
the mitochondrion is a common factor that can serve in plant and animal cell death
as a stress sensor and as a dispatcher of programmed cell death.
U
ntil recently, research in programmed
cell death (PCD) in animals focused
more on the role of the proteolytic cascade of caspases than on the morphotype of
death. Caspases play a pivotal role in animal
cell death, serving both as the amplifiers of
extracellular death signals and the means by
which specific cellular components are disassembled by apoptosis. However, PCD can also
occur independently of caspases; this finding
led to a re-examination of the role played by
these proteases in relation to cell death. In
studies where the activity of caspases was
blocked with general caspase inhibitors, apoptotic death was either prevented1,2 or only
partially executed3. Genetic knockouts of
apoptotic factors of the caspase cascade
[including Apaf-1 (Ref. 4) , caspase 3 (Ref. 5)
and caspase 8 (Ref. 6)] did not prevent death
even though some or all the apoptotic features
1360 - 1385/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01605-8
29 Hatcher, P.E. et al. (1997) Added soil nitrogen
does not allow Rumex obtusifolius to escape the
effects of insect–fungus interactions. J. Appl.
Ecol. 34, 88–100
30 Hatcher, P.E. et al. (1994) The effect of an
insect herbivore and a rust fungus individually,
and combined in sequence, on the growth of
two Rumex species. New Phytol.
128, 71–78
Nigel D. Paul* and Jane E. Taylor are at the
Dept of Biological Sciences, Institute of
Environmental and Natural Sciences,
University of Lancaster, Lancaster, UK
LA1 4YQ; Paul E. Hatcher is at the Dept of
Agricultural Botany, School of Plant Sciences,
University of Reading, 2 Earley Gate,
Whiteknights, Reading, UK RG6 6AU
(tel 144 118 931 6369; fax 144 118 935
1804; e-mail [email protected]).
*Author for correspondence
(tel 144 1524 593406; fax 144 1524
843854; e-mail [email protected]).
were blocked. In spite of this apoptosis inhibition, the cells died – they exhibited a type of
death termed oncosis7, which is generally considered to be unprogrammed. The observation
that cells can inducibly die, even when another
programmed cell death is blocked, provokes
the question, what is the basis of this oncoticlike death and can it be described at the molecular level? What is emerging from more
critical analyses of cell death is that caspasedriven death might not be the central execution
pathway8, but instead operates on top of an
underlying oncotic-like pathway. When caspase-driven death is blocked, the underlying
death pathway is revealed as being oncotic.
PCD occurs in other organisms, including
plants, but without the apoptotic morphotype
where corpse morphology is the result of caspase action9. Thus, it is not altogether surprising that there is neither sequence similarity to
caspase genes in yeast and plant sequence databases nor unequivocal evidence for caspase
activity in these organisms. Is there a primordial death mechanism that is universal in all
eukaryotes, but upon which the caspase proteolytic cascade operates in animal cells? Do
caspases represent an evolved sophistication of
cell death control and corpse management in
animals? Analyses of death control and corpse
management in organisms such as plants and
yeast should answer these questions.
What features are shared among
different cells?
Programmed cell death, which occurs independently of a central role for caspases, has returned
May 2000, Vol. 5, No. 5
225
Perspectives
Coping with multiple enemies:
an integration of molecular
and ecological perspectives
Nigel D. Paul, Paul E. Hatcher and Jane E. Taylor
How plants respond to attack by the range of herbivores and pathogens that
confront them in the field is the subject of considerable research by both
molecular biologists and ecologists. However, in spite of the shared focus of
these two bodies of research, there has been little integration between them. We
consider the scope for such integration, and how greater dialogue between
molecular biologists and ecologists could advance understanding of plant
responses to multiple enemies.
R
ecent research has highlighted how
plants respond to the diversity of organisms that attack them in the field1–4.
Much of this current interest derives from cellular studies of plant defence pathways. These
studies take a reductionist approach and attempt
to identify individual signalling pathways that
link wounding or attack by a pathogen or herbivore to specific gene expression events. For
example, in Arabidopsis this approach has been
successful in identifying key signalling molecules and their relative position in particular
pathways. It has also identified possible points
of cross-talk between pathways and synergistic
or antagonistic events. Recently, there has been
increasing interest in the possible interactions
between plant signalling pathways, including
those involved in responses to disease and
herbivory4.
Induced defence is also an important area of
ecological study5. In contrast with molecular
studies, the ecological view of induced defence
is broad, and is seen within the context of the
overall ‘defence strategy’ of the host plant. This
wider strategy includes constitutive defences,
the ‘quality’ of host tissues as a resource for
consumers, and tolerance (i.e. the capacity of
the plant to adjust its physiology to buffer the
effects of herbivory or disease). With few
exceptions6, ecologists have paid little attention
to plant responses to multiple enemies, and
studies of non-crop hosts have concentrated
largely on defence against herbivores.
Here we consider induced resistances from
the contrasting perspectives of molecular biology and ecology, and whether integration of
these different standpoints offers any novel
insights into plant defence against multiple
Box 1. Rumex –Gastrophysa–Uromyces: an ecological model for
host–herbivore–pathogen interactions
The beetle Gastrophysa viridula and the biotrophic rust fungus Uromyces rumicis (Fig. 1) both
attack the leaves of Rumex species (docks). The physiology and ecology of this system have
been intensively studied. Rust infection of dock reduces the growth, survivorship and fecundity of G. viridula (Fig. 2), which is associated with changes in the nutritional status of host
tissues following infection26. Conversely, G. viridula feeding induces resistances to the rust
fungus, both locally and systemically (Fig. 2)27. These effects are not confined to controlled
environments: rust infection is significantly and substantially reduced in the field by both beetle grazing28 and artificial wounding27. Thus, the combined effect of the beetle and the rust fungus on the host might be expected to be inhibited (i.e. the effect less than that of one of the
agents alone). However, the effect is usually either equivalent (in Rumex obtusifolius, damage
is equivalent to that of the agents; Fig. 2) or additive (in R. crispus, damage is equivalent to the
sum of the effects of each agent alone; Fig. 2)29,30, illustrating the difficulty in extrapolating
from the effects of the agents on each other to their effect on the host. This effect is robust even
when plants are grown with differing levels of nitrogen fertilization29. Tripartite interactions in
dock are not confined to Gastrophysa and Uromyces, because herbivory also significantly
reduced field infection by two other fungal pathogens of dock, the necrotroph Ramularia
rubella, and the hemibiotroph Venturia rumicis28. We are currently applying molecular techniques to the Rumex system to try to identify genes that are up-regulated in response to attack
by either the pathogen or the herbivore, and then to examine their spatial and temporal expression patterns in response to multiple attack. Our aim is to apply these techniques to investigate
the mechanisms of interactions between Rumex, Gastrophysa and Uromyces in the field.
220
May 2000, Vol. 5, No. 5
enemies. We will concentrate on three broad
issues: cross-talk between pathways, specificity of induced resistances and the evolution
of induced resistances in the context of attack
by multiple enemies.
Cellular perspective on systemic
induced resistance
Of the several pathways of systemic induced
defence in plants the best characterized are systemic acquired resistance (SAR), which provides enhanced resistance to pathogen
infection, and the wound response pathway,
which provides resistance to insect feeding.
Both pathways have been shown to induce the
expression of distinct sets of genes. SAR has
been shown to be induced by salicylic acid and,
in at least some systems, salicylic acid appears
to be both necessary and sufficient for inducing SAR. By contrast, the wound response
pathway appears to be induced by jasmonic
acid and ethylene without salicylic acid.
Although it has been widely accepted that the
pathways responsive to pathogen or herbivore
attack are distinct, some recent evidence suggests that jasmonic acid also mediates induced
resistance to some pathogens7,8. Thus, the strict
division of induced responses might be incorrect. In addition, the salicylic acid and jasmonic acid signalling pathways can interact1.
Although the nature of the cross-talk is not yet
fully defined, recent data suggest that induction of the salicylic acid pathway often inhibits
the systemic induction of the jasmonic acid
pathway by later herbivory. Conversely, the
induction of the jasmonic acid pathway can
reduce induction of the salicylic acid pathway
following subsequent pathogen attack9–12.
Ecological perspective on systemic
induced resistance
The molecular biology of induced resistances
has been investigated mostly in herbaceous
species. More than 100 species are known to
express induced resistance following herbivory, but two-thirds of these are trees or
shrubs5. The vast majority of ecological
research into induced resistance deals with herbivores, ranging from mites to large mammals,
and it is clear that induced systemic resistances
are widespread and of sufficient magnitude to
significantly reduce herbivore populations5,13.
In long-lived plants, induced changes can
persist to reduce herbivore populations in the
years following induction14, and in several forest systems the decay of induced resistance can
be over several herbivore generations14.
Where the mechanisms underlying these
ecologically significant induced defences
have been defined, it is clear that different
plant species use a great diversity of
responses. As well as induced chemical and
morphological changes, responses include
elements that are more closely linked to
1360 - 1385/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01603-4
trends in plant science
Perspectives
tolerance (for example, altered protein or
nitrogen content in regrowth leaves) than
resistance per se5. Ecologists have also considered the broader issue of the selective
advantages of induced resistance, in particular the relationship between the benefits and
costs of induction15. This in turn leads to questions concerning the population genetics and
evolution of induced defence. Current theories
are based on the concept that current herbivory
or disease is correlated with the risk of future
attack16. In this sense, current or past attack
can be seen as having the potential to provide
information about the future environment. On
this basis, induced resistances will be selected
for only if (i) current attack is a reliable predictor of future attack, and (ii) if attack reduces
plant fitness.
This body of ecological information on
induced defence against herbivores is in stark
contrast with the limited knowledge of comparable responses to pathogens. Given the
limited understanding of plant pathogens in
natural systems it is not surprising that
responses to multiple enemies are also poorly
understood. One of the few examples of multiple defence to have been studied in a noncrop system – Rumex spp. attacked by the
beetle Gastrophysa viridula and the rust fungus Uromyces rumicis – highlights the potential complexity of such interactions (Box 1;
Figs 1 and 2). Several whole plant studies
using crops also highlight the diversity of herbivore–pathogen interactions: with examples
of attack by one organism inducing resistance
to attack by another, but there are also many
examples of exceptions6.
Defence against multiple enemies
Cross-talk
Molecular biologists have considered
‘defence on multiple fronts’ in terms of interactions between different pathways of induced
resistance1,3,17. For example, whether induction of the salicylic acid pathway influences
subsequent induction of the jasmonic acid
pathway and vice versa. In this respect there
is evidence that resistance induced via one
pathway can suppress induced responses
(b)
Number of eggs laid by
G. viridula
1000
900
800
700
600
500
400
300
200
100
0
Control
plants
Rusted
leaves
mediated via another pathway9–11. However,
this is only part of the story, and there are
examples of induction by one agent resulting
in increased resistance to another12. It is also
increasingly clear that the concept of specific
pathogen and wounding (herbivore)-induced
pathways mediated by salicylic acid and
jasmonic acid, respectively, is simplistic1.
Jasmonic acid also mediates resistances to certain pathogens, and both the salicylic acid and
jasmonic acid pathways induce a spectrum of
(c)
35
120
Plant dry weight (g)
U. rumicis pustules (cm–2)
(a)
Fig. 1. (a) Rumex obtusifolius foliage attacked by the rust fungus Uromyces rumicis.
(b) Rumex obtusifolius foliage attacked by the chrysomelid beetle Gastrophysa viridula,
which attacks Rumex spp. throughout its larval stages (illustrated), and as an adult.
30
25
20
15
10
5
100
80
60
40
20
0
0
Control
plants
Ungrazed Grazed
leaf
leaf
Infected plant
Control Beetle
Rust
RB
Trends in Plant Science
Fig. 2. (a) Fecundity of Gastrophysa viridula reared on Rumex obtusifolius (white) and R. crispus (grey), infected by Uromyces rumicis (rust).
Adults had been fed on the no-choice treatments since hatching in Petri dishes, given excess food, and reared at constant temperature. Data for
control and infected plants are means of 9–15 plants 6 1 SE. Modified from Ref. 26. (b) Infection of Rumex obtusifolius (white) and R. crispus
(grey) by Uromyces rumicis (rust) seven days after feeding by the beetle Gastrophysa viridula. Data for control plants, and grazed and ungrazed
leaves of grazed plants are means of 14–15 plants 6 1 SE. Modified from Ref. 27. (c) The effects of grazing by the beetle Gastrophysa viridula
and infection by Uromyces rumicis, alone and in combination, on the growth of Rumex obtusifolius (white) and R. crispus (grey). Key:
control 5 plants with neither beetle nor rust; beetle 5 plants with beetles only; rust 5 plants with rust only; RB 5 plants infected by U. rumicis that were then grazed by G. viridula. Data are means of 15 plants 6 1 SE. Modified from Ref. 30.
May 2000, Vol. 5, No. 5
221
trends in plant science
Perspectives
Herbivore
Pathogen
b
d
e
d
e
Recognition
Signalling pathways
Gene products (+ or —)
a
c
f
g
h
Effects on attackers
Effects on host plant
Overlap between responses
Trends in Plant Science
Fig. 3. Specificity and overlap in induced resistances: a question of perspective? There is no
question that the host plant has the capacity to recognize attack by different organisms with
a high degree of specificity. At the level of recognition, it is likely that there is no ‘overlap’
or ‘cross-talk’ between responses to different attackers. The signal transduction pathways
involved in induction following attack by herbivores or pathogens also appear to be largely
specific, but recent evidence suggests that there are elements of cross-talk, both positive and
negative. It is clear that different attackers can affect the expression of distinct sets of genes,
and result in different sets of gene products, but such ‘fine-tuning’ might also include individual elements that are less specific17. For example, a herbivore might induce the expression of genes abcde and a pathogen might induce genes defgh. This could be taken as
evidence of specificity or overlap. A more concrete example is that both herbivores and
pathogens might induce specific sets of genes related to the synthesis of phenolics, but
induction of PAL, and/or the resultant general up-regulation of phenolic metabolism is common to a wide range of attackers. The effects of these gene products on attackers can be even
less specific: the major classes of secondary chemicals that are inducible by arthropods have
been shown to have anti-microbial properties and vice versa6. In addition, ‘non-defence’
changes, such as alterations in tissue nitrogen might have wider-ranging effects. Finally,
pathogen and herbivore-induced changes viewed from the perspective of their effects on the
overall functioning and fitness of the host plant, including both defence and ‘civilian’
responses are likely to show considerable overlap.
gene products that although distinct to the
pathway show considerable overlap between
the two (Fig. 3). The relationship between specific gene products and resistance is an area of
research that could benefit from closer liaison
between molecular biologists and ecologists.
We echo the recent suggestion that the value
of plant defence investigations would be
increased if they were to concentrate on how
a putative defence actually contributes to
resistance18.
Likewise, the literature dealing with herbivores and pathogens, rather than altered levels
222
May 2000, Vol. 5, No. 5
of gene products, frequently indicates that the
effects of induced resistances are often not specific to any particular attacker. Thus, although
induced resistances might show some degree
of ‘fine-tuning’ (Fig. 3), it might be expected
that resistance induced by one attacker would
be effective against a range of other potential
attackers. Molecular studies focusing on biochemical mechanisms of ‘cross-talk’ between
pathways would benefit from complementary
studies considering, for example, the extent to
which induced changes are specific in terms of
their effects on a range of attackers.
Specificity
As we have discussed, many of the biochemical responses induced by resistance pathways
are effective against a wide range of organisms. This specificity of effect, or lack of, must
be distinguished from the plant’s specificity of
response (Fig. 3). The plant’s specificity of
response is the ability of the plant to distinguish between different types of biotic challenge and to produce different sets of
chemicals in response to each challenge.
Research suggests that plants do possess
specificity of response10,19.
Reviews of the molecular biology of
induced resistance suggest that plants might
‘fine-tune’ induced resistances to specific
attackers3,17. These arguments seem consistent
with current evolutionary theory. Induced
resistance is selected for when past or current
attack is correlated with the risk of future
attack. Such correlation is likely to be greatest when considering current and future attack
by the same organism, favouring selection of
a ‘fine-tuned’ induced response. However,
selection reflects the interplay of benefits and
costs of induced resistances, which should be
considered when determining the selective
advantage of ‘fine-tuned’ resistances. There
are many questions here, such as:
• Is the cost of a fine-tuned resistance less
than that of a broad-spectrum resistance?
• Is induced resistance finely tuned to a particular attacker likely to confer more or less
benefit than a broad-spectrum induced
resistance?
Although these questions remain unanswered
it is notable that the suggestion of fine-tuning
appears to be at odds with much of the literature that shows that resistance induced by one
organism is often effective against a range of
potential attackers. This might be explained
partly by a wider range of induced changes
than are often considered as resistance mechanisms. Systemic changes in the nutritional
quality of host tissues (e.g. nitrogen chemistry) following pathogen infection might
have rather non-specific effects on herbivores,
and potentially modify responses to induced
allelochemicals. There is less evidence that
herbivore-induced changes in ‘non-defence’
chemistry have systemic effects on pathogens.
However, these broader changes in host chemistry should not be overlooked when considering the biology of plant responses to
multiple enemies.
Evolution of induced resistances in the
context of attack by multiple enemies
There has been considerable debate over
whether a characteristic of the host affecting a
particular attacker has necessarily resulted
from selection pressure imposed by that
attacker5. Of course, this also applies to characteristics acting against multiple enemies.
trends in plant science
Perspectives
Herbivore attack
Pathogen
Civilian
Defence
Systemin
JA-dependent
SA-dependent
Damage
Jasmonic acid
or ethylene
Proteinase
inhibitors
Basic PR
proteins
Salicylic acid
Acidic PR proteins,
defensins,
thionins
Wound-induced SA-independent
SAR
SAR
?
?
Tolerance
SA-induced
SAR
Host growth and development
Host fitness
Trends in Plant Science
Fig. 4. ‘Defence’ and ‘civilian’ elements of induced responses and the possible interactions between them. The range of responses in host
metabolism and physiology that result from attack by herbivores and attack by pathogens can be divided into ‘defence’ and ‘civilian’ reactions.
Induced defence includes responses mediated via the salicylic acid and jasmonic acid pathways, which are well defined. Civilian responses
reflect the interplay of damage and tolerance – physiological and metabolic adjustments made in response to damage. The nature of damage
varies markedly with the attacking organism, but adjustment at the whole-plant level shares a degree of commonality between, for example, the
range of organisms that attack the foliage. The regulation of the balance between damage and tolerance remains poorly defined. Even less
defined are the interactions between the civilian and defence responses. Possible interactions include the effects of salicylic acid and jasmonic
acid on civilian responses independent of defence, and the effects of metabolic shifts necessary for induced resistance on civilian responses.
Equally, the shifts in metabolism and physiology inherent in civilian responses might influence the expression of induced resistances (at the
level of the production of resistance compounds and/or the responses of attackers to those compounds). Host growth, development and, ultimately, fitness cannot be seen as a function of defence or civilian responses, but as the outcome of the interaction between both sets of responses.
Further, because the products of induced
resistance(s) might be active against a wide
range of organisms, it is possible that broadspectrum induced resistance is the result
entirely of selection by a single attacker, rather
than integrated selection by diverse enemies.
However, given that attack by multiple and
diverse enemies is commonplace in the field,
can current models for the selection of induced
resistances be applied to this situation? Such
models are based on ‘information’, defined as
the ability of current attack to predict future
attack. Selection for induced resistance is predicted when past or current attack provides
‘information’ concerning (i.e. is well correlated with) the risk of future attack. Such correlation has been shown experimentally in a
few systems20 and might be expected for many
others. However, is it likely that attack by a
herbivore contains any information about
future attack by pathogens, or vice versa?
Arguably, herbivory, by creating wounds that
might be vulnerable to invasion by microorganisms2,5 predicts an increased risk of future
pathogen attack. However, this is a localized
response, consistent with the induction of
resistance in the vicinity of wounds. Is it likely
that herbivory predicts the risk of pathogen
attack in ungrazed tissues? Equally, on a
whole-plant basis, is past or current pathogen
attack likely to be closely correlated to future
herbivory? If such correlations between current and future attack are unlikely across multiple enemies, this suggests that selection of
broad-spectrum induced resistance is unlikely.
Selection based on ‘information’ is also difficult to relate to negative interactions between
induced responses, such that the response to
one attacker might constrain the plant’s ability to respond to another. Indeed, recent argu-
ments suggest that such interactions reflect the
manipulation of pathways by attackers21,
although this appears to be based on the concept of specific herbivore (jasmonic acid) and
pathogen (salicylic acid) pathways, which is
at odds with some recent evidence.
An alternative view of selection arises if we
develop the suggestion5 that induced resistance should be considered in the context of the
overall response of a plant to attack. This
includes what have been called ‘civilian’
responses in host physiology5. For example,
following foliar attack, civilian responses
include remobilization of reserves from storage tissues, increased partitioning of assimilate to leaves rather than roots and
up-regulation of photosynthesis in undamaged
leaves. Mechanisms of tolerance appear common to attack by both leaf-feeding herbivores
and foliar pathogens. These physiological
May 2000, Vol. 5, No. 5
223
trends in plant science
Perspectives
responses to attack buffer the host from the
effects of an initial attack but can increase
vulnerability to the effects of future attack.
Taking up this theme, a more phytocentric
viewpoint of selection for induced resistance
might be that current attack (regardless of the
organisms doing the attacking) conveys
information about the vulnerability of the
plant to the physiological effects of any
future attack. We hypothesize that attack by
any (foliar) attacker is a reliable predictor of
the physiological effect of future attack by
any (other) foliar attacker(s). This is also
consistent with the suggestion that induced
resistances allow maximum expression of
the plant’s potential to tolerate herbivory or
disease5, because the spatial distribution of
attack over the plant is altered, maintaining
undamaged tissues capable of compensatory
physiological changes5.
Viewing current attack as providing information about the physiological effects of
future attacks requires integration between
induced resistance and changes in the
primary physiology of the plant following
attack. Physiological responses to salicylic
acid and jasmonic acid are known to extend
far beyond induced resistance, but their
effects on ‘civilian’ responses are poorly
understood22–24. However, it is notable that
exogenous applications of both salicylic acid
and jasmonic acid have been shown to inhibit
photosynthesis, reducing levels of Rubisco
and chlorophylls24. Exogenous applications
of salicylic acid or jasmonic acid lack the
temporal and spatial organization of
responses to attack by consumer organisms.
However, these effects of exogenous jasmonic acid and salicylic acid are in marked
contrast with the increase in photosynthesis
in undamaged leaves that is a common
response to herbivory and disease. This paradox highlights the need for a greater understanding of post-induction changes in
‘civilian’ genes that have no direct relationship to resistance.
Salicylic acid and jasmonic acid also interact with other plant growth regulators, including abscisic acid and ethylene, which are
central to the coordination of responses to the
abiotic environment4. The relationships
between plant responses to biotic and abiotic
stress have recently been discussed in terms of
how abiotic stress might influence induced
resistances deployed in agriculture1 or whether
the abiotic environment provides ‘information’
about the risk of future herbivory or disease16.
We argue that responses to biotic and abiotic
stress are linked because optimizing induced
responses to minimize the physiological
effects of attack is highly dependent on the
abiotic environment. This includes civilian
responses such as altered partitioning between
shoot and root, as well as chemical resistance6.
224
May 2000, Vol. 5, No. 5
Conclusions
Acknowledgements
We have approached the issue of defence
against multiple enemies from the contrasting
backgrounds of molecular biology and plant
ecology. Once the initial hurdle of the lack of a
common terminology was overcome, it became
clear that the major limitation to integrating
these viewpoints was the gap between the level
of organization we normally dealt with. A key
We thank NERC for financial support, and
Peter Ayres and John Whittaker for many
valuable discussions.
“…tolerance and induced
resistances can be considered
as complementary elements
of phenotypic plasticity”
constraint on the integration of the details of
signalling pathways and specific gene products
in the few model systems with field data was
the relative scarcity of whole plant investigations of induced responses to herbivory and/or
disease. There is a risk that quantifying the
effects of induced responses on the herbivores
or on the pathogens rather than on the host plant
has falsely emphasized mechanisms that induce
resistance, but fail to benefit plants, while
underestimating the significance of tolerance
and other induced ‘civilian’ responses21.
Consideration of this ‘middle ground’ has
highlighted the need to integrate the available
information from this ‘phytocentric’ viewpoint. We echo the view5 that defence and
civilian aspects of induced responses to attack
should be considered together (Fig. 4). By
analogy with the term used in the context of
abiotic stress, induced resistance and tolerance
can be treated as elements of acclimation to
attack by herbivores and/or pathogens.
Although tolerance and (constitutive) resistance might be alternative adaptive strategies25,
tolerance and induced resistances can be considered as complementary elements of phenotypic plasticity. Thus, we would argue that it
would be fruitful for molecular biologists,
ecologists and plant physiologists to consider
the functional links between tolerance and
induced resistance.
Clearly, progress in this area will benefit
from the use of appropriate model systems. We
agree with the recent suggestion that resolving
uncertainties about signalling conflicts and
synergies in induced resistance will benefit
from the use of systems that are convenient for
genetic, molecular, biochemical and physiological studies1. However, the use of standard
laboratory models can provide little information about the ecology of plant responses to
multiple enemies. We would argue that molecular biologists should not shy away from
applying well-established techniques to native
plants whose responses to multiple attackers
have been defined in the field.
References
1 Bostock, R.M. (1999) Signal conflicts and
synergies in induced resistances to multiple
attackers. Physiol. Mol. Plant Pathol. 55, 99–109
2 Heil, M. (1999) Systemic acquired resistance:
available information and open ecological
questions. J. Ecol. 87, 341–346
3 Maleck, K. and Dietrich, R.A. (1999) Defense on
multiple fronts: how do plants cope with diverse
enemies? Trends Plant Sci. 4, 215–219
4 Genoud, T. and Metraux, J.P. (1999) Crosstalk in
plant cell signalling: structure and function of the
genetic network. Trends Plant Sci. 4, 503–507
5 Karban, R. and Baldwin, I.T. (1997) Induced
Responses to Herbivory, University of Chicago
Press
6 Hatcher, P.E. (1995) Three-way interactions
between plant pathogenic fungi, herbivorous
insects and their host plants. Biol. Rev. 70,
639–694
7 Schweizer, P. et al. (1998) Induced systemic
resistance in wounded rice plants. Plant J. 14,
475–481
8 Terras, F.R.G. et al. (1998) Evidence that the role
of plant defensins in radish defense responses is
independent of salicylic acid. Planta 206,
117–124
9 Felton, G.W. et al. (1999) Inverse relationship
between systemic resistance of plants to
microorganisms and to insect herbivory. Curr.
Biol. 9, 317–320
10 Stout, M.J. et al. (1998) Specificity of induced
resistance in the tomato, Lycopersicon
esculentum. Oecologia 113, 74–81
11 Thaler, J.S. et al. (1999) Trade-offs in plant
defense against pathogens and herbivores: a field
demonstration of chemical elicitors of induced
resistance. J. Chem. Ecol. 25, 1597–1609
12 Stout, M.J. et al. (1999) Plant-mediated
interactions between pathogens and herbivores of
the tomato, Lycopersicon esculentum. Physiol.
Mol. Plant Pathol. 54, 115–130
13 Underwood, N.C. (1999) The influence of plant
and herbivore characteristics on the interactions
between induced resistance and herbivore
population dynamics. Am. Nat. 153, 282–294
14 Haukioja, E. and Neuvonen, S. (1987) Insect
population dynamics and induction of plant
resistance: the testing of hypotheses. In Insect
Outbreaks (Barbosa, P. and Schultz, J.C., eds),
pp. 411–432, Academic Press
15 Simms, E.L. (1992) Costs of plant resistance to
herbivores. In Plant Resistance to Herbivores and
Pathogens, Ecology, Evolution and Genetics
(Fritz, R.S. and Simms, E.L., eds), pp. 392–425,
University of Chicago Press
16 Karban, R. et al. (1999) Induced plant responses
and information content about risk of herbivory.
Trends Ecol. Evol. 14, 443–447
trends in plant science
Perspectives
17 Pieterse, C.M.J. and van Loon, L.C. (1999)
Salicylic acid-independent plant defence
pathways. Trends Plant Sci. 4, 52–58
18 Hammerschmidt, R. and Nicholson, R.L. (1999)
A survey of plant defense responses to pathogens.
In Induced Plant Defenses Against Pathogens
and Herbivores: Biochemistry, Ecology and
Agriculture (Agrawal, A.A. et al., eds),
pp. 55–71, APS Press
19 Felton, G.W. et al. (1994) Oxidative responses in
soybean foliage to herbivory by bean leaf beetle
and three-cornered alfalfa hopper. J. Chem. Ecol.
20, 639–649
20 Karban, R. and Adler, F.R. (1996) Induced
resistance to herbivores and the information
content of early season attack. Oecologia 107,
379–385
21 Karban, R. and Kuc, J. (1999) Induced resistance
against pathogens and herbivores: an overview. In
Induced Plant Defenses Against Pathogens and
Herbivores: Biochemistry, Ecology and Agriculture
(Agrawal, A.A. et al., eds), pp. 1–16, APS Press
22 Hammerschmidt, R. and Smith-Becker, J.A. (1999)
The role of salicylic acid in disease resistance. In
Induced Plant Defenses Against Pathogens and
23
24
25
26
27
28
Herbivores: Biochemistry, Ecology and Agriculture
(Agrawal, A.A. et al., eds), pp. 37–53, APS Press
Sembdner, G. and Parthier, B. (1993) The
biochemistry and the physiological and molecular
actions of jasmonates. Annu. Rev. Plant Physiol.
Plant Mol. Biol. 44, 569–589
Reinbothe, S. et al. (1994) JIPs and RIPs: the
regulation of plant gene expression by jasmonates
in response to environmental cues and pathogens.
Plant Cell 6, 1197–1209
van der Meijden, E. et al. (1988) Defence and
regrowth: alternative plant strategies in the
struggle against herbivores. Oikos 51, 355–363
Hatcher, P.E. et al. (1994) The effect of a foliar
disease (rust) on the development of Gastrophysa
viridula (Coleoptera: Chrysomelidae). Ecol.
Entomol. 19, 349–360
Hatcher, P.E. et al. (1994) Interactions between
Rumex spp., herbivores and a rust fungus:
Gastrophysa viridula grazing reduces subsequent
infection by Uromyces rumicis. Funct. Ecol. 8,
265–272
Hatcher, P.E. and Paul, N.D. Beetle grazing
reduces natural infection of Rumex obtusifolius
by fungal pathogens. New Phytol. (in press)
Does the plant mitochondrion
integrate cellular stress and
regulate programmed cell
death?
Alan Jones
Research on programmed cell death in plants is providing insight into the
primordial mechanism of programmed cell death in all eukaryotes. Much of the
attention in studies on animal programmed cell death has focused on determining
the importance of signal proteases termed caspases. However, it has recently been
shown that cell death can still occur even when the caspase cascade is blocked,
revealing that there is an underlying oncotic default pathway. Many programmed
plant cell deaths also appear to be oncotic. Shared features of plant and animal
programmed cell death can be used to deduce the primordial components of
eukaryotic programmed cell death. From this perspective, we must ask whether
the mitochondrion is a common factor that can serve in plant and animal cell death
as a stress sensor and as a dispatcher of programmed cell death.
U
ntil recently, research in programmed
cell death (PCD) in animals focused
more on the role of the proteolytic cascade of caspases than on the morphotype of
death. Caspases play a pivotal role in animal
cell death, serving both as the amplifiers of
extracellular death signals and the means by
which specific cellular components are disassembled by apoptosis. However, PCD can also
occur independently of caspases; this finding
led to a re-examination of the role played by
these proteases in relation to cell death. In
studies where the activity of caspases was
blocked with general caspase inhibitors, apoptotic death was either prevented1,2 or only
partially executed3. Genetic knockouts of
apoptotic factors of the caspase cascade
[including Apaf-1 (Ref. 4) , caspase 3 (Ref. 5)
and caspase 8 (Ref. 6)] did not prevent death
even though some or all the apoptotic features
1360 - 1385/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01605-8
29 Hatcher, P.E. et al. (1997) Added soil nitrogen
does not allow Rumex obtusifolius to escape the
effects of insect–fungus interactions. J. Appl.
Ecol. 34, 88–100
30 Hatcher, P.E. et al. (1994) The effect of an
insect herbivore and a rust fungus individually,
and combined in sequence, on the growth of
two Rumex species. New Phytol.
128, 71–78
Nigel D. Paul* and Jane E. Taylor are at the
Dept of Biological Sciences, Institute of
Environmental and Natural Sciences,
University of Lancaster, Lancaster, UK
LA1 4YQ; Paul E. Hatcher is at the Dept of
Agricultural Botany, School of Plant Sciences,
University of Reading, 2 Earley Gate,
Whiteknights, Reading, UK RG6 6AU
(tel 144 118 931 6369; fax 144 118 935
1804; e-mail [email protected]).
*Author for correspondence
(tel 144 1524 593406; fax 144 1524
843854; e-mail [email protected]).
were blocked. In spite of this apoptosis inhibition, the cells died – they exhibited a type of
death termed oncosis7, which is generally considered to be unprogrammed. The observation
that cells can inducibly die, even when another
programmed cell death is blocked, provokes
the question, what is the basis of this oncoticlike death and can it be described at the molecular level? What is emerging from more
critical analyses of cell death is that caspasedriven death might not be the central execution
pathway8, but instead operates on top of an
underlying oncotic-like pathway. When caspase-driven death is blocked, the underlying
death pathway is revealed as being oncotic.
PCD occurs in other organisms, including
plants, but without the apoptotic morphotype
where corpse morphology is the result of caspase action9. Thus, it is not altogether surprising that there is neither sequence similarity to
caspase genes in yeast and plant sequence databases nor unequivocal evidence for caspase
activity in these organisms. Is there a primordial death mechanism that is universal in all
eukaryotes, but upon which the caspase proteolytic cascade operates in animal cells? Do
caspases represent an evolved sophistication of
cell death control and corpse management in
animals? Analyses of death control and corpse
management in organisms such as plants and
yeast should answer these questions.
What features are shared among
different cells?
Programmed cell death, which occurs independently of a central role for caspases, has returned
May 2000, Vol. 5, No. 5
225