ANRV283-PY44-21 ARI 13 June 2006 13:56
Climate Change Effects on
Plant Disease: Genomes
to Ecosystems
K. A. Garrett, S. P. Dendy, E. E. Frank,
M. N. Rouse, and S. E. Travers
Department of Plant Pathology, Kansas State University, Manhattan, Kansas 66506;
email: kgarrett@ksu.edu; sdendy@ksu.edu; efrank@ksu.edu; mrouse@ksu.edu;
travers@ksu.edu
Annu. Rev. Phytopathol.
2006. 44:489–509
First published online as a
Review in Advance on
May 23, 2006
The Annual Review of
Phytopathology is online at
phyto.annualreviews.org
doi: 10.1146/
annurev.phyto.44.070505.143420
Copyright
c
2006 by
Annual Reviews. All rights
reserved
0066-4286/06/0908-
0489$20.00
Key Words
climate variability, disease ecology, ecological genomics,
epidemiology, global warming
Abstract
Research in the effects of climate change on plant disease contin-
ues to be limited, but some striking progress has been made. At the
genomic level, advances in technologies for the high-throughput
analysis of gene expression have made it possible to begin discrim-
inating responses to different biotic and abiotic stressors and po-
tential trade-offs in responses. At the scale of the individual plant,
enough experiments have been performed to begin synthesizing the
effects of climate variables on infection rates, though pathosystem-
specific characteristics make synthesis challenging. Models of plant
disease have now been developed to incorporate more sophisticated
climate predictions. At the population level, the adaptive potential
of plant and pathogen populations may prove to be one of the most
important predictors of the magnitude of climate change effects.
Ecosystem ecologists are now addressing the role of plant disease in
ecosystem processes and the challenge of scaling up from individual
infection probabilities to epidemics and broader impacts.
489
Annu. Rev. Phytopathol. 2006.44:489-509. Downloaded from arjournals.annualreviews.org
by Kansas State University on 08/09/06. For personal use only.
ANRV283-PY44-21 ARI 13 June 2006 13:56
IPCC:
Intergovernmental
Panel on Climate
Change
INTRODUCTION
Eight years ago, Coakley et al. (38) reviewed
the implications of climate change for plant
disease management in the Annual Review of
Phytopathology series. They pointed out sev-
eral challenges for evaluating the likely ef-
fects of climate change. Most experiments
considering climate change effects include
only one or two of the changing climatic
factors, experiments tend to be performed
under conditions very different from those
in the field, and experiments are generally
short-term. But there were already enough re-
sults in hand to indicate that climate change
could “alter stages and rates of develop-
ment of the pathogen, modify host resis-
tance, and result in changes in the physiol-
ogy of host-pathogen interactions.” Coakley
et al. (38) concluded that the effects of climate
change on plant disease management may
be less important than changes in land-use
patterns, transgenic technologies, and avail-
ability of chemical pesticides. Another gen-
eral conclusion was that the effects of climate
change will tend to be different for different
pathosystems in different locations, so that
generalization is a challenge. Here we con-
sider multiple scales of host-pathogen inter-
action (Figure 1) and review factors that con-
tribute to determining how and when climate
change could have important effects on plant
disease.
Since the review by Coakley et al. (38),
what has changed? Consensus has contin-
ued building among climatologists that global
warming is occurring and linked to human
activity (62). Scientists have also continued
to evaluate the effects of climate change on
disease risk across systems (63). More stud-
ies of the “fingerprint” of global warming
have appeared as interest in the effect grows
and as trends become more distinct (66, 85,
108, 120, 140). More climate change simula-
tion experiments have been put in place (28).
Ecologists working outside agricultural sys-
tems have turned more attention to the ecol-
ogy of disease (58). There has been an explo-
sion in the development of genomics tools and
their application (reviewed in 56). And, with
the turn of the millennium, groups such as
the UN have reevaluated progress toward so-
cietal goals through the formulation of Mil-
lennium Development Goals and the Millen-
nium Ecosystem Assessment, while the U.S.
National Research Council has formulated a
list of Grand Challenges for the environmen-
tal sciences, which includes climate change as
well as infectious disease (96).
CLIMATE CHANGE
The Intergovernmental Panel on Climate
Change (IPCC), which was jointly established
by the World Meteorological Organization
(WMO) and the United Nations Environ-
ment Program (UNEP) in 1988, has respon-
sibility for assessing information relevant to
climate change and summarizing this infor-
mation for policy makers and the public. It
has published major assessment reports most
recently in 1995 and 2001 (69). A new as-
sessment is scheduled for publication in 2007,
and updated predictions are available in other
publications (e.g., 142). Since the 1995 report,
there have been a number of advances, includ-
ing improvements in the Atmosphere-Ocean
General Circulation Models (AOGCM) used
to predict climate change. Other improve-
ments include better regionalization tech-
niques, a better understanding of the physical
processes underlying the models, and better
availability of paleoclimate data for evaluating
long-term temperature change and historic
climate data for evaluating preindustrial
atmospheric concentrations of greenhouse
gases.
Climate change predictions are based on
scenarios that describe greenhouse gas emis-
sions from potential resource use patterns,
technological innovations, and demographics.
The results from modeling experiments based
on these emissions scenarios give a range of
predictions, depending on the assumptions
quantified by each scenario. Sources of un-
certainty in predictions include inability to
490 Garrett et al.
Annu. Rev. Phytopathol. 2006.44:489-509. Downloaded from arjournals.annualreviews.org
by Kansas State University on 08/09/06. For personal use only.
ANRV283-PY44-21 ARI 13 June 2006 13:56
Host plant
Host-pathogen
interactions
Host-pathogen
interactions
Community
and ecosystem
dynamics
Regional
ecosystem
dynamics
Biosphere
E
nvir
o
nm
e
nt
Mi
cr
o
c
lim
a
t
e
Mi
cr
o
c
lim
a
t
e
L
ocal
c
lim
a
t
e
L
ocal
c
lim
a
t
e
R
e
g
io
n
l
c
li
t
e
R
e
g
io
n
a
l
li
m
a
t
e
S
ub
c
on
t
i
e
n
t
al
G
l
o
b
al
c
lim
a
t
e
P
atho
g
e
n
Intra
p
o
p
ulati
o
n
dynamics
Examples of potential
climate change effects
Research needs
Heating of
montane prairie
had mixed
effects (121)
Increased
CO
2
increased
fungal pathogen
load in prairie
(91)
Higher fecundity of
Colletotrichum
gloeosporioides
under increased
CO
2
(29)
Good models of
interspecific
interactions like
competition and
facilitation
Pathogen role in
long-term
ecological
processes
Needle blight moving
north as precipitation
patterns change (148)
P. cinnamomi
predicted expansion
in Europe due to
temperature
change (16)
Soybean rust
pathogen
immigration
potentially via
hurricane
Downregulation of
HR and other
genes in tallgrass
prairie grass in
response to
precipitation
change (Travers et al.
In preparation)
Peanut gene
expression
response to
drought and
Aspergillus (81)
Proteomic
and/or
metabolomic
studies of host
and pathogen
responses
Stomatal
closure and
leaf growth
inhibition
during
drought, e.g.
(32)
Gene
expression in
plants and
pathogens in
response to
climatic
factors
Multifactor
studies of
climate
change
effects
Better models of
adaptation rates
Better data and
models related to
dispersal, current
levels of intraspe-
cific diversity,
strength of selection
under different
climate change
scenarios, and
heritability of traits
Long-term large--
scale records of
pathogen and host
distributions
Models of regional
processes that
incorporate disease
Data and models
regarding dispersal
of propagules and
vectors
Integrated
multi-
disciplinary
international
networks for
data collection
and synthesis
Plant
structural
changes in
response to
CO
2
(117)
G
eno
m
e Physiology
C
ellula
r
processes
Intrapopulation
dynamics
Genome Physiology
Cellular
processes
Figure 1
Examples of potential climate change effects and research needs across biological scales. Arrows indicate
propagation of effects from smaller to larger processes, but feedbacks will also link across scales.
fully predict human resource use and incom-
plete understanding of climate processes. In
addition to the predicted increases in tem-
perature for much of the world, changes in
extremes are also predicted. For tempera-
ture, more frequent extreme high temper-
atures and less frequent extreme low tem-
peratures are predicted. Likewise, increased
intensity of precipitation events is predicted in
some regions. Although the IPCC (69) con-
cluded in 2001 that there was no compelling
evidence that characteristics of tropical and
extratropical storms have changed, more re-
cent analyses have concluded that there have
been changes in storm patterns in recent years
(47, 145), which could influence the global
movement of pathogens (26).
Additional predictive variability comes
into play for modeling of regional climates
(69). All the forms of uncertainty about global
www.annualreviews.org
•
Climate Change Effects on Plant Disease 491
Annu. Rev. Phytopathol. 2006.44:489-509. Downloaded from arjournals.annualreviews.org
by Kansas State University on 08/09/06. For personal use only.
ANRV283-PY44-21 ARI 13 June 2006 13:56
processes are still a factor, with additional
uncertainty due to lack of data from some
regions. Meteorological stations in some
regions are sparse, particularly in remote re-
gions with complex topography that may
produce rapid climatic variation over small ar-
eas. While water vapor, evaporation, and pre-
cipitation are predicted to increase on aver-
age, predictions about increased or decreased
precipitation are region specific. In general,
precipitation is predicted to increase in both
summer and winter in high-latitude regions.
In northern mid-latitudes, Antarctica, and
tropical Africa, precipitation is predicted to
increase in winter. In southern and eastern
Asia, precipitation is predicted to increase in
summer. Decreases in winter rainfall are pre-
dicted for southern Africa, Central America,
and Australia. Supplementary material on
IPCC websites supplies finer scale predic-
tions. Decreased snow cover and land-ice ex-
tent are expected to follow from the trend in
increasing temperature.
PLANT RESPONSES TO
CLIMATE CHANGE
Plant Responses in General: At the
Level of the Individual
The direct effects of climate change on indi-
vidual plants and plant communities may oc-
cur in the absence of pathogens, but may also
bring about changes in plants that will affect
their interactions with pathogens. Changes
in plant architecture may affect microclimate
and thus risks of infection (27). In general,
increased plant density will tend to increase
leaf surface wetness and leaf surface wet-
ness duration, and so make infection by fo-
liar pathogens more likely (65). But, of course,
how abiotic stress factors interact to affect
plants will be key to understanding climate
change effects on plants (92); abiotic stress
such as heat and drought may contribute to
plant susceptibility to pathogens or it may in-
duce general defense pathways which increase
resistance.
Elevated CO
2
levels tend to result in
changed plant structure. At multiple scales,
plant organs may increase in size: Increased
leaf area, increased leaf thickness, higher
numbers of leaves, higher total leaf area per
plant, and stems and branches with greater
diameter have been observed under elevated
CO
2
(117). Enhanced photosynthesis, in-
creased water use efficiency, and reduced
damage from ozone are also reported un-
der elevated CO
2
(139). Since many foliar
pathogens benefit from denser plant growth
and the resulting more humid microclimate
(27), there is the potential for these changes in
plant architecture to increase infection rates,
all else being equal. But interactions with
other changing climatic variables may compli-
cate the effects of elevated CO
2
. For example,
in a California annual grassland, warming, al-
tered precipitation, addition of nitrogen, and
elevated CO
2
each increased net primary pro-
ductivity when applied as single factors; but
in multifactor treatments, elevated CO
2
ap-
peared to suppress the positive effects of the
other factors (131).
The effects of elevated temperature on
plants will tend to vary greatly throughout the
year. During colder parts of the year, warming
may relieve plant stress, whereas during hotter
parts of the year it may increase stress. When
high-temperature stress is exacerbated, plant
responses may be similar to those induced by
water stress, with symptoms including wilt-
ing, leaf burn, leaf folding, and abscission, and
physiological responses including changes in
RNA metabolism and protein synthesis, en-
zymes, isoenzymes, and plant growth hor-
mones (34). These changes will certainly af-
fect susceptibility to pathogens, though the
wide range of changes may make interactions
difficult to predict. As a striking example of the
potential effect on the yield of crop plants in
response to elevated temperature, rice yield in
the Philippines was estimated to decline 10%
for each 1
◦
C increase in the minimum tem-
perature during the dry season (110).
Elevated ozone concentrations can change
the structure of leaf surfaces, altering the
492 Garrett et al.
Annu. Rev. Phytopathol. 2006.44:489-509. Downloaded from arjournals.annualreviews.org
by Kansas State University on 08/09/06. For personal use only.
ANRV283-PY44-21 ARI 13 June 2006 13:56
physical topography as well as the chemical
composition of surfaces, including the struc-
ture of epicuticular wax (74). These changes in
leaf structure may alter leaf surface properties
such as leaf wettability and the ability of leaves
to retain solutes, all influencing the ability of
pathogens to attach to leaf surfaces and in-
fect (74). Ozone exposure has been proposed
to enhance attacks on plants by necrotrophic
fungi, root-rot fungi, and bark beetles
(123).
Gene expression studies of plant responses
to drought stress have expanded rapidly, al-
lowing a more mechanistic understanding of
responses and comparison between responses
to drought and other stressors. As an exam-
ple of expression responses, Way et al. (144)
found that, under both short- and long-term
stress, genes significantly up-regulated in-
cluded those coding for aldehyde dehydroge-
nase (associated with osmoregulation), delta
pyrroline-5-carboxylate synthetase (with a
role in biosynthesis of proline, which acts
to protect plant cells from dehydration), and
fatty acid alpha-oxidase (involved in repair-
ing stress-induced damage in membranes and
regulating fluidity of membrane and perme-
ability to toxic ions). Bray (24) summarized
expression responses to drought stress across
gene classes. Up-regulated genes included
those involved in cellular metabolism, cellular
transport, signal transduction, and transcrip-
tional regulation, as well hydrophilic, heat-
soluble proteins. Down-regulated genes in-
cluded those involved in cell wall synthesis, as
well as cellulases, and germin-like proteins.
These results can be linked to well-known
processes occurring at a larger scale within a
plant, such as stomatal closure and the inhibi-
tion of leaf growth, changes in leaf architec-
ture, and change in root:shoot ratio (32, 34).
It is now possible to measure gene ex-
pression responses to environmental changes
in natural plant populations. For example,
Travers et al. (S.E. Travers, M.D. Smith,
J. Bai, S.H. Hulbert, J.E. Leach, et al.,
manuscript submitted) studied the effects of
simulated changes in predicted precipitation
patterns in tallgrass prairie, where one pre-
diction is for increased intervals between rain
events even if total precipitation is not re-
duced. This experiment focused on the tall-
grass prairie dominant plant species Andro-
pogon gerardii. Using maize microarrays, gene
expression was studied in the natural popu-
lation of A. gerardii to which rainout exclu-
sion shelters were applied to impose the differ-
ent precipitation patterns. Increased intervals
between precipitation events decreased tran-
scription of genes related to photosynthesis
and carbon fixation and increased transcrip-
tion of a variety of heat shock proteins and
kinases. A gene associated with a hypersensi-
tive reaction, HIR1, was significantly down-
regulated under the treatment with increased
intervals, suggesting a defensive cost associ-
ated with climate change. Whereas these re-
sults are of interest in and of themselves for
understanding the tallgrass prairie ecosystem,
they also are important as an illustration of ad-
vances in microarray technologies to the point
where highly variable natural field systems can
be sampled and statistically significant differ-
ences in gene expression observed in response
to climate change simulations.
Host Resistance
Detecting the effects of drought stress on
plant resistance to infection is complicated by
the fact that foliar pathogens will tend to have
lower infection success under dry conditions
(65). But plant pathologists have studied the
interactions between pathogen and drought
stress at the scale of pathogen populations for
some time. For example, Pennypacker et al.
(111) found that alfalfa plants inoculated with
Verticillium albo-atrum exhibited fewer symp-
toms under drought stress. For some host-
pathogen systems, however, resistance is ap-
parently reduced under drought conditions
(34).
Temperature may have important reper-
cussions on the effectiveness of resistance
genes, though it may generally be chal-
lenging to discriminate between temperature
www.annualreviews.org
•
Climate Change Effects on Plant Disease 493
Annu. Rev. Phytopathol. 2006.44:489-509. Downloaded from arjournals.annualreviews.org
by Kansas State University on 08/09/06. For personal use only.