Plant disease epidemiology: concepts and techniques for rice

1
Plant disease epidemiology concepts and
techniques for rice disease management through
breeding and crop husbandry
S. Savary and I. Pangga, R.Hijmans, L.
Willocquet, N.P. Castilla, T.W. Mew
2
Why plant disease epidemiology?

• Importance of plant diseases because of epidemics
• Need to understand epidemics as we know them
  today
• to implement effective control tools (incl. HPR)
• to deploy efficient, durable control tools
  (ditto)
• Need to predict epidemics which will inevitably
  occur (possibly new diseases) tomorrow
• climate change (incl. water scarcity)
• labor resource, natural resources, energy
  shortage
• drivers of agricultural change affect crop
  health (1) the relative importance of diseases
  is changing, (2) factors underpinning epidemics
  are changingmechanisms remain
• Many diseases need for a framework, for
  methodology
• New approaches (general theory, R0)

3
Plant disease epidemiology can

• Help control epidemics (good)
• host plant resistance - 1 (partial HPR)
• host plant resistance - 2 (deployment complete
  HPR, partial HPR over space and/or time)
• tactical decisions crop (health) management
• Help prevent epidemics (better)
• host plant resistance - 3 (complete HPR)
• disease exclusion techniques (e.g., seed health)
• Provide a conceptual framework so many diseases,
  some barely known, biologically

4
seven questions

• Why do some diseases take off, whereas others do
  not?
• Why do some strains, races, or pathotypes die
  out, some coexist, and others come to dominate
  pathogen populations?
• How does the inherent variability associated with
  epidemics translate into risk?
• Given that new infections occur at the small
  scale but epidemics are manifest at the large
  scale, how can we scale from individual to
  population behavior?
• How can this information be used to identify
  control methods?
• How can this information be used to optimize the
  efficient deployment and durability of control
  methods?
• How does the way we grow and protect our crops or
  manage our natural and seminatural environment
  affect these outcomes?

from Gilligan van den Bosch, 2008. Annu. Rev.
Phytopathol. 246385-418.
5
Diversity of rice diseases (pathogens, biological
cycles)
6
Shapes of a few rice disease epidemics
disease severity (fraction leaf surface diseased)
disease incidence (fraction of leaves diseased)
disease severity (fraction leaf surface diseased)
disease incidence (fraction of plants diseased)
disease incidence (fraction of tillers diseased)
7
the SEIR model

• SEIR Suscepts, Exposed, Infectious, Removed
• or (Plant Pathology)
• H, Healthy sites
• L, Latent sites
• I, Infectious sites, and
• P, post-infectious sites
• One key rate infection rate
• Two delays latency period, infectious period

8
the SEIR model applications

• medical epidemiology
• measles
• HIV
• influenza
• tuberculosis
• animal epidemiology
• Pseudorabies virus in pigs
• Mouse typhoid
• computer viruses?
• and botanical epidemiology

9
infection rate (RI) - equation

• dL/dt RI Rc I Ca
• L latent sites I infectious sites
• Rc basic infection rate corrected for removals
  number of new infections, per unit time, per
  infectious site (I)
• C correction factor fraction of healthy
  sites (H), relative to the total number of sites
  in the system
• a disease aggregation coefficient

10
infection rate (RI) over time

• dL/dt RI Rc I Ca

11
some additional detail

• growth of the host crop growth growth of
  healthy sites
• senescence physiological (or/and) pathological
• additional effects (on rate of infection only)
• plant age (variable susceptibility)
• temperature
• canopy moisture

12
spatial scales of plant disease epidemics

• local infections on the foliage
• 1 lesion a small fraction of leaf area
• ex. leaf blast brown spot

• rapidly expanding infections on the foliage
• 1 lesion a leaf
• ex. bacterial blight

• infections affecting entire tillers
• 1 lesion a tiller
• ex. sheath blight

• systemic infections
• 1 lesion a plant
• ex. tungro

13
scaling the model structure to address different
diseases

• definition for a site (a lesion)
• sites levels of hierarchy chosen
• portion of leaf area leaf blast, brown spot
• a leaf bacterial blight
• a tiller sheath blight
• a plant tungro

14
the SEIR model in plant pathology
C
D
a
RG
H
L
I
P
RS
RI
S
Rc
Vanderplank JE. 1963. Plant Diseases. Epidemics
and Control. Academic Press, New York. Zadoks JC.
1971. Systems analysis and the dynamics of
epidemics. Phytopathology 61600-610
15
the SEIR model in plant pathology
Forrester, J.W., 1961. Industrial Dynamics. The
Massachusetts Institute of Technology Press,
Cambridge (Mass.) 464p.
16
SEIR system of differential ordinal equations

• H number of healthy individuals
• L number of latent individuals
• I number of infectious individuals
• R number of post-infectious (removed)
  individuals
• 1/? mean latent period
• 1/µ mean infectious period
• ß per capita transmission rate (new diseased
  individuals per diseased individual per healthy
  individual per unit time).

17
A few rice disease epidemics simulated
18
A few rice disease epidemics simulated
19
From genes to landscapes epidemiological
concepts bridging host plant resistance concepts
genes
epidemics
in many cases (but wrong in the case of leaf
blast)
20
Simulated effects of components of partial
resistance to leaf blast
definition of parameters relative resistance (rr)
parameters s susceptible check t test
variety 0 (s) RR(t) 1 (complete
resistance) rrE Et / Es rrN St / Ss rri
it / is rrp 1 - (pt / ps)
Note Challenge and opportunity of linking
relative resistance (rr) parameters to QTLs and
genes, e.g., in the case of leaf blast, where
partial resistance is in the process of being
well characterized (Ballini et al., 2008. A
genome-wide meta-analysis of rice blsat
resistance and quantitative trait loci provides
new insights into partial and complete
resistance. Molecular Plant-Microbe Interactions
21859-868).
21
Simulated effects of components of partial
resistance to leaf blast
rrp
rri
rrE and rrN
22
Simulated effects of aggregation on sheath blight
epidemics
definition of parameter a coefficient for
disease aggregation (i.e., pathogen and host, and
infection) a 1 (default) random distribution
of diseased sites amongst host sites (each site
has an even chance of becoming infected). a gt 1
disease aggregation within a landscape of
susceptible suscepts e.g., Sheath Blight a gtgt 1
(varies with crop establishment and disease
spread)
23
Simulated effects of aggregation on sheath blight
epidemics
24
Simulated effects of onset time on tungro
epidemics
definition of parameters onset time of onset of
epidemic (date of establishment of the 1st
infection), relative to crop establishment date.
25
Simulated effects of onset time on tungro
epidemics
26
Potential epidemics

• why
• IRRI needs to have an outlook of its
  phytopathological research (partners, donors,
  NARES, etc.)
• impact of what could happen without control
  (recurrent diseases -- invasion)
• extent of potential epidemics (persistence)
• climate change shifts in agricultural practices
  leading to new crop health contexts (persistence
  invasion)
• control options take years to develop (e.g.,
  breeding HPR 10 yrs)
• how
• Simulation modelling mapping/GIS capacities
• (Note not a substitute, but complements, eyes
  in the fields)

27
AUDPC of leaf blast severity in Tropical Asia (
day)
28
AUDPC of brown spot severity in Tropical Asia (
day)
29
invasion and persistence

• invasion the potential for a pathogen to cause
  an epidemic
• persistence the ability a pathogen may have to
  survive over successive host cycles (endemicity
  polyetic processes)

30
Perspectives (invasion I, persistence P)

• I crop health management with (variable,
  evolving) crop management
• crop health as a whole multiple diseases
• I P optimize disease management (control
  points)
• HPR crop management
• variable spatial scales (plot? field?agric.
  landscape), depending on disease
• variable temporal scale (crop stage?season?multipl
  e seasons), depending on disease
• P emerging diseases (FSm, Viruses, Spikelet Rot
  Disease)
• specifc disease
• often important (biological) knowledge gaps
• I P anticipatory research climate change and
  disease epidemics
• potential epidemics
• link with large-scale characterization, global
  ag. change natural resource management
• outcome breeding priorities
• I design of durable resistance in environmental
  contexts where HPR can be sustained
• design of reliable screening procedures
• landscapes (different scales) for control