The Origins of Plant Pathogens in Agro-Ecosystems: Further

Annu. Rev. Phytopathol. 2008.46:75-100. Downloaded from arjournals.annualreviews.
org
ANNUAL
REVIEWS Further The Origins of
by International Potato Center-Lima,Peru on 07/21/09. For personal use only.
Click here for quick links to

Annual Reviews content online,
including:
Plant Pathogens in
• Other articles in this volume
• Top cited articles
• Top downloaded articles
Agro-Ecosystems
• Our comprehensive search
Eva H. Stukenbrock and Bruce A. McDonald

Plant Pathology, Institute of Integrative Biology, ETH Zurich, CH-8092 Zurich,
Switzerland; email: bruce.mcdonald@agrl.ethz.ch, eva.stukenbrock@agrl.ethz.ch
Annu. Rev. Phytopathol. 2008. 46:75–100 Key Words

First published online as a Review in Advance on disease emergence, agriculture, coevolution, host shift, lateral gene
April 28, 2008
transfer
The Annual Review of Phytopathology is online at
phyto.annualreviews.org Abstract
This article’s doi: Plant pathogens can emerge in agricultural ecosystems through several
10.1146/annurev.phyto.010708.154114
mechanisms, including host-tracking, host jumps, hybridization and
Copyright c 2008 by Annual Reviews. horizontal gene transfer. High-throughput DNA sequencing coupled
All rights reserved
with new analytical approaches make it possible to differentiate among
0066-4286/08/0908/0075$20.00 these mechanisms and to infer the time and place where pathogens first
emerged. We present several examples to illustrate the different mech-
anisms and timescales associated with the origins of important plant
pathogens. In some cases pathogens were domesticated along with their
hosts during the invention of agriculture approximately 10,000 years
ago. In other cases pathogens appear to have emerged very recently
and almost instantaneously following horizontal gene transfer or hy-
bridization. The predominant unifying feature in these examples is the
environmental and genetic uniformity of the agricultural ecosystem in
which the pathogens emerged. We conclude that agro-ecosystems will
continue to select for new pathogens unless they are re-engineered to
make them less conducive to pathogen emergence.
75
INTRODUCTION questions is the realization that the nature of
the agro-ecosystem itself has played a critical
Advances in DNA sequencing and analyti-
role in the recent emergence and spread of plant
Agro-ecosystem: the cal approaches have combined with genome
ecosystem that pathogens.
projects to greatly increase our understanding
develops on farmed Pathogens have challenged farmers since the
of processes affecting the evolution of plant
land and which first crop plants were domesticated during the
pathogens in agricultural systems. Compared
includes both the crop transition to agriculture that occurred globally
Annu. Rev. Phytopathol. 2008.46:75-100. Downloaded from arjournals.annualreviews.org
species and its to the long timescales needed to evolve the

between 2000 to 12,000 years BP (9, 28). Many
associated complex biochemical machinery that underlies
of today’s most important crop species orig-
microorganisms and pathogen attack and plant defense in pathogen-

the indigenous inated in the Fertile Crescent in the Middle
plant interactions, agriculture is very young
microorganisms, East. Archaeological findings provide ample ev-
and domestication of crop species and their
plants, and animals idence of the profound changes that took place
pathogens has occurred only recently, mainly
BP: before present in Mesopotamia and other Neolithic cultures
during the past 10,000 years (9). Because of
following the first cultivation of crop plants
their interdependence, selection pressures act-
10,000 to 12,000 years BP (37, 71, 118). Genetic
ing on hosts and parasites are coupled and
data now support archaeology as genome-wide
evolutionary changes may occur faster in an
measures of genetic similarity have traced the
agricultural ecosystem (agro-ecosystem) com-
origin of several domesticated crop species to
pared to a natural ecosystem (69, 91, 113).
wild populations of naturally occurring plants
Thus plant pathogens in agro-ecosystems pro-
that persist in the Middle East to this day (85).
vide rapidly evolving model systems for evo-
The eight most important Neolithic founder
lutionary biologists interested in host-parasite
crops in the Fertile Crescent were the wild pro-
coevolution, speciation processes, and popula-
genitors of emmer wheat, einkorn, barley, flax,
tion dynamics. Plant pathologists also benefit
chick pea, pea, lentil, and bitter vetch. Inten-
from integrating principles of evolutionary bi-
sive selection and cultivation of selected pheno-
ology into plant pathology. An understanding of
types altered populations of these wild progen-
the evolutionary processes leading to the emer-
itor species into the domesticated varieties of
gence of new pathogens is essential to assess the
the crop species that we know today. The first
risks posed by potential future pathogens. Ap-
Mesopotamian crops were likely species mix-
plication of evolutionary principles enable us
tures harvested directly from wild populations.
to improve the development and implementa-
These mixtures maintained a high degree of
tion of sustainable control strategies based on
genetic and environmental heterogeneity com-
new or existing pesticides and host resistance
pared to modern crops. Over time Neolithic
genes.
farmers began growing crops as species mono-
In this chapter we focus on the where, when,
cultures and these monocultures spread even-
and how of pathogen emergence. Where did
tually to other regions in Europe and Asia
the important pathogens that currently dom-
(28).
inate our agro-ecosystems originate? When
The development of new crop species and
did they first emerge and over what tempo-
agricultural practices led simultaneously to the
ral scale did they spread across the global
emergence of new pathogens and to signifi-
agro-ecosystem? What mechanisms drove the
cant changes in pathogen populations that al-
emergence of these pathogens and why were
ready existed on the wild ancestors of the culti-
they successful? Did domestication of crops
vated crops. The new agro-ecosystem provided
lead to domestication of their corresponding
a more dense and genetically more uniform host
pathogens? Or did already existing pathogens
population that enabled easier pathogen trans-
jump onto crop hosts after the crops were intro-
mission between infected and uninfected plants
duced into new regions? The most significant
and created an environment that was more
finding that emerges from considering these
76 Stukenbrock · McDonald
conducive to host specialization compared to density in agro-ecosystems, with identical host
the natural ecosystem. The agricultural field plants often touching their neighbors, transmis-
provided a more homogenous environment sion of a pathogen between plants is very effi-
Sympatric speciation:
characterized by soil mixing (tillage), irrigation, cient, which allows the development of higher the formation of two
and fertilization and was less prone to envi- virulence in the pathogen population infecting or more descendant
ronmental fluctuations that could lead to re- the agro-ecosystem compared to the pathogen species from a single
duced pathogen population size or even com- population in the natural ecosystem (100). ancestral species all
occupying the same

plete extinction. This new agro-ecosystem thus The emergence of a new pathogen can oc-
geographic location
imposed a very different type of selection on cur either sympatrically or allopatrically. Sym-
Allopatric speciation:
pathogens compared to the selection operat- patric speciation represents the genetic diver-
the formation of two
ing in more heterogeneous natural vegetation gence of populations that inhabit the same or more species from a
(Figure 1). The ability of a pathogen to adapt to geographic region, whereas allopatric specia- single ancestral species
this new human-engineered environment de- tion occurs between populations that are geo- following geographical
pended largely on its life history traits and its graphically separated (68). New pathogens that isolation of
subpopulations
evolutionary potential (69). emerge through host specialization, changes in
Pathogen populations in modern agro- virulence or adaptation to uniform host pop- Center of origin: the
geographical place of
ecosystems are regularly challenged by pesti- ulations may become genetically differentiated
origin of a species
cides, resistance genes, crop rotations, and a from their source population even when they
variety of cultural practices aimed at reducing occur in the same location with an overlapping
plant infections. Selected genotypes that can range of spore dispersal. Selection will main-
adapt to these management practices, includ- tain the separation between sympatric popula-
ing the ability to overcome fungicides and resis- tions if hybrids of the two populations exhibit
tance genes, may increase rapidly in frequency a lower fitness than “pure” progeny resulting
and be dispersed to neighboring fields. Popula- from within-population crosses. This sympatric
tion genetic studies have greatly increased our divergence can lead to speciation, creating a
knowledge regarding the evolutionary poten- new pathogen adapted to a different niche than
tial of different pathogens in agro-ecosystems the source population, e.g., intersterility groups
and have allowed us to characterize the most in the Heterobasidion species complex (54). Al-
important processes that govern pathogen evo- lopatric speciation may occur by the introduc-
lution (69). The dense and uniform host popu- tion and successful establishment of a pathogen
lations in an agro-ecosystem enabled the evolu- in a new environment that is separated from the
tion of increased virulence in the corresponding source population. For example, global travel
pathogen populations. The degree of virulence and trade of agricultural products have moved
in a pathogen is thought to be affected by a crops and pathogens away from their orig-
trade-off between pathogen replication within inal environments at their center of origin.
hosts and transmission between hosts (4, 67). The introduction of plants or pathogens into
If a pathogen grows very quickly within an in- new environments creates novel host-pathogen
fected host so that the host is killed quickly interactions where introduced pathogens can
(i.e., the pathogen is highly virulent) before the cause severe damage in naı̈ve host populations,
pathogen can be transmitted to an uninfected e.g., the introduction of the potato late blight
host, then the pathogen will not spread and an pathogen Phytophthora infestans into Ireland in
epidemic will not occur. If within-host replica- the mid-nineteenth century (36) caused the
tion is slower and the host suffers fewer symp- Irish potato famine, and the movement of the
toms of infection (i.e., the pathogen is less vir- wheat stripe rust pathogen Puccinia striiformis
ulent), more opportunities are provided for the f. sp. graminis into the United States 100 years
pathogen to be transmitted to an uninfected ago caused severe economic losses for wheat
host. Because plants are grown at such high farmers (20).
www.annualreviews.org • Origins of Plant Pathogens 77

Figure 1
An illustration
of the main differences
between natural
and agricultural
ecosystems.
Agro-ecosystems
(bottom panel ) exhibit
greater environmental
homogeneity,
lower species diversity,
and higher host
density that facilitate
pathogen transmission
to uninfected
neighboring
plants. These
properties enable the
development of epi-
demics characterized
by several cycles of
pathogen reproduction
every year and
drive the emergence
of highly virulent,
host-specialized
plant pathogens.
Natural ecosystems
(top panel ) have
greater environmental
heterogeneity, higher
species diversity,
and lower host density
that favor less virulent,
generalist pathogens
that can infect
many host species.
INFERRING EVOLUTIONARY crop domestication. But new methods based
PROCESSES AND TIMESCALES on population genetics, coalescence methods,
FOR FUNGAL PATHOGENS and phylogeography can provide insight into
RFLP: restriction
the processes that led to the emergence of fragment length
The Russian botanist Nikolai Vavilov was the pathogens associated with the first cultivated polymorphism
first to identify the center of origin of culti- crops. For example, population genetic stud- Coalescent theory: a
vated plants. In his work based on morpho- ies of different cereal pathogens such as the retrospective
logical characters, he concluded that each crop powdery mildew pathogen Blumeria graminis population genetic
has a primary center of diversity that is also its and the Septoria tritici blotch pathogen My- model based on
genealogies of the
center of origin (105). In recent years, DNA cosphaerella graminicola suggest that the cen-
alleles of a gene. It
analyses have been used to study diversity and ter of diversity of these pathogens coincides uses statistical analyses
origin of crop species (e.g., 7, 90). In general, with the center of origin of their hosts in the to infer the
these studies have confirmed Vavilov’s theory Middle East (11, 109). These population stud- coalescence of lineages
that the centers of origin are the areas of great- ies used geographical collections of pathogen back in time to the
most recent common
est diversity. However, increased diversity of a isolates and analyses of DNA markers and se-
ancestor (TMRCA)
cultivated plant in a certain region may also quences. In the case of the wheat infecting
result from long and intense cultivation, eco- M. graminicola, phylogenetic analyses and hap-
logical diversity, and/or introgression of wild lotype networks were used to show a higher
crop relatives (45). In these cases knowledge diversity at RFLP (restriction fragment length
about the cultivation history of the crop is also polymorphism) loci and DNA sequences in
needed. By combining the tools of population Israeli populations (11, 116). Haplotype net-
genetics with the tools of archaeology and ar- works of polymorphic DNA loci suggested that
chaeobotany, it is possible to identify the place ancestral haplotypes originated in the Middle
and time of crop domestication with greater East (10). But these population genetic anal-
precision. Two independent lines of evidence yses did not tell when the wheat pathogen
indicate that agriculture originated in the Fer- emerged and did not confirm that the center
tile Crescent 10,000 to 12,000 years BP. First, of diversity actually represented the center of
the current distribution of wild progenitors of origin of the pathogen. Just as wild relatives
modern cereals including wild wheat (Triticum of modern cereals in the Fertile Crescent had
urartu, T. boeoticum, and T. dicoccoides), wild bar- to be analyzed to confirm their ancestral rela-
ley (Hordeum spontaneum), and wild rye (Secale tionship, the wild progenitors of M. gramini-
vavilovii) intersects in this region. Second, the cola were needed to conduct a coalescence
seeds of the wild species occur in early archae- analysis of DNA polymorphisms in pathogen
ological sites of the region and can be dated by populations on both wild and cultivated hosts
radiocarbon analyses. By using population ge- (95).
netic tools to compare wild and domesticated Coalescence methods use population-based
species and combining these data with arche- DNA sequence data to infer the merging of an-
ological findings, it is possible to identify the cestral lineages going back in time. These meth-
specific valleys in the Fertile Crescent where ods simulate all possible gene trees backwards
different cereals were first cultivated and also to in time to the most recent common ancestor
generate a precise timescale of archaeobotani- (TMRCA) of two populations and provide like-
cal events (85). lihood estimates for each tree. The coalescent
Though rust spores have been found in 3300 theory is a powerful statistical tool that can be
year-old Israeli granaries (59), the great ma- used to derive estimators of population parame-
jority of plant pathogens were not preserved ters, such as rates of mutation or migration and
in these archaeological sites, and thus archae- time of population splitting. A number of dif-
ology cannot assist us in the reconstruction ferent programs and models have been devel-
of pathogen evolution during the process of oped to analyze population samples of different

organisms. These models and the statistics be- vides a usable alternative (22). It is also impor-
hind the coalescence methods are described in tant to consider whether the number of gen-
a number of comprehensible reviews that are erations per year has been constant over the
recommended for an introduction to the field evolutionary history of a pathogen. We con-
(6, 49, 82). sider it likely that “cultivated” pathogen popu-
When inferring coalescence parameters for lations growing in irrigated and fertilized host
pathogen populations, it is important to take monocultures characteristic of agro-ecosystems
into account the population genetics of the go through many more cycles of reproduction
species to ensure that assumptions of the model every year than the ancestral pathogen popu-
are suitable. Many population processes influ- lations that spent most of their evolutionary
ence the divergence of two populations, such history in a natural ecosystem. Year-round pro-
as the amount and the direction of gene flow duction of crops such as rice and barley cre-
or population expansion in one or both pop- ated green bridges that enabled pathogens to
ulations. One model that allows for changes propagate continuously without the need for an
in population size and directional gene flow overseasoning stage. As a result, pathogen pop-
between diverging populations is the Isolation ulations may go through more cycles of repro-
with Migration model (50, 74). This model production per year and effective population sizes
vides an ideal framework for coalescence analy- can be larger in modern agricultural popula-
ses of pathogens because emerging populations tions than in ancestral wild populations, as il-
are likely to experience fluctuations in popula- lustrated using Bayesian skyline plots (30) for
tion size and/or gene flow from or to the source the barley scald pathogen Rhynchosporium secalis
population. (111). Furthermore, the balance between sex-
Other coalescence approaches allow the re- ual and asexual reproduction may be quite dif-
construction of gene trees showing the dis- ferent in agro-ecosystems and natural ecosys-
tribution of sequence haplotypes and specific tems because fungal reproduction is affected
mutations going backwards in time (38). The by environmental conditions. In general, low
timescales for these analyses are given in coa- stress conditions (e.g., with lots of fertilizers and
lescence units that must be scaled by mutation higher humidity due to irrigation) favor asex-
rate to give the actual time of population di- ual propagation and shorter generation times
vergence. The overall mutation rates per base whereas high stress conditions (e.g., senesc-
per replication are surprisingly uniform across ing leaves, depleted nutrients, lower humid-
all DNA-based genomes, including prokary- ity) favor sexual reproduction and longer gen-
otes and eukaryotes (29). However, mutation eration times (21). Contemporary populations
rate can vary greatly in different parts of the of pathogens such as P. infestans and Magna-
genome and even between different lineages. porthe oryzae are mainly clonal, but the loss
Therefore, a mean value of mutation rate rep- of sexual reproduction seems to be associated
resenting the mutation rate of several genes and with their adaptation to an agricultural host
eventually a model that allows for variable mu- (36, 112) as wild relatives of both pathogens
tation rates among lineages must be used to in- still undergo regular cycles of sexual reproduc-
fer the actual time for a species split. Another tion (60, 101). The occurrence of sexual repro-
crucial input parameter in coalescence analy- duction in the “wild” populations may reflect
ses is generation time. Because most fungi have the fact that the sexual cycle produces a dor-
mixed reproductive systems, the first question mant resting stage (e.g., oospore or cleistothe-
to address is how many sexual cycles occur per cium) that can survive in the absence of a living
year. Some pathogens reproduce mainly asex- host. These survival structures were especially
ually and in these cases the definition of gen- important in natural ecosystems characterized
eration time becomes complicated because it is by heterogeneous environments and low host
not clear if the number of asexual cycles pro- densities.
In conclusion, to apply coalescence analy- Pathogen Domestication Along
ses in studies of pathogen evolution it is essen- with the Host
tial to consider the population dynamics of a
We call this host-tracking in analogy to a similar Host-tracking: the
given species and to recognize possible changes coevolution of a
coevolutionary process occurring over longer
in generation time during the course of evolu- pathogen with its host.
time frames in natural ecosystems (84). Under
tion. Ideally, direct observation of sexual fruit- By host-tracking the
this scenario, the host and pathogen coevolved
ing structures developing over time could be pathogen is likely to be
during the process of host domestication and younger than the host,
used to determine the number of sexual cycles
the development of the agro-ecosystem spe- in contrast to
per year for a fungal species. But sexual fruit-
cific to the host crop (e.g., the agro-ecosystem cospeciation where

ing structures can be difficult to recognize and host and pathogen
for wheat is quite different from the agro-
quantify for many pathogenic fungi. Alterna- have diverged
ecosystem for apples). In this case the expecta-
tively, population genetic studies can be used simultaneously
tion is that the center of origin of the pathogen
to quantify the relative contributions of sexual Host shift/host jump:
will correspond with the center of origin of the
and asexual reproduction to a pathogen pop- the process by which a
host. During domestication, the selection and pathogen infects a new
ulation over short periods of time (114, 115),
cultivation of desirable host genotypes simul- previously unaffected
providing important knowledge on the popula- host species. In a host
taneously selects for pathogen genotypes that
tion dynamics of a species, i.e. whether sexual shift, the alternate host
are adapted to the selected individuals and their
reproduction occurs frequently or rarely. Com- is a close relative of the
specialized agro-ecosystem. As the new agro-
parative population genetic studies of closely former host, whereas a
ecosystem develops and spreads, the growing host jump involves a
related species occuring in either natural or
host population provides a highly conducive en- new host that is
human managed ecosystems may also shed taxonomically very
vironment for selected pathogens with high fit-
light on differences in population dynamics distant from the
ness that can expand with the new agricultural
in different environments and could indicate former host, e.g., from
host. To properly infer the evolutionary history
possible changes in generation time since the another class or order
of the agricultural pathogen under this scenario,
adaptation to an agricultural system (i.e. if the
we need to sample pathogen populations from
agricultural pathogen exhibits a higher degree
both the domesticated and the undomesticated
of sexual reproduction than its sister species in
hosts in order to reconstruct the evolutionary
a natural ecosystem). To calculate a completely
history backwards in time to their common an-
accurate generation time of a pathogen may be
cestor. Finally, we need analytical tools to test
impossible, but the points mentioned above can
evolutionary models of population divergence.
be applied to arrive at a reasonable estimate of
The domestication of pathogens will not neces-
generation time.
sarily date back thousands of years to the start of
agriculture. Pathogen domestication is poten-
MODELS TO EXPLAIN THE tially a recent occurrence wherever new crop
EMERGENCE OF NEW species, such as kiwi fruit or triticale, and their
PATHOGENS IN associated agro-ecosystems are developed.
AGRO-ECOSYSTEMS
The integration of population genetics, evolu-
tionary biology, and bioinformatics can allow Host Shifts and Host Jumps
us to test specific hypotheses regarding when A new pathogen can emerge by adaptation to
and where pathogens emerged on domesticated a new host following a host shift or host jump.
crops. Though pathogens can emerge in many Here we define a host shift as involving a new
different ways, we focus here on what we con- host that is genetically not far removed from
sider the four most likely scenarios (Table 1). In the old host (e.g., shifting from wild barley to
this section we briefly summarize the different domesticated barley), whereas a host jump in-
models and the expected pattern of evolution volves a new host that is genetically distant from
that would result under each scenario. the original host (e.g., jumping from clover

Table 1 Examples of evolutionary mechanisms by which plant pathogens have emerged in agro-ecosystems over different
time scales
Evolutionary mechanism Plant pathosystem Time scale Reference
Domestication/host-tracking
Mycosphaerella graminicola on wheat 10–12,000 years BP 95
Magnaporthe oryzae on rice 7000 years BP 24
Phytophthora infestans on potato 7000 years BP 34
Ustilago maydis on maize 8000 years BP 72

Host jump/host shift

Magnaporthe oryzae from Setaria millet Abrupt evolutionary change, approx. 24
to rice 7000 years BP
Rhynchosporium secalis from wild grasses Abrupt evolutionary change, approx. 111
to barley and rye 2,000 years BP
Phytophthora infestans from wild Abrupt evolutionary change, 35, 39
Solanum species to potato <500 years BP
Horizontal gene transfer
ToxA from Phaeosphaeria nodorum into Abrupt evolutionary change, approx. 32
Pyrenophora tritici-repentis 60 years BP
PEP cluster in Nectria haematococca Unknown 43
Host specific toxins in Alternaria Unknown Reviewed in (99)
alternata
Hybridization
Hybrid of Phytophthora cambivora and Abrupt evolutionary change within this 53
relative of P. fragariae on alder century
Hybrids of Ophistoma ulmi and Abrupt evolutionary change within this Reviewed in (14)
O. novo-ulmi on elm trees century
to maize). In this case, the geographical ori- were brought to North America, leading to the
gin of the host need not correspond with the chestnut blight epidemic (2). This case high-
geographical origin of the pathogen. A new lights an important concern associated with
pathogen in an agro-ecosystem can originate host shifts. As a result of the lack of coevolution
from either a wild host population or from an- in a new host-pathogen combination, a newly
other crop species. Wild plant species growing acquired host may exhibit a lower level of resis-
within or adjacent to agricultural fields, e.g., tance and the pathogen an excessive degree of
weeds, likely represent an important source of virulence.
new pathogens. Pathogen emergence may also To infer the evolutionary history of an agri-
follow the introduction of new crops into natu- cultural pathogen that emerged by a host jump
ral ecosystems that were not previously planted we need to compare isolates from the different
to agriculture, such as the introduction of soy- host species which can represent the new or the
beans into the Amazon rainforest basin or wheat old hosts. Coalescence analyses can be used to
into the former subtropical forest of south- date the time and determine the direction of the
ern Brazil. The global transport of infected host shift or jump by differentiating the source
plant material also increases the potential for and founder populations of the pathogen. Over
host shifts/jumps of pathogens, for example, by timescales of millions of years, host jumps,
moving already coevolved pathogens into naı̈ve even between plant and animal kingdoms (75,
populations of related hosts. This appears to 89), are probably common and have been the
have happened when infected Asian chestnuts source of many new pathogen species. One
example involves the grass symbionts Clavicip- in animals and fungi (5). Other studies suggest
itaceae belonging to the ascomycete family that HGT may be much more important than
Hypocreaceae. Claviceps purpurea is the best char- previously thought for fungal pathogens (83).
Horizontal gene
acterized species of Clavicipitaceae and causes er- In a recent review, van der Does & Rep (103) transfer (HGT): also
got of rye. The common ancestor of Clavicipi- discuss the role of horizontal gene transfer in called lateral gene
taceae was an animal pathogen, but during the creating and distributing clusters of virulence transfer. A process by
course of evolution a minimum of 5–8 inter- genes among fungal pathogens. They suggest which a genomic
region is transferred to
kingdom host jumps occurred, including 3–5 that fungal pathogens often have gene clusters
another organism, e.g.,
jumps to fungi, 1–2 to animals, and 1 to plants analogous to pathogenicity islands in bacteria
by a transposon. HGT
(89). We focus here on host shifts or jumps that and that these clusters can be exchanged be- may occur between
have occurred within the recent time frame of tween species or between vegetatively incom- different species or
agriculture, i.e., during the past 10,000 years. patible lineages of fungi. between different
vegetatively
incompatible lines
Horizontal Gene Transfer Hybridization
Horizontal gene transfer (HGT) is the ex- While horizontal gene transfer involves the ex-
change of specific genes or genomic regions change of a limited number of genes between
between species that are normally reproduc- different species, interspecific hybridization in-
tively isolated. As in the previous section, we volves entire genomes and can lead to changes
focus here on recent events, i.e., those that in chromosome number, ploidy levels, or exten-
are thought to have occurred since the devel- sive genomic rearrangement (87). A survey of
opment of agriculture. Under this scenario, a genetically similar species is needed to identify
nonpathogen or a weak pathogen is converted the origin of a hybrid. If hybridization or back-
into a virulent pathogen after acquiring genes crossing has occurred several times, the hybrid
needed for pathogenicity from a different population may exhibit a large genetic and phe-
species. Criteria generally used to identify can- notypic variation and a broad population sam-
didate horizontally transferred genes are un- ple will be needed to infer the evolution of the
usual GC content and unusual codon usage hybrid species. The definition of hybrids de-
(96). HGT can also result in unexpected gene pends on the definition of species within a given
homologies among organisms and, in the case fungal clade. In many fungal clades species def-
of pathogens, in a patchy distribution of vir- initions are unclear, which also complicates the
ulence genes among different phylogenetic identification of interspecific hybrids.
lineages. To identify the source and the sink Reproductive barriers normally evolve be-
organisms involved in HGT, a genetic compar- tween closely related fungal species to limit
ison of both species is needed, including the the amount of outcrossing and protect the ge-
transferred region as well as other loci. In most netic integrity of the species. Several studies
cases, the source species is not known and infer- have shown that reproductive barriers are more
ence is therefore needed to suggest a possible likely to evolve between sympatrically diverg-
origin for a gene that may have been acquired ing species than between allopatric species (3,
through horizontal transfer. Comparisons of 27, 92). This is due to the lack of selection
complete genomes coming from the rapidly in- for reproductive isolation between populations
creasing bacterial and fungal genome projects or species that are geographically separated.
will likely uncover many new cases of HGT. Hybridization is therefore more likely to oc-
A number of well-documented examples il- cur between allopatric species that are brought
lustrate how bacterial adaptation and speciation together, for example, by the global trade of
was influenced by HGT (26), but HGT has crop and forest plants. Interspecific hybridiza-
only rarely been postulated between eukary- tion between pathogens can create hybrid lin-
otes (58) and is thought to be especially rare eages with properties very different from the

parents. If a successful hybrid offspring estab- species to infer the coalescence of pathogen
lishes itself on a new host different from the lineages. Parsimony, maximum likelihood, and
one occupied by either parental population, it Bayesian analyses showed a single origin of
is possible that the hybrid offspring will have a a pandemic rice-infecting M. oryzae lineage.
higher fitness than either of the parental species Two different coalescence approaches were ap-
and emerge as a pathogen on the new host. plied to date the emergence of the pathogen.
Both methods supported an early divergence
It is possible that more than one of the above time 5000 to 7000 years BP coinciding with
mentioned mechanisms may be involved in the rice domestication. Couch et al. (24) also per-
emergence of a new pathogen. In fact all four formed tests of pathogenicity on rice and tor-
mechanisms may come into play for a given pedo grass (Panicum repens), a common weed
pathogen, but may occur over different time in rice agro-ecosystems. Isolates from nonrice
scales, e.g., a host shift may have occurred 1000s hosts were either not pathogenic or were much
of years ago while a horizontal gene trans- less pathogenic toward rice compared to rice-
fer may have occurred 100 years ago. In the infecting lineages. Host specialization of the
next section, we categorized different exam- rice blast pathogen could be due to loss of the
ples of pathogen evolution in accordance with avirulence gene AVR-Co39 that was widespread
the most important mechanism shown in each in haplotypes from other hosts. Genetic isola-
study. However, as the reader will recognize, tion and divergence of an epidemic lineage from
most of these examples involve more than one other populations on grasses may have resulted
evolutionary mechanism leading to the emer- from rapid clonal propagation and strong se-
gence of the pathogen. lection mediated by the new domesticated host
and its associated agro-ecosystem. The inten-
EXAMPLES OF PATHOGEN sification and spread of rice cultivation allowed
DOMESTICATION THROUGH propagation and global dispersal of clones of
HOST-TRACKING the rice-infecting blast pathogen.
COEVOLUTION
Domestication of the Rice Blast Domestication of the Wheat Pathogen

Pathogen Magnaporthe oryzae in Asia Mycosphaerella graminicola in the
The blast pathogen Magnaporthe oryzae is one Fertile Crescent
of the most important pathogens of rice (Oryzae The emergence of the pathogen Mycosphaerella
sativa). Reproduction of M. oryzae is mainly graminicola causing the Septoria tritici leaf
asexual, giving rise to a clonal population struc- blotch disease on wheat was analyzed using
ture (112). However, more diverse populations genealogical and model-based coalescent ap-
with equal mating-type frequencies and ev- proaches applied to 7 genes in 184 isolates (95).
idence for frequent sexual reproduction are Two sister populations of the wheat-adapted
found in the Himalayas (60). This suggests an M. graminicola were identified on wild grasses
Asian center of diversity of the blast pathogen collected in northwest Iran. Two genetically
that coincides with the center of origin of its differentiated wild grass-infecting populations,
host. Archaeological evidence suggests that the named S1 and S2, were both encountered on
earliest domestication of rice from its wild rel- three different weedy grass species growing
atives took place in China ∼7000 years BP next to fields of cultivated wheat. The split
(104). Couch et al. (24) tested the hypothesis between the most closely related wild grass-
that a rice-infecting population of M. oryzae infecting population S1 and M. graminicola oc-
emerged simultaneously with the initial cul- curred approximately 10,000 to 12,000 years
tivation of its host. The study used 497 iso- BP. This time coincides with the domestication
lates of M. oryzae from rice and other grass of wheat and suggests a cospeciation of host
and pathogen. In this case, it appears that se- of maize suggest that the pathogen also had its
lection maintained a broader host range in the origin in Mexico. Most Ustilago species are host
undomesticated pathogen population whereas specific, and the host range of U. maydis is lim-
the domesticated pathogen has become highly ited to Z. mays subspecies including species of
host specific. As in the case of M. oryzae, the teosinte, the progenitors of today’s maize and
wheat pathogen likely diverged from its an- endemic in Mexico.
cestral population as a result of strong di- Munkacsi and coworkers studied the origin
rectional selection and a rapid accumulation of U. maydis using geographical isolate collec-
of mutations following an increase in popu- tions and DNA sequence data from 5 genes and
lation size. Only a small fraction of the di- 10 microsatellite markers in 1088 isolates. They
versity in the ancestral population moved to initially tested the hypothesis that U. maydis
the population infecting wheat, but the result- and three other closely related smut pathogens,
ing M. graminicola population rapidly expanded, U. scitaminea, Sporisorium reilianum, and S.
coinciding with the rapid expansion of the sorghi (pathogens of sugarcane, sorghum, and
wheat agro-ecosystem in the Fertile Crescent. wild ancestors of these species) emerged as
The size of the undomesticated S1 pathogen new species following domestication of their
populations declined, also consistent with the respective host species (72). The divergence of
conversion of natural grasslands into wheat these four species was shown to have occurred
agro-ecosystems during the intensification of millions of years ago, and the authors concluded
agriculture in Mesopotamia. Wheat was intro- that the differentiation of these pathogens could
duced into Europe 6000 to 7000 years BP, which not be related to the onset of agriculture.
was the most likely time frame for introduc- But more recently, maize domestication
tion of M. graminicola to the European con- has had a significant impact on the population
tinent. Ancient patterns of migration inferred structure of U. maydis. Munkacsi et al. (73) used
from DNA sequence data show an asymmetri- microsatellite markers to characterize popula-
cal migration of M. graminicola from European tion genetic structure of the maize pathogen
and Israeli populations to New World coun- and to determine the time of population diver-
tries (10). Estimates of recent migration pat- gence among geographical populations of the
terns using microsatellites support an asymmet- smut pathogen in South and North America.
rical migration of the pathogen from the Old They identified five major populations based
World, but additionally include North Amer- on genetic similarity between individuals: two
ica as a contemporary major donor of pathogen in Mexico (where one population consisted
immigrants (10). This pattern of contemporary of isolates from a wild population of Z. m.
migration is consistent with the intensive culti- ssp. parviglumis and the other isolates were
vation of wheat on the North American conti- from different maize fields), two populations
nent today and global wheat trading patterns. in South America and one in North America.
The Mexican U. maydis population did not
show the greatest diversity even though
Domestication of the Corn Smut Mexico is considered the center of origin of
Pathogen Ustilago maydis in Mexico the pathogen. Instead, one South American
Maize (Zea mays) was domesticated by pre- population had the largest heterozygosity while
Columbian cultivators in Mexico about 8000 the North American population had the largest
years BP (66). An important pathogen of maize number of rare and private alleles. But estimates
is the corn smut fungus Ustilago maydis. Corn of divergence time confirmed that the Mexican
galls produced by U. maydis were considered a populations were older than other American
delicacy by the Aztecs and are still popular in populations. The oldest divergence date, co-
Mexico. The early occurrence of U. maydis in inciding with the time of maize domestication
the Aztec cuisine as well as the Mexican origin 6000 to 10,000 years BP, was between the two

Mexican populations. Population genetic South America” migration. The coalescence-
analyses suggested that geographic U. maydis based approach showed both the geographical
populations have been largely isolated since distribution of mutations and a relative age or-
they were established in maize populations in dering of mutations and haplotypes. Two an-
South and North America and that recent pop- cestral haplotypes giving rise to contemporary
ulation bottlenecks and expansions have shaped potato pathogens were present in South Amer-
the distribution of genetic diversity. This sup- ican populations whereas only one ancestral
ports the finding of higher diversity in North haplotype was found in Mexico. Older muta-
America, where expansion of the pathogen tions in the mitochondrial genealogy all origi-
population may have been greater due to more nated in South America, where the divergence
intensive cultivation of maize. The studies of ancestral haplotypes also occurred. Although
of Munkacsi and coworkers provide evidence a number of unique isolates were found in the
for the important impact of agriculture on Mexican sample, these were all descendants of
pathogen demography and evolution. Although only one of the ancient lineages whereas the
host domestication did not lead to speciation of South American haplotypes were descended
a new pathogen, it greatly influenced the pop- from both ancestral lineages. The study also
ulation genetic structure of the pathogen and inferred the directional migrations of haplo-
also led to panglobal distribution of U. maydis types between South American and non-South
with the spread of the maize agro-ecosystem. American populations and presented evidence
of an “out-of-South America” migration in the
past and a more balanced modern-day migra-
Evidence for Domestication tion between South American and non-South
of the Potato Late Blight American populations. The authors did not
Pathogen in the Andes date the origin of P. infestans, but their coales-
Potato (Solanum tubersosum) was domesticated cence analyses allowed a relative age ordering of
by pre-Columbian cultures approximately 7000 haplotypes and thus identified the most ances-
years BP in northern Peru (28, 90). The diploid tral lineages among the strains included in the
oomycete Phytophthora infestans is the causal analysis. Their interpretation was that P. infes-
agent of late blight of tomato and potato and tans most likely emerged through cospeciation
was responsible for the Irish potato famine with its host during the domestication of potato
starting in 1846. The origin of this devastat- 7000 years BP (90).
ing pathogen has been an issue of much debate.
Several studies suggested that the center of ori-
gin of P. infestans was the central highlands of EXAMPLES OF HOST SHIFTS
Mexico because populations found in these re- AND JUMPS
gions (a) show a higher level of genetic diversity,
including Avr-gene diversity, (b) are diploid, and
A Host Jump in Northern Europe
(c) have equal mating-type ratios (33, 36, 39).
for the Barley Scald Pathogen
However, in a recent article, Gómez-Alpizar
Rhynchosporium secalis
et al. (35) used coalescence methods to sug- The scald pathogen Rhynchosporium secalis in-
gest that P. infestans did not originate in fects barley (Hordeum vulgare), rye (Secale ce-
Mexico but instead emerged in the Andes fol- reale), and a number of other grasses. The
lowing the domestication of potatoes. The au- barley and rye hosts both originated in the
thors used gene genealogies of 2 nuclear and Fertile Crescent approximately 10,000 to
2 mitochondrial genes from 94 isolates from 12,000 years BP (7, 71). Zaffarano et al. (110)
8 geographical locations to test the hypothe- analyzed the RFLP allelic diversity of 1366 ge-
sis of codomestication of P. infestans and pota- ographical isolates and found that the center of
toes in South America followed by an “out-of- diversity of the pathogen did not coincide with
the center of origin of barley. The center of that the rice-infecting lineage emerged follow-
pathogen diversity was instead located in Scan- ing a host shift from Setaria millet to rice. Both
dinavia. A similar pattern was found for the avir- millet and rice were domesticated in China and
ulence gene NIP1 (19). NIP1 has a dual func- probably co-occurred during early cultivation
tion as both an elicitor of plant defense and as a (23, 25). Close proximity between crop plants
toxin-encoding gene (81). Due to its role as an can facilitate a host jump of pathogens associ-
elicitor for plant defense, NIP1 is under positive ated with either one of the species. Couch et al.
diversifying selection and is often deleted (88). (24) analyzed the evolutionary relationship be-
Brunner et al. (19) used a coalescence approach tween M. oryzae haplotypes on rice and on
to determine the origin of the NIP1 gene. Con- weeds of rice. The more ancestral haplotypes
sistent with the finding of Zaffarano et al. (110), originated from rice whereas haplotypes from
they found a Scandinavian origin of NIP1 hap- weeds of rice were found at the tips of a haplo-
lotypes. Estimates of TMRCA (time to most type network. The authors concluded that after
recent common ancestor) of NIP1 haplotypes host specialization of the rice-infecting lineage,
were between 2500–5000 years BP, much later additional host jumps occurred to common
than the domestication of barley and rye. weeds of rice, including cutgrass (Leersia hexan-
Because nonneutral evolution in the NIP1 dra) and torpedo grass (Panicum repens). To
gene may have biased coalescence estimates, se- test whether pathogen populations on rice and
quences from six additional neutral and house- weeds of rice still share a common gene pool, in-
keeping genes were collected from ∼500 iso- compatible sites of a multigene alignment were
lates to infer the origin of R. secalis and to identified. The presence of incompatible sites
reveal phylogenetic relationships between dif- would support conflicting phylogenies and pro-
ferent host-related lineages (111). These anal- vide evidence of recombination and genetic ex-
yses confirmed a later emergence of the scald change. However, the analyses showed no sign
pathogen, most likely between 1200–3600 years of conflicts between blast pathogens on differ-
BP following the introduction of the barley ent host populations, indicating that these pop-
agro-ecosystem into northern Europe. Phylo- ulations no longer exchange genes. Comparison
genetic analyses additionally revealed a clear of host and pathogen phylogenies did not sup-
differentiation between Rhynchosporium popu- port the hypothesis of cospeciation of host and
lations infecting rye, barley, and wild grasses. pathogen lineages. These findings confirm the
Zaffarano et al. (111) conclude that the scald occurrence of multiple host shifts during the
pathogen on barley emerged through a host evolution of Magnaporthe pathogens and sug-
shift following the introduction of the barley gest that similar processes may be responsible
agro-ecosystem into northern Europe. Collec- for the 1989 emergence of wheat blast in south-
tions of R. secalis on its original host in northern ern Brazil.
Europe or elsewhere will be needed to more
precisely reconstruct the evolutionary history
of this important barley pathogen. Evidence for a Host Jump in Mexico
for the Potato Late Blight Pathogen
Phytophthora infestans
Divergence of Magnaporthe oryzae Although coalescence analyses described above
Lineages on Weeds of Rice through placed the origin of ancestral Phytophthora hap-
Host Shifts and Specialization lotypes in South America, other evidence points
As described above, the emergence of Magna- to a Mexican origin of the potato-infecting
porthe oryzae fits a host-tracking coevolutionary lineage. Central Mexico is considered a cen-
scenario with rice; however, the emergence of a ter of diversity for the genus Solanum with
rice-infecting lineage also involved a number of the presence of several endemic tuber-bearing
host shifts. Couch et al. (24) presented evidence species (48, 51). P. infestans is able to infect

most of the common native Solanum species progression by inducing cell death in the host
(63, 80). Additionally, more complex races in- tissue.
cluding isolates containing all known virulence ToxA was initially characterized in P. tritici-
factors were found among isolates collected repentis (8, 117), but during annotation of the P.
from wild and cultivated potatoes in Mexico nodorum genome (44) an almost identical gene
(39). Co-existing relatives of P. infestans are sequence was discovered. Knock-out and trans-
host specific on endemic plant species of dif- formation experiments confirmed that ToxA
ferent plant families; however, they show great was expressed in the same way in both species.
morphological and genetic similarity to P. in- Further sequence comparisons between P. nodo-
festans on Solanum hosts. Based on phyloge- rum and P. tritici-repentis showed that the un-
netic analyses and a survey of host specificity, usually high sequence similarity involved a re-
Grünwald & Flier (39) hypothesized that host gion of 11 kb and included a transposase-like
switching followed by reproductive isolation re- gene downstream of the ToxA sequence, con-
sulted in the sympatric speciation of Phytoph- sistent with a hypothesis of horizontal genetic
thora species in central Mexico. The emergence exchange between the two pathogens.
of sibling pathogen species on different hosts The limited number of interspecific poly-
could have resulted from chance mutations as morphisms in both coding and noncoding
well as from hybridization. As presented below, DNA suggested that the HGT occurred re-
hybridization between closely related Phytoph- cently. A population survey of the ToxA gene
thora species could result in hybrid species with in both species showed that the level of in-
host specificities distinct from parental lineages. traspecific ToxA diversity was much higher in
The conflicting conclusions on the emer- P. nodorum than in P. tritici-repentis. The in-
gence of P. infestans will likely be resolved us- terpretation of this finding was that the gene
ing additional coalescence-based analyses that moved very recently from P. nodorum into P.
include a large number of closely related Phy- tritici-repentis, without sufficient time yet for
tophthora species from both South and Central mutations to occur in the recipient species. It
America as well as P. infestans from wild and is notable that P. tritici-repentis is a very recently
cultivated potatoes in the same regions. emerged pathogen. The tan spot disease caused
by P. tritici-repentis was first reported in 1941
and today represents a very serious disease of
EXAMPLES OF HORIZONTAL wheat. The authors’ interpretation was that the
GENE TRANSFER acquisition of the ToxA gene by HGT led to the
emergence of a new, highly damaging race of P.
Interspecific Transfer between Two tritici-repentis that emerged and became glob-
Wheat Pathogens of the Fungal ally distributed only during the past 60 years.
Virulence Gene ToxA
Friesen et al. (32) recently presented evidence
for horizontal transfer of the gene encoding A Cluster of Genes Determining
the host-specific toxin ToxA between the Pathogenicity to Pea on a
wheat pathogens Phaeosphaeria nodorum and Conditionally Dispensable
Pyrenophora tritici-repentis. Both pathogens Chromosome in Nectria haematococca
can cause similar foliar symptoms in major In Nectria hematococca (anamorph Fusarium
wheat-growing regions worldwide. The ToxA solani) the genes required for pathogenicity to-
gene contributes to pathogenicity on suscep- wards pea are located on a 1.6-Mb conditionally
tible wheat cultivars through a gene-for-gene dispensable (CD) chromosome (70, 107). The
interaction with the host toxin-sensitivity gene three virulence genes PEP1, PEP2, and PEP5
Tsn1 (31, 62). A compatible interaction between and the pisatin demethylase gene PDA1 are all
the products of the two genes facilitates disease located within a 25-kb region, forming a cluster
of pathogenicity determinants (43). Han and with host range than the phylogenetic relation-
coworkers (43) also found four open reading ships among pathogens. This parallels the find-
frames within the cluster having high similarity ings for ToxA in P. nodorum and P. tritici-repentis.
to known fungal transposases. The presence of
mobile elements suggests that the gene cluster
could have been acquired by HGT. Supporting Host-Specific Toxins in
this hypothesis is that the cluster of N. hemato- Alternaria alternata
cocca appears distinct from other regions of the Pathogenic isolates of the asexual fungus Al-
genome with respect to GC content and codon ternaria alternata are grouped into different
usage bias. pathotypes according to host specificity and the

The clustering of genes involved in production of host-specific toxins (61, 76). Al-
pathogenicity is well known in bacteria, where though the different host-specific toxins all in-
they have been called pathogenicity islands. duce cell death, their mode of action and chemi-
Pathogenicity islands are characterized by cal structure differ greatly between pathotypes.
genetic instability, discontinuous distribution For example, the AM-toxin produced by the
among bacterial species, proximity to mobile A. alternata apple pathotype affects both the
elements and associated repeated sequences, host plasma membrane and the chloroplasts
and dissimilarities to other regions of the chro- whereas the ACT-toxin produced by the tan-
mosome based on GC content (40). Horizon- gerine pathotype affects only the host mito-
tal transfer of pathogenicity islands has been chondria (78). Several of the genes required for
invoked to explain the rapid emergence of the production of the A. alternata toxins have
new pathogenic strains and the discontinu- been cloned and they all were organized in clus-
ous distribution of pathogenicity genes among ters, with some located on CD chromosomes
bacterial strains (77). Han et al. (43) under- (46, 55, 98). Hatta et al. (47) identified five
lined the striking similarity between bacte- disrupted transposable elements near the toxin
rial pathogenicity islands and the gene clus- gene cluster on a CD chromosome in the A. al-
ter of N. hematococca and suggested that fungal ternata strawberry pathotype. The clustered ar-
pathogens could also acquire disease-causing rangement and the presence of mobile elements
capabilities through horizontal transfer of coad- resemble the organization of N. hematococca and
apted pathogenicity genes arranged in clusters. U. maydis pathogenicity genes. These findings
Clusters of genes that are up-regulated dur- suggest that A. alternata pathotypes have ac-
ing infection are also seen in U. maydis, where quired the genes needed for production of host-
20% of the genes for secreted proteins are or- specific toxins through HGT. The emergence
ganized in clusters of 3–26 genes (56). Like the of new pathotypes may thus occur through the
PEP cluster of N. hematococca, a disrupted trans- acquisition of a coadapted gene complex that
posase sequence was identified in a cluster of enables the fungus to infect a previously un-
genes involved in virulence in U. maydis, sug- infected host. In support of this hypothesis
gesting that transposase activity and HGT may is the observation that homologous genes for
have played a role in the genomic organiza- toxin biosynthesis are not found in strains from
tion of pathogenicity genes in the corn smut other pathotypes or nonpathogenic A. alternata
pathogen. isolates (46, 65, 97).
Besides being involved in pathogenicity, the In their review, van der Does & Rep (103)
clustered genes of N. hematococca may also be de- discuss the origin of clusters of virulence genes.
terminants of host range. The finding of highly They suggest two main scenarios: one in which
similar pisatin demethylase genes in strains of a combination of gene shuffling and selection
Fusarium oxysporum pathogenic on peas sug- links virulence genes together into clusters, and
gests that particular pathogenicity genes or a second scenario where certain genomic re-
gene clusters may be more likely to correlate gions are more receptive for either de novo

emergence or insertion of virulence genes. Both survey was conducted to characterize the new
scenarios are plausible and not mutually exclu- alder pathogen. Brasier et al. (15) compared
sive. The high density of transposable elements phenotypic and genetic characters of P. cam-
ITS: internal
transcribed spacer of and repetitive sequences in the gene clusters bivora and the alder Phytophthora and found
the ribosomal RNA described to date could facilitate the accumula- that the pathogens clearly differed from each
gene cluster tion of additional virulence genes. An important other. Isolates of the alder pathogen exhibited
selective advantage favoring the maintenance a much larger variation than expected within
of virulence gene clusters in pathogens is the a single Phytophthora taxon, including differ-
propensity for them to be transmitted as a sin- ences in chromosome number, colony develop-
gle block during HGT. Comparative genomics ment, morphology, and optimal growth tem-
will likely provide significant insight into the perature. Single isolates of the alder pathogen
mechanisms underlying the evolution of clus- harbored a number of different ITS sequences–
ters of virulence genes. a phenomenon also known from allopolyploid
HGT has been proposed as an important hybrids of plants where concerted evolution
evolutionary mechanism in fungi for several has not yet homogenized ribosomal sequences.
years (83, 86, 106). Are the examples pre- Brasier et al. (18) concluded that the new
sented in this section unique or is the interspe- alder pathogen was composed of a range of
cific exchange of pathogenicity-related genes heteroploid hybrids originating from several
much more common than previously believed? hybridization events. The new Phytophthora
A successful exchange of genes between fun- species was formally named P. alni but was fur-
gal species may not be common in natural ther divided into three subspecies: P. alni subsp.
ecosystems. But the environmental homogene- alni (Paa), P. alni subsp. uniformis (Pau), and
ity, dense host populations, and large pathogen P. alni subsp. multiformis (Pam) (16). To infer
population sizes inherent in agro-ecosystems the origins of the three P. alni subspecies, Ioos
increase the chances for a successful interspe- and coworkers (53) used a combination of nu-
cific genetic exchange. The global trade in agri- clear and mitochondrial markers to study a large
cultural products moves pathogens around the European collection of Phytophthora. They
world, thereby increasing the probability that found that the three subspecies differed consid-
different species will meet and possibly ex- erably in their allelic diversity as well as genomic
change genes. The examples presented in this structure, consistent with different origins for
section suggest that HGT can be an important Paa, Pam, and Pau. The Paa subspecies con-
factor in the emergence of new pathogens or tained alleles specific to both Pam and Pau, and
new pathogen specificities and should therefore the authors therefore suggested that Paa could
be considered a potential risk affecting disease represent a hybrid between Pam and Pau gen-
management in agro-ecosystems. erated over several hybridization events. The
Pau subspecies showed the highest similarity to
P. cambivora and may be a progenitor of this
EXAMPLES OF INTERSPECIFIC species. The origin of Pam was less clear, but
HYBRIDIZATION the authors proposed that this subspecies was
generated through ancient hybridization events
Hybridization of Allopatric Taxa of or by autopolyploidization. Contemporary iso-
Phytophthora Led to the Emergence lates of Pam and Pau are far less common on
of a New Pathogen of Alder Trees alder and are less aggressive. Pam was initially
In 1993 a new disease on alder trees was re- isolated from soil in a natural alder stand where
ported (18). The disease-causing pathogen was no disease symptoms occurred, whereas Pau
recorded as Phytophthora cambivora, a common was known from asymptomatic alder seedlings.
pathogen of trees. Phytophthora was not pre- Therefore, Ioos et al. (53) hypothesized that
viously known to infect Alnus, and a genetic Pam and Pau existed on or in the vicinity of
alder trees for a long time without causing no- sive than their parents. Although hybrids are
table disease on the trees. However, the hybrid transient, they can act as a genetic bridge be-
progeny of the two subspecies contained a com- tween the two parental species. Bates et al. (12)
bination of genes that enabled Paa to cause dis- showed that rare O. ulmi polymorphisms were
ease on alder and to become a much more ag- present in the mtDNA of O. novo-ulmi isolates.
gressive pathogen. Introgression of O. ulmi genes into O. novo-
A similar example of a Phytophthora hybrid ulmi also included genes that were involved in
population proliferating on new hosts was re- pathogenicity. Abdelali et al. (1) investigated
cently reported in the Netherlands (64). Hy- an EAN isolate of O. novo-ulmi that showed
brids formed between P. cactorum and the intro- an unusually low level of pathogenicity. The
duced P. hedraiandra differed genetically from authors identified a single gene Pat1 control-
both parent species and were able to infect ling pathogenicity and found that this gene was
a range of new hosts including onions, leeks, linked to RAPD markers that were specific to
and other horticultural species. Future stud- O. ulmi. The authors hypothesized that the Pat1
ies should follow the establishment and evo- gene was introgressed into O. novo-ulmi from O.
lution of these Phytophthora hybrids on Alnus ulmi, leading to a change in pathogenicity in the
and Allium and other horticultural hosts–not EAN isolate. Backcrossing of hybrid progeny
only to develop risk assessments for the new with parental strains of O. novo-ulmi and O. ulmi
pathogens but also to study the on-going host- could provide a transmission route for other
specialization and speciation of these Phytoph- genes involved in pathogenicity or increased
thoras occurring within a known time frame. fitness.
Interspecific Hybridization as a UNIFYING THEMES FOR

Genetic Bridge between Species of the PATHOGEN EMERGENCE
Dutch Elm Disease Pathogens ACROSS AGRO-ECOSYSTEMS
Ophiostoma ulmi and O. novo-ulmi As illustrated by these examples, plant
Dutch elm disease is caused by ascomycete pathogens can emerge through many different
fungi of the genus Ophiostoma. The two se- mechanisms and across different timescales
vere pandemics in Europe and North America (Table 1). Most worrisome are the examples
in the twentieth century were caused by the suggesting that new pathogens can emerge
two species O. ulmi and O. novo-ulmi (14). almost instantaneously following host jumps,
The origin of the two species is not known horizontal gene transfers, or hybridization
but they exhibit different ecological strategies, events. As we compare the examples and
suggesting an adaptation to different environ- look for common underlying mechanisms, it
ments. O. novo-ulmi is further divided into appears these examples have little in common.
the Eurasian EAN and the North American Pathogen classes included Ascomycetes, Ba-
NAN subspecies. Although phylogenetic anal- sidiomycetes, and Oomycetes. Hosts included
yses group O. ulmi and O. novo-ulmi into dis- cereals, potatoes, fruits, vegetables, and trees.
tinct taxa, they are not totally reproductively The common factor uniting all of these
isolated. Like the allopatric species of Phytoph- examples is that the disease emerged in a
thora mentioned above, it is possible that repro- human-managed ecosystem, with the most
ductive barriers did not evolve between the two prominent examples occurring in the agro-
Ophiostoma species if they were separated for ecosystem. In this section we discuss which
a long time by geographical barriers. Rare hy- properties of the agro-ecosystem are most likely
brids are now encountered in nature (17). These driving the emergence of plant pathogens.
hybrids show a range of nonparental pheno- One of the most important factors driv-
types but are generally less fit and less aggres- ing the emergence of plant pathogens is the

highly homogeneous environment of agro- netic diversity, probably reflecting limited host
ecosystems. An agricultural field is tilled, availability and/or genetic bottlenecks that oc-
graded, fertilized, watered, and weeded to cre- curred as wild grasslands were converted to
ate an environment that is much more homoge- wheat fields. Thus the adaptation to an agri-
neous than a natural ecosystem (Figure 1). This cultural environment was followed by large
homogeneous environment is then planted to changes in the population dynamics of the
a dense and genetically uniform plant popu- pathogen.
lation that generates a uniform microclimate Large and dense host populations grow-
and provides a highly conducive environment ing in a homogeneous environment also in-
for pathogen propagation and dispersal and crease the potential for new pathogens to
for the evolution of more virulent genotypes. emerge through genetic exchanges and inter-
The lack of genetic variation in the host pop- actions with other pathogen species coexisting
ulation makes it highly vulnerable to disease on the same host plant. HGT and hybridiza-
outbreak because virulent pathogen genotypes tion are probably rare events in natural ecosys-
adapted to a particular host genotype can in- tems where the environment is heterogeneous,
crease very rapidly in frequency, quickly gener- disease tends to be patchily distributed, and
ating a degree of host specificity or race speci- pathogen population density is low. As a result,
ficity that rarely occurs in natural ecosystems. coinfection by different pathogens may be rare,
These host specificities can be generated as a and plants of the same species may be too far
result of strong directional selection operating apart to facilitate transmission of the rare hy-
on only one or two genes, such as the ToxA gene brids to uninfected plants (Figure 1). But in
of Phaeosphaeria nodorum and Pyrenophora tritici- agro-ecosystems it is common to find plants
repentis, host-specific toxins in Alternaria alter- colonized by several different pathogens (e.g.,
nata, or NIP1 in Rhynchosporium secalis. P. tritici-repentis and P. nodorum are frequently
A second important unifying factor is the found on the same wheat leaf ), infections are
vast scale of the agro-ecosystem. Approximately often found at a high density, and any rare hy-
50% of the global landmass has been con- brids that might form will have a high proba-
verted to cultivated crops or grazed land dur- bility of making a successful transmission to a
ing the past 10,000 years (57). Individual fields neighboring plant. Thus agro-ecosystems are
planted to a monoculture can be 100s to 1000s likely to provide more effective incubators for
of hectares in size and contain millions of plants. HGT and pathogen hybridization than natural
As a result, the corresponding pathogen popu- ecosystems.
lations also can be very large. Large pathogen A third important unifying factor is the
populations increase genetic diversity through global expansion of trade in crops and
an accumulation of mutations and they expe- germplasm coupled with inefficiency in quar-
rience less inbreeding than small populations. antine and monitoring systems. New pathogens
As a result of increased genetic diversity, the may be emerging continuously in agro-
pathogen is able to respond more rapidly to ecosystems around the world, but their dam-
the deployment of control measures, such as age will be limited to the field or localized re-
the introduction of new fungicides or resis- gion where they emerge and they may not be
tance genes into the agro-ecosystem (102). In noticed unless they escape and spread to new
the wheat pathogen Mycosphaerella graminicola, areas where the same crop is grown. The es-
the adaptation to an agricultural host was fol- cape is most likely to occur for pathogens that
lowed by a tremendous increase in population infect or infest seed or other plant parts that
size consistent with the increased host popu- are propagated and traded (e.g., tubers, cut-
lation size (95). In contrast, the sibling species tings). Examples include Panama disease of ba-
occurring on wild grasses showed a much lower nanas (13) and potato late blight (34). Such
population size as well as a lower level of ge- long-distance movement of pathogens from
one environment to another is probably rare Given the similarities between pathogens of hu-
in natural ecosystems. The global transport of mans and plants, the approaches and methods
pathogens has created cosmopolitan pathogens used to study human pathogens can also provide
as well as meta-populations of locally adapted insight into the evolution of plant pathogens
populations. As natural habitats continue to and serve as inspiration for future research.
be converted into agro-ecosystems in order
to feed the growing human population, new
pathogens are likely to emerge via host jumps, CONCLUSIONS

horizontal gene transfer, or hybridization in- Plant pathogens will continue to emerge in
volving pathogen species that already existed agro-ecosystems around the world. Advances
in the natural habitat. This process may ex- in DNA analyses allow us to reconstruct the
plain the emergence of wheat blast in Brazil evolutionary history of pathogens and to un-
(42). In turn, some pathogens that have already derstand the processes that caused their emer-
adapted to the agro-ecosystem may be able to gence. The timescales involved in pathogen
jump from the cultivated crop and onto wild emergence range from thousands of years for
relatives neighboring the newly installed agro- pathogens that emerged through codomesti-
ecosystem. This appears to have happened with cation of host and pathogen to practically in-
the barley scald pathogen R. secalis, which now stantaneous for new pathogen races or species
infects Hordeum spontaneum, the wild ances- that emerged through horizontal gene trans-
tor of cultivated barley, in the Fertile Crescent fer or interspecific hybridization. The unifying
(C.C. Linde & B.A. McDonald, unpublished). driving force for these different evolutionary
In this chapter we have focused on the evo- processes is the nature of the agro-ecosystem.
lution of plant pathogens, but similar evolu- The genetic and environmental uniformity of
tionary processes operate for human-associated monocultures coupled with the vast scale of
pathogens. The emergence and evolution of agro-ecosystems select for highly specialized
human diseases is favored by dense and large and aggressive pathogens that rapidly respond
human populations, by close contact with other to selective agents such as fungicides or resis-
host species such as domesticated or wild an- tance genes. It is important to acknowledge that
imals, and by the global movement of peo- new plant pathogens will continue to emerge as
ple. Many new diseases, including measles, long as agro-ecosystems maintain their present
smallpox, tuberculosis, malaria, and pertussis, structure.
emerged with the onset of agriculture and So what can we do to slow or prevent the
the resulting development of agro-ecosystems emergence of new pathogens? An important
and increase in human population density (79). contribution would be to improve monitoring
Some diseases emerged by host jumps from abilities and capacities to detect new pathogens
wild or domesticated animals [e.g., the measles when they emerge and then try to prevent or
virus Morbillivirus from cattle (108)], while oth- slow their spread. This approach is already used
ers proliferated in the agro-ecosystem [e.g., the for emerging human diseases with monitoring
malaria parasite Plasmodium falciparum and its coordinated by the Center for Disease Con-
vector Anopheles gambiae (52)]. More recently a trol in the United States and globally through
host shift led to the emergence of HIV through the World Health Organization. A useful model
multiple interspecific transmissions of viruses for implementing plant disease containment is
from chimpanzee to humans (41). Interdisci- presented by the Global Rust Initiative (GRI)
plinary research cutting across the disciplines that formed in 2001 to counteract the spread
of pathology, ecology, evolutionary biology, and of a new race of the wheat stem rust fungus
bioinformatics is used to determine the under- Puccinia graminis that emerged in Uganda in
lying causes of human disease emergence and to 1999 (93). Such initiatives can achieve success
control the spread of human infectious diseases. when new diseases are recognized early in their

emergence and before they have spread be- genetic heterogeneity into the agro-ecosystem
yond a reasonable containment zone that can to make it less conducive to pathogen emer-
be managed by quarantines or eradication ef- gence. For example, crop species diversity can
forts. Unfortunately, GRI-style approaches are be increased spatially and temporally through
very expensive and require intensive surveil- faster and more complex crop rotations, plant-
lance networks. The general trend in the USA ing of species mixtures, and by decreasing
(with a few exceptions) is for less surveillance average field size. Genetic diversity within
of diseases in the field, rather than more. This monocultures can be increased spatially and
is tied to the high cost of effective surveil- temporally by growing cultivar mixtures and
lance and the decreased number of extension multilines or by growing several different cul-
specialists. tivars of the same host in patches within the
Many plant pathogens are limiting factors same farm. Environmental heterogeneity can
in food production throughout the world. In be increased by combining agriculture and
industrialized countries the impact of plant dis- forestry or with other mixed cropping systems.
ease is mostly an economic issue, but in devel- It is obvious that such re-engineering of ex-
oping countries plant pathogens can be a pri- isting agriculture and forestry systems would
mary cause of starvation (94). Agro-ecosystems present formidable challenges, with many con-
will need to be re-engineered to prevent straints imposed by political, sociological and
the continuous emergence of new pathogens. economic considerations. However, the result
Several approaches can be used to slow or pre- of re-engineering the agro-ecosystem will be
vent the emergence of new pathogens. The to develop more sustainable and reliable food
goal of these approaches will be to reintroduce production that will be needed to support the
a combination of environmental, species, and human population for the next 10,000 years.
SUMMARY POINTS
1. Pathogens emerge through many different mechanisms.
2. These mechanisms can be differentiated.
3. Time scales for emergence can differ considerably.
4. DNA sequencing and statistical analyses can be used to infer evolutionary processes
including:
1) long-term host-tracking of pathogens since crop domestication.
2) instant emergence of pathogens by host shift, horizontal gene transfer or hybridiza-
tion.
5. Structure of the agro-ecosystem is the main driver of pathogen emergence.
DISCLOSURE STATEMENT
The authors are not aware of any biases that might be perceived as affecting the objectivity of this
review.
ACKNOWLEDGMENTS
This work was supported in part by grants from the Swiss National Science Foundation (3100A0-
104145) and the Swiss Federal Institute of Technology (ETH Zurich, TH-23/02-4). We thank
Georgiana May for providing a manuscript under review prior to publication and Linda Kohn for
inspiring conversations.
LITERATURE CITED
1. Abdelali E-T, Brasier CM, Bernier L. 1999. Localization of a pathogenicity gene in Ophios-
toma novo-ulmi and evidence that it may be introgressed from O. ulmi. Mol. Plant Microbe
Interact. 12:6–15
2. Anagnostakis S. 1987. Chestnut blight: the classical problem of an introduced pathogen.
Mycologia 79:23–37
3. Anderson JB, Korhonen K, Ullrich RC. 1980. Relationships between European and North
American biological species of Armillaria mellea. Exp. Mycol. 4:78–86
4. Anderson RM, May RM. 1982. Coevolution of hosts and parasites. Parasitology 85:411–12
5. Andersson JO. 2005. Lateral gene transfer in eukaryotes. Cell. Mol. Life Sci. 62:1182–97
6. Arbogast BS, Edwards SV, Wakeley J, Beerli P, Slowinski JB. 2002. Estimating divergence
time from molecular data on phylogenetic and population genetic timescales. Annu. Rev.
Ecol. Syst. 33:707–40
7. Badr A, Müller K, Schäfer-Pregl R, Rabey HE, Effgen S, et al. 2000. On the origin and
domestication history of barley (Hordeum vulgare). Mol. Biol. Evol. 17:499–510
8. Ballance GM, Lamari L, Bernier CC. 1989. Purification and characterization of a host-
selective necrosis toxin from Pyrenophora tritici-repentis. Physiol. Mol. Plant Pathol. 35:203–13
9. Balter M. 2007. Seeking agriculture’s ancient roots. Science 316:1830–35
10. Banke S, McDonald BA. 2005. Migration patterns among global populations of the
pathogenic fungus Mycosphaerella graminicola. Mol. Ecol. 14:1881–96
11. Banke S, Peschon A, McDonald BA. 2004. Phylogenetic analysis of globally distributed
Mycosphaerella graminicola populations based on three DNA sequence loci. Fungal Genet.
Biol. 41:226–38
12. Bates MR, Buck KW, Brasier CM. 1993. Molecular relationships of the mitochondrial DNA
of Ophiostoma ulmi and the NAN and EAN races of O. novo-ulmi determined by restriction
fragment length polymorphisms. Mycol. Res. 97:1093–100
13. Bentley S, Pegg KG, Moore NY, Davis RD, Buddenhagen IW. 1998. Genetic variation
among vegetative compatibility groups of Fusarium oxysporum f. sp. cubense analyzed by DNA
fingerprinting. Phytopathology 88:1283–93
14. Brasier CM. 2001. Rapid evolution of introduced plant pathogens via interspecific hybridiza-
tion. BioScience 51:123–33
15. Brasier CM, Cooke DEL, Duncan JM. 1999. Origin of a new Phytophthora pathogen through
interspecific hybridization. Proc. Natl. Acad. Sci. USA 96:5878–83
16. Brasier CM, Kirk SA, Delcan J, Cooke DEL, Jung T, Man in’t Veld WA. 2004. Phytophthora
alni sp nov and its variants: designation of emerging heteroploid hybrid pathogens spreading
on Alnus trees. Mycol. Res. 108:1172–84
17. Brasier CM, Kirk SA, Pipe ND, Buck KW. 1998. Rare interspecific hybrids in natural
populations of the Dutch elm disease pathogens Ophiostoma ulmi and O. novo-ulmi. Mycol.
Res. 102:45–57
18. Brasier CM, Rose J, Gibbs JN. 1995. An unusual Phytophthora associated with widespread
alder mortality in Britain. Plant Pathol. 44:999–1007
19. Brunner PC, Schurch S, McDonald BA. 2007. The origin and colonization history of the
barley scald pathogen Rhynchosporium secalis. J. Evol. Biol. 20:1311–21
20. Carleton MA. 1915. A serious new wheat rust in this country. Science 42:58–59

21. Carlile MJ, Watkinson SC, Gooday GW. 2001. The Fungi. London: Academic
22. Ceplitis A. 2003. Coalescence times and the Meselson effect in asexual eukaryotes. Genet.
Res. Camb. 82:183–90
23. Chang KC. 1986. The Archaeology of Ancient China. New Haven, CT: Yale Univ. Press
24. Couch BC, Fudal I, Lebrun M-H, Tharreau D, Valent B, et al. 2005. Origins of host-
A detailed study
showing both specific populations of the blast pathogen Magnaporthe oryzae in crop domestication
with subsequent expansion of pandemic clones on rice and weeds of rice. Genetics
host-tracking and
host-shifting during 170:613–30
the emergence of the
25. Crawford GW. 1992. Prehistoric plant domestication in East Asia. In The Origins of Agri-
rice blast pathogen
at the onset of
culture: An International Perspective, ed. CW Cowan PJ Watson, pp. 7–38. Washington, DC:
agriculture in Asia. Smithsonian Inst. Press
26. de la Cruz F, Davies J. 2000. Horizontal gene transfer and the origin of species: lessons from
bacteria. Trends Microbiol. 8:128–33
27. Dettman JR, Jacobson DJ, Taylor JW. 2003. A multilocus genealogical approach to phyloge-
netic species genealogical recognition in the model eukaryote Neurospora. Evolution 57:2703–
20
28. Diamond J. 1997. Guns, Germs, and Steel: the Fates of Human Societies. New York: Norton
29. Drake JW, Charlesworth B, Charlesworth D, Crow JF. 1998. Rates of spontaneous muta-
tion. Genetics 148:1667–86
30. Drummond AJ, Rambaut A, Shapiro B, Pybus OG. 2005. Bayesian coalescent inference of
past population dynamics from molecular sequences. Mol Biol Evol 22:1185–92
31. Faris J, Anderson JA, Francl LJ, Jordahl JG. 1996. Chromosomal location of a gene
conditioning insensitivity in wheat to a necrosis-inducing culture filtrate from Pyrenophora
tritici-repentis. Phytopathology 86:459–63
32. Friesen TL, Stukenbrock EH, Liu Z, Meinhardt S, Ling H, et al. 2006. Emergence of
The first study to
show direct evidence a new disease as a result of interspecific virulence gene transfer. Nat. Genet. 38:953–56
for a HGT event in 33. Fry WE, Goodwin SB, Matuszak JM, Spielman LJ, Milgroom MG, Drenth A. 1992. Popula-
pathogenic fungi, tion genetics and intercontinental migrations of Phytophthora Infestans. Annu. Rev. Phytopathol.
leading to the 30:107–30
emergence of the tan
spot disease on 34. Fry WE, Goodwin SB. 1997. Resurgence of the Irish potato famine fungus. BioScience
wheat. 47:363–71
35. Gomez-Alpizar L, Carbone I, Ristaino JB. 2007. An Andean origin of Phytophthora infes-
tans inferred from mitochondrial and nuclear gene genealogies. Proc. Natl Acad. Sci. USA
104:3306–11
36. Goodwin SB, Cohen BA, Fry WE. 1994. Panglobal distribution of a single clonal lineage of
the Irish potato famine fungus. Proc. Natl Acad. Sci. USA 91:11591–95
37. Gopher A, Abbo S, Lev-Yadun ST. 2002. The ‘when’, the ‘where’ and the ‘why’ of the
Neolithic revolution in the Levant. Doc. Praehist. 28:49–62
38. Griffiths RC, Tavaré S. 1995. Unrooted genealogical tree probabilities in the infinitely-
many-sites model. Math. Biosci. 127:77–98
39. Grünwald NJ, Flier WG. 2005. The biology of Phytophthora infestans at its center of origin.
Annu. Rev. Phytopathology 43:171–90
40. Hacker J, BlumOehler G, Muhldorfer I, Tschape H. 1997. Pathogenicity islands of virulent
bacteria: structure, function and impact on microbial evolution. Mol. Microbiol. 23:1089–97
41. Hahn BH, Shaw GM, De Cock KM, Sharp PM. 2000. AIDS as a zoonosis: scientific and
public health implications. Science 287:607–14
42. Hamer JE. 1991. Molecular probes for rice blast disease. Science 252:632–33
43. Han Y, Liu X, Benny U, Kistler HC, VanEtten HD. 2001. Genes determining pathogenicity
to pea are clustered on a supernumerary chromosome in the fungal plant pathogen Nectria
haematococca. Plant J. 25:305–14
44. Hane JK, Lowe RG, Solomon PS, Tan KC, Schoch CL, et al. 2007. Dothideomycete-plant
interactions illuminated by genome sequencing and EST analysis of the wheat pathogen
Stagonospora nodorum. Plant Cell 19:3347–68
45. Harlan JR. 1951. Anatomy of gene centers. Am. Nat. 85:97–103
46. Hatta R, Ito K, Hosaki Y, Tanaka T, Tanaka A, et al. 2002. A conditionally dispensable
chromosome controls host-specific pathogenicity in the fungal plant pathogen Alternaria
alternata. Genetics 161:59–70

47. Hatta R, Shinjo A, Ruswandi S, Kitani K, Yamamoto M, et al. 2006. DNA transposon fossils
present on the conditionally dispensable chromosome controlling AF-toxin biosynthesis and
pathogenicity of Alternaria alternata. J. Gen. Plant Pathol. 72:210–19
48. Hawkes J. 1990. The Potato. Evolution, Biodiversity and Genetic Resources. London: Belhaven
49. Hey J, Machado CA. 2003. The study of structured populations—new hope for a
Review with a clear
difficult and divided science. Nat. Rev. Genet. 4:535–43 presentation of the
50. Hey J, Nielsen R. 2007. Integration within the Felsenstein equation for improved Markov isolation with
chain Monte Carlo methods in population genetics. Proc. Natl. Acad. Sci. USA 104:2785–90 migration model for
51. Hijmans RJ, Spooner DM. 2001. Geographic distribution of wild potato species. Am. J. Bot. coalescent analysis.
88:2101–12
52. Hume JCC, Lyons EJ, Day KP. 2003. Human migration, mosquitoes and the evolution of
Plasmodium falciparum. Trends Parasitol. 19:144–49
53. Ioos R, Andrieux A, Marcais B, Frey P. 2006. Genetic characterization of the natural
Infers the possible
hybrid species Phytophthora alni as inferred from nuclear and mitochondrial DNA evolutionary history
analyses. Fungal Genet. Biol. 43:511–29 of Phytophthora
54. Johannesson H, Stenlid J. 2003. Molecular markers reveal genetic isolation and phylogeog- hybrids by genetic
raphy of the S and F intersterility groups of the wood-decay fungus Heterobasidion annosum. analyses of hybrid
Mol. Phylogenet. Evol. 29:94–101 strains.
55. Johnson L, Johnson R, Akamatsu H, Salamiah A, Otani H, et al. 2001. Spontaneous loss of
a conditionally dispensable chromosome from the Alternaria alternata apple pathotype leads
to loss of toxin production and pathogenicity. Curr. Genet. 40:65–72
56. Kamper J, Kahmann R, Bolker M, Ma L-J, Brefort T, et al. 2006. Insights from the genome
of the biotrophic fungal plant pathogen Ustilago maydis. Nature 444:97–101
57. Kareiva P, Watts S, McDonald R, Boucher T. 2007. Domesticated nature: shaping landscapes
and ecosystems for human welfare. Science 316:1866–69
58. Kidwell MG. 1993. Lateral transfer in natural populations of eukaryotes. Annu. Rev. Genet.
27:235–56
59. Kislev ME. 1982. Stem rust of wheat 3300 years old found in Israel. Science 216:993–94
60. Kumar J, Nelson RJ, Zeigler RS. 1999. Population structure and dynamics of Magnaporthe
grisea in the Indian Himalayas. Genetics 152:971–84
61. Kusaba M, Tsuge T. 1995. Phologeny of Alternaria fungi known to produce host-specific
toxins on the basis of variation in internal transcribed spacers of ribosomal DNA. Curr. Genet.
28:491–98
62. Lamari L, Bernier CC. 1989. Toxin of Pyrenophora tritici-repentis: host-specificity significance
in disease, and inheritance of host reaction. Phytopathology 79:740–44
63. Lozoya-Saldaña H, Hernández A, Flores R, Bamberg J. 1997. Late blight on wild Solanum
species in the Toluca valley in 1996. Am. Potato J. 74:445
64. Man in ‘t Veld W, de Cock A, Summerbell R. 2007. Natural hybrids of resident and intro-
duced Phytophthora species proliferating on multiple new hosts. Eur. J. Plant Pathol. 117:25–33

65. Masunaka A, Tanaka A, Tsuge T, Peever TL, Timmer LW, et al. 2000. Distribution
and characterization of AKT homologs in the tangerine pathotype of Alternaria alternata.
Phytopathology 90:762–68
66. Matsuoka Y, Vigouroux Y, Goodman MM, Sanchez GJ, Buckler E, Doebley J. 2002. A single
domestication for maize shown by multilocus microsatellite genotyping. Proc. Natl Acad. Sci.
USA 99:6080–84
67. May RM, Anderson RM. 1983. Epidemiology and genetics in the coevolution of parasites
and hosts. Proc. R. Soc. London Ser. B 219:281–313

68. Mayr E. 1982. The Growth of Biological Thought. Cambridge, MA: Belknap
69. McDonald BA, Linde C. 2002. Pathogen population genetics, evolutionary potential, and
durable resistance. Annu. Rev. Phytopathol. 40:349–79
70. Miao VPW, Matthews DE, VanEtten HD. 1991. Identification and chromosomal locations
of a family of cytochrome P-450 genes for pisatin detoxification in the fungus Nectria haema-
tococca. Mol. Gen. Genet. 226:214–23
71. Moore AMT, Hillman GC, Legge AJ. 2000. Village on the Euphrates: From Foraging to Farming
at Abu Hureyra. New York: Oxford Univ. Press
72. Munkacsi AB, Stoxen S, May G. 2007. Domestication of maize, sorghum, and sugarcane did
not drive the divergence of their smut pathogens. Evolution 61:388–403
73. Munkacsi AB, Stoxen S, May G. 2008. Ustilago maydis populations tracked maize
through domestication and cultivation in the Americas. Proc. R. Soc. London Ser. B.
doi:10.1098/rspb.2007.1636
74. Nielsen R, Wakeley J. 2001. Distinguishing migration from isolation: a Markov chain Monte
Carlo approach. Genetics 158:885–96
75. Nikoh N, Fukatsu T. 2000. Interkingdom host jumping underground: phylogenetic analysis
of entomoparasitic fungi of the genus Cordyceps. Mol. Biol. Evol. 17:629–38
76. Nishimura S, Kohmoto K. 1983. Host-specific toxins and chemical structures from Alternaria
species. Annu. Rev. Phytopathol. 21:87–116
77. Ochman H, Moran NA. 2001. Genes lost and genes found: evolution of bacterial pathogen-
esis and symbiosis. Science 292:1096–99
78. Otani H, Kohmoto K, Kodama M. 1995. Alternaria toxins and their effects on host plants.
Can. J. Bot. 73:S453–58
79. Pearce-Duvet J. 2006. The origin of human pathogens: evaluating the role of agriculture
and domestic animals in the evolution of human disease. Biol. Rev. 81:369–82
80. Rivera-Peña A. 1990. Wild tuber-bearing species of Solanum and incidence of Phytophthora
infestans (Mont.) de Bary on the Western slopes of the volcano Nevado de Toluca. 2. Distri-
bution of Phytophthora infestans. Potato Res. 33:341–47
81. Rohe M, Gierlich A, Hermann H, Hahn M, Schmidt B, et al. 1995. The race specific elicitor,
NIP1 from the barley pathogen Rhyncosporium secalis determines avirulence on host plants of
Provides a the RRS1 resistance genotype. EMBO J. 14:4168–77
comprehensible
82. Rosenberg NA, Nordborg M. 2002. Genealogical trees, coalescent theory and the
review of
coalescence analyses. analysis of genetic polymorphisms. Nat. Rev. Genet. 3:380–90
83. Rosewich UL, Kistler HC. 2000. Role of horizontal gene transfer in the evolution of fungi.
Annu. Rev. Phytopathol. 38:325–63
84. Roy BA. 2001. Patterns of association between crucifers and their flower-mimic pathogens:
Presents a number of
Host jumps are more common than coevolution or cospeciation. Evolution 55:41–53
studies analyzing the
origin of cereal crops 85. Salamini F, Ozkan H, Brandolini A, Schafer-Pregl R, Martin W. 2002. Genetics and
in the Fertile geography of wild cereal domestication in the Near East. Nat. Rev. Genet. 3:429–41
Crescent. 86. Sanders IR. 2006. Rapid disease emergence through horizontal gene transfer between eu-
karyotes. Trends Ecol. Evol. 21:656–58
87. Schardl CL, Craven KD. 2003. Interspecific hybridization in plant-associated fungi and
oomycetes: a review. Mol. Ecol. 12:2861–73
88. Schurch S, Linde CC, Knogge W, Jackson LF, McDonald BA. 2004. Molecular population
genetic analysis differentiates two virulence mechanisms of the fungal avirulence gene NIP1.
Mol. Plant-Microbe Interact. 17:1114–25
89. Spatafora JW, Sung GH, Sung JM, Hywel-Jones NL, White JF. 2007. Phylogenetic evi-
dence for an animal pathogen origin of ergot and the grass endophytes. Mol. Ecol. 16:1701–11
90. Spooner DM, McLean K, Ramsay G, Waugh R, Bryan GJ. 2005. A single domestication
for potato based on multilocus amplified fragment length polymorphism genotyping. Proc.
Natl. Acad. Sci. USA 102:14694–99
91. Stahl EA, Bishop JG. 2000. Plant-pathogen arms races at the molecular level. Curr. Opin.
Plant Biol. 3:299–304
92. Stenlid J, Karlsson JO. 1991. Partial intersterility in Heterobasidion annosum. Mycol. Res.
95:1153–59
93. Stokstad E. 2007. Deadly wheat fungus threatens World’s breadbaskets. Science 315:1786–87
94. Strange RN, Scott PR. 2005. Plant disease: a threat to global food security. Annu. Rev.
Phytopathol. 43:83–116
95. Stukenbrock EH, Banke S, Javan-Nikkhah M, McDonald BA. 2007. Origin and do-
Different
mestication of the fungal wheat pathogen Mycosphaerella graminicola via sympatric coalescence
speciation. Mol. Biol. Evol. 24:398–411 approaches were
96. Syvanen M. 1994. Horizontal gene transfer: evidence and possible consequences. Annu. Rev. used to describe the
Genet. 28:237–61 emergence of a new
wheat pathogen
97. Tanaka A, Shiotani H, Yamamoto M, Tsuge T. 1999. Insertional mutagenesis and cloning through
of the genes required for biosynthesis of the host-specific AK-toxin in the Japanese pear host-tracking and
pathotype of Alternaria alternata. Mol. Plant-Microbe Interact. 12:691–702 the subsequent
98. Tanaka A, Tsuge T. 2000. Structural and functional complexity of the genomic region con- pathogen expansion.
trolling AK-toxin biosynthesis and pathogenicity in the Japanese pear pathotype of Alternaria
alternata. Mol. Plant-Microbe Interact. 13:975–86
99. Thomma BPHJ. 2003. Alternaria spp.: from general saprophyte to specific parasite. Mol.
Plant Pathol. 4:225–36
100. Thrall PH, Burdon JJ. 1999. The spatial scale of pathogen dispersal: consequences for disease
dynamics and persistence. Evol. Ecol. Res. 1:681–701
101. Tooley PW, Fry WE, Gonzalez MJV. 1985. Isozyme characterization of sexual and asexual
Phytophthora infestans populations. J. Hered. 76:431–35
102. van der Bosch F, Gilligan CA. 2008. Models of fungicide resistance dynamics. Annu. Rev.
Phytopathol. 46:123–47
103. van der Does HC, Rep M. 2007. Virulence genes and the evolution of host specificity
Review summarizes
in plant-pathogenic fungi. Mol. Plant-Microbe Interact. 20:1175–82 evidence for islands
104. Vaughan DA, Morishima H, Kadowaki K. 2003. Diversity in the Oryza genus. Curr. Opin. of pathogenicity
Plant Biol. 6:139–46 genes and horizontal
gene transfer in
105. Vavilov NI. 1926. Studies on the origin of cultivated plants. Bull. Appl. Bot. Plant Breed. (St.
pathogenic fungi.
Petersburg, Russia: Leningrad, USSR) 16:139–248.
106. Walton JD. 2000. Horizontal gene transfer and the evolution of secondary metabolite gene
clusters in fungi: an hypothesis. Fungal Genet. Biol. 30:167–71
107. Wasmann CC, Vanetten HC. 1996. Transformation-mediated chromosome loss and disrup-
tion of a gene for pisatin demethylase decrease the virulence of Nectria haematococca. Mol.
Plant-Microbe Interact. 9:793–803

108. Westover KM, Hughes AL. 2001. Molecular evolution of viral fusion and matrix protein
genes and phylogenetic relationships among the Paramyxoviridae. Mol. Phylogenet. Evol.
21:128–34
109. Wyand RA, Brown JKM. 2003. Genetic and forma specialis diversity in Blumeria graminis
of cereals and its implications for host-pathogen coevolution. Mol. Plant Pathol. 4:187–98
110. Zaffarano PL, McDonald BA, Zala M, Linde CC. 2006. Global hierarchical gene diversity
analysis suggests the Fertile Crescent is not the center of origin of the barley scald pathogen
Rhynchosporium secalis. Phytopathology 96:941–50

111. Zaffarano PL, McDonald BA, Linde CC. 2008. Rapid speciation following recent host
This study infers the

evolution of the scald shifts in the plant pathogenic fungus Rhynchosporium. Evolution 62–6:1418–36
pathogen of barley 112. Zeigler RS. 1998. Recombination in Magnaporthe grisea. Annu. Rev. Phytopathol. 36:249–75
and rye through 113. Zhan J, Mundt CC, Hoffer ME, McDonald BA. 2002. Local adaptation and effect of host
multiple host shifts. genotype on the rate of pathogen evolution: an experimental test in a plant pathosystem.
J. Evol. Biol. 15:634–47
114. Zhan J, Mundt CC, McDonald BA. 2000. Estimating rates of recombination and migration
in populations of plant pathogens–a reply. Phytopathology 90:324–26
115. Zhan J, Mundt CC, McDonald BA. 1998. Measuring immigration and sexual reproduction
in field populations of Mycosphaerella graminicola. Phytopathology 88:1330–1337
116. Zhan J, Pettway RE, McDonald BA. 2003. The global genetic structure of the wheat
pathogen Mycosphaerella graminicola is characterized by high nuclear diversity, low mito-
chondrial diversity, regular recombination, and gene flow. Fungal Genet. Biol. 38:286–97
117. Zhang H, Francl LJ, Jordahl JG, Meinhardt SW. 1997. Structural and physical properties
of a necrosis-inducing toxin from Pyrenophora tritici-repentis. Phytopathology 87:154–60
118. Zohary D, Hopf M. 2000. Domestication of Plants in the Old World. Oxford, UK: Oxford Univ.
Press. 3rd ed.

The Origins of Plant Pathogens in Agro-Ecosystems: Further

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

The Origins of Plant Pathogens in Agro-Ecosystems: Further

Transféré par

Droits d'auteur :

Formats disponibles

Annu. Rev. Phytopathol. 2008.46:75-100. Downloaded from arjournals.annualreviews.

Click here for quick links to

Eva H. Stukenbrock and Bruce A. McDonald

Annu. Rev. Phytopathol. 2008. 46:75–100 Key Words

species and its to the long timescales needed to evolve the

microorganisms and pathogen attack and plant defense in pathogen-

occupying the same

www.annualreviews.org • Origins of Plant Pathogens 77

www.annualreviews.org • Origins of Plant Pathogens 79

ciﬁc to the host crop (e.g., the agro-ecosystem cospeciation where

www.annualreviews.org • Origins of Plant Pathogens 81

Ustilago maydis on maize 8000 years BP 72

Host jump/host shift

www.annualreviews.org • Origins of Plant Pathogens 83

Domestication of the Rice Blast Domestication of the Wheat Pathogen

www.annualreviews.org • Origins of Plant Pathogens 85

www.annualreviews.org • Origins of Plant Pathogens 87

usage bias. pathotypes according to host speciﬁcity and the

www.annualreviews.org • Origins of Plant Pathogens 89

Interspecific Hybridization as a UNIFYING THEMES FOR

www.annualreviews.org • Origins of Plant Pathogens 91

pathogens are likely to emerge via host jumps, CONCLUSIONS

www.annualreviews.org • Origins of Plant Pathogens 93

www.annualreviews.org • Origins of Plant Pathogens 95

alternata. Genetics 161:59–70

www.annualreviews.org • Origins of Plant Pathogens 97

and hosts. Proc. R. Soc. London Ser. B 219:281–313

www.annualreviews.org • Origins of Plant Pathogens 99

Rhynchosporium secalis. Phytopathology 96:941–50

This study infers the

100 Stukenbrock · McDonald

Vous aimerez peut-être aussi