• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Trends Ecol Evol. Author manuscript; available in PMC Jul 1, 2011.
Published in final edited form as:
PMCID: PMC2885483

Linking emergence of fungal plant diseases and ecological speciation


Emerging diseases represent a growing worldwide problem accompanying global environmental changes, and there is tremendous interest in identifying the factors controlling the appearance and spread of these diseases. Here, we discuss emerging fungal plant diseases, and argue that they often result from host shift speciation, a particular case of ecological speciation. We consider the factors controlling local adaptation and ecological speciation and show that certain life-history traits of many fungal plant pathogens are conducive for rapid ecological speciation, thus favoring the emergence of novel pathogen species adapted to new hosts. We argue that placing the problem of emerging fungal diseases of plants within the context of ecological speciation can significantly improve our understanding of the biological mechanisms governing emergence of such diseases.

The problem of emerging diseases in the context of ecological speciation

During the last century, human activities have strongly contributed to environmental and ecosystem changes at a global scale, breaking down many natural barriers to dispersal and causing an unprecedented redistribution of organisms [1]. There is growing evidence that these global changes play a key role in the emergence of infectious diseases in humans [2], wildlife [3], domestic animals [4], and plants [5]. Diseases are qualified as emergent when they have recently become a cause for concern due to an increase in virulence, infection of a novel host and/or occurrence in a new area. Pathogenic fungi in particular have been responsible for emerging diseases in plants and animals, and even in humans. Consequently, there is a greatly increased interest in identifying the factors driving the emergence of new fungal diseases [5-7].

Here, we focus on emerging fungal diseases in plants, because of their importance for humans and because of the abundance of documented cases. We discuss the importance of novel fungal diseases of plants in agricultural production and for natural ecosystems, and we argue that host shift speciation appears to be a major route for their emergence. Host shift speciation is the formation of a novel species adapted to a novel host, and is a particular case of ecological speciation [8, 9]. We argue that the problem of emerging fungal diseases is therefore best understood in the context of ecological speciation. There is a well-established body of mathematical theory dealing with the population dynamics of infectious diseases [10, 11]. However, in this theoretical framework, adaptation and divergence among pathogens are largely neglected despite recognition that infection of a novel host can create strong selective pressures and raise the risk of devastating emerging diseases.

We argue that understanding the emergence of fungal diseases requires integration with established theories of ecological speciation, while taking into account the life-history characteristics particular to pathogens. We review the factors that usually limit the potential for adaptation and speciation and show that some life-history features of fungal pathogens can be conducive to rapid ecological speciation by host shifts. Recognizing the link between emerging diseases and ecological speciation should lead to more relevant theoretical models that incorporate the features of fungal pathogens, as well as novel approaches to studying emerging pathogens. Together these efforts can yield important advances in our understanding of emerging fungal plant diseases and have practical applications for designing more efficient and sustainable control programs.

Importance of emerging plant fungal diseases

Fungi (sensu lato, i.e. including Oomycetes) are responsible for about 30% of emerging diseases in plants [5]. Epidemics caused by invasive pathogens have frequently been reported to alter natural ecosystems [5, 6]. Well-documented examples include chestnut blight fungus (Cryphonectria parasitica), the spread of which eliminated nearly 100% of the native chestnut trees throughout eastern American forests during the 20th century, and the Dutch elm disease fungus (Ophiostoma ulmi), which led to the destruction of American elms. A more recent example is Phytophthora cinnamomi, which currently threatens native forests throughout Australia [5, 6]. Such devastating diseases affect not only the host plants, but also the whole associated fauna including insects, birds and mammals that depend upon them.

Our primary food production is also at risk due to emerging crop diseases [12]. The most dramatic historical example is the epidemic caused by Phytophthora infestans on cultivated potato in Ireland in the 1840s [13]. The infamous Irish Potato Famine led to mass emigration and more than 1 million deaths from starvation or famine-related diseases. More recent examples include the severe blast disease of wheat that appeared in Brazil in the 1980s and then spread to other South American countries [14], and the Ug99 fungal pathotype causing stem rust disease of wheat, first identified in 1998 in Uganda and now threatening North Africa, Middle East and Asia [15].

Emergence of plant fungal diseases via host shift speciation

Given their importance, there has been an increasing focus on identifying the factors that drive the emergence of new fungal diseases [5-7]. Emerging diseases can result from a sudden increase in virulence and/or expansion of the geographic range of a previously unnoticed pathogen [16], but infection of a novel host is the most frequent cause of fungal emerging diseases [7]. Infection of a novel host can result from spillover, host range expansion or host shift. Much interest has been devoted to the extrinsic factors that promote contact between a pathogen and a novel host, such as climate change and worldwide trade [5, 6]. Less attention has been paid to the intrinsic genetic changes in the pathogen needed for successful infection of a novel host in cases of host range expansion and host shift. Although the pathogen can be pre-adapted for infection and transmission onto a new type of host, most often successful disease emergence requires the pathogen to adapt, as witnessed by the usual failure of experimental inoculations in novel host-pathogen combinations [17]. Horizontal gene transfer and inter-specific hybridization have been invoked to explain how pathogens might achieve an enhanced adaptive potential that allow the emergence of new fungal diseases on novel hosts [7], but most often such adaptation is the result of selection among existing genetic variants or novel mutations within a fungal population [7, 18]. Existing evolutionary theory tells us that adaptation to a new host will be most efficient if the flow of ancestral genes into the population adapting to a new host has ceased completely or is significantly reduced [19]. This makes the issue of the evolution of reproductive isolation (and speciation) important for understanding many emergent diseases.

The focus of this paper is the mechanism of fungal disease emergence by host shift speciation. There are many examples of formation of sibling pathogen species on different hosts, and these radiations most often involved hosts shifts [20]. An example is the anther smut fungi, constituting a complex of sibling species sterilizing Caryophyllaceae plants [21]. Comparisons of host and pathogen phylogenies have indeed shown that host shift speciation was frequent in the history of this association [22]. In fact, several incipient host shifts have been detected in natural populations of anther smut fungi. In these cases, populations that already appeared differentiated while still sub-optimally adapted to their novel host plants were found to be sympatric with their original hosts [23, 24].

Host shifts have also been involved in cases of recent emerging diseases due to introductions of fungal pathogens into new continents [25, 26]. Examples include the introduction of C. parasitica and Phytophthora cinnamomi in the USA and devastation of the American chestnut (Castanea dentata), the most probable source of which was Japanese chestnut trees (C. crenata) that were imported and sold throughout the country [27, 28]. These diseases are very recent, having occurred within the last century. The pathogens might thus have had no time to differentiate from their population of origin yet, although this has not been investigated in many cases. Differentiation has in fact been found when looked for, as in the case of C. parasitica between North America and Europe [27].

Other historical cases of disease emergence, associated with plant domestication, are also consistent with ecological speciation by host shifts. Well-documented cases are provided by the wheat fungal pathogen Mycosphaerella graminicola, that originated in the Fertile Crescent at the time of wheat domestication [29], and the fungus Rhynchosporium secalis, causing a disease called scald on rye, barley and other grasses [30].

The plethora of cases of closely related, but well-recognized species of fungal pathogens infecting different host species after host shift speciation leads to the question of whether some specificities of parasites render them particularly prone to adaptation to new niches and so to ecological speciation.

What factors do usually restrict the possibility of speciation?

The existing theory of ecological speciation [9, 19, 31-33] can be used to better understand factors constraining or promoting the adaptation of pathogens to a new host. According to this theory, alleles providing an advantage on a new host need to greatly increase in frequencies in the local population on that new host. This increase is accomplished by strong selection for local adaptation. The theory also tells us that locally advantageous combinations of alleles need to be protected from being “diluted” by ancestral alleles brought by immigrants. Such a “dilution” is expected if mating is random and immigration is recurrent. Therefore, protection of locally adapted allele combinations is required through the evolution of assortative mating (which can be achieved by mate choice, or habitat choice if mating occurs within habitats) or by strongly reduced viability of immigrants [34, 35]. The evolution of assortative mating and speciation can be prevented by a number of factors, including a lack of genetic variation, immigration of ancestral genes, or the costs of being choosy in mate and/or habitat choice. Moreover, recombination and segregation will work against the establishment of locally advantageous allele and trait combinations by continuously destroying them. Therefore the overall success of ecological speciation in the presence of gene flow depends on a delicate balance of several different factors, as discussed above.

Applying the above theory to fungi suggests that some characteristics of fungal pathogens can act to constrain the possibility of adaptation and divergence. For instance, fungi lack active means of dispersal, spores are usually air borne, and mycelium is also passively dispersed. This means they cannot choose their most suitable hosts, a factor that should produce a high genetic load. Fungi cannot move toward their most suitable mates either, which should increase the cost of waiting for a mate. Factors therefore exist that tend to oppose ecological speciation in fungi, but there are other, more powerful, mechanisms that strongly promote it.

Features of fungal pathogens of plants promoting ecological speciation

Several features of life-history traits in fungal pathogens are indeed conducive to ecological speciation by reducing the constraints that usually impair speciation. We detail these features below and examine their consequences for the possibility of ecological speciation.

Very large numbers of spores: population persistence and large mutational input

Pathogenic fungi can produce thousands of spores per lesion per day [36, 37] , and multiple asexual cycles on the same individual plant can yield hundreds of separate infections. This means that billions of spores can be released from a given plant during an infection by a single fungal genotype. Such large numbers of spores can allow the population to persist on a new host even when selection against allele combinations adapted to infection of the ancestral host is extremely strong and the initial degree of adaptation to a new host is very low. Moreover, such large numbers of spores allow the rapid and recurrent creation of genetic variation by mutations. Empirical examples support this logic. For instance, consider evolution of virulent strains able to overcome resistance in crops. New cultivars of plants with resistance genes conferring complete resistance often become susceptible in just a few years when deployed over large areas, due to the rapid appearance of new fungal genotypes by mutations that can infect the hitherto resistant plants. Such rapid evolution occurs even for clonal fungal pathogens with very low genetic diversity [38]. When the number of offspring is relatively small, host and/or mate choice are crucial for providing the offspring with the best possible genetic and environmental background [19, 39]. Producing a very large number of spores represents an alternative reproductive strategy, which can make the evolution of the mechanisms for host and mate choice unnecessary.

Mating within hosts: pleiotropy between host adaptation and assortative mating

Many pathogenic fungi can disperse over large distances either after mating or by asexual spores, but they lack the ability to disperse between infection of the host and mating. This is the case for many ascomycete fungal pathogens (Boxes 1 and 2), responsible for most of the devastating crop diseases. In cases of obligate biotrophs undergoing sex within their host plant, mating occurs only between individuals able to grow on the same host. This means that mutations providing adaptation to a new host will pleiotropically affect both local adaptation and mating patterns. This “magic trait” scenario is one of the most favourable for ecological speciation [19]. A theoretical model has shown that, because of this characteristic of the life cycle of some fungal pathogens, adaptation to a new host can significantly restrict gene flow even in sympatry, without the need for mate choice or host-choice [18, 35, 40] (Boxes 1 and 3). In this model the barrier to gene flow is the reduced viability of immigrants [34, 35]. With very strong selection this can completely prevent neutral gene flow. A number of studies provide empirical support for the generality of this mechanism in natural fungal plant pathogens (Box 2).

BOX 1Life cycle and the possibility of ecological speciation by host shift

Whether a plant pathogen mates within or outside its host has important consequences for the level of gene flow between populations adapted to different hosts. Consider two populations of an obligate biotroph fungal pathogen adapted to two different hosts. Assume there exists a single diallelic haploid locus involved in a gene-for-gene relationship, with allele A1 preventing infection of host 2 (because it codes for an effector recognized by the plant and inducing a defense reaction), and allele A2 preventing infection of host 1. Consider also a neutral locus with alleles B1 and B2. The figures figuresII and andIIII depict different kinds of barriers that could act to restrict neutral gene flow between the pathogen populations adapted to the alternative hosts:, i.e. host choice, host adaptation, geographic barriers between the two hosts, mate choice (i.e. assortative mating between the host races), and post-zygotic barriers. The arrows show potential gene flow. The red crosses indicate the steps in the life cycle where the barriers to gene flow act to prevent gene exchange at the neutral locus. The figures illustrate that, for obligate biotrophs mating within their hosts, host adaptation alone can be a barrier to gene flow, even at neutral loci, because only individuals able to grow on the same host can eventually mate. In contrast, host adaptation cannot substantially reduce gene flow at neutral loci when mating occurs outside the host. A life cycle of a pathogen mating outside of its host is represented in Fig. I. Geographic barriers, mate choice, and post-zygotic barriers can prevent gene flow at a neutral B locus (B1 and B2 alleles). Host choice and host adaptation can decrease the frequencies of allele A1 on host 2 and allele A2 on host 1 but cannot prevent gene flow at the neutral locus B. This is a typical life cycle of some basidiomycete fungal pathogens like rusts and smuts. A life cycle of an obligate biotroph mating within its host is represented in Fig. II. Host choice and host adaptation can prevent gene flow at the host choice or specialization loci (A1 and A2 alleles) but also at the neutral locus (B1 and B2 alleles), as can geographic barriers, mate choice and post-zygotic barriers. Host adaptation can therefore pleiotropically cause both specialization and reproductive isolation when mating occurs within hosts, which is not the case when mating occurs outside the host (Fig. I). This is a typical life cycle of ascomycete plant pathogens (Box 3).

We thank Antoine Orry for help in illustrations.

Fig. I

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0001.jpg

A life cycle of a pathogen mating outside of its host

Figure II

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0002.jpg

A life cycle of a pathogen mating within its host

BOX 2Host specialization as the only barrier to gene flow

The model of speciation described in Box 1 and Box 3 [35, 40] predicts that pathogens mating within their hosts can become different species just by host specialization. In contrast, additional reproductive barriers (e.g. intrinsic prezygotic isolation) are required to prevent gene flow in species mating outside of their host or substrate. Data on reproductive isolation in fungi support these predictions.

A comparative analysis has been conducted on the type and degree of reproductive isolation found in vitro among closely related fungal species [55]. Basidiomycetes (e.g. mushrooms, smuts, rusts) mate outside of their substrate after having dispersed their spores, while most Ascomycetes, which include many crop pathogens, mate on their host or substrate after mycelial development. Results showed that closely related species of Basidiomycetes have all evolved strong intrinsic reproductive isolation at any genetic distance unless they were allopatric. For instance, sympatric species pairs of Pleurotus and of Lentinula (Figs. (Figs.II and andII)II) show complete intersterility, while allopatric species pairs show a high degree of compatibility [55].

In contrast, closely related species of Ascomycetes could be as interfertile in vitro in sympatry as in allopatry, indicating that host adaptation can be an efficient barrier to gene flow. A good example is provided by Ascochyta pathogens. Recent multilocus phylogenetic analyses of Ascochyta fungi causing blights of chickpea, faba bean, lentil, and pea have revealed that fungi causing disease on each of these hosts form distinct species [56] (Fig. III). Experimental inoculations demonstrated that each species was highly host-specific, yet experimental crosses showed that the species were completely inter-fertile. The host specificity of these fungi might therefore constitute the sole reproductive barrier [56].

Another excellent example of host adaptation acting as an efficient barrier to gene flow is provided by Venturia inaequalis, an ascomycete responsible for the scab disease on apples [57, 58] (Fig. IV) and pyracantha [59]. The two host plants grow in sympatry but the populations on apples and pyracantha remain highly differentiated [60], indicating lack of gene flow, despite interfertility [59]. Reproductive isolation might thus be due to a mere host specialization, as cross-inoculations are unsuccessful [59] and as the fungus mates within leaves of its host tree. Similarly, an emerging disease caused by a V. inaequalis population carrying a virulence gene allowing infection of apple trees with the resistance gene Vf has remained genetically separated from the avirulent population since its appearance 10 years ago. This is despite their co-occurrence in cider orchards containing both resistant and susceptible trees [57, 58], and despite interfertility [61].

We thank T. Peever, B Le Cam, and Jacques Guinberteau (INRA MycSA) for pictures.

Fig. I

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0003.jpg

Pleurotus ostreatus

Fig. II

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0004.jpg

Lentinula edodes, the shiitake mushroom.

Fig. III

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0005.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0006.jpg

Lesions on a pod and on a leaf of chickpea caused by the fungus Ascochyta

Fig. IV

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0007.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0008.jpg

Lesion on a leaf and fruit of apple tree caused by the fungus Venturia inaequalis

BOX 3A model of speciation by host specialization

Mathematical models have shown that specificities in the life cycles of pathogens can favour ecological speciation [35, 40]. Consider sexual, haploid individuals of pathogens with four possible genotypes at two unlinked, diallelic loci A and B: A1B1, A1B2, A2B1, and A2B2. Assume that the population on the old host is close to fixation for the locally adapted genotype A1B1. The genotype fitnesses on the new host are (1 – s)2, 1 – s, 1 – s and 1, respectively. That is, genotype A2B2 has the highest fitness on the new host while the ancestral genotype A1B1has the smallest fitness. Each spore (both immigrating from the old host and produced locally) survives viability selection and becomes an adult with a probability equal to its fitness. If the number of surviving spores is larger than the carrying capacity K, they are randomly culled to have the population of adults of size K. Mating is random between adults surviving on the new host. Each mating results in a random number of spores taken from a Poisson distribution with mean c. Their genotypes are determined in a stochastic way subject to recombination at rate r and mutation at rate μ. A proportion m of locally produced spores immigrates. The parameters K, c, m and μ are assumed to be the same for both hosts.

In numerical simulations the dynamics of the colonization of new hosts proceed in two stages. During the first stage, all immigrating spores and surviving adults (if any) on the new host have the ancestral genotype A1B1 and thus have a very low fitness (Figure Ia and Ib). This stage ends when a mutation in one of the loci enters the local population and then becomes quickly fixed. The increase in the frequency of the mutation is then fast due to the assumption of a very large number of spores produced by each individual. The duration of the first stage mostly depends on the overall probability of an advantageous mutation entering the population (which is approximately K*c* μ). The second stage (during which the individuals present on the new host have intermediate genotype A1B2 or A2B1) ends when an advantageous mutation in the other locus enters the population and then becomes quickly fixed. The second stage is much shorter than the first one (Figure Ic and Id). This is so because the fixation of the first advantageous mutation results in a significant increase in the population size, which in turn increases the overall number of advantageous mutations.

Figure I

An external file that holds a picture, illustration, etc.
Object name is nihms-195244-f0009.jpg

The dynamics of adaptation to a new host. (a) Number of adults and (b) the number of spores on the new host. (c) Genotype densities and allele frequencies. The blue line represents the density of A1B2 or A2B1 genotypes, and the red line the frequencies of A2B2 genotypes. (d) Frequencies of the A2 and B2 alleles, allowing infection of the new host.

These arguments about plant pathogenic fungi are similar to those made about host shift speciation in phytophagous insects mating on their host plants [39]. However, there are important differences between the two types of organisms. Because insects are able to disperse at any stage of their life cycle, divergence in both local adaptation traits and host plant preference traits is required for substantial ecological differentiation. Selection must therefore most often generate linkage disequilibrium between the genes controlling these two types of traits [19,39]. In contrast, in many plant pathogens there is no dispersal stage between selection for local adaptation and mating. This implies that mating always happens only between individuals who survived selection on the same host, removing the need to evolve strong linkage disequilibrium and thus making conditions for speciation easier. It has however been suggested that, in some phytophagous insects, host choice traits can be “magic traits” pleiotropically controlling both local adaptation and mating [41, 42].

Strong disruptive selection by host plants acting on a small number of pathogenicity-related genes

Another factor that could facilitate ecological speciation in fungal pathogens is strong selection imposed by their host plants. Strong selection is effective in both increasing the frequency of locally advantageous alleles and preventing the immigration of locally deleterious ancestral alleles [19, 43]. Many plants have resistance genes involved in immunity while pathogens have effector genes that contribute to pathogenicity. Recognition of the pathogen effectors by a resistant plant activates a defence reaction, which completely prevents the establishment of a harmful infection. This resistance mechanism imposes a maximum fitness cost: fitness of a fungal strain carrying an “avirulent” allele on a plant carrying the resistant allele is simply zero, as the resistant plants are able to recognize these pathogen genotypes and completely prevent their growth (see Glossary).

Such resistance mechanisms can promote ecological speciation not only because of the strength of fitness costs, but also because they involve a single diallelic locus in the host and a single diallelic locus in the pathogen (thus being called “gene-for-gene” relationships) [44]. In general, selection is indeed most efficient when it acts on a small number of genes [43]. Gene-for-gene relationships are usually identified based on within-species polymorphisms, but the genetics underlying the ability to infect a particular plant species by a given fungal species, as compared to a closely related fungal species unable to parasitize this plant species, looks similar to gene-for gene relationships, at least in some cases [45-47]. Furthermore, even when the ability of infecting particular host species does not correspond to the canonical gene-for-gene relationship, there are several examples in plant pathogenic fungi where infection of a particular host is due to a change in a single locus or in a cluster of tightly linked loci. In the fungal plant pathogen Alternaria alternata, a cluster of several genes involved in the production of toxins controls host-specific pathogenicity [48]. Similarly, several genes that contribute to the ability of the fungus Nectria haematococca to cause disease on pea plants are linked on a supernumerary chromosome, and are therefore not present in all strains [49]. In these cases, lack of the matching genes will result in zero fitness on particular hosts. Genes involved in quantitative features of pathogenicity, such as those encoding secondary metabolites [50], are also frequently tightly linked in fungi.

Asexuality and selfing: reducing gene immigration

Many fungal pathogens can have multiple asexual cycles and most often only a single event of sexual reproduction per year. Ecological speciation should be facilitated by multiple cycles of asexual reproduction since these correspond to multiple cycles of selection for local adaptation without recombination breaking down locally advantageous allelic combinations and introducing locally deleterious immigrant alleles. The mixture of sexual and asexual reproduction present in most fungal pathogens allows both the creation of new genetic combinations and the rapid amplification by selection of those combinations that promote infection of a new host [51]. Selfing in some pathogenic fungi can also facilitate ecological speciation by reducing the probability of mating with pathogens adapted to other hosts [52].


We argue that host shift speciation is one of the main routes for emergence of new fungal diseases of plants and that some life-history traits of fungal plant pathogens can facilitate rapid ecological divergence. These are (1) strong disruptive selection imposed by the hosts, (2) large number of spores increasing both the possibility of survival on a new host and levels of adaptive variation created by mutation, (3) mating within host, creating pleiotropy between host adaptation and assortative mating, (4) a low number of genes underlying the specificity of host-pathogen interactions, and (5) frequent asexual reproduction with rare events of sexual recombination. These specificities of plant pathogenic fungi make them prone to rapid ecological speciation. These features have been neglected but require more focused attention, especially given their importance in emerging diseases on plants.

We have focused here on plant fungal pathogens, but the same arguments can be applied to many pathogens including nematodes, bacteria, and viruses. These indeed share many traits with plant pathogenic fungi that should render them prone to ecological speciation by host shifts; namely the production of numerous propagules, gene exchange occurring within hosts, linkage of traits experiencing selection and strong selection imposed by the hosts [53, 54].

In conclusion, we argue that existing models of ecological speciation are not appropriate for understanding how pathogens form new species on novel hosts, thereby causing emerging diseases, since they ignore the specificities of the life cycle of plant pathogens. We need to think differently about life history traits to tailor models based on specificities of pathogens if we are to fully understand emerging diseases. Linking ecological speciation with emergent diseases can also have practical implications (Box 4). In case studies, we should focus on the mechanisms allowing rapid adaptation and those preventing gene flow between emerging populations and their population-of-origin.

BOX 4Some implications of linking emerging diseases and ecological speciation

Recognizing that emerging diseases caused by fungal plant pathogens often result from host shift speciation, and that several characteristics of fungal plant pathogens render them conducive to this kind of ecological speciation, will improve our knowledge of the mechanisms responsible for both disease emergence and the biodiversity of fungi. In addition, this recognition has several important consequences for our understanding of disease dynamics and evolution as well as for designing more efficient and sustainable control programs.

First, if host adaptation alone can be sufficient for speciation, then intersterility, one of the most commonly applied criteria for delimiting species, will not be an appropriate criterion. Interfertility can be retained long after gene flow has ceased between plant pathogens species when the sole reproductive barrier is host adaptation. Failure to recognize that host adaptation can be an efficient barrier facilitating speciation by host shifts means that two distinct pathogenic species could be considered as one. This can lead to the development of control measures that overlook the specificities of each species, such as specific fungicide resistance [62]. Accurate species delimitation and identification is also fundamental for making sound quarantine decisions and policies, and for the implementation of strategies specifically designed to target the right taxa. Furthermore, since host adaptation alone can allow for very rapid speciation via host shift, rapidly evolving markers would be needed to delimit the species [63].

Second, it is important to link emerging diseases with ecological speciation to assess whether new diseases are due to spillover, to host range expansion, or to host shift speciation. These different scenarios will impact the control measures to be taken, e.g. whether one or multiple hosts should be targeted by fungicides. Also, the dynamics and evolution of the disease will be different if the pathogen is adapted to a single versus multiple hosts [64] or if the newly attacked host is only a reservoir for the second host.

Finally, if host adaptation is sufficient for speciation onto a new host via host shifts, then disease emergence can be relatively rapid, something which should be taken into account in theoretical models aiming to understand and predict disease emergence. Furthermore, models of the evolution of fungal pathogens characteristics, e.g. virulence, should take into account the specificities of the pathogen life cycle. For instance, the software package Quantinemo [65] in which no dispersal between selection and mating is allowed, can be used for modeling pathogens mating within their hosts. Also, the life cycle should be considered when predicting which pathogens are the most serious threats: plant fungal pathogens mating within their hosts are expected to cause disease emergence on novel hosts more easily. This feature should be accounted for in quarantine policies, control design and plant breeding programs.


We thank M. Fisher, J. Shykoff and the Emerfundis group for discussions, S. Sadedine, M.E. Hood, and anonymous reviewers for comments on the manuscript. Supported by grants ANR-06-BLAN-0201 and ANR-07-BDIV-003, NIH-GM56693, and Invited Professorship (University Paris-Sud).


Assortative mating
positive assortative mating takes place when sexually reproducing organisms mate with individuals that are similar to themselves in a given trait. This can be achieved by mate choice, but also by habitat choice (e.g., for phytophagous insects mating on their host plants), or by pleiotropy between genes controlling adaptation and mating.
Ecological speciation
a process by which barriers to gene flow between populations evolve as a result of ecologically-based divergent selection [9]. Examples include speciation of fish in a crater lake [66], palms on an oceanic island [67], and hybrid speciation of butterflies in jungles [68]. In alternative models of speciation, the evolution of reproductive isolation involves processes other than ecologically based divergent selection (e.g., polyploidization, genetic drift, population bottlenecks and sexual selection).
Emerging disease
recent disease on a new host and/or a new area, or having recently increased in virulence.
Gene-for-Gene (GFG) model
this assumes that infection ability is controlled by a single, diallelic locus in the fungal pathogen and a single, diallelic locus in the plant. Plant immunity can only occur when the plant carries the resistant allele and the pathogen carries the “avirulent” allele. All three other allelic combinations lead to infection, either because the plant does not carry the resistant allele allowing it to recognize effectors produced by the pathogen and to induce a defense reaction [44], or because the pathogen carries the “virulent” allele, that escapes host recognition. Pathogen alleles are called “virulent” and “avirulent” because “virulence” refers to the qualitative ability to infect a host genotype in plant pathology.
Host range expansion
evolution of the ability to exploit a novel host in addition to the host of origin.
Host shift speciation
speciation by specialization onto a novel host. A subset of the fungal population speciates on a new host, becoming incapable of infecting its host of origin, with cessation of gene flow from the population parasitizing the host of origin.
Obligate biotroph
pathogen using a living host for obtaining resources.
Formation of a new species, in sexual organisms by cessation of gene flow in most parts of the genome.
Until quite recently, the most commonly used species criterion for fungi has been morphology. However, many cryptic species have been discovered within morphological species, using the intersterility [69], or the GCPSR (Genealogical Concordance Phylogenetic Species Recognition) [70]. This latter species criterion uses the phylogenetic concordance of multiple unlinked genes to indicate a lack of genetic exchange and thus evolutionary independence of lineages. The GCPSR criterion has proved immensely useful in fungi, because it is more finely discriminating than the other criteria in many cases, or more convenient (e.g. for species that we are not able to cross) [18]. We consider here exclusively fungi able to undergo sexual reproduction, at least occasionally.
cross-species disease transmission without the establishment of a self sustainable population onto the new host


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Kolar CS, Lodge DM. Progress in invasion biology: predicting invaders. Trends Ecol. Evol. 2001;16:199–204. [PubMed]
2. Tatem AJ, et al. Global transport networks and infectious disease spread. Adv. Parasitol. 2006;62:293–343. [PMC free article] [PubMed]
3. Daszak P, et al. Emerging infectious diseases of wildlife-- Threats to biodiversity and human health. Science. 2000;287:443–449. [PubMed]
4. Cleaveland S, et al. Diseases of humans and their domestic mammals: pathogen characteristics, host range and the risk of emergence. Phil. Trans. R. Soc. Lond. B. 2001;356:991–999. [PMC free article] [PubMed]
5. Anderson PK, et al. Emerging infectious diseases of plants: pathogen pollution, climate change and agrotechnology drivers. Trends Ecol. Evol. 2004;19:535–544. [PubMed]
6. Desprez-Loustau MR, C, et al. The fungal dimension of biological invasions. Trends Ecol. Evol. 2007;22:472–480. [PubMed]
7. Stukenbrock EH, McDonald BA. The origins of plant pathogens in agro-ecosystems. Ann. Rev. Phytopathol. 2008;46:75–100. [PubMed]
8. Norton DA, Carpenter MA. Mistletoes as parasites: host specificity and speciation. Trends Ecol. Evol. 1998;13:101–105. [PubMed]
9. Rundle HD, Nosil P. Ecological speciation. Ecol. Lett. 2005;8:336–352.
10. Anderson RM, May RM. Infectious Diseases of Humans: Dynamics and Control. Oxford Univ. Press; 1991.
11. Keeling MJ, Rohani P. Modeling Infectious Diseases in Humans and Animals. Princeton Univ. Press; 2008.
12. Strange R, Scott P. Plant Disease: A threat to global food security. Ann. Rev. Phytopath. 2005;43:83–116. [PubMed]
13. Birch PRJ, Whisson SC. Phytophthora infestans enters the genomics era. Mol. Plant Pathol. 2001;2:257–263. [PubMed]
14. Urashima AS, et al. Host range, mating-type, and fertility of Pyricularia grisea from wheat in Brazil. Plant Dis. 1993;77:1211–1216.
15. Singh RP, et al. Will stem rust destroy the world’s wheat crop? Adv. Agronomy. 2008;98:271–309.
16. Fisher MC, et al. Global emergence of Batrachochytrium dendrobatidis and amphibian chytridiomycosis in space, time, and host. Annu. Rev. Microbiol. 2009;63:291–310. [PubMed]
17. de Vienne D, et al. Phylogenetic determinants of potential host shifts in fungal pathogens. J. Evol. Biol. 2009;22:2532–2541. [PubMed]
18. Giraud T, et al. Speciation in fungi. Fung. Genet. Biol. 2008;45:791–802. [PubMed]
19. Gavrilets S. Fitness landscapes and the origin of species. Princeton University Press; 2004.
20. Tellier A, et al. Host-Pathogen Interactions: Genetics, Immunology and Physiology. Nova Science Publishers; 2010. Theory and examples of host-parasite reciprocal influence: from short-term to long term interactions, from coevolution to cospeciation and host shifts.
21. Le Gac M, et al. Phylogenetic evidence of host-specific cryptic species in the anther smut fungus. Evolution. 2007;61:15–26. [PubMed]
22. Refrégier G, et al. Cophylogeny of the anther smut fungi and their Caryophyllaceous hosts: Prevalence of host shifts and importance of delimiting parasite species. BMC Evol. Biol. 2008;8:100. [PMC free article] [PubMed]
23. Lopez-Villavicencio M, et al. The anther smut disease on Gypsophila repens: a case of parasite sub-optimal performance following a recent host shift? J. Evol. Biol. 2005;18:1293–1303. [PubMed]
24. Antonovics J, et al. The ecology and genetics of a host-shift: Microbotryum as a model system. Am. Nat. 2002;160:S40–S53. [PubMed]
25. Slippers B, et al. Emerging pathogens: fungal host jumps following anthropogenic introduction. Trends Ecol. Evol. 2005;20:420–421. [PubMed]
26. Parker IM, Gilbert GS. The evolutionary ecology of novel plant-pathogen interactions. Annu. Rev. Ecol. Evol. Syst. 2004;35:675–700.
27. Milgroom MG, et al. Intercontinental population structure of the chestnut blight fungus, Cryphonectria parasitica. Mycologia. 1996;88:179–190.
28. Anagnostakis S. The effect of multiple importations of pets and pathogens on a native tree. Biol. Invasions. 2001;3:245–254.
29. Stukenbrock EH, et al. Origin and domestication of the fungal wheat pathogen Mycosphaerella graminicola via sympatric speciation. Mol. Biol. Evol. 2007;24:398–411. [PubMed]
30. Zaffarano PL, et al. Rapid speciation following recent host shifts in the plant pathogenic fungus Rhynchosporium. Evolution. 2008;62:1418–1436. [PubMed]
31. Gavrilets S, Vose A. Case studies and mathematical models of ecological speciation. 2. Palms on an oceanic island. Mol. Ecol. 2007;16:2910–2921. [PubMed]
32. Gavrilets S, et al. Case studies and mathematical models of ecological speciation. 1. Cichlids in a crater lake. Mol. Ecol. 2007;16:2893–2909. [PubMed]
33. Rundell RJ, Price TD. Adaptive radiation, nonadaptive radiation, ecological speciation and nonecological speciation. Trends Ecol. Evol. 2009;24:394–399. [PubMed]
34. Nosil P, et al. Perspective: reproductive isolation caused by natural selection against immigrants from divergent habitats. Evolution. 2005;59:705–719. [PubMed]
35. Giraud T. Selection against migrant pathogens: The immigrant inviability barrier in pathogens. Heredity. 2006;97:316–318. [PubMed]
36. Andrivon D, et al. Adaptation of Phytophthora infestans to partial resistance in potato: evidence from French and Moroccan populations. Phytopathology. 2007;97:338–343. [PubMed]
37. Alexopoulos CJ, et al. Introductory Mycology. John Wiley & Sons, Inc; 1996.
38. Enjalbert J, et al. Genetic evidence of local adaptation of wheat yellow rust (Puccinia striiformis f. sp tritici) within France. Mol. Ecol. 2005;14:2065–2073. [PubMed]
39. Johnson PA, et al. Conditions for sympatric speciation: a diploid model incorporating habitat fidelity and non-habitat assortative mating. Evol. Ecol. 1996;10:187–205.
40. Giraud T, et al. Importance of the life cycle in host race formation and sympatric speciation in parasites. Phytopathology. 2006;96:280–287. [PubMed]
41. Bush GL. Sympatric speciation in animals: new wine in old bottles. Trends Ecol. Evol. 1994;9:285–288. [PubMed]
42. Rice WR. Disruptive selection on habitat preference and the evolution of reproductive isolation: a simulation study. Evolution. 1984;38:1251–1260.
43. Gavrilets S, Losos JB. Adaptive radiation: contrasting theory with data. Science. 2009;323:732–737. [PubMed]
44. Bent AF, Mackey D. Elicitors, effectors, and R genes: The new paradigm and a lifetime supply of questions. Annu. Rev. Phytopathol. 2007;45:399–436. [PubMed]
45. Kamoun S. Nonhost resistance to Phytophthora: novel prospects for a classical problem. Curr. Opin. Plant Biol. 2001;4:295–300. [PubMed]
46. Atienza SG, et al. Accumulation of genes for susceptibility to rust fungi for which barley is nearly a nonhost results in two barley lines with extreme multiple susceptibility. Planta. 2004;220:71–79. [PubMed]
47. Jafary H, et al. Innate nonhost immunity in barley to different heterologous rust fungi is controlled by sets of resistance genes with different and overlapping specificities. Mol. Plant-Microb. Int. 2006;19:1270–1279. [PubMed]
48. Hatta R, et al. A conditionally dispensable chromosome controls host-specific pathogenicity in the fungal plant pathogen Alternaria alternata. Genetics. 2002;161:59–70. [PMC free article] [PubMed]
49. Han YN, et al. Genes determining pathogenicity to pea are clustered on a supernumerary chromosome in the fungal plant pathogen Nectria haematococca. Plant J. 2001;25:305–314. [PubMed]
50. Aguileta G, et al. Genome evolution in pathogenic and symbiotic fungi. Adv. Bot. Res. 2009;49:151–193.
51. McDonald BA, Linde C. Pathogen population genetics, evolutionary potential, and durable resistance. Annu. Rev. Phytopathol. 2002;40:349–379. [PubMed]
52. Giraud T, et al. The mating system of the anther smut fungus, Microbotryum violaceum. Euk. Cell. 2008;7:765–775. [PMC free article] [PubMed]
53. Duffy S, et al. Evolution of host specificity drives reproductive isolation among RNA viruses. Evolution. 2007;61:2614–2622. [PubMed]
54. Dobrindt U, et al. Genomic islands in pathogenic and environmental microorganisms. Nat. Rev. Microbiol. 2004;2:414–424. [PubMed]
55. Le Gac M, Giraud T. Existence of a pattern of reproductive character displacement in Basidiomycota but not in Ascomycota. J. Evol. Biol. 2008;21:761–772. [PubMed]
56. Peever T. Role of host specificity in the speciation of Ascochyta pathogens of cool season food legumes. Eur. J. Plant Pathol. 2007;119:119–126.
57. Guérin F, et al. Origin and colonization history of newly virulent strains of the phytopathogenic fungus Venturia inaequalis. Fung. Genet. Biol. 2007;44:284–292. [PubMed]
58. Guérin F, Le Cam B. Breakdown of the scab resistance gene Vf in apple leads to a founder effect in populations of the fungal pathogen Venturia inequalis. Phytopathology. 2004;94:364–369. [PubMed]
59. Le Cam B, et al. Evidence of two formae speciales in Venturia inaequalis, responsible for apple and pyracantha scab. Phytopathology. 2002;92:314–320. [PubMed]
60. Gladieux P, et al. Host-specific differentiation among populations of Venturia inaequalis causing scab on apple, pyracantha and loquat. Fung. Genet. Biol. 2010 in press doi:10.1016/j.fgb.2009.12.007. [PubMed]
61. Benaouf G, Parisi L. Genetics of host-pathogen relationships between Venturia inaequalis races 6 and 7 and Malus species. Phytopathology. 2000;90:236–242. [PubMed]
62. Giraud T, et al. RFLP markers show genetic recombination in Botrytis cinerea and transposable elements reveal two sympatric species. Mol. Biol. Evol. 1997;14:1177–1185. [PubMed]
63. Petit R, Excoffier L. Gene flow and species delimitation. Trends Ecol. Evol. 2009;24:386–393. [PubMed]
64. Gandon S. Evolution of multihost parasites. Evolution. 2004;58:455–469. [PubMed]
65. Neuenschwander S, et al. QuantiNEMO: an individual-based program to simulate quantitative traits with explicit genetic architecture in a dynamic metapopulation. Bioinformatics. 2008;24:1552–1553. [PubMed]
66. Barluenga M, et al. Sympatric speciation in Nicaraguan crater lake cichlid fish. Nature. 2006;439:719–723. [PubMed]
67. Savolainen V, et al. Sympatric speciation in palms on an oceanic island. Nature. 2006;441:210–213. [PubMed]
68. Mavarez J, et al. Speciation by hybridization in Heliconius butterflies. Nature. 2006;441:868–871. [PubMed]
69. Anderson JB, Ullrich RC. Biological species of Armillaria mellea in North America. Mycologia. 1978;71:402–414.
70. Taylor JW, et al. Phylogenetics species recognition and species concepts in Fungi. Fung. Genet. Biol. 2000;31:21–32. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...