Evolutionary Interaction Networks of Insect Pathogenic Fungi

EN59CH24-Boomsma

ARI

ANNUAL REVIEWS

4 December 2013

16:22

Further

Annu. Rev. Entomol. 2014.59:467-485. Downloaded from www.annualreviews.org by Copenhagen University on 01/08/14. For personal use only.

Click here for quick links to Annual Reviews content online, including: • Other articles in this volume • Top cited articles • Top downloaded articles • Our comprehensive search

Evolutionary Interaction Networks of Insect Pathogenic Fungi Jacobus J. Boomsma,1,∗ Annette B. Jensen,2 Nicolai V. Meyling,2 and Jørgen Eilenberg2 1 Centre for Social Evolution, Department of Biology, University of Copenhagen, 2100 Copenhagen, Denmark; email: [email protected] 2 Centre for Social Evolution, Department of Plant and Environmental Sciences, University of Copenhagen, 1871 Frederiksberg, Denmark; email: [email protected]

Annu. Rev. Entomol. 2014. 59:467–85

Keywords

First published online as a Review in Advance on October 23, 2013

life histories, specialization, host shifts, toxins, Onygenales, Hypocreales, Entomophthorales, Neozygitales

The Annual Review of Entomology is online at ento.annualreviews.org This article’s doi: 10.1146/annurev-ento-011613-162054 c 2014 by Annual Reviews. Copyright All rights reserved ∗

Corresponding author

Abstract Lineages of insect pathogenic fungi are concentrated in three major clades: Hypocreales (several genera), Entomophthoromycota (orders Entomophthorales and Neozygitales), and Onygenales (genus Ascosphaera). Our review focuses on aspects of the evolutionary biology of these fungi that have remained underemphasized in previous reviews. To ensure integration with the better-known domains of insect pathology research, we followed a conceptual framework formulated by Tinbergen, asking complementary questions on mechanism, ontogeny, phylogeny, and adaptation. We aim to provide an introduction to the merits of evolutionary approaches for readers with a background in invertebrate pathology research and to make the insect pathogenic fungi more accessible as model systems for evolutionary biologists. We identify a number of questions in which fundamental research can offer novel insights into the evolutionary forces that have shaped host specialization and life-history traits such as spore number and size, somatic growth rate, toxin production, and interactions with host immune systems.

467

EN59CH24-Boomsma

ARI

4 December 2013

16:22

FOUR COMPLEMENTARY DOMAINS OF QUESTIONS ABOUT INSECT PATHOGENIC FUNGI Life history: the timing of major developmental and reproductive transitions in an organism’s life as shaped by natural selection


Pathogen: a (usually microbial) agent that causes disease Virulence: a quantitative measure of the severity of a disease, for example, in terms of mortality or reduced reproductive success Specialization: adaptation of pathogens or parasites to a narrow taxonomic diversity of hosts Prevalence: the number or proportion of infected hosts at a given point in time Pathogenicity: a qualitative measure of the ability of a pathogen or parasite to cause disease in a host

468

The abundance of fungi as mycorrhizae, phytopathogens, and endophytes in plants (11) suggests that modular growth forms with indeterminate iteration of semi-independent units are particularly amenable for evolving interactions with symbionts that grow in similar ways. Unitary hosts with determinate growth toward adulthood are more challenging. They are rather ephemeral moving targets to infect and, for parasitic symbionts, more laborious to exploit as they usually need to be killed before spores can be produced. In association with arthropod hosts, some fungi have overcome these transmission constraints by internalization as gut symbionts (e.g., yeasts), but this often imposed selection for mutualistic rather than parasitic life histories (74). Most of the known pathogenic fungi that managed to adapt to arthropods realize infections via penetration of the cuticle (80) and less commonly through the gut (73). Only a limited number of fungal lineages pathogenic to insects and arachnids have been relatively well studied, and our recent understanding of their classification, general biology, ecology, influence on host behavior, and usefulness as biocontrol agents has been summarized in several comprehensive reviews (14, 67, 82). This review focuses on the evolutionary adaptations that characterize the major lineages of insect pathogenic fungi, including both convergent traits that have evolved repeatedly and idiosyncratic characteristics that are lineage specific. To our knowledge, such an explicitly functional approach has not been pursued previously, such that evolutionary insights have often appeared scattered across discussion paragraphs of primary research papers, with the consequence that they remained underemphasized in earlier reviews. Another reason for choosing an explicitly evolutionary focus is that several recent studies have indicated that insect pathogenic, plant endosymbiotic, and plant saprotrophic life histories are less distinct than previously assumed, with at least some lineages expressing several complementary phenotypes. For example, Beauveria species that were thought to be exclusive insect pathogens are also known to be plant endophytes (80), and the genus Ascosphaera encompasses a range of life histories from virulent bee pathogens, via low-virulence bee commensals, to saprotrophs feeding on feces or pollen provisions of bees (86). In what follows, we therefore omit the usual justifications for studying insect pathogenic fungi, i.e., their agroecosystem importance as biological control agents (14) or bee pathogens (44). Instead, we concentrate on life-history trade-offs that may explain host specialization, the prevalence of sexuality in life cycles, the selection pressures that may constrain pathogenicity, and the rationale of the same gene pool of some of these fungi interacting with both insects and plants. We are aware that such evolutionary considerations have elements of informed speculation and often lack the matter-of-fact validity that research on physiological interaction mechanisms can offer. However, we believe that these evolutionary evaluations have merit because they ask why specific traits appear logical as products of natural selection, i.e., as adaptations that appear to have been designed for maximizing reproductive success (4, 22). Evolutionary why questions complement mechanistic what and how questions, and an active interaction between these two approaches has greatly benefitted the fields of evolutionary ecology (20) and evolutionary developmental biology (15). To maintain complementarity with the more familiar directions in insect pathology, we have adopted a modified version of Nobel Laureate (Physiology or Medicine 1973) Nico Tinbergen’s (79) scheme (Figure 1) for seeking four complementary types of explanations. This approach laid the foundation for modern studies in animal behavior but was later highlighted as a quartet of approaches required in its entirety to fully explain any biological phenomenon (54, 59). The overview in Figure 1 uses terminology that is somewhat different from previous versions to make it explicitly relevant for insect pathology while retaining the conceptual generality that was originally intended (54, 59, 79). Figure 1a begins with questions of mechanism that concern the

Boomsma et al.

EN59CH24-Boomsma

ARI

4 December 2013

16:22

Snapshots in time Mechanism (Figure 2) Structural traits and mechanistic processes in a lifetime or life cycle


a

What morphological structures and physiological mechanisms make single phenotypes functional at any particular point in time? Adaptation (Figure 5)

Current adaptations and the origin of lineages over evolutionary time

Sequential developments

b

Ontogeny (Figure 3)

What changes happen during growth and maturation and how are sequential phenotypes of individuals linked to (epi)genetic processes of gene expression?

d

Why has natural selection produced particular syndromes of traits? Can we understand how complexes of phenotypic traits may represent adaptations?

c

Phylogeny (Figure 4)

How do extant lineages relate to each other and to the fossil record? Can increases (speciation, radiation) and losses of lineage diversity be understood?

Figure 1 Schematic overview of the four complementary domains of biological questions to be asked to fully understand both proximate and ultimate causation of biological phenomena, adjusted from previous versions of the same concept (54, 59, 79). Each of the four domains (a–d ) is explained in more detail (Figures 2–5).

structural traits and physiological processes familiar to most biologists. These questions address what happens in a biological process or interaction: what tissues or other structures are involved, what molecular mechanisms apply, and how metabolic processes can be understood in terms of physiology or cell biology. In other words, what is it that makes single interacting phenotypes of fungal pathogens and insect hosts functional at any particular point in time? Figure 1b extends our focus to entire life cycles and thus includes a series of subsequent phenotypes of host and pathogen that partly interact but also include stages in which both maintain themselves independently. In the ontogeny domain, research typically continues to ask how cellular and molecular processes work, but with more explicit emphasis on how individual development requires different solutions during growth, maturation, and dispersal that should somehow all be achievable via (epi)genetic processes of differential gene expression. Separating mechanism and ontogeny into separate domains remains somewhat arbitrary when considering life cycles rather than single animal life spans as Tinbergen (79) did. Therefore, our section on mechanism addresses interactions between insect hosts and fungal pathogens at the single organism level and the ontogeny section reviews mostly the non-insect life cycle phases, including spore production and dispersal. Together, the mechanism and ontogeny domains of biological inquiry ask proximate questions. Intricate knowledge of physiology, cell biology, and, increasingly, molecular biology are crucial elements that allow researchers in these fields to obtain observational answers and design key experiments. Panels c and d of Figure 1 address evolutionary questions and thus consider orders-ofmagnitude-longer timescales than single phenotypes or life cycles. Figure 1c addresses questions related to phylogeny, a perspective that became generally accessible for fungi and insects only after gene sequencing allowed their trees of life to be reconstructed. Initial questions are often also of the what and how type; i.e., what lineages are monophyletic, in what order did they branch off, and how quickly did they become reproductively isolated? However, now that phylogenies become increasingly detailed and validated by fossil evidence (31, 45), researchers can also map traits of extant species on trees to ask why certain lineages became more diverse than others; how that might relate to their habitats or niches; and why certain evolutionary innovations evolved only once, whereas others originated and were lost repeatedly. The statistical methodology of this so-called comparative method was developed in the 1980s and www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

469

ARI

4 December 2013

16:22

1990s (36) and has now reached a remarkable level of sophistication and statistical power (e.g., http://mesquiteproject.org/mesquite/mesquite.html). As extant species are all descendants of ancestors that somehow managed to avoid extinction, the modern phylogeny-based comparative method is in fact a reconstruction of adaptive states that arose and vanished over evolutionary time. Figure 1d addresses adaptations of extant phenotypes that can be investigated under field and laboratory conditions. These approaches are least familiar to many entomologists and mycologists. They mostly generate testable hypotheses on why combinations of individual or social traits of organisms make sense as products of natural selection and whether they do so in most or only some extant populations (78). Together, the phylogeny and adaptation domains are often referred to as the ultimate domain of biological inquiry, encompassing both long-term and short-term evolutionary questions that need to be answered to complement the proximate explanations of the mechanism and ontogeny domains. Now that we have come full circle in Figure 1, it is clear we have tracked the general history of biology, as mechanistic questions have the longest history of biological inquiry, developmental biology (embryology) arose in the eighteenth century, evolutionary thinking arose in the nineteenth century, and the gene’s eye view of adaptation arose in the second half of the twentieth century. The objectives of the present review are to (a) offer a summary of the merits of considering evolutionary angles to the field of insect pathology, which has traditionally focused mostly on the proximate domains of inquiry, and (b) offer evolutionary biologists an accessible introduction to the insect pathogenic fungi that provide many suitable model systems for testing hypotheses of host specificity, life-history trade-offs, pathogenicity, virulence, and, possibly, (co)adaptation. Throughout this review we focus on the three major lineages of insect pathogenic fungi that have been sufficiently studied to allow at least tentative generalizations: Ascosphaera from the order Onygenales (phylum Ascomycota), genera from the order Hypocreales (phylum Ascomycota), and genera from the orders Entomophthorales and Neozygitales (phylum Entomophthoromycota). To maximize clarity we use specific primary colors for these three lineages when comparing their biology in Figures 2–4. Here, entomophthoralean refers to both Entomophthorales and Neozygitales.


EN59CH24-Boomsma

MECHANISM DOMAIN: HOW TO ATTACH, PROLIFERATE, AND REPRODUCE Insect pathogenic fungi have to meet several host challenges to produce enough new infectious spores in every generation to maintain viable populations. First, successful transmission often requires the release of massive spore numbers and/or sticky spore surfaces or substances that maximize adhesion in other ways (17, 80). Second, spores should germinate and initiate penetration of the solid insect exoskeleton relatively quickly (17) or survive digestion after oral uptake (Ascosphaera) (52). Third, the fungal cells must proliferate inside the hemocoel, muscles, or other tissues of the host body to collapse the host’s immune system so that the host dies shortly after. Fourth, the fungal pathogen should manage the host cadaver to optimize spore production and dispersal under prevailing environmental conditions (67). Meeting these sequential challenges requires specific syndromes of structural traits and physiological mechanisms that have indeed evolved in interestingly convergent ways across the three major lineages of insect pathogenic fungi (Figure 2). Inoculation occurs by direct contact between infectious cadavers and susceptible hosts or indirectly via airborne spores or spores deposited on vegetation or soil particles (37). Laboratory infection experiments show that some minimum number of spores is generally needed to achieve predictable infections (43). However, these results are not necessarily representative of field 470

Boomsma et al.


EN59CH24-Boomsma

ARI

4 December 2013

16:22

a

b

c

INFECTION

GROWTH

REPRODUCTION

Pathogenic fungi

Entomophthoromycota (Entomophthorales, Neozygitales) Ascomycota (Hypocreales) Ascomycota (Onygenales, Ascosphaera)

Figure 2 Diagrams illustrating the ways in which pathogenic fungi infect arthropod hosts by asexual or sexual spores, proliferate, and disperse. Hosts are represented by outlines and a gut with a dotted line indicating that the hindgut of bee larvae is closed. Sizes of colored symbols are approximately proportional to mass but are drawn much larger than they are relative to host size. (a) Infection: Entomophthoralean fungi infect mainly by large, sticky conidia penetrating the cuticle directly, and hypocrealean fungi infect by small conidia that produce appressorial structures. Ascosphaera spores are also small, enter orally, and infect through the gut epithelium. (b) Growth: Most Hypocreales and Onygenales proliferate through hyphal growth, and most entomophthoralean fungi proliferate through protoplasts without cell walls. (c) Reproduction: Asexual conidia of entomophthoralean fungi and sexual ascospores of the Hypocreales are primarily forcibly discharged from the surface of cadavers (upward arrows at right), whereas (very large and thick-walled) sexual resting spores of entomophthoralean fungi and asexual conidia of Hypocreales are passively released (upward and downward arrows at left). Ascosphaera produces only sexual spores that are passively released (downward arrow at right).

conditions (see below), where successful infections must normally be achieved by lower-dose inoculations that kill rather few host individuals but still produce enough new spores to maintain viable pathogen populations and in some cases even initiate epizootics (37). Social insect hosts offer even more significant challenges to pathogenic fungi because of allogrooming and the ensuing social immunity of colonies (19). Adhesion of small asexual hypocrealean conidia (e.g., genera Beauveria and Metarhizium) is often mediated by specific proteins and appressorial structures (Figure 2a) (17). The spores initiate germ tube growth on the cuticular surface to find suitable sites for penetration, for example, next to setae. This infection route has been documented many times (17, 23, 70), but the interesting details that differ between fungal lineages have seldom been emphasized. Conidia of Entomophthorales are large and typically attach themselves firmly to the cuticle via extra cell wall layers or mucus that assists the adhesion process (23). After adhesion, representatives of both orders face similar challenges to penetrate the multiple cuticular layers of the exoskeleton (Figure 2a) by some combination of physical force and enzymatic degradation. Whereas a single or a few entomophthoralean conidia are assumed to be sufficient for achieving infection (35), hypocrealean species likely need a larger number of conidia (43). In contrast, species of Ascosphaera infect their hosts (larvae of honey bees Apis mellifera or solitary bees such as Megachile rotundata) only after spores are ingested. Because the larval gut of bees remains closed until pupation (Figure 2), spores remain in the gut, germinate, and penetrate the epithelium to enter the hemocoel (16). Details of this process remain largely unknown as gut penetration is much harder to study microscopically than cuticular penetration. www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

Conidia: asexual nonmotile spores of fungi

471

EN59CH24-Boomsma

ARI

Toxin: a poisonous substance produced by living cells or organisms


Epitope: a section of a molecule (often an antigen) to which an antibody binds so the molecule can be recognized by the immune system Teleomorph: the sexual reproductive stage of a fungus, typically a fruiting body Resistance: a host’s ability to defend itself against cells, organic particles, or toxins produced by pathogens

472

4 December 2013

16:22

Once inside the insect body the fungus must overwhelm the immune system of the host while continuing to take up nutrients for further proliferation. To do this, representatives of Onygenales and Hypocreales grow as septated hyphae (Figure 2b) that can access the interstitial space between muscle fibers (70). Some hypocrealean species also produce toxins during this part of the infection process to help sedate or kill host cells and thus facilitate fungal resource acquisition (80). However, a delicate trade-off balance between fungal growth and toxin production may exist, because high levels of toxin production may lead to host death before the fungus has accumulated sufficient cellular mass to develop all the structures necessary for ultimate spore production and dispersal (see below). In contrast, many entomophthoralean fungi proliferate by yeast-like protoplasts, i.e., single-cell structures without a cell wall (12, 21) (Figure 2b). The advantages of this cell form are that nutrient acquisition rates are probably higher and that fungal cells can multiply in the hemocoel without being detected by the insect immune system that uses cell wall epitopes as cues (21). However, producing infective conidia requires that clonal protoplasts switch to producing cell walls and hyphal bodies during the final stages of infection (13). When that happens, the host immune system is quickly overwhelmed by the massive presence of pathogen biomass. Entomophthoralean fungi have thus evolved a strategy to weaken host immune responses very different from that used by many hypocrealean fungi (see below). The time from infection to becoming infectious (incubation period) typically varies from 3 to 20 days, depending on host body mass, fungal species, spore production dynamics, and environmental conditions (35). A few Entomophthorales (e.g., the genus Strongwellsea; 41) release infective spores from weakened but still living hosts, but hosts are otherwise usually killed before sporulation. This suggests that comparative studies on the timing of fungal growth and spore production relative to host immune collapse may be worthwhile (see below). Asexually produced conidia can be either passively released or actively discharged (42). Also in this respect we find interesting differences between entomophthoralean and hypocrealean fungi that appear adaptive when combined with other traits (Figure 2c). Entomophthoralean fungi produce a moderate number of medium to large (15–40 μm) (42) conidia per host, often after they have manipulated hosts into climbing plant structures so conidia can be dispersed from an elevated location (67). With few exceptions, these conidia are actively discharged by mechanisms that likely evolved to directly or indirectly increase inoculation success (5, 23). In contrast, hypocrealean conidia (e.g., Isaria, Lecanicillium, Metarhizium, Beauveria) are small (40 μm) globular spores (42) with very thick cell walls that are resistant to harsh environmental conditions (33) (Figure 2c). These spores are passively released from the dissolving insect cadaver and end up on soil or vegetation surfaces (33). This observation suggests that they can function as winter survival devices and seems to confirm that size/number trade-offs in spore production may also apply across entomophthoralean and hypocrealean lineages.

Anamorph: the asexual reproductive stage of a fungus


ONTOGENY AND LIFE CYCLES Hypocrealean insect pathogens have complex life cycles that include stages outside their hosts (Figure 3a), as both asexual conidia (produced by anamorphic stages) and sexual ascospores (produced by teleomorphic stages) can persist for prolonged periods in the soil or on plants with which the insect hosts interact. Genetically identical but phenotypically different, anamorphs and teleomorphs have increasingly been associated with each other via DNA sequencing (75). These findings have revealed impressive amounts of stage-specific phenotypic plasticity in morphology and life-history syndromes, as it is now clear that Beauveria and Lecanicillium anamorphs belong to the same lineage as Cordyceps teleomorphs, that Metarhizium anamorphs are concordant with Metacordyceps teleomorphs, and that Hirsutella anamorphs correspond to Ophiocordyceps teleomorphs (80). Anamorphic states are characterized by mass production of conidia, usually in single bursts of semelparous reproduction a few days after host death. These conidia can survive in the environment for prolonged periods depending on abiotic and biotic conditions. Considerable environmental reservoirs of hypocrealean propagules are therefore found both below- and aboveground (56, 57) and new hosts can become infected directly from such inocula (37) (Figure 3a). The anamorphic stages of these fungi, including the well-known genera Beauveria and Metarhizium, are often generalists infecting a wide range of host taxa. In contrast to conidia, which are released shortly after host death, the sexual ascospores are produced and released by the same host individual over several months from a slowly developing stroma (3, 58). The host ranges of these teleomorphs appear to be generally much narrower than those of the corresponding anamorphs, and some appear to be outright host specialists (49). Many temperate ecosystems apparently lack the mating type diversity to allow sexual reproduction of insect pathogenic Hypocreales in any appreciable frequency. Although Cordyceps has a worldwide distribution, hotspots tend to be concentrated in tropical habitats in East Asia and the Americas (68, 75, 80). This could be due to anamorphs of both mating types being present in more even proportions, but direct studies connecting anamorph mating type distributions and the prevalence of teleomorphs have never been carried out. Such studies would seem worthwhile, as skewed mating type representation disproportionally decreases the likelihood of a successful sexual event after conidia coinfect the same host. Isolates of Beauveria bassiana can persist and possibly be propagated in aboveground live plant tissues, and the number of plant species on which this occurs is growing rapidly (55, 81) (Figure 3a). Other hypocrealean species have cryptic lives as asymptomatic plant endophytes (80), but it remains unclear whether their presence implies a net cost or benefit for host plants (18, 26). In cacao (Theobroma cacao), B. bassiana tends to become established in seedlings and may produce epiphytic conidia afterward (63). Negative effects (but no mortality) of feeding on plants with such endophytes have been shown for aphids (Aphis fabae, Acyrthosiphon pisum, 1; Aphis gossypii, 32) and locusts (Chortoicetes terminifera, 32) and plant damage by aphid infestations tends to be www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

473

EN59CH24-Boomsma

ARI

4 December 2013

16:22

a

Ascomycota (Hypocreales) 9


6

5 2

4

Asexual cycle

Sexual cycle

1

3

8

7

10

b

Entomophthoromycota (Entomophthorales, Neozygitales)

Asexual cycle

Sexual cycle

c

Ascomycota (Onygenales, Ascosphaera)

Sexual cycle

Figure 3 Generalized life cycles for insect pathogenic fungi. (a) Ascomycota (Hypocreales): Following death of an infected host, () the fungus produces asexual conidia on the outside of the cadaver. () Conidia are released passively into the environment to infect new hosts. Occasionally, infected cadavers produce sexual fruiting bodies, stromata (), which produce asci-carrying ascospores after meiosis that are actively dispersed to infect new hosts (), presumably after conidiation. Conidia of some taxa can also infect live plant tissues () in which the fungus can become established and grow as an endophyte (). Conidia in the soil can colonize plant roots () followed by proliferation in the rhizosphere ( ), but it remains to be clearly documented that fungal propagules produced rhizospherically or endophytically can infect insects feeding on plants ( ,). Well-documented life cycle links are indicated by dark blue arrows, those recently documented by light blue arrows, and those remaining speculative by gray arrows. (b) Entomophthoromycota (Entomophthorales and Neozygitales): Recurrent infection cycles usually occur via the production of asexual conidia, but many species produce thick-walled resting spores sexually when conditions are unfavorable. (c) Ascomycota (Onygenales, Ascosphaera): Only sexually produced spores are known, so that dual infections with spores of different mating types are required for infection to result in spore production.

474

Boomsma et al.


EN59CH24-Boomsma

ARI

4 December 2013

16:22

generally reduced when insect pathogenic fungi are established as endophytes (1). However, reports of newly initiated infections from plants back into insects are still few. They concern, respectively, European corn borer, Ostrinia nubilalis, and tomato fruitworm, Helicoverpa zea, larvae feeding on maize (Zea mays) and tomato (Solanum lycopersicum) with B. bassiana endophytes (8, 64), but the mechanisms by which such reverse infections may take place remain unclear. Endophytic colonization of plant tissues by hypocrealean insect pathogenic fungi may occur both above- and belowground (81) (Figure 3a), adding further complexity to the life cycles of these fungi. Plants might benefit from the presence of these fungi (18, 26, 32) if they facilitate nutrient acquisition or conversion, but explicit data are lacking. Belowground Metarhizium spp. are rhizosphere competent in cabbage Brassica oleracea (39) and some perennial plants with which they associate (27, 87), whereas Metarhizium robertsii may penetrate root cells (69), so future work may well reveal specific adaptations for growing in association with plants. In particular, the intimate associations of some lineages with plant roots suggest some form of nutrient translocation from infected insect hosts in the soil to the benefit of plants (7). Such functions would resemble those of mycorrhizal associations, possibly including direct protection against root herbivory (Figure 3a). For species in the Entomophthoromycota, life cycles involve both asexual conidia and resting spores that are either sexually or asexually produced (5, 47) (Figure 3b). Complete life cycles including both spore types have been documented for several host-pathogen systems in temperate ecosystems (25, 34, 66). Conidia of entomophthoralean fungi are short-lived and must either infect a new host within hours or produce secondary conidia (35). The primary conidia of these fungi may therefore be mere dispersal structures, with secondary conidia being the most common infective units (29). The environmental transmission stage of Ascosphaera bee pathogens remains somewhat elusive (Figure 3c). Only sexual ascospores are known and they can either act immediately as infective units or stay dormant in the environment for years. A final group worth mentioning is the genus Coelemomyces (Chytridiomycota) (77), which has species whose life cycle shifts between a dipteran and a copepod host; each host is infected by a different type of spore.

PHYLOGENY DOMAIN: DIVERSIFICATION OF LIFE HISTORIES Insect pathogens have evolved in most of the major fungal lineages. There are only a few exceptions, such as the Glomeromycota, that contain only mycorrhizal clades (Figure 4a). Nutritional modes of the earliest fungi remain poorly understood, but the ability to infect insects is clearly represented in several basally diverging clades, e.g., Blastocladiomycota, Chytridiomycota, and Entomophthoromycota. Whereas molecular dating and fossil records have estimated that fungi evolved 0.5–1.5 bya, the fossil records of the Ascomycota and Basidiomycota only date back to the Devonian some 400 mya (6, 51). The earliest known fossil of an insect pathogenic fungus, a scale insect (Hemiptera) infected by an Ophiocordyceps-like fungus, is from Myanmar amber (100–110 mya) (76), and later fossils include a termite (Isoptera) infected by an Entomophthora-like fungus and an ant (Hymenoptera) infected by Beauveria, both from Dominican amber (20–30 mya) (61, 62). Figure 4a (38, 45, 84) summarizes present knowledge of the distribution of insect pathogenic species across the major fungal lineages, indicating both known species richness per clade and the approximate number of insect pathogenic species among them (higher proportions have darker shades of grey). The approximate number of described insect pathogenic species varies from just a few in the Chytridiomycota, Blastocladiomycota, Kickxellomycotina, and Basidiomycota to a substantial number in the Ascomycota and almost complete dominance in the Entomophthoromycota. The basal fungal lineages are still poorly resolved (83), but the presence of pathogenic lifestyles in several of them suggests that these associations evolved rapidly once www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

475

EN59CH24-Boomsma

ARI

4 December 2013

16:22

b Taphrinomycotina Saccharomycotina

Entomophthoromycota (Entomophthorales, Neozygitales) Ascomycota (Hypocreales)


Ascomycota (Onygenales, Ascosphaera)

a Microsporidia

Approximate number of species

Chytridiomycota

700

Neocallimastigomycota

20

Blastocladiomycota

180

Kickxellomycotina

270

Zoopagomycotina

190

Mucoromycotina

330

Entomophthoromycota

280

Glomeromycota

170

Ascomycota

64,000

Basidiomycota

30,000

Pezizomycotina Orbiliomycetes Pezizomycetes Dothideomycetes Arthoniomycetes Chaetothyriomycetidae Eurotiomycetes Eurotiales Coryneliales; Eurotiomycetidae Onygenales Mycocaliciomycetidae Laboulbeniomycetes Lichinomycetes Lecanoromycetes Leotiomycetes Calosphaeriales Sordariomycetes Lulworthiales Meliolales Phyllachorales Trichosphaeriales Xylariales Coronophorales Hypocreales Microascales Melanosporales Boliniales Chaetosphaeriales Diaporthales Ophiostomatales Sordariales

Figure 4 Phylogenetic positions of insect pathogenic fungi. (a) The fungal kingdom (38, 45, 84), highlighting phyla with insect pathogens. Figures represent the approximate number of described species within each (sub)phylum and shading intensities of boxes represent relative proportions of known insect pathogenic species (higher proportion when darker grey). Red, yellow, and blue dots mark branches with the most common and better-known insect pathogens (color scheme as in Figures 2 and 3). Four other branches have some insect pathogenic species, but few of them have been studied in detail: Chytridiomycota, Blastocladiomycota, Kickxellomycotina (includes orders previously placed in the Trichomycetes), and Basidiomycota. (b) Ascomycota that have both the highest number of described species and the highest diversity of life histories, including a variety of saprophytic, mutualistic, and pathogenic lifestyles. Bold lines represent orders with insect pathogenic fungi and colored dots mark those with the better-known insect pathogens: Hypocreales (yellow, e.g., Cordyceps sensu lato, Metarhizium, Beauveria) and Onygenales (blue, Ascosphaera).

insects became available as hosts. The phylogenetic positions of Ascomycota that harbor most of the known insect pathogens, the Hypocreales and the Onygenales, are detailed in Figure 4b, illustrating that insect pathogenic lifestyles evolved repeatedly in the Ascomycota (72). The extremely low frequency of insect pathogenic lineages in the Basidiomycota suggests that their adaptive radiation has been driven by ecological functionalities other than parasitizing insects, marking a fundamental difference with the Entomophthoromycota, in which almost all 280 known species are pathogens of arthropod hosts (5). Species-level specialization patterns appear to be mostly lineage specific, with the Entomophthoromycota (46) generally having rather narrow host ranges and Ascosphaera (Onygenales) 476

Boomsma et al.


EN59CH24-Boomsma

ARI

4 December 2013

16:22

being associated with typical bee nest environments (10). The sexual (teleomorphic) stages of the Hypocreales also appear to be host specialists, whereas their asexual (anamorphic) counterparts are generalists. This offers an interesting contrast to the obligatorily insect pathogenic entomophthoralean fungi, which tend to remain host specialists also in their asexual stages. However, some entomophthoralean species have a completely different nutritional mode as parasites of fern gametophytes (Completoria) or desmids, nematodes, and tardigrades (Meristacraceae and Ancylistaceae), whereas others become opportunistic vertebrate pathogens or saprotrophs (e.g., Conidiobolus coronatus), but whether these conditions are derived from an ancestral insect pathogenic lifestyle remains unresolved (5). Inclusion of these deviating entomophthoralean species in explicitly phylogenetic analyses will ultimately reveal which characters (e.g., protoplast growth and/or host manipulation for optimal spore dispersal) are basal and which are evolutionarily derived. If basal, such characters would suggest a long history of host-pathogen coevolution. Although the Hypocreales belong to the derived Sordariomycetes, several basal lineages of the ecologically diverse Pezizomycotina have also evolved insect pathogenic lineages (Figure 4b). These include some species in the Dothideomycetes (50) that are scale insect pathogens, and the orders Onygenales and Eurotiales, which belong to the Eurotiomycetes. The development of osmophilic traits (e.g., the ability to grow on high sugar content substrates) may have driven the need for habitat specificity (bee brood cells and later bee guts) in the genus Ascosphaera. Members of the Eurotiales mostly appear to have opportunistic life histories as facultative pathogens without much host specialization (71). The best-studied genera, Aspergillus and Penicillium, are notorious for their production of toxic secondary compounds, which may represent an explicit strategy for poisoning host tissues (53) for later saprophytic exploitation. The Eurotiales may thus have struck a different trade-off balance than the asexual Hypocreales, where moderate toxin production may allow more proficient hyphal growth in live host tissues (see below). The opportunistic lifestyle of Eurotiales species may be beneficial during coinfections; i.e., rapid host tissue exploitation may allow Aspergillus to outcompete Metarhizium (40). Substantial novel information about the key drivers of insect pathogenic life histories is expected to emerge from novel omics approaches that will allow insight into the structural and regulatory genes that have mediated lifestyle transitions over evolutionary time. These techniques will complement recent findings of multiple events of interkingdom host jumping in the Hypocreales between animals, fungi, and plants (see above) (60). Such explicit genetic underpinning will allow further testing of the hypothesis that these host shifts have been facilitated primarily by ecological proximity (the host habitat hypothesis) rather than relatedness among hosts (host relatedness hypothesis) (48). Another relevant example is M. anisopliae, which has now been divided into several species (9). The pathogen M. robertsii, which has a broad host range, has a higher number of pathogenicity genes than M. acridum, which has a more restricted host range (28). Another study has produced hints of transcriptomic evolutionary convergence by showing that insect pathogenic fungi may secrete a variety of trypsin-like proteases that are likely to be important for successful host colonization (30). Finally, transcriptomic analyses of B. bassiana grown on locust wings, cotton bollworm Helicoverpa armigera blood, and corn root exudates have revealed that adaptation to these different niches is mediated by well-defined gene sets (88).

Parasite: a (usually macroscopic) organism that relies on infecting a host to complete its life cycle, reducing host survival or reproductive success Host shift: colonization and possible successful reproduction of pathogens or parasites on a new host species or lineage

ADAPTATION: EVOLUTIONARY HYPOTHESES AND QUESTIONS WORTH ASKING A number of questions regarding evolution have already been mentioned, illustrating that hypotheses about adaptation emerge naturally when working with mechanistic, ontogenetic, or phylogenetic problems. Obtaining answers of this kind varies in feasibility, and some of the most www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

477

EN59CH24-Boomsma

ARI


Nonhost resistance: non-(co)evolved resistance due to the pathogen not having been under selection to exploit a particular host

478

4 December 2013

16:22

general questions will remain out of empirical reach for years to come. For example, our present knowledge is insufficient to refute the hypothesis that at least some insect pathogenic life histories have evolved in all major fungal lineages, and it may be only a matter of time before these rare cases are discovered. However, in spite of limited phylogenetic resolution, the number of possible insect pathogenic lineages within the major fungal clades differs widely, as does host specialization. It would thus be interesting to acquire more general answers to questions such as, Are switches from saprotrophic to parasitic lifestyles as common as secondary reversals back to saprotrophy? Is the degree of radiation in insect pathogenic lineages related to the extent of host specialization? Is the evolutionary success of insect pathogenic lineages related to the prevalence of sexual versus asexual reproduction? One evolutionary generalization that deserves to be tested is the expectation by Ebert & Hamilton (22) that virulence does not tend to increase over evolutionary time when hosts remain genetically diverse. For insect pathogenic fungi we should thus expect disease pressure on host populations to be low when pathogens remain generalists and to be higher when they specialize on a specific host lineage. Such specialization may produce so-called nonhost resistance (selection no longer targeting the use of some hosts) as an incidental by-product of the pathogen narrowing its niche, a type of resistance that is fundamentally different from unilateral or bilateral coevolved resistance in response to successful infections compromising host fitness. Insect pathogenic fungi differ considerably in their degree and type of host specialization, across lineages, within lineages, and even between closely related species (28, 47). We suspect that these differences will be correlated at least partly with spore production (many spores in generalists; fewer spores in specialists; Figure 2c). In addition, limiting factors for realizing reproductive success will likely be different, with generalists killing the weakest hosts that would not have realized any (further) reproductive success anyway (hence no selection for resistance) and specialists killing healthy hosts provided infection doses reach some critical threshold. The logic of this idea is summarized in Figure 5a, where specialists and generalists have different allometric relationships between naturally present spore doses and the extent to which focal host condition needs to be compromised before infection can be successful. The (1 and 1, implying that each additional spore increases infection success more than proportionally and that host condition does not matter much once the spore dose threshold is reached. Such specialists tend to infect a larger fraction of their hosts but are also more likely to be epidemic rather than endemic because the number of spores produced per cadaver remains low, making stochastic local extinctions more likely unless spores can persist in the habitat for longer periods (2). Many extant lineages of insect pathogens will not be specialists or generalists in all aspects of their life history, so intermediate positions will probably be common. The conceptual framework of Figure 5a illustrates that only specialized host-pathogen interactions are likely to maintain directional selection for virulence and partial resistance when hosts are involved in coevolutionary arms races with their pathogens (4), whereas www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

Epidemic: outbreak of disease in a host population causing high prevalence followed by a period of low or negligible prevalence Endemic: a disease that (chronically) coexists with its host population, often with low prevalence

479

EN59CH24-Boomsma

ARI


Iteroparous: a reproductive life history characterized by a sufficiently long life span to allow recurrent bouts of reproduction, in contrast to a semelparous life history, in which reproduction inevitably leads to death

480

4 December 2013

16:22

generalist interactions may be more comparable to predators that cull the weakest prey irrespective of species. Apart from degrees of host specialization affecting optimal number and size of spores (Figure 2a), fungal pathogens also differ in their exploitation of infected hosts (Figure 2b). From the viewpoint of fungal cells that have just managed to gain access to the body cavity of a potential host, there will be a number of adaptive decision-making trade-offs to maximize reproductive success. The first hurdle for the pathogen to overcome in this phase is to overwhelm the host immune system, as it will not realize any reproductive success if it fails to reach that milestone because the host has somehow managed to control the infection. Many, but not all, insect pathogenic fungi produce toxins during this phase, which weaken the host and increase the likelihood of immune collapse but may at the same time make host tissues less suitable for rapid proliferation. This may be an advantage for the fungal pathogen when competing with other microorganisms (40), but a disadvantage when mostly its own growth is inhibited (see above). The second hurdle is to ensure that the cadaver can be maximally exploited for the pathogen’s own reproductive success rather than the reproductive success of competing fungi and bacteria. At this stage, toxins may help sterilize cadavers for anything other than self growth (3), but producing toxins is likely to be costly and reduce growth rate toward sporulation, and such increase in generation time may negatively affect reproductive success unless there are compensating gains elsewhere. The very first exploration of the possible interactions between these factors is given in Figure 5b, based on what is known from hypocrealean and entomophthoralean fungal pathogens of insects. Toxins are typically produced by pathogens that are not particularly host specific, sometimes to such degree that it is difficult to decide whether they are pathogens or merely saprotrophs realizing most of their growth after the host has been killed by intoxication (53). Whereas most insect pathologists tend to focus on the time between inoculation and sporulation (i.e., incubation time), Figure 5b divides this total time into two periods that are represented by the left and right vertical axes, following the logic of the previous paragraph where we outlined the two crucial hurdles that may each preclude pathogen reproductive success. Entomophthoralean fungi generally have very fast life cycles and no toxin production (35, 80). They accumulate fungal cells that are not detected by the host immune system (see above), and typically a short time (measured in hours) separates host immune collapse and sporulation (in some cases extensive sporulation already occurs before host death). The Hypocreales, however, have slower life cycles and more variable toxin production, so that their anamorphic forms may well be systematically variable for all three parameters along the axes of Figure 5b. It would thus be worthwhile to compare closely related species or strains of Metarhizium or Beauveria that differ in toxin production to determine whether low toxin producers do indeed accumulate more fungal biomass before the host immune system collapses than do high toxin producers, and whether these differences are reflected in the remaining time needed to achieve sporulation. Although the logic of Figure 5b is largely speculative, there are some indications that expecting a generally inverse relationship between toxin production and pathogen growth to realizing reproductive success may be approximately correct. First, there are facultative insect pathogenic fungi such as Aspergillus and Penicillium that are renowned for their excessive toxin production and low prevalence under natural conditions (68). We know too little about their biology to formally include them in Figure 5b, but we suspect that they overwhelm the host immune system rapidly, which would place them in the bottom right corner of Figure 5b. The teleomorphic Cordyceps-like Hypocreales also combine high toxin production (65) with relatively slow life cycles, even to the extent that they produce long-lasting iteroparous fruiting bodies (stromata) that produce sexual and asexual spores after the remaining fungal tissue has sealed off the cadaver from colonization by Boomsma et al.

EN59CH24-Boomsma

ARI

4 December 2013

16:22

competing microorganisms (3). Ascosphaera is perhaps unlikely to have life histories shaped by the same set of evolutionary trade-offs, because it infects per os and likely faces different competition from other fungi or bacteria in the guts of bee larvae. However, the anamorphic stages of many hypocrealean arthropod pathogenic genera (e.g., Gibellula pathogens of spiders; 68) have remained severely understudied, so it remains to be seen whether they all will ultimately turn out to fit the diagram of Figure 5b. As pointed out by Blackwell (11), there are likely to be thousands of insect pathogenic fungal species, so modifications of our hypotheses may be needed as more data come to light.


CONCLUSION With this brief and necessarily incomplete review we hope to have demonstrated that explicit studies of adaptations that allow lineages of insect pathogenic fungi to maintain their populations may offer interesting insights that complement the approaches that have mostly shaped the insect pathology field so far. Many of these questions emerge naturally in programs that primarily address proximate questions concerning mechanism or life cycle (Figures 2 and 3) but are often not pursued further. We hope to have shown that there may be considerable long-term benefits associated with a better understanding of the evolutionary biology of insect pathogenic fungi. First, a more detailed appreciation of why specific pathogenic life-history traits emerge uniquely or convergently in the various branches of the fungal tree of life (Figure 4) may allow a better a priori evaluation of biological control strategies. Combining insights from ecological and evolutionary theory with the four strategies of biological control (classical biological control, inundation biological control, inoculation biological control and conservation biological control; 24) might reveal that inoculation strategies can sometimes replace inundation strategies, and that evolutionary inferences may allow better predictions of factors that determine epidemic development. It might also reveal that classical biological control requires that the general conditions for maintaining endemic presence have to be met first, and that conservation strategies should primarily help create the environmental conditions that make endemic dynamics most likely. Such conceptual underpinning from evolutionary ecology might also imply that conservation biological control should focus on the creation of metapopulations in which epidemic lineages of insect pathogens go extinct locally but can restart naturally via fungal propagules entering from other subpopulations where epidemics peak at different points in time. Second, if some of the logic that we outlined in Figure 5 proves to be correct, it may lead to a more coherent understanding of the interplay between somatic growth and toxin production and thus better informed selection criteria for strains with potential for inundation biological control. Third, the integrated approaches advocated in Figure 1 will allow more explicit collaborative synergies between the fundamental research agendas of evolutionary biologists interested in insect pathogenic fungi and those of applied insect pathologists who use fungi for pest control or for the prevention of fungal infections in beneficial insects such as bees.

SUMMARY POINTS 1. Insect pathogenic fungi cause important diseases of beneficial and pest insects, but their evolutionary biology has not been systematically studied. 2. We highlight how synergies between insect pathology and evolutionary biology can be developed using a modified version of Tinbergen’s four complementary types of questions in biology. www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

481

EN59CH24-Boomsma

ARI

4 December 2013

16:22

3. These questions address proximate explanations of mechanism and life cycle ontogeny, and ultimate explanations of the phylogenetic history and current adaptive value of traits in hosts and pathogens. 4. We focus on three clades of insect pathogenic fungi that have been studied in sufficient detail to allow tentative generalizations: Hypocreales, Entomophthoromycota, and Onygenales. 5. We emphasize that evolutionary and ecological theories about life-history traits and hostparasite interactions may add useful perspectives to applied research on insect pathogenic fungi.


6. We review insights into physiological interaction mechanisms, life cycle interactions with plants and non-insect environments, and fungal phylogeny to provide an accessible summary of insect pathology to evolutionary biologists.

DISCLOSURE STATEMENT The authors are not aware of any affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS Sylvia Cremer, Leila Masri, and Line V. Ugelvig commented on a previous version of the manuscript, and Line V. Ugelvig also helped with the final design of the figures. The authors were supported by a grant from the Danish National Research Foundation (DNRF57).

LITERATURE CITED 1. Akello J, Sikora R. 2012. Systemic acropedal influence of endophyte seed treatment on Acyrthosiphon pisum and Aphis fabae offspring development and reproductive fitness. Biol. Control 61:215–21 2. Andersen SB, Ferrari M, Evans HC, Elliot SL, Boomsma JJ, Hughes DP. 2012. Disease dynamics in a specialized parasite of ant societies. PLoS One 7(5):e36352 3. Andersen SB, Gerritsma S, Yusah KM, Mayntz D, Hywel-Jones NL, et al. 2009. The life of a dead ant: the expression of an adaptive extended phenotype. Am. Nat. 174:424–33 4. Antonovics J, Boots M, Ebert D, Koskella B, Poss M, Sadd BM. 2013. The origin of specificity by means of natural selection: evolved and nonhost resistance in host-pathogen interactions. Evolution 67:1–9 5. Balazy S. 1993. Entomophthorales. Krakow, Poland: Polish Academy of Sciences 6. Barbee Ml, Taylor JW. 2010. Dating the molecular clock in fungi: How close are we? Fungal Biol. Rev. 24:1–16 7. Behie SW, Zelisko PM, Bidochka MJ. 2012. Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants. Science 336:1576–77 8. Bing LA, Lewis LC. 1993. Occurrence of the entomopathogen Beauveria bassiana (Balsamo) Vuillemin in different tillage regimes and in Zea mays L. and virulence towards Ostrinia nubilalis (Hubner). Agric. Ecosyst. Environ. 45:147–56 9. Bischoff JF, Rehner SA, Humber RA. 2009. A multilocus phylogeny of the Metarhizium anisopliae lineage. Mycologia 101:512–30 10. Bissett J. 1988. Contribution toward a monograph of the genus Ascosphaera. Can. J. Bot. 66:2541–60 11. Blackwell M. 2011. The fungi: 1, 2, 3 . . . 5.1 million species? Am. J. Bot. 98:426–38 482

Boomsma et al.


EN59CH24-Boomsma

ARI

4 December 2013

16:22

12. Butt TM, Beckett A, Wilding N. 1981. Protoplasts in the in vivo life cycle of Erynia neoaphidis. J. Gen. Microbiol. 127:417–21 13. Butt TM, Hajek AE, Humber RA. 1996. Gypsy moth immune defenses in response to hyphal bodies and natural protoplasts of entomophthoralean fungi. J. Invertebr. Pathol. 68:278–85 14. Butt TM, Jackson C, Magan N, eds. 2001. Fungi as Biocontrol Agents: Progress, Problems and Potential. Oxford, UK: CABI Publ. 15. Carroll SB. 2006. Endless Forms Most Beautiful: The New Science of Evo Devo and The Making of the Animal Kingdom. New York: Norton 16. Chorbinski P. 2004. The development of the infection of Apis mellifera larvae by Ascosphaera apis. Electron. J. Polish Agric. Univ. 7(2):3. http://www.ejpau.media.pl/volume7/issue2/veterinary/art-03.html 17. Cole GT, Hoch HC, eds. 1991. The Fungal Spore and Disease Initiation in Plants and Animals. New York: Plenum 18. Cory JS, Ericsson JD. 2010. Fungal entomopathogens in a tritrophic context. BioControl 55:75–88 19. Cremer S, Armitage SAO, Schmid-Hempel P. 2007. Social immunity. Curr. Biol. 17:R693–702 20. Davies NB, Krebs JR, West SA. 2012. An Introduction to Behavioural Ecology. Oxford, UK: Wiley-Blackwell. 4th ed. 21. Dunphy GB, Chadwick JM. 1985. Strains of protoplasts of Entomophthora egressa in spruce budworm larvae. J. Invertebr. Pathol. 45:255–59 22. Ebert D, Hamilton WD. 1996. Sex against virulence: the coevolution of parasitic diseases. Trends Ecol. Evol. 11:A79–82 23. Eilenberg J, Bresciani J, Latge JP. 1986. Ultrastructural studies of primary spore formation and discharge in the genus Entomophthora. J. Invertebr. Pathol. 48:318–24 24. Eilenberg J, Hajek A, Lomer C. 2001. Suggestions for unifying the terminology in biological control. BioControl 46:387–400 25. Eilenberg J, Michelsen V. 1999. Natural host range and prevalence of the genus Strongwellsea (Zygomycota: Entomophthorales) in Denmark. J. Invertebr. Pathol. 73:189–98 26. Elliot SL, Sabelis MW, Janssen A, van der Geest LPS, Beerling EAM, Fransen J. 2000. Can plants use entomopathogens as bodyguards? Ecol. Lett. 3:228–35 27. Fisher JJ, Rehner SA, Bruck DJ. 2011. Diversity of rhizosphere associated entomopathogenic fungi of perennial herbs, shrubs and coniferous trees. J. Invertebr. Pathol. 106:289–95 28. Gao QA, Jin K, Ying SH, Zhang YJ, Xiao GH, et al. 2011. Genome sequencing and comparative transcriptomics of the model entomopathogenic fungi Metarhizium anisopliae and M. acridum. PLoS Genet. 7(1):e1001264 29. Glare TR, Chilvers GA, Milner RJ. 1985. Capilliconidia as infective spores in Zoophthora phalloides (Entomophthorales). Trans. Br. Mycol. Soc. 85:463–70 30. Grell MN, Jensen AB, Olsen PB, Eilenberg J, Lange L. 2011. Secretome of fungus-infected aphids documents high pathogen activity and weak host response. Fungal Genet. Biol. 48:343–52 31. Grimaldi D, Engel MS. 2005. Evolutions of the Insects. Cambridge, UK: Cambridge Univ. Press 32. Gurulingappa P, Sword GA, Murdoch G, McGee PA. 2010. Colonization of crop plants by fungal entomopathogens and their effects on two insect pests when in planta. Biol. Control 55:34–41 33. Hajek AE, Bauer L, McManus ML, Wheeler MM. 1998. Distribution of resting spores of the Lymantria dispar pathogen Entomophaga maimaiga in soil and on bark. BioControl 43:189–200 34. Hajek AE, Humber RA, Walsh SRA, Silver JC. 1991. Sympatric occurrence of two Entomophaga aulicae (Zygomycetes, Entomophthorales) complex species attacking forest Lepidoptera. J. Invertebr. Pathol. 58:373–80 35. Hajek AE, Papierok B, Eilenberg J. 2012. Methods for study of the Entomophthorales. In Manual of Techniques in Invertebrate Pathology, ed. LL Lacey, pp. 285–316. Amsterdam: Academic 36. Harvey PH, Pagel MD. 1991. The Comparative Method in Evolutionary Biology. Oxford, UK: Oxford Univ. Press 37. Hesketh H, Roy HE, Eilenberg J, Pell JK, Hails RS. 2010. Challenges in modelling complexity of fungal entomopathogens in semi-natural populations of insects. BioControl 55:55–73 38. Hibbett DS, Binder M, Bischoff JF, Blackwell M, Cannon PF, et al. 2007. A higher-level phylogenetic classification of the Fungi. Mycol. Res. 111:509–47 www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

483

ARI

4 December 2013

16:22

39. Hu G, St Leger J. 2002. Field studies using a recombinant mycoinsecticide (Metarhizium anisopliae) reveal that it is rhizosphere competent. Appl. Environ. Microb. 68:6383–87 40. Hughes WOH, Boomsma JJ. 2004. Let your enemy do the work: within-host interactions between two fungal parasites of leaf-cutting ants. Proc. R. Soc. B 271:S104–6 41. Humber RA. 1976. Systematics of genus Strongwellsea (Zygomycetes-Entomophthorales). Mycologia 68:1042–60 42. Humber RA. 2012. Identification of entomopathogenic fungi. In Manual of Techniques in Invertebrate Pathology, ed. LL Lacey, pp. 151–87. Amsterdam: Academic 43. Inglis GD, Enkerli J, Goettel MS. 2012. Laboratory techniques used for entomopathogenic fungi: Hypocreales. In Manual of Techniques in Invertebrate Pathology, ed. LL Lacey, pp. 189–253. Amsterdam: Academic 44. James RJ. 2012. From silkworms to bees: diseases of beneficial insects. In Insect Pathology, ed. FE Vega, HK Kaya, pp. 425–59. Amsterdam: Academic 45. James TY, Kauff F, Schoch CL, Matheny PB, Hofstetter V, et al. 2006. Reconstructing the early evolution of Fungi using a six-gene phylogeny. Nature 443:818–22 46. Jensen AB, Thomsen L, Eilenberg J. 2001. Intraspecific variation and host specificity of Entomophthora muscae sensu stricto isolates revealed by random amplified polymorphic DNA, universal primed PCR, PCR-restriction fragment length polymorphism, and conidial morphology. J. Invertebr. Pathol. 78:251–59 47. Keller S. 2002. The genus Entomophthora (Zygomycetes, Entomophthorales) with a description of five new species. Sydowia 54:157–97 48. Kepler RM, Sung GH, Harada Y, Tanaka K, Tanaka E, et al. 2012. Host jumping onto close relatives and across kingdoms by Tyrannicordyceps (Clavicipitaceae) gen. nov. and Ustilaginoidea (Clavicipitaceae). Am. J. Bot. 99:552–61 49. Kobmoo N, Mongkolsamrit S, Tasanathai K, Thanakitpipattana D, Luangsa-Ard JJ. 2012. Molecular phylogenies reveal host-specific divergence of Ophiocordyceps unilateralis sensu lato following its host ants. Mol. Ecol. 21:3022–31 50. Kodsueb R, Jeewon R, Vijaykrishna D, McKenzie EHC, Lumyong P, et al. 2006. Systematic revision of Tubeufiaceae based on morphological and molecular data. Fungal Divers. 21:105–30 51. Lucking R, Huhndorf S, Pfister DH, Plata ER, Lumbsch HT. 2009. Fungi evolved right on track. Mycologia 101:810–22 ¨ 52. Maurizio A. 1934. Uber die Kaltbrut (Pericystis-Mykose) der Bienen. Archiv. Fur ¨ Bienenkd. 15:165–95 53. Maurya P, Mohan L, Sharma P, Srivastava CN. 2011. Evaluation of larvicidal potential of certain insect pathogenic fungi extracts against Anopheles stephensi and Culex quinquefasciatus. Entomol. Res. 41:211–15 54. Mayr E. 1981. Adaptation and selection. Ital. J. Zool. 48:66–67 55. Meyling NV, Eilenberg J. 2006. Isolation and characterisation of Beauveria bassiana isolates from phylloplanes of hedgerow vegetation. Mycol. Res. 110:188–95 56. Meyling NV, Eilenberg J. 2007. Ecology of the entomopathogenic fungi Beauveria bassiana and Metarhizium anisopliae in temperate agroecosystems: potential for conservation biological control. Biol. Control 43:145–55 57. Meyling NV, Thorup-Kristensen K, Eilenberg J. 2011. Below- and aboveground abundance and distribution of fungal entomopathogens in experimental conventional and organic cropping. Biol. Control 59:180–86 58. Mongkolsamrit S, Kobmoo N, Tasanathai K, Khonsanit A, Noisripoom W, et al. 2012. Life cycle, host range and temporal variation of Ophiocordyceps unilateralis/Hirsutella formicarum on formicine ants. J. Invertebr. Pathol. 111:217–24 59. Nesse RM, Stearns SC. 2008. The great opportunity: evolutionary applications to medicine and public health. Evol. Appl. 1:28–48 60. Nikoh N, Fukatsu T. 2000. Interkingdom host jumping underground: phylogenetic analysis of entomoparasitic fungi of the genus Cordyceps. Mol. Biol. Evol. 17:629–38 61. Poinar GO, Thomas GM. 1982. An entomophthoralean fungus from Dominican amber. Mycologia 74:332–34


EN59CH24-Boomsma

484

Boomsma et al.


EN59CH24-Boomsma

ARI

4 December 2013

16:22

62. Poinar GO, Thomas GM. 1984. A fossil entomogenous fungus from Dominican amber. Experientia 40:578–79 63. Posada F, Vega FE. 2005. Establishment of the fungal entomopathogen Beauveria bassiana (Ascomycota: Hypocreales) as an endophyte in cocoa seedlings (Theobroma cacao). Mycologia 97:1195–200 64. Powell WA, Klingeman WE, Ownley BH, Gwinn KD. 2009. Evidence of endophytic Beauveria bassiana in seed-treated tomato plants acting as a systemic entomopathogen to larval Helicoverpa zea (Lepidoptera: Noctuidae). J. Entomol. Sci. 44:391–96 65. Putri SP, Kinoshita H, Ihara F, Igarashi Y, Nihira T. 2010. Ophiosetin, a new tetramic acid derivative from the mycopathogenic fungus Elaphocordyceps ophioglossoides. J. Antibiot. 63:195–98 66. Ramoska WA, Hajek AE, Ramos ME, Soper RS. 1988. Infection of grasshoppers (Orthoptera, Acrididae) by members of the Entomophaga grylli species complex (Zygomycetes, Entomophthorales). J. Invertebr. Pathol. 52:309–13 67. Roy HE, Steinkraus DC, Eilenberg J, Hajek AE, Pell JK. 2006. Bizarre interactions and endgames: entomopathogenic fungi and their arthropod hosts. Annu. Rev. Entomol. 51:331–57 68. Samson RA, Evans HC, Latgé JP. 1988. Atlas of Entomopathogenic Fungi. Berlin: Springer 69. Sasan RK, Bidochka MJ. 2012. The insect-pathogenic fungus Metarhizium robertsii (Clavicipitaceae) is also an endophyte that stimulates plant root development. Am. J. Bot. 99:101–7 70. Schreiter G, Butt TM, Beckett A, Vestergaard S, Moritz G. 1994. Invasion and development of Verticillium lecanii in the western flower thrips, Frankliniella occidentalis. Mycol. Res. 98:1025–34 71. Sosa-Gomez DR, Lastra CCL, Humber RA. 2010. An overview of arthropod-associated fungi from Argentina and Brazil. Mycopathologia 170:61–76 72. Spatafora JW, Sung GH, Sung JM, Hywel-Jones NL, White JF. 2007. Phylogenetic evidence for an animal pathogen origin of ergot and the grass endophytes. Mol. Ecol. 16:1701–11 73. Steinhaus EA. 1963. Insect Pathology: An Advanced Treatise, I and II. New York: Academic 74. Suh SO, McHugh JV, Pollock DD, Blackwell M. 2005. The beetle gut: a hyperdiverse source of novel yeasts. Mycol. Res. 109:261–65 75. Sung GH, Hywel-Jones NL, Sung JM, Luangsa-Ard JJ, Shrestha B, Spatafora JW. 2007. Phylogenetic classification of Cordyceps and the clavicipitaceous fungi. Stud. Mycol. 57:5–59 76. Sung GH, Poinar GO, Spatafora JW. 2008. The oldest fossil evidence of animal parasitism by fungi supports a Cretaceous diversification of fungal-arthropod symbioses. Mol. Phylogenet. Evol. 49:495–502 77. Tanada Y, Kaya HK. 1993. Insect Pathology. San Diego, CA: Academic 78. Thompson JN. 2005. The Geographic Mosaic of Coevolution. Chicago: Univ. Chicago Press 79. Tinbergen N. 1963. On aims and methods of ethology. Z. Tierpsychol. 20:410–33 80. Vega FE, Meyling NV, Luangsa-ard JJ, Blackwell M. 2012. Fungal entomopathogens. In Insect Pathology, ed. FE Vega, HK Kaya, pp. 171–220. Amsterdam: Academic 81. Vega FE, Posada F, Aime MC, Pava-Ripoll M, Infante F, Rehner SA. 2008. Entomopathogenic fungal endophytes. Biol. Control 46:72–82 82. Vincent C, Goettel MS, Lazarovitz G, eds. 2007. Biological Control: A Global Perspective. Case Studies from Around the World. Wallingford, UK: CAB International 83. Wang H, Xu Z, Gao L, Hao B. 2009. A fungal phylogeny based on 82 complete genomes using the composition vector method. BMC Evol. Biol. 9:195 84. White MM, James TY, O’Donnell K, Cafaro MJ, Tanabe Y, Sugiyama J. 2006. Phylogeny of the Zygomycota based on nuclear ribosomal sequence data. Mycologia 98:872–84 85. Wilson-Rich N, Spivak M, Fefferman NH, Starks PT. 2009. Genetic, individual, and group facilitation of disease resistance in insect societies. Annu. Rev. Entomol. 54:405–23 86. Wynns AA, Jensen AB, Eilenberg J, James R. 2012. Ascosphaera subglobosa, a new spore cyst fungus from North America associated with the solitary bee Megachile rotundata. Mycologia 104:108–14 87. Wyrebek M, Huber C, Sasan RK, Bidochka MJ. 2011. Three sympatrically occurring species of Metarhizium show plant rhizosphere specificity. Microbiology 157:2904–11 88. Xiao GH, Ying SH, Zheng P, Wang ZL, Zhang SW, et al. 2012. Genomic perspectives on the evolution of fungal entomopathogenicity in Beauveria bassiana. Sci. Rep. 2:483

www.annualreviews.org • Evolutionary Interaction Networks of Insect Pathogenic Fungi

485