Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

I. Introduction

The nematodes or roundworms constitute the diverse animal phylum Nematoda. Over 25,000 species have been described of which more than half are parasitic (Zhang 2013). They have successfully established in nearly all-ecological niches. Plant parasitic nematodes, which obtain their food from plant foliage and roots, cause a total global agricultural damage of more than $100 billion USD per year (Nordmeyer 1992). The most important nematodes thereby are the so-called “root-knot nematodes” (e.g., Meloidogyne spp.), which are globally distributed and infect more than 2000 plant species and thus reduce global crop yields by about 5 % (Park et al. 2014). Many nematodes that directly or indirectly affect plant growth have developed parasitic strategies to more efficiently exploit their source of food (Sijmons et al. 1994). Plant parasitic nematodes affect the plant directly by altering the morphology of the root system as a result of their feeding activities or by invasion of the plant tissue. The most specialized level has been reached by the sedentary endoparasitic nematodes that invade the root and partially reorganize the root function to satisfy their demand of nutrients (Jung and Wyss 1999). The nematode penetrates root tissues by means of its stylet and injects secretory fluids, produced in esophageal glands; these fluids modify the plant cytoplasm prior to food removal. Some species feed on the root tips of their host plants, which thereby become transformed into terminal galls, which contain necrotic cells and enlarged multinucleated cells that are essential for nematode development, growth, and reproduction. The most drastic alterations in root architecture are generated by cyst and root-knot nematodes. Root-knot nematodes are generally polyphagous, and each species can infect a large variety of plants species, from grasses to trees, by generating galls in the root. Due to their broad host range, these nematodes cannot be controlled by crop rotation. In contrast, cyst nematodes are highly host-specific parasites and can be effectively controlled by crop rotation using non-host plants (Timper 2014).

Since the origins of agriculture, men have used diverse strategies to eliminate pests that attack crop plants. Today the most common plant parasitic nematodes are controlled with chemical nematicides, cultural practices, and by the use of resistant cultivars (Timper 2014). In spite of the “successful” use of chemicals to efficiently control plant pests, it has been determined that these compounds are highly hazardous to human health and the environment. Another disadvantage of chemical pesticides is their persistence in the environment, which favors the selection of resistant pests, leading to the use of more aggressive chemicals. These actions have generated concerns around the world. Consequently, there is a heightened scientific interest on the establishment of integrated pest management strategies in order to reduce the application of chemical pesticides, that should be more effective, and less pollution, such as traps, and other means of biological control.

Yet there are still other means, such as the direct introduction of a biological control agent, i.e., a specific organism that will rapidly reduce nematode populations and/or protect the growing seedling from damage (Flint and Dreistadt 1998). Nematophagous fungi are on the top of the list of natural enemies against nematodes, because they are found in diverse environments and have been shown to be very effective as biocontrol agents in early and recent studies (cf. Kerry 2000). They have therefore attracted the attention of scientists as model organisms for “carnivorous and/or eaters” of nematodes and the deciphering of the mechanism used by them to hunt their prey (Van Ooij 2011).

II. Biology of Nematophagous Fungi

Nematophagous fungi are a diverse group of fungal species that use refined mycelial structures or their conidia to capture their prey. In addition, a large group of opportunistic fungi can parasitize the eggs and cysts of these worms (Niu et al. 2010). More than 200 species of parasitic fungi known differ in saprophytic/parasitic ability (Nordbring-Hertz et al. 2006). Therefore, they are classified in three main groups on the basis of their mechanism of interaction with the animal:

  • Endoparasites: They are mostly obligate parasites and in most cases have a restricted host range. Their use for biocontrol is therefore limited (Stirling 1992). They infect nematodes by their spores (Fig. 13.1), either through ingestion or their attachment to the cuticle (Morton et al. 2004). Some endoparasites produce zoospores that are attracted to the nematodes before adhesion and encystment on the cuticle surface. Esteya vermicola, an endoparasite of the pine wood nematode Bursaphelenchus xylophilus (Wang et al. 2010), produces α-pinene, β-pinene, and camphor—volatile compounds that are also emitted by the pine—for attraction of B. xylophilus (Lin et al. 2013).

    Fig. 13.1
    figure 1

    Scanning electron micrograph of the head region of a nematode, fixed in glutaraldehyde, heavily infected with conidia. Arrow: adhesive bud of conidia. Bar: 2

    Full size image

  • Predators or trap-forming fungi: This group of predatory fungi is the most studied and comprises the most widely used species for the biological control of nematodes. They effectively reduce the respective nematode population both under laboratory and field conditions. The trap-forming fungi include the genera Arthrobotrys, Duddingtonia, and Monacrosporium. To catch nematodes, they produce special extracellular structures of adhesive networks and tridimensional traps, which can be differentiated into six types: (a) non-differentiated adhesive hyphae; (b) ramifications of hyphae that undergoes anastomosis, forming three-dimensional adhesive networks; (c) adhesive ramifications, which, at times, can join together forming simple two-dimensional adhesive networks; (d) adhesive nodules; (e) constricting rings; and (f) non-constricting rings (Yang et al. 2007a, b). The constricting ring is the only one that actively captures nematodes (Fig. 13.2). The most common trap in predatory fungi is the one composed of adhesive networks, which has undergone significant specialization during evolution (Yang et al. 2007a, b). Despite the fact that the morphology of the traps can vary extensively, nematode-trapping fungi are generalists, and they can infect many different nematode species (Nordbring-Hertz et al. 2011).

    Fig. 13.2
    figure 2

    Nematode captured by the constricting rings of the predatory fungus Arthrobotrys anchonia. Note that the ring cells “cushion” around the body of the victim but have not yet constricted the body. This is a very early stage after capture. Scanning Electron Micrograph N. Allin and G.L. Barron

    Full size image

  • Opportunistic fungi that parasitize eggs, cysts, and female nematodes: this group of fungi has been studied for a long time, as they were considered the most promising agents for the reduction of nematode and helminth populations because they reduce levels of viable eggs in the soil (Frassy et al. 2010; Mello et al. 2013; Braga and de Araújo 2014). They are opportunistic saprophytes and do not depend on the presence of the parasite in the soil for their survival. Their hyphae penetrate the eggshell through the small pores in the vitelline layer, causing changes in permeability of the shell and expanding its volume. The hyphae then move through the adjacent layer of chitin and lipid. They thus colonize the interior of the egg and also the developing larvae (Frassy et al. 2010; Dallemole-Giaretta et al. 2012; Araujo et al. 2013). Fungi, for which this ability has been shown, include Pochonia chlamydosporia (syn. Verticillium chlamydosporium Goddard), several Trichoderma spp. (particularly T. longibrachiatum and T. harzianum), Paecilomyces lilacinus, and Dactyella ovoparasitica. Some of them, like P. chamydosporia, grow much better on nematode infested roots than on healthy roots or in the soil (Kerry 2000) and show a genetic variability closely related to the host from which they were isolated (Morton et al. 2003), suggesting that the nematodes may be more important to these fungi, than just an eventual source of nutrients.

Trichoderma spp. significantly reduce nematode infection in plants as well as the number of egg masses per plant. In vitro experiments have shown a significant effect on nematode hatching due to egg weakness and mortality due to penetration of the eggs. Additionally, nematode eggs induced growth and chitinase production in Trichoderma (Sahebani and Hadavi 2008). The reduction in nematode infection in plants could also be attributed to the induction of defense-related genes and priming by Trichoderma spp. (Sahebani and Hadavi 2008; Salas-Marina et al. 2011; Velazquez-Robledo et al. 2011; Contreras-Cornejo et al. 2011).

Although successful biological control of plant pathogens has been reported, a single fungal or bacterial biological control agent often results insufficient for effective control. Thus, the combination of two or more biological control agents could enhance their activity to protect plants; however, this combination could result in incompatibility or even antagonism. In this regard, recently, 18 strains belonging to five species of the genus Trichoderma (T. harzianum, T. virens, T. atroviride, T. rossicum, and T. tomentosum) were tested against six strains of four nematode-trapping fungal species (Arthrobotrys oligospora, A. tortor, Monacrosporium haptotylum, and M. cionopagum). Interestingly, T. harzianum and M cionopagum showed nearly identical growth rate, and no coiling around the hyphae of the nematode-trapping fungi by Trichoderma was observed (Szabó et al. 2012). The antagonism assays for each Trichoderma species against Caenorhabditis elegans revealed that T. harzianum parasitized the most eggs during the time course of the examination.

Taxonomically, the nematode-trapping fungi are exclusively found in the lineage of the Orbiliomycetes, which consists of a single order (Orbiliales) and one family (Orbiliaceae). Phylogenetic analysis placed Orbiliomycetes as a basal branch among the Pezizomycotina (James et al. 2006), the largest subphylum of the fungi that includes the vast majority of filamentous growing and fruiting-body-producing species. The time of divergence of nematode-trapping fungi from the other Pezizomycotina species is not completely clear: Yang et al. (2007a, b, 2012) estimated 400–520 Mya, whereas Meerupati et al. (2013) arrived at a much more recent time frame (198–208 Mya). The reason for this difference could be that the latter authors used 9632 orthologous genes (in contrast to five used by Yang et al. 2012). Also the calibration of the time scale differs: Meerupati et al. (2013) used the split between ascomycetes and basidiomycetes (500–650 Mya; Lücking et al. 2009), whereas Yang et al. (2007a, b) used two fossil records of carnivorous fungi, dated to 100 Mya and 24 Mya, respectively, which however is complicated by the uncertainties in the identification of the trap structures and the assignment of the taxa in fossils (Meerupati et al. 2013). Constricting rings are probably the most ancient device for nematode trapping because the fungi displaying this type of structure form a basal branch in the tree of nematode-trapping fungi (Ahrén et al. 1998).

Genotypification of 228 isolates from the nematode-trapping fungus A. oligospora from different ecological niches and geographical locations, by means of 12 single nucleotide polymorphic loci located at eight random DNA fragments, showed that ecological niche separation contributed significantly, whilst geographic separation contributed relatively little to genetic variation. The differences found between strains isolated from polluted zones versus those from non-contaminated locations suggested that environmental stress might have contributed to ecological divergence for populations. Thus, these data confirm the relevance of local adaptation and ecological niche specialization. Additionally, the lower level of differentiation among geographical populations suggested a long-distance dispersal and frequent gene flow because of the predominant clonal reproduction between populations. It was also found that those strains isolated from stressful niches presented more variability, an unambiguous evidence for recombination at the same geographic areas (Zhang et al. 2012).

To investigate the genetic variation over the geographic and ecological contexts, two virulence associated genes and two housekeeping gene fragments of 80 natural Paecilomyces lilacinus strains were analyzed. Using this approach, it was found that 32 and 19 multilocus genotypes were represented by a single isolate. Several multilocus genotypes were shared by multiple isolates, all of them from different geographical locations. Various degrees of polymorphisms and haplotypes were determined among the six genes analyzed. The analysis showed that P. lilacinus has a clonal reproduction mode in natural populations. These results suggest that these fungal isolates have limited geographic distribution and might have undergone strong genetic differentiation during their adaptation to environments in different geographic regions (Li et al. 2013).

III. Biology of Trap Formation and Nematode Infection

There are distinct differences in the mechanism of trapping with adhesive cells and the constricting ring. Fungi that feature adhesive traps capture nematodes by secretion of extracellular polymers that accumulate at the site of infection (Tunlid et al. 1992), whereas those with constricting rings ensnare the nematode by rapid swelling of a ring formed by three cells (Meerupati et al. 2013). When the nematode enters the ring, the cells inflate and the nematode is trapped. This closure occurs very rapidly (0.1 s) and is triggered by pressure of the nematode on the constricting-ring cells (Higgins and Pramer 1967). Ultrastructural examinations revealed that the cell wall of the constricting-ring cells is folded; when the cells inflate, the folded cell wall balloons out and forms the new cell wall (Heintz and Pramer 1972; Liu et al. 2012). Interestingly, nematodes may still be able to escape trap formation: the constricting rings of Drechslerella doedycoides catch early larval stages with a diameter, which is similar to the trap opening. Yet there is a short delay between the ring entry and ring closure, which allows the animal to withdraw from the trap before being caught. Mutants that fail to suppress head movements in response to touch are caught more efficiently than the wild-type. This demonstrates that the coordination of motor programs allows C. elegans to smoothly retract from a fungal noose and evade capture (Maguire et al. 2011).

In contrast, the adhesive trap is surrounded by a layer of fibrillar, extracellular polymers, which becomes reorganized during the attachment of the traps to the nematode cuticle (Tunlid et al. 1992). Following the trapping of nematodes, the infection mechanisms appear to be rather similar in the species with constricting rings and adhesive traps: the fungus forms a penetration tube that pierces the nematode cuticle and paralyzes the nematode. Subsequently, the internal tissues are rapidly colonized and digested by fungal hyphae (Nordbring-Hertz et al. 1995).

One striking feature of the nematode-trapping fungi is that they can sense the presence of their prey and only then form the traps. Earlier work demonstrated that nematodes secrete a morphogenic substance that induces trap formation in the fungi (Pramer and Stoll 1959). Hsueh et al. (2013) have recently shown that these substances are in fact ascarosides, nematode pheromones that are composed of the dideoxy-sugar ascarylose linked to a fatty acid-like side chain (Fig. 13.3). More than 100 different ascarosides have meanwhile been identified from nematodes and function as inter-organismal signals that play a central role in regulating nematode development and behavior (Butcher et al. 2007; Srinivasan et al. 2008, 2012). In both adhesive and constricting-ring types, the cuticles of the captured nematodes are then penetrated and finally an infection bulb is formed inside the nematode. After the nematode is killed, the fungus grows inside and feeds on it (Nordbring-Hertz et al. 1995). Interestingly, Li et al. (2011) showed that attachment of the soil bacterium Chryseobacterium spp. to the hyphae of A. oligospora also induced trap formation in the absence of a nematode host. However, the mechanism has not been determined.

Fig. 13.3
figure 3

Chemical structure of the ascaroside hormones from nematodes, according to Hsueh et al. (2013)

Full size image

Andersson et al. (2014) used comparative transcriptomics to investigate the molecular events during the infection process of two nematodes (the root-knot nematode M. hapla and the sugar beet cyst nematode H. schachtii) and of three nematode-trapping fungi that display different trapping mechanisms, i.e., adhesive nets (A. oligospora), adhesive branches (M. cionopagum), and constricting rings (A. dactyloides). They studied the phases of adhesion, penetration, and digestion stages. Their data showed that the divergence in interspecific gene expression between the three fungi was significantly larger than that inferred by the nematode host used. The core set of genes, identified by the Pfam domains of the encoded proteins, that was significantly expressed by all three fungi included serine endoproteases belonging to the subtilisin family, aspartyl peptidases, proteins containing a CFEM domain (a fungal-specific cysteine-rich domain that is found in some proteins with proposed roles in fungal pathogenesis), proteins involved in fungal stress response, cell signaling, organization of the cytoskeleton, vesicular transport and membrane transport, as well as several families of calcium-binding proteins and transcription factors.

Among the species-specific transcripts were those encoding proteins of metallopeptidase families M1 and M24, lectins, tyrosinase, as well as some transcription factors and cell-signaling components, and proteins containing the WSC (cell wall integrity and stress response component) domain and the DUF3129 domain. The latter is a domain of unknown function that is found in the GAS1 protein of Magnaporthe grisea, which participates in appressorial penetration and lesion formation (Xue et al. 2002). The DUF3129 domain protein-encoding gene was highly expressed during infection among the species that forms adhesive branches and adhesive knobs. As for lectins, a fruiting-body lectin and a d-mannose-binding lectin were only highly expressed in A. oligospora and not in the other two fungi. Also, the WSC domain proteins, which are one of those gene families that are expanded in nematode-trapping fungi, are highly—yet differently—expressed during pathogenesis in different nematode-trapping fungi, suggesting that they contribute to the specialization of the trapping mechanisms. Two DUF3129-domain proteins, whose orthologs in other fungi were proven as virulence genes, were differently expressed by the three fungi (Andersson et al. 2014). Finally, a hydrophobin-like protein (AOL_s00006g570) was upregulated more than 12-fold in A. oligospora. Hydrophobins are able to assemble spontaneously into amphipathic monolayers at hydrophobic–hydrophilic interfaces (Linder et al. 2005). In Beauveria bassiana, a nonspecific hydrophobic interaction between the fungal spore coat hydrophobin and the insect epicuticle was found to be essential for the pathogenicity of the fungus (Zheng et al. 2011).

Trap formation has been shown to be favored under poor nutrient conditions. Chen et al. (2013) showed that the presence of the nematodes induces autophagy—a process characterized by the degradation of unnecessary or dysfunctional cellular components that are involved in morphogenesis and morphology in fungi—by nematodes during the early stage of trap formation in A. oligospora. This is illustrated by the high expression of the atg8 gene, which encodes an essential protein in the autophagic pathway (Nakatogawa et al. 2007). Disruption of a homolog of this gene in A. oligospora leads to reduce trap formation (Chen et al. 2013). During the early stage of trap formation, the expression of genes encoding enzymes involved in amino acid biosynthesis and the general regulator of amino acid biosynthesis GCN4 are induced, suggesting that nematodes induce autophagy probably by triggering intracellular amino acid starvation.

A transcriptomic analysis of Drechslerella stenobrocha, which mechanically traps nematodes using a constricting ring also shed some light on the signal transduction cascade that is involved: like in entomopathogenic fungi during insect infection (Zheng et al. 2011), trap formation may involve the protein kinase C (PKC) pathway, as suggested by the strong upregulation of the pkc1 gene during trap formation. The authors proposed that one of the Gα-proteins (DRE_07451) could be the first step in this PKC pathway, which could be activated by a signal from nematodes (see above). The authors also detected several putative transcriptional regulators of the fungal-specific Zn(2)Cys(6) type that could regulate the downstream genetic responses. Finally, they found that the transcription of genes encoding a cell division protein and a cyclin peaked during the phase of trap formation, suggesting that the formation of constricting ring would involve cell division processes. In a similar approach, global patterns of gene expression in traps and mycelium of the fungus Monacrosporium haptotylum were compared. In this case, the trap is a unicellular spherical structure called the knob, which develops on the apex of a hyphal branch. Substantial differences in the patterns of genes expressed in the two cell types were found, with about 23 % of the putative genes preferentially expressed in knobs. Various differentially expressed genes were similar to genes known to be involved in regulating morphogenesis and cell polarity in other fungi. This set of differentially expressed genes included several putative small GTPases, such as rho1, rac1, and ras1, and a rho GDP dissociation inhibitor (rdi1). Genes involved in stress responses, protein synthesis and protein degradation, transcription, and carbon metabolism were also among this set. A number of the differentially expressed genes are also differentially regulated during infection structure formation in plant-pathogenic fungi. Interestingly, gks1 a homologue of the Magnaporthe grisea (gas1/mas3) gene, which is specifically expressed in appressoria, was found (Ahrén et al. 2005).

The initial phase of penetration is believed to be associated with recognition mediated by a lectin–carbohydrate interaction (Nordbring-Hertz 1983). These lectins are located on fungal traps or adhesive conidia that can specifically bind a carbohydrate on the nematode cuticle. It is has been suggested that after the recognition event, the fungus immobilizes the nematode and secretes extracellular enzymes at the point of contact that allow the posterior parasitism (Tunlid et al. 1994). The nematode cuticle consists mainly of proteins, including collagens (Cox 1992), and the nematode eggshell contains chitin fibrils embedded in a protein matrix, with the chitin complex as a major barrier against fungal infections (Warton 1980). Extracellular enzymes that are capable of digesting the main chemical constituents of the nematode cuticle and eggshell (protein, chitin, and lipids) have been isolated and identified in various nematophagous fungi (e.g., Lopez-Llorca 1990; Tunlid et al. 1994; Yang et al. 2005a, b; for review see Yang et al. 2013a, b). When the hyphae from the endoparasitic fungi Drechmeria coniospora and Hirsutella rhossiliensis reach the eggshell, they form appressoria from which these extracellular enzymes are then secreted (Lopez-Llorca and Robertson 1992). This formation of appressoria depends on the recognition of the host surface, and surface hydrophobicity is considered an important recognition factor in this process (Lopez-Llorca et al. 2002).

V. Virulence Factors

Virulence factors are molecules produced by pathogens, which are essential for major contribution to their pathogenicity, by enabling them to attach to the host, escape the defense mechanisms, and finally feed on it. The previous chapter on the molecular mechanisms that take place during attack of the nematode by the fungus has already pointed to potential candidates for virulence factors. In this chapter, we will discuss those components for which detailed scientific information is already available.

A. Proteases

Being saprophytes, most Pezizomycota are known to possess a rich arsenal of proteolytic enzymes. According to the MEROPS database, the main groups found are aspartyl proteases, cysteine proteases, metalloproteases, and serine proteases, the latter two making up for the bulk of secreted proteases (cf. Lai et al. 2014). Comparative genomics indeed showed that some families of the metalloproteases (particularly those that show collagenase activity) and of the subtilisin-type of the serine proteases are strongly enriched in the nematophagous fungi (Meerupati et al. 2013; Lai et al. 2014; Fig. 13.4).

Fig. 13.4
figure 4

Number of selected protease genes in the genomes of nematophagous and insect pathogenic fungi. Fungal species are indicated by color bars: blue: A. oligospora, red: M. haptotylum, grey: D. stenobrocha, yellow: H. minnesotensis, black: M. anisopliae, green: B. bassiana. Abbreviations: AspP aspartyl proteases, CysP cysteine proteases, MeP metalloproteases, M10 MEROPS family M10 metalloproteases within MeP, SerP serine proteases, S8 MEROPS family S8 subtilisins within SerP, Total all protease genes encoded in the respective genome. Data taken from Yang et al. (2011a, b), Meerupati et al. (2013), Liu et al. (2014), Lai et al. (2014), Gao et al. (2011), and Xiao et al. (2012)

Full size image

1. Collagenases

The ascarid cuticle is a three-layered, fibrous structure, which contains nematode-specific types of collagen and keratin (Bird 1971). Collagens are among the most complex of proteins and are slowly degraded in natural soils and waters (Weiss 1976). During the infection of nematodes, nematophagous fungi must penetrate the nematode cuticle, and particularly collagenases have been considered important enzymes involved in the pathogenicity of nematophagous fungi (Dackman et al. 1992; Tunlid et al. 1994). Meerupati et al. (2013) showed that peptidases belonging to the MEROPS family M10, which display strong collagenolytic activity, are strongly enriched in A. oligospora and M. haptotylum, whereas these genes are almost absent from insect pathogenic fungi. This is in agreement with earlier studies that nematode-trapping fungi produced collagenase in the growth media of all tested species (Schenck et al. 1980).

The M10 metallopeptidases belong to a group of peptidases known as the “metzincins” due to a conserved methionine C-terminal to the zinc ligands. Both subfamilies of M10 contain mosaic proteins, which contain, e.g., glycine-rich C-terminal domains that can bind calcium ions or domains homologous to hemopexin and vitronectin that aid in binding to the extracellular matrix. The gelatinases have acquired three domains homologous to type II segments of fibronectin nested within the peptidase unit. Details about the structure of the fungal M10 metalloproteases, as well as genetic proof for their action as virulence factors, are however yet lacking.

2. Subtilisins

Almost all the proteases from entomopathogenic and nematophagous fungi that were identified and shown to have nematicidal activity in the pre-genomic era belonged to the large subtilisin family of endopeptidases MEROPS M8 found only in fungi and bacteria. This family is enriched in nematophagous fungi, but also in insect pathogens (Meerupati et al. 2013; Lai et al. 2014). The role of serine proteases as virulence factors was first demonstrated in invertebrate pathogenesis by the entomopathogenic fungi Metarhizium anisopliae (St. Leger et al. 1987) and B. bassiana (Bidochka and Khachatourians 1987), where a 30 kDa serine protease (Pr1) was found to play an important role in the infectious process (Morton et al. 2004). Proteases that share characteristics with Pr1, such as size, sensitivity to inhibitors, and substrate pattern (thus called Pr1-like), were first purified and characterized from the opportunistic nematophagous fungi Paecilomyces lilacinus (Bonants et al. 1995), P. rubescens (Lopez-Llorca 1990), and P. chlamydospora (Segers et al. 1994).

In A. oligospora, subtilisins appear to play a key role in the early stages of infection, including immobilization of the captured nematode (Tunlid and Jansson 1991; Åhman et al. 2002; Yang et al. 2011a, b). The A. oligospora subtilisin PII immobilizes the active stages of Panagrellus redivivus and hydrolyzes its cuticle (Tunlid et al. 1994). The enzyme is expressed under starvation conditions (Åhman et al. 1996). Another subtilisin from A. oligospora (Aoz1), whose amino acid sequence is 97 % similar to that of PII, also produced dramatic structural changes in the nematode cuticle (Minglian et al. 2004). Deletion of the PII gene had only a limited effect on pathogenicity (decreased adhesion and immobilization of nematodes and formation of less traps), probably due to the presence of aoz1. Overexpression of the PII-encoding gene, however, resulted in a higher capacity to kill nematodes and formation of more traps (Åhman et al. 2002).

More recently, a genome-wide transcriptional analysis of A. oligospora revealed that in fact the gene encoding yet another subtilisin (P186) was more than 40-fold upregulated, while that encoding PII was even downregulated, suggesting that P186 may be the main protease required for penetration of the nematode cuticle. In fact, a protease belonging to the same subfamily in S8 as P186 was also found upregulated in the knob proteome of M. haptotylum (Andersson et al. 2013).

To better understand the cellular functions of adhesive traps, the proteome and transcriptome of trap cells versus mycelia of the fungus Monacrosporium haptotylum were assessed. The comparison of protein expression between mycelia and knobs revealed that 54 out of 336 detected proteins were highly expressed in the knobs compared with mycelia. Secreted proteins were overrepresented: secretion signals were predicted in 26 sequences (48 % of de proteins identified), including Small secreted cysteine-rich proteins (SSCRPs), peptidases and carbohydrate-binding proteins containing WSC and GLEYA domains, and proteins involved in stress response. WSC is a cysteine-rich domain with eight conserved cysteine residues that are required for its function (Heinisch et al. 2010; Dupres et al. 2011). All the upregulated WSC domain proteins belong to a large expanded cluster of paralogs in M. haptotylum. Various peptidases and homologs of experimentally verified proteins in other pathogenic fungi were also upregulated in the knob proteome. The expression of only six of the upregulated knob genes was reflected in increased protein levels. These proteins included a putative surface protein of the PA14_2/GLEYA family, a glutathion S-transferase, an alcohol dehydrogenase, and two hypothetical proteins with predicted secretion signals. In agreement with the upregulated proteome, the upregulated transcriptome was also enriched in sequences predicted to have a signal peptide (20 %). Therefore, the traps of M. haptotylum seem to have the necessary proteins for the early stages of infection (Andersson et al. 2013).

Subtilisins also appear to play dominant roles in the infection by non-trap forming nematophagous fungi: infection of nematode eggs by D. coniospora was blocked by the addition of the serine protease inhibitor, chymostatin, indicating the possible role of serine proteases in the infection process (Jansson and Friman 1999). Addition of serine protease inhibitors reduced egg penetration by the fungi Lecanicillium lecanni and P. chlamydosporia, further supporting the relevance of proteases at the early stages of the infection process (Lopez-Llorca et al. 2002). A subtilisin named P32 was immunolocalizated in appressoria of the fungus P. rubescens that infects eggs of the beet cyst nematode Hetrodera schachtii (Lopez-Llorca and Robertson 1992).

The opportunistic fungus P. chlamydosporia produces an alkaline subtilisin (VCP1) during the infection of nematode eggs. The incubation of nematodes eggs with purified VCP1 resulted in the removal of the outer vitellin membrane from eggs of Meloidogyne incognita (Segers et al. 1994). Subsequent infections of these eggs by P. chlamydosporia degraded extensively the eggshell to the degree of generating large holes in the structure, with no evident formation of appressoria while this was not the case when eggs of G. pallida were treated with VCP1, which points to the importance of the different composition of nematode eggshells (Morton et al. 2004).

Sequence analysis of the vcp1 upstream region from 30 different isolates of Pochonia chlamydospora revealed that this region is highly conserved, ranging from 91 to 97 % identity, and contains putative regulatory motifs for carbon (CREA and CREB) and nitrogen repression (GATA), and pH regulation (PacC). Indeed, the presence of glucose, ammonium, and changes in pH affected the expression of this gene. For instance, addition of glucose to the growth medium significantly repressed VCP1 enzyme and mRNA levels, whereas the presence of M. incognita eggs did not downregulate neither the VCP1 enzyme nor the mRNA. Furthermore, the presence of ammonium chloride significantly reduced the VCP1 mRNA and proteins levels; however, at longer times (24 h), the enzyme and the mRNA levels were considerably upregulated (Ward et al. 2012). Cryo-scanning electron microscopy revealed that VCP1 production occurred only when the fungus and P. chlamydospora eggs are in close contact. These results indicated that the presence of preferable carbon sources and unfavorable pH in the rhizosphere/egg-mass environment might negatively affect the nematode parasitism by P. chlamydosporia. On the contrary, the presence of ammonium nitrate may favor the biocontrol of this nematode by the fungus at longer times (Ward et al. 2012).

The regulation of the expression of subtilisins has also been studied in the case of prC in Clonostachys rosea, whose encoded protein immobilized nematodes and hydrolysed proteins of the nematode cuticle (Li et al. 2006): the presence of putative transcription control sites in the promoter for nitrogen regulation (5′-GATA), carbon regulation (5′-SYGGRG), pH regulation (5′-GCCARG), and stress response element (STRE) (5′-AGGGG) suggested that the expression of prC may be regulated by nitrogen sources, environmental pH, and/or other stress conditions (Zou et al. 2010a, b). To study the effect of pH, the C. rosea orthologue of the pH transcriptional regulator PacC was deleted. The expression of prC was downregulated in ΔpacC mutants, and the prC transcript levels were significantly higher under alkaline growth conditions than under acidic growth conditions. Induction of prC expression by nematode cuticles was significantly suppressed by glutamine, ammonia, and serine protease inhibitors (Zou et al. 2010c).

Aside from pH and nematode cuticle-induced changes of gene expression, the expression of prC was also upregulated by oxidants (H2O2 or menadione) and heat shock, probably through a stress response pathway. Interestingly, the addition of nematode cuticle significantly attenuated the production of reactive oxygen species induced by oxidants and heat shock in the wild-type strain but not in the ΔprC mutants (Zou et al. 2010b). This suggests that PrC is not only involved in the degradation of nematode cuticles but also plays a role in the adaptation to environmental stresses.

The serine protease Ver112 from the nematophagous fungus Lecanicillium psalliotae is capable of degrading the nematode cuticle and killing nematodes effectively (Yang et al. 2005a, b). The Ver112 gene was used to genetically transform P. lilacinus. Protease activity of the transformants was higher than in the wild-type and correlated with a stronger ability to immobilize, infect, and degrade the nematode Panagrellus redivivus and Caenorhabditis elegans than the wild-type. The crude protein extract of the transformants showed enhanced nematicidal activity compared to the wild-type (Yang et al. 2011a, b).

The evolution of subtilisin-like serine proteases in Pezizomycotina has been analyzed (Li et al. 2010): molecular phylogeny divided the serine proteases from nematophagous fungi into two clades with neutral proteases from nematode-trapping fungi clustering in clade A and the alkaline ones from nematode-parasitic and entomopathogenic fungi clustering in clade B. Both share a high degree of sequence identity, have very similar molecular structure, and play a similar role in degrading host cuticle during fungal infection of nematodes. However, their structure reveals interesting differences in the 3D structure of the substrate-binding regions and some neighboring loops and turns (Liang et al. 2010; Yang et al. 2010a, b): disulfide bridges, which contribute to the stabilization of the local/global structures and enhance the structural flexibility of two of the substrate sites, were only present in the alkaline, but not in the neutral protease. This may explain why the alkaline proteases have higher substrate affinity and catalytic activity than neutral proteases. Since nematode-parasitic fungi, which contain only alkaline proteases, do not produce trapping devices, they likely rely mainly on the extracellular enzymes as virulence factors to help them penetrate and digest nematode cuticles (Huang et al. 2004; Yang et al. 2007a). It is likely that their subtilisins evolved towards increased activity and broad substrate specificity. In contrast, the nematode-trapping fungi degrade the trap-captured nematode without time constraints because it is already paralyzed by the trap. This interpretation is also supported by the fact that the subtilisin-like serine proteases of the nematode-trapping fungi were found to be under positive selection, suggesting co-evolution of trapping structures and proteolytic enzymes (Li et al. 2010).

B. Chitinases

Nematophagous fungi that parasitize nematode eggs must penetrate the eggshell during the infection (Lopez-Llorca and Duncan 1988; Lýsek and Krajcí 1987). As mentioned above, the structure of the eggshell is formed by several layers, including one formed by chitin (Warton 1980), which is the thickest and probably the major barrier for infection (Bird and Bird 1991).

Most fungi are able to degrade chitin, and two main enzyme classes cooperate in its degradation: chitinases (belonging to the glycoside hydrolase family GH18) and N-acetyl-β-d-glucosaminidases (belonging to GH20). The action of the former leads to soluble chitooligosaccharides with a chain length of at least two amino sugar units, which are subsequently further hydrolysed to NAcGln by the N-acetyl-β-d-glucosaminidases. Furthermore, chitin can be deacetylated by chitin deacetylases (EC 3.5.1.31) found in carbohydrate esterase family 4 (CE 4) in the Carbohydrate Active Enzymes database (CAZy) classification (Hartl et al. 2012).

Fungal chitinases can be further divided into three different subgroups, A, B, and C, based on the amino acid sequences of their GH18 modules. These subgroups differ in the architectures of their substrate-binding cleft and thus their enzymatic activities (exo vs. endo), and those from subgroups B and C contain different carbohydrate-binding modules (CBM 18 and CBM50; see also www.cazy.org; Gruber and Seidl-Seiboth 2011; Seidl 2008). Class A chitinases comprise some of the most frequently described fungal chitinases. They typically have a Mr (Molecular weight range) between 40 and 60 kDa, their active center is located in a deep cleft, and they are exo-acting. Class B enzymes are usually somewhat smaller (30–50 kDa), endo-acting, their active center is located close to the proteins surface, and they typically contain a carbohydrate-binding domain at their C-terminus (frequently of the cellulose-binding CBM1 type). Class C chitinases, in contrast, are large proteins (120–200 kDa) that act in an exo-type with an active center in a deep and narrow cleft. Most typical for them is the presence of a CBM18 chitin-binding or CBM50 peptidoglycan-binding (LysM) domain. The presence of CBMs in class B or C chitinases enables them to bind more tightly to insoluble substrates (Eijsink et al. 2008). Bacterial proteins with LysM domains have been reported to be involved in specific recognition events between nitrogen fixing bacteria and their plant hosts (Knogge and Scheel 2006). Besides nutritional purposes, subgroup B chitinases appear also to be involved in mycoparasitic and entomopathogenic functions.

The first report for chitinase activity in nematophagous fungi was in Verticillium spp., isolated from infected nematode eggs, both in screening on solid media with colloidal chitin and in liquid media. Several chitinases have also been identified from egg-parasitic fungi, which were found to serve as a nematicidal factor in infecting nematode eggs (Tikhonov et al. 2002; Khan et al. 2004; Gan et al. 2007). On the basis of their properties, most of them seem to belong to the A subgroup.

The Arthrobotrys oligospora genome contains 16 Open Reading Frames encoding putative chitinases that belong to the glycoside hydrolases (GH) family 18. These chitinases vary considerably in their functional domains, size, and pI. Based on the phylogenetic relationship, these were grouped into four clades: I, II, III, and IV, that include an A. oligospora-specific subclade (Clade IV-B), which includes chitinases ≥100 kDa. Most of the A. oligospora chitinases genes are downregulated in absence of carbon; conversely nitrogen starvation upregulates all chitinase-encoding genes. Nonetheless, chitinase AO-190 was upregulated in both, carbon and/or nitrogen starvation. Furthermore, chitinases AO-59, AO-190, and AO-801 increased their transcription in the presence of colloidal chitin or R. solani cell wall. This suggests a role of A. oligospora chitinases in biocontrol. The expression patterns of A. oligospora chitinases suggest that they play different roles in growth, differentiation, and infection (Yang et al. 2013a, b).

When five Trichoderma species (T. harzianum, T. virens, T. atroviride, T. rossicum, and T. tomentosum) were tested against C. elegans, T. harzianum was more successful at parasitizing eggs. During egg parasitism, the expression levels of chi18-12 and chi18-5 were significantly higher than controls, which suggest a role of these endochitinases in the infection process (Szabó et al. 2012).

C. Lectins

It has for a long time been assumed that the adhesion of nematophagous fungi to their host might be mediated by the interaction between lectins on the surface of trapping devices or adhesive spores and carbohydrate ligands on the nematode cuticle (Nordbringhertz and Mattiasson 1979). Lectins are carbohydrate-binding proteins that are present in all organisms. A comparison of the genomic inventory of trap-forming fungi and nematode pathogenic or insect pathogenic fungi revealed that the trap-forming species indeed had a much higher number of lectin-encoding genes than other fungi. A detailed analysis showed that the most abundant lectin family—like in other fungi—are the concanavalin A-like lectins that bind α-d-glucose and α-d-mannose (Fig. 13.5). However, the ricin B-type lectins (PF14200), the H-type lectin (binding to N-acetyl-β-d-galactosamine), the fucose-specific lectin, and the bulb-type lectin were all significantly more abundant in the trap-forming species A. oligospora, D. stenobrocha, and M. haptotylum (Lai et al. 2014).

Fig. 13.5
figure 5

Number of selected lectin-encoding genes in the genomes of nematophagous and insect pathogenic fungi. Blue Concanavalin A lectin; red ricin B-like lectin; grey H-type lectin; yellow fucose-specific lectin; black bulb-type lectin; black total number of lectins. Abbreviations: AOL A. oligospora, MHA M. haptotylum, DST D. stenobrocha, HMI H. minnesotensis, MAN M. anisopliae, BBA B. bassiana. Data taken from Yang et al. (2011a, b), Meerupati et al. (2013), Liu et al. (2014), Lai et al. (2014), Gao et al. (2011) and Xiao et al. (2012)

Full size image

During trap formation and infection, all trap-forming fungi expressed transcripts encoding RicinB_lectins, which are ribosome-inactivating proteins (RIPs) consisting of a catalytic A-chain and a sugar-binding B-chain (Michiels et al. 2010). The effects of these lectins from trap-forming fungi on the nematode have not yet been studied, but a RicinB_lectin_2 domain-containing protein (MOA) of the basidiomycete Marasmius oreades displayed nematotoxic activity against C. elegans (Wohlschlager et al. 2011). This nematotoxicity was dependent on the cysteine protease activity of MOA and the binding of its lectin domain to glycosphingolipids in the worm intestine. A Sclerotinia sclerotiorum agglutinin (SSA) also contains a RicinB-lectin_2 domain and shows insecticidal properties when fed to the pea aphid Acyrthosiphon pisum (Hamshou et al. 2010). Most recently, a ricin B-like single-domain lectin (MpL) has been isolated and characterized from the parasol mushroom Macrolepiota procera (Žurga et al. 2014). MpL exhibits highest specificity for terminal N-acetyllactosamine and related β-d-galactosides and contains a second putative carbohydrate-binding site with a low affinity for d-galactose. MpL was shown to be toxic to C. elegans. Summarizing, there is now accumulating evidence that the RicinB-lectins exhibit toxicity to nematodes, and they could well be the agent that paralyzes the host after forming the trap. This is a challenging topic of further work with the trap-forming fungi.

D. Small Secreted Cysteine-rich Proteins

One of the largest groups of proteins secreted by mycoparasitic fungi like Trichoderma spp. are the so-called “small secreted cysteine-rich proteins (SSCPs).” They were identified by the criteria that their M r should be ≤300 amino acids long and containing four or more cysteine residues. Among them, hydrophobins, hydrophobin-like proteins, and elicitor-like proteins make up for a major part, but many others are found for which no function has been predicted. In Trichoderma spp., some members of this cluster contain CFEM domains or consensus sequences for glycosylphosphatidylinositol (GPI anchors), suggesting that they could be cell surface proteins with important roles in the interaction with other organisms, as in C. albicans (Druzhinina et al. 2012). Meerupati et al. (2013) detected that “SSPs” make up for a large amplified group of orphan genes of the “knob-forming” fungus M. haptotylum—but not in the “net-forming” A. oligospora. 27.6 % of them were actually clustered in the genome, of which 34 genes that were >10-fold upregulated in M. haptotylum during early infection were located in clusters. This suggests that the SSPs could play an important role in knob formation in this species.

Proof for a function of SSPs has been obtained in the case of cerato-platanins, a group of small, secreted, cysteine-rich proteins that have been implicated in virulence of certain plant pathogenic fungi and also shown to stimulate plant defense against pathogenic fungi (Djonovic et al. 2007; Salas-Marina et al. 2015; Pazzagli et al. 2014).

The nematophagous fungus Dactylellina cionopaga, which is a known parasite of the nematode plant pathogens Meloidogyne javanica and Heterodera schachtii, develops adhesive columns and two-dimensional networks (Khan et al. 2006; Jaffee and Muldoon 1995). The cerato-platanin family of proteins plays roles in parasitism, recognition, adhesion, cell-wall morphogenesis, fungal growth and development, and induction of the systemic resistance to pathogens in plants (Djonovic et al. 2007; Salas-Marina et al. 2015). The transcription levels of D. cionopaga snodprot increased in the presence of nematodes and was induced during the development of traps and conidia. The recombinant protein changed the chemotaxis and increased the body-bend frequency of C. elegans, but did not induce immunity in plants. In agreement with the parasitism mechanisms of nematophagous fungi, chemotaxis, and locomotion mechanism of C. elegans, the possible targets of snodprot could be ASE and ASI neurons, which are involved in the process of chemotaxis to NaCl, the response to serotonin, and the locomotion of C. elegans (Yu et al. 2012). Together these results indicated that snodprot is a novel parasitism-related protein of nematophagous fungi with a non-described activity.

E. Secondary Metabolites

In order to antagonize or kill their competitors, many microorganisms produce toxic metabolites, such as antibiotics. Toxins are particularly important for parasitic microorganisms, because they facilitate infection by debilitating the host (Morton et al. 2004). In addition, P. lilacinus produces acetic acid to paralyze juvenile nematodes (Djian et al. 1991).

So far, the majority of nematicidal secondary metabolites characterized are those produced by opportunistic fungi. Fusarium equiseti produces compounds that reduce hatch of root-knot nematode eggs and immobilize infective juveniles (Nitao et al. 1999); a metabolite with nematicidal activity against infective juvenile, phomalactone, was isolated from P. chlamydosporia (Hellwig et al. 2003); and culture filtrates from several fungi grown in malt extract broth were toxic to infective juveniles and eggs (Chen et al. 2000). Several secondary metabolites have been isolated from the nematode egg-parasite P. chlamydosporia, including radicicol (=monorden; a resorcylic acid lactone), tetrahydromonorden, pseurotin A, pochonins A to J (Hellwig et al. 2003 and Shinonaga et al. 2009; Zhou et al. 2010), and various aurovertin-type metabolites (Niu et al. 2010). Radicicol biosynthesis has been studied in detail, as usually found in fungal genomes, the genes encoding the corresponding biosynthetic pathways are clustered (rdc1–rdc5; Zhou et al. 2010) and encode two fungal iterative polyketide synthases (PKS). Rdc5, the highly reducing IPKS, and Rdc1, the nonreducing IPKS, are required for the biosynthesis of radicicol. The biochemical pathway, by which Rdc1 and Rdc5 catalyze the biosynthesis of radicicol, and how the remaining genes of the cluster contribute, has been elucidated (Zhou et al. 2010). During endophytic root colonization, P. chlamydosporia expressed 56 % of the secondary metabolism pathway genes found, including seven of the radicicol cluster.

So far, 179 nematicidal compounds belonging to diverse chemical groups have been identified from nematophagous fungi, of which only three (oligosporon, 4′,5′-dihydro-oligosporon, and linoleic acid) were from A. oligospora (Li et al. 2007). A genomic comparison of PKS, nonribosomal peptide synthases, and terpene synthases (TPS) revealed that the trap-forming fungi contained the lowest number of these genes (Fig. 13.6), whereas the nematode endoparasite H. minnesotensis exhibited the highest number of secondary metabolite synthases, even in comparison with insect pathogens, particularly of the PKS-1, NRPS, and TPS class (Lai et al. 2014). Six NRPS and two TPS genes were unique to the nematode endoparasitic fungus, suggesting lineage-specific expansion of these families in the H. minnesotensis genome. Thus, while trap-forming fungi do not seem to make strong use of secondary metabolites (possibly because the lectins already paralyzed the nematodes; vide supra), the endoparasites heavily rely on these metabolites to kill the host.

Fig. 13.6
figure 6

Number of secondary metabolite synthases in the genomes of nematophagous and insect pathogenic fungi. Red: polyketide synthases (PKS), green: non-ribosomal peptide synthases (NRPS), grey: hybrid PKS-NRPS, yellow: terpene synthases, blue: total number of synthases lectins. Abbreviations: AOL A. oligospora, MHA M. haptotylum, DST D. stenobrocha, HMI H. minnesotensis, MAN M. anisopliae, BBA B. bassiana. Data taken from Yang et al. (2011a, b), Meerupati et al. (2013), Liu et al. (2014), Lai et al. (2014), Gao et al. (2011), and Xiao et al. (2012)

Full size image

Trichoderma spp. are biocontrol agents widely used in plant protection due to their capacity to antagonize phytopathogenic fungi. Nevertheless, it is known that some Trichoderma spp. produce secondary metabolites with nematicidal activity, including trichodermin (Yang et al. 2010a, b), acetic acid (Djian et al. 1991), gliotoxin (Watanabe et al. 2004; Anitha and Murugesan 2005), and the peptide cyclosporin A. Recently, the volatile organic compound 6-pentyl-2H-pyran-2-one from Trichoderma spp. was shown to kill >85 % of Panagrellus redivivus, Bursaphelenchus xylophilus, and C. elegans in 48 h at 200 mg/l (Yang et al. 2012).

VI. Plant Endophytism by Nematophagous Fungi

Several nematophagous fungi may be found as endophytes of plant roots. A plant endophyte is a plant microbial endosymbiont, which lives part of its live cycle in the plant, without provoking negative effects to its host. On the contrary, the presence of endophytes frequently results in positive effects in plant growth and development. Nematophagous fungi such as P. chlamydosporia and A. oligospora can present endophytic life style for both monocots (Lopez-Llorca et al. 2002) and dicots (Bordallo et al. 2002). Both fungi were capable of growth inter- and intracellularly and form appresoria when penetrating the cell wall of epidermis and cortex cells, from tomato and barley. However, these fungi never penetrate the vascular tissue (Lopez-Llorca et al. 2002; Bordallo et al. 2002). Plants endophytically colonized by nematophagous fungi show enhanced defense responses and biomass gaining (Maciá-Vicente et al. 2009a, b). LopezLlorca and coworkers detected the production of serine protease P32, VCP1, and SCP1 from a nematophagous fungus in roots colonized endophytically by this microorganism. As mentioned above, these proteases are produce by nematophagous fungi on their nematode host. Thus, the expression of these proteins in the absence of their host would imply that plants colonized by these fungi could be protected from nematode attack before contact (Lopez-Llorca et al. 2010).

Lopez-Llorca and coworkers showed that fungi other than A. oligospora and P. chlamydosporia grow endophytically in roots (Lopez-Llorca et al. 2006). For instance, the endoparasitic basidiomycete Nematocnus robustus, which infects nematodes through adhesive conidia, penetrated and colonized barley roots and formed clamp connections, whereas N. pachysporus did not colonize the root system, but colonized the root surface (Lopez-Llorca et al. 2006). The nematode endoparasitic fungus Hirsutella rhossiliensis has a similar behavior since it only colonizes the root surface. Furthermore, fungi belonging to the basidiomycota, such as Pleurotus djamor that immobilizes nematodes with a toxin (Kwok et al. 1992) prior to infection and digestion of its prey, also colonizes and penetrates barley roots (Lopez-Llorca et al. 2006). The nematode-trapping fungus Arthrobotrys dactyloides is also a root colonizer that penetrates the epidermal cells and forms coiling structures in barley root cells (Lopez-Llorca et al. 2006).

To better understand the endophytic process, the P. chlamydospora genome was recently sequenced and consists of 41.2 Mb, from which 12,122 gene models were predicted. Under the endophytic relationship with barley roots, 63 % (7586 genes) of the genome was expressed. From the 1432 predicted secreted proteins, 57 % were expressed under this condition. 663 predicted genes did not exhibit any homologue in the NCBI database and almost a half of them (277) were expressed during P. chlamydospora endophytic lifestyle. Phylogenetic analysis of genome-encoded orthologous showed that P. chlamydospora is most closely related to Metarhizium anisopliae and M. acridium. A search for pathogenesis-related genes in the Pathogen–Host Interaction (PHI) database that collects pathogenesis-related genes of fungi, bacteria, and oomycetes (Winnenburg et al. 2008) showed that 1981 genes (16 %) shared homology with genes included in the PHI database, 24 % (468 genes) of which encoded putatively secreted proteins. The majority of genes putatively associated with pathogenesis and endophytism encode hydrolytic enzymes and signal transduction proteins. The hydrolases found in PHI included metalloproteases and chitinases, whilst those expressed under endophytism include serine and rhomboid protease families and a protein phosphatase. The P. chlamydospora genome contains a wide set of genes encoding hydrolytic enzymes, from which almost a half were expressed during endophytism. In addition it contains 15 PKS and 12 putative non-ribosomal peptide synthases (PRPS), together with a number of PKS- and NRPS-like proteins, and 4 NRPS–PKS hybrid genes. A radicicol gene cluster was also found. P. chlamydospora expressed 56 of the secondary metabolism pathway genes identified. From these seven genes belonging to the radicicol cluster were expressed. The P. chlamydospora genome contains 290 transporters of the major facilitator superfamily (MFS), 58 ATP-binding cassette (ABC) transporters, and 113 general transporters, most of which exhibited homologs included in PHI database. P. chlamydospora expressed drug resistance, sugar/inositol, oligopeptide, and amino acid transporters during endophytism. Furthermore, a number of genes encoding oxidoreductases related to detoxification were also detected. Genes involved in cell wall biosynthesis and modification, including chitin synthesis activators, chitin synthases, lipopolysaccharide modifying proteins, and hydrophobins, were also found expressed under endophytism. Additionally, the P. chlamydospora genome encodes 409 putative transcription factors (TFs), grouped in six families, of which the Zn2Cys6 fungal-specific type TF contains the highest number of genes expressed under endophytism. In addition, the P. chlamydospora genome encodes G-protein subunits (8), putatively involved in vegetative growth, conidiation conidium attachment, appresorium formation, mating, and pathogenicity, from which six were expressed in endophytism. The genome of P. chlamydospora also contains 54 homologues to Pth11-like G protein-coupled receptors (GPCR), 27 small GTPase regulators, and 12 Rab GTPase activators, all of which presumably regulate its endophytic behavior. Similarly to entomopathogens or plant pathogens, the P. chlamydospora genome contains 153 genes for protein kinases involved in the regulation of cell and metabolic processes, from which 75 had matches in the PHI database. Together with PKs, genes coding for histidine kinases (HKs) were identified, nearly all which had homologous in the PHI database. From these 14 were expressed during endophytism (Larriba et al. 2014). All together these results provide information for understanding the molecular mechanism involved in the multitrophic lifestyle of P. chlamydospora.

Recently, a fusion PCR-based deletion method was developed for P. chlamydosporia, using the split marker strategy and PEG-mediated protoplast transformation. The authors were capable of generating three stable deletion mutants resistant to neomycin (G418 sulfate) of genes induced during infection of nematode eggs by P. chlamydosporia: one chitinase (VFPPC_01099) and two protease genes (VFPPC_10088 and VFPPC_06535) (Shen et al. 2014). After screening ~100 mutant candidates by PCR, the average rate of gene knockout was 13 %. This method resulted in an efficient homologous gene knockout strategy for P. chlamydosporia, which together with the availability of the genome sequences opens the opportunity for high-throughput genetic analysis in this fungus (Shen et al. 2014).

VII. Concluding Remarks

The last decade has resulted in a burst of scientific research on nematophagous fungi, fueled by the advance in techniques for genome sequencing and transcriptome analysis. Thus, the knowledge that has previously accumulated in relation to the structural characteristics of the specialized structures produced by nematode-trapping fungi during the interaction with their hosts and even in the life cycle and ecology of this organisms, and the now obtained complement by “-omics” data on several species, renders the nematophagous fungi an attractive subject to investigate their molecular physiology in detail. Despite all the above-mentioned success, clear-cut evidence for the involvement of the described virulence factors is still lacking. In this regard, it will be of crucial importance to develop a toolbox for the genetic manipulation of nematophagous fungi, as is now the standard for other filamentous fungi such as the model systems Neurospora crassa and Aspergillus nidulans, but not for many plant pathogenic and industrially used fungal species. The availability of techniques for high throughput preparation of genetically manipulated strains would stimulate the analysis of whole gene groups, and the results to be obtained thereby could be used to generate improved biological control fungi against nematodes by genetic engineering. In addition, the analysis of different isolated structures such as knobs, trapping nets, appressoria, and other structures will help us better understand their role in the parasitic process as infection structures.