Species of Cryphonectria include some of the world's most important and devastating tree pathogens. Largely through the application of DNA sequence phylogenies, the taxonomy of these fungi has undergone major changes in recent years. Cryphonectria, including the chestnut blight pathogen Cryphonectria parasitica, has been restricted to species that have semi-immersed stromata, orange and pulvinate conidiomata, and one-septate ascospores. Other species of Cryphonectria with different morphological characteristics have been transferred to new genera that are strongly supported by phylogenetic data. This review represents a summary of the taxonomic changes to species of Cryphonectria sensu lato, and we discuss the impact that these changes might have on the understanding of their ecology, pathology and worldwide distribution.
Species in Cryphonectria (Diaporthales) are easily recognized on the bark of trees by their large and conspicuous orange stromata (Shear et al., 1917; Barr, 1978). These stromata (Fig. 1a–c) are semi-immersed and erumpent; conidiomata are also orange, pulvinate and stromatic. Species are further identified based on their fusoid to ellipsoid, one-septate ascospores and aseptate, cylindrical conidia (Myburg et al., 2004a). Cryphonectria spp. occur on a wide variety of woody hosts and they have a worldwide distribution (Table 1), which included the tropics as well as more temperate areas in the past (Shear et al., 1917; Kobayashi, 1970; Roane, 1986a). Many species in this group are economically significant and include some of the world's most serious tree pathogens.
Fruiting structures of Cryphonectria and related genera. (a) Ascostroma of Cryphonectria. (b) Longitudinal section through ascostroma of Cryphonectria. (c) Conidioma of Cryphonectria. (d) Ascostromata of Chrysoporthe. (e) Perithecial neck of Chrysoporthe (arrow) with brown surrounding tissue. (f) Conidioma of Chrysoporthe. (g) Conidioma of Amphilogia. (h) Conidioma of Rostraureum. (i) Conidiomata of Aurapex. Scale bars: a–h=100 μm, i=200 μm.
CMW, Forestry & Agricultural Biotechnology Institute (FABI), University of Pretoria, Pretoria; ATCC, American Type Culture Collection, Manassas, USA; CBS, Centraalbureau voor Schimmelcultures, Utrecht, the Netherlands; TFM:FPH, Forestry and Forest Products Research Institute, Danchi-Nai, Ibaraki, Japan (E or Ep refers to an isolate); MAFF, Microorganisms Section, MAFF GENEBANK, National Institute of Agrobiological Sciences (NIAS), Ibaraki, Japan; ITS, internal transcribed spacer; n/a, not available.
↵* Accession numbers given as sequences from the ITS region, and two regions from the β-tubulin genes amplified with primers 1a/1b and 2a/2b, respectively.
One of the best known species in Cryphonectria is Cryphonectria parasitica (Murrill) M. E. Barr (Fig. 2). It was introduced into North America where it devastated the American chestnut (Castanea dentata, Fagaceae, Fagales) throughout its natural range (Anagnostakis, 1987; Heiniger & Rigling, 1994). Cryphonectria parasitica was also introduced into Europe, where it resulted in a serious canker disease on Castanea sativa or the European chestnut (Heiniger & Rigling, 1994). Damage caused by this pathogen in Europe has, however, been much less severe than in North America due to reduced virulence imparted by a hypovirus (Heiniger & Rigling, 1994). The fungus and its associated viruses have been widely studied due to the significance of C. parasitica and the success of the biocontrol brought about by the hypoviruses (Nuss, 1992; Hillman & Suzuki, 2004; Milgroom & Cortesi, 2004).
Disease symptoms of chestnut blight caused by Cryphonectria parasitica. (a) A relic Castanea dentata tree still standing long after being killed by chestnut blight after its introduction into the USA. (b) Diffuse canker on Castanea dentata. (c) Die-back of branches with dead leaves still attached. (d) Sunken canker on Castanea dentata with fruiting structures visible. (e) Numerous stump sprouts from Castanea dentata roots of tree killed earlier by chestnut blight. (f) Multiple cankers on stump sprouts. (g) One of the few large and surviving specimens of Castanea dentata in the USA today.
Cryphonectria represents a relatively small genus of fungi. Until recently, the genus included only 10 species. Besides C. parasitica, these included C. gyrosa (Berk. & Broome) Sacc., which was the type species, C. radicalis (Schwein.: Fr.) M. E. Barr, C. nitschkei (G. H. Otth) M. E. Barr, C. macrospora (Tak. Kobay. & Kaz. Itô) M. E. Barr, C. havanensis (Bruner) M. E. Barr, C. longirostris (Earle) Micales & Stipes, C. coccolobae (Vizioli) Micales & Stipes, C. cubensis (Bruner) Hodges (Micales & Stipes, 1987) and C. eucalypti M. Venter & M. J. Wingf. (Venter et al., 2002). Other than C. parasitica, species such as C. cubensis (Fig. 3) and C. eucalypti are serious canker pathogens of trees (Old et al., 1986; Gryzenhout et al., 2003; Wingfield, 2003), but the majority of the species are saprophytes on wood (Roane, 1986b).
Disease symptoms of Chrysoporthe canker caused by various Chrysoporthe species. (a and b) Cankers on the trunks and bases of Eucalyptus trees. (c) Cross section of canker showing killed vascular tissue. (d) Stem breakage due to girdling cankers. (e) Die-back of Tibouchina spp. (f).
Species of Endothia (Diaporthales) have commonly been confused with Cryphonectria. This is due to their similar orange fruiting structures and a shared anamorph genus, Endothiella (Barr, 1978). Members of Cryphonectria and Endothia have also been recorded to share the same hosts and to occur in the same countries. For example, E. gyrosa (Schwein.: Fr.) Fr., E. singularis (Syd. & P. Syd.) Shear & N. E. Stevens, C. parasitica and C. radicalis all occur in North America (Table 1) on members of the Fagaceae (Roane, 1986a). Cryphonectria has also been treated as a synonym of Endothia for a large portion of its taxonomic history (Shear et al., 1917; Kobayashi, 1970; Barr, 1978; Roane, 1986a).
Based on DNA sequence comparisons and phylogenetic analyses, species of Endothia can be distinguished (Fig. 4) from those in Cryphonectria (Venter et al., 2002; Myburg et al., 2004a). This is strongly supported by morphological differences such as the aseptate ascospores and superficial stromata, which are characteristic in Endothia but not in Cryphonectria (Myburg et al., 2004a). Currently, Endothia includes only two species, E. gyrosa (type) and E. singularis (Myburg et al., 2004a). A third species with green stromata, E. viridistroma Wehm., most likely resides in a genus other than Endothia (Myburg et al., 2004a).
A phylogenetic tree showing the grouping of Cryphonectria and related genera. The tree was obtained from a combined DNA sequence data set of the ITS1, 5.8S rRNA gene and ITS2 regions of the ribosomal operon, and β-tubulin genes, respectively. Alignment was obtained using the web interface (http://timpani.genome.ad.jp/%7Emafft/server/) of the alignment program MAFFT ver. 5.667 (Katoh et al., 2002). Distance analyses using the Tamura–Nei model, which was shown by Modeltest to be the appropriate model, were employed. The following parameters were used: G=0.2301, freqA=0.1936, freqC=0.3312, freqG=0.2287, freqT=0.2465; rate matrix 1, 3.1964, 1, 1, 3.8818, 1. Bootstrap confidence levels (>70%, 1000 replicates) are indicated on the branches. Diaporthe ambigua, another species in the Diaporthales, was used as an outgroup taxon.
The application of DNA sequence comparisons revealed the fact that the taxonomy of Cryphonectria required revision. For example, Cryphonectria was shown not to be monophyletic (Fig. 4), but consisting of many species residing in newly recognized genera (Myburg et al., 2004a; Gryzenhout et al., 2005a). These newly recognized groups were supported by morphological features that were inordinantly diverse (Fig. 1) to warrant retaining these taxa in a single genus (Myburg et al., 2004a; Gryzenhout et al., 2005a). The primary aim of this review is to provide a summary of the recent changes to the taxonomy of species in Cryphonectria sensu lato, and to re-evaluate the host range, ecology and distribution of Cryphonectria species.
Phylogenetic comparisons have revealed that isolates labeled as Cryphonectria havanensis in Japan are identical to those of C. nitschkei (Myburg et al., 2004b). It has, furthermore, been confirmed (Myburg et al., 2004b) that C. nitschkei has a wide host range including five plant orders (Table 1). This fungus is restricted to the Far East and is known to occur in Japan, China (Myburg et al., 2004b) and Siberia, Russia (Vasilyeva, 1998).
European isolates originally labeled as C. radicalis represent two different species of Cryphonectria (Myburg et al., 2004a, b). This was determined independently based on morphology and DNA sequence comparisons (Fig. 4). Cryphonectria radicalis, defined by the type specimen from North America, corresponds morphologically to a phylogenetic group containing isolates from Greece, Italy, Switzerland and Japan (Myburg et al., 2004b). This has not been confirmed using phylogenetic analyses, because isolates that can be linked to C. radicalis in North America, do not exist. The other species has ascospores longer than those of C. radicalis sensu stricto, but specimens had also previously been labeled as C. radicalis (Myburg et al., 2004b). This species could possibly be linked to a second phylogenetic group consisting of isolates labeled as C. radicalis from Italy, France and Portugal, but this is also speculative as there are no isolates linked to herbarium specimens for this species. As the results of the DNA sequence-based and morphological comparisons cannot be linked, this species has not yet been described as a unique taxon.
Additional collections of C. radicalis sensu lato are clearly needed to resolve questions regarding its taxonomy. The phylogenetic placement of North American C. radicalis isolates has yet to be determined. It is also possible that another species of Cryphonectria, similar to C. radicalis sensu lato and referred to as E. radicalis mississippiensis Shear and N. E. Stevens (Shear et al., 1917), occurs in North America (Myburg et al., 2004b). Additional collections from Japan will also be required to determine whether the other, undescribed subclade encompassing isolates labelled as C. radicalis coexists with C. radicalis in Japan. This question arose because only a single isolate of C. radicalis from Japan, which was grouped in the subclade representing C. radicalis sensu stricto, has been available for study (Myburg et al., 2004b). Isolates of C. radicalis sensu lato will, however, be difficult to detect in nature because they are possibly displaced by or their occurrence is masked by C. parasitica, especially in Europe and North America (Hoegger et al., 2002).
One of the fungi associated with cankers on Eucalyptus that has come into consideration during taxonomic studies of Cryphonectria is C. eucalypti. This fungus was previously known as Endothia gyrosa, which is a well-known associate of stem and branch cankers on native tree species in the United States (Stipes & Phipps, 1971; Roane et al., 1974). Rather unusually, it is also a name that was applied to the causal agent of stem cankers on Eucalyptus in Australia (Walker et al., 1985; Old et al., 1986) and South Africa (Van der Westhuizen et al., 1993). One of the reasons that the fungus on Eucalyptus was treated as E. gyrosa is that it has orange stromata and aseptate ascospores, which made it similar to Endothia spp. (Walker et al., 1985). Phylogenetic and morphological studies of isolates from Eucalyptus and those of E. gyrosa from the USA (Venter et al., 2001, 2002) showed that these fungi are not the same, and that the fungi from Eucalyptus are more closely related to Cryphonectria. This led to a name being provided for the fungus on Eucalyptus in Cryphonectria as C. eucalypti (Venter et al., 2002).
Although phylogenetic studies have grouped C. eucalypti closely with Cryphonectria, the fungus is unlike other species of Cryphonectria, which all have single septate ascospores (Venter et al., 2002; Myburg et al., 2004a). Cryphonectria eucalypti has aseptate ascospores, and, thus, based on morphological characteristics, has been suspected to represent a distinct genus (Myburg et al., 2004a). This hypothesis is confirmed in the new phylogenetic tree presented in the present study (Fig. 4), showing that C. eucalypti groups separately from species in Cryphonectria sensu stricto, and a new genus should thus be erected for this species.
New type for Cryphonectria
A group of isolates from Elaeocarpus spp. in New Zealand, labeled as C. gyrosa and C. radicalis, was shown in DNA sequence-based phylogenetic analyses (Fig. 4) to group separately from other Cryphonectria spp. (Myburg et al., 2004a). Their distinct grouping was defined morphologically by the one- to three-septate ascospores, conical and superficial conidiomata (Fig. 1g), and conidia of variable size (Myburg et al., 2004a; Gryzenhout et al., 2005b). These isolates are also unique in being associated with root cankers on Elaeocarpus spp. in New Zealand (Pennycook, 1989).
In resolving the identity of the fungus defined by the isolates from New Zealand, it was realized that the fungus shared the same morphology as specimens of C. gyrosa from Sri Lanka (Myburg et al., 2004a; Gryzenhout et al., 2005b, c). Because C. gyrosa was commonly recognized as the type of Cryphonectria (Barr, 1978), this implied that isolates residing in the new clade from New Zealand should have had the name Cryphonectria, rather than those of the clade defining currently known Cryphonectria spp., including C. parasitica (ICBN, Art. 7.2, Greuter et al., 2000).
Studies on the type status of C. gyrosa showed that C. gyrosa was erroneously cited as the type of Cryphonectria (Gryzenhout et al., 2005c). The error arose because C. gyrosa was mechanically selected as type at the time, while the species included in the original sub-genus Cryphonectria, namely C. abscondita Sacc. and C. variicolor Fuckel, were ignored as choice for type (Gryzenhout et al., 2005c). This erroneous lectotypification of C. gyrosa and the separate grouping of the isolates similar to C. gyrosa from Cryphonectria spp. prompted a proposal to conserve the name Cryphonectria against a new type (Gryzenhout et al., 2005c). The proposal was accepted by the International Association of Plant Taxonomists (IAPT) Nomenclature Committee for Fungi (Gams, 2005). The original species C. variicolor and C. abscondita were not suitable as new types and C. parasitica was chosen as the type for the genus. This was justified largely because of its importance as a pathogen and the viruses that have been characterized from it (Gryzenhout et al., 2005c).
The clade containing C. gyrosa and the isolates from New Zealand could thus be described as a distinct genus with the new name Amphilogia (Gryzenhout et al., 2005b). Amphilogia gyrosa (Berk. & Broome) Gryzenh. & M. J. Wingf. occurs on Elaeocarpus spp. in both New Zealand and Sri Lanka. This genus includes a second species, A. major Gryzenh. & M. J. Wingf., which also occurs on Elaeocarpus spp., but is currently known only from the South Island of New Zealand (Gryzenhout et al., 2005b).
New genera for Cryphonectria spp.
Only C. parasitica, C. radicalis sensu lato, C. nitschkei and C. macrospora group in the well-supported clade that represents Cryphonectria (Fig. 4), and they all have the morphological characteristics that define Cryphonectria (Myburg et al., 2004a, b). The taxonomic position of C. havanensis and C. coccolobae remains confused because no isolates exist for these species that can be used in phylogenetic studies based on DNA sequence comparisons (Myburg et al., 2004a). The remaining species described in Cryphonectria have, however, now been transferred to newly described genera, or will be transferred soon. These new genera have largely been recognized based on DNA sequence data (Fig. 4). Robust morphological characters also support the phylogenetic grouping of these new genera (Fig. 1).
Chrysoporthe is a new genus that has been described to house isolates of C. cubensis from various parts of the world (Gryzenhout et al., 2004). The fungi previously collectively known as C. cubensis are some of the most serious canker pathogens (Wingfield, 2003) of commercially grown Eucalyptus (Myrtaceae, Myrtales), and they are also pathogenic to other plant genera in the Myrtales (Fig. 3; Table 1). Species in Chrysoporthe are all characterized by limited ascostromatic tissue covering the perithecial bases, long perithecial necks covered in black stromatic tissue, and superficial, generally pyriform and fuscous black conidiomata (Fig. 1d–f; Gryzenhout et al., 2004; Myburg et al., 2004a). The anamorph genus Chrysoporthella has been described for asexual structures of Chrysoporthe (Gryzenhout et al., 2004).
Chrysoporthe cubensis isolates reside in two distinct phylogenetic subclades (Fig. 4, Gryzenhout et al., 2004). One subclade consists of isolates from South and Central America, and Eastern Africa (Gryzenhout et al., 2004), while the other subclade includes isolates from South East Asian countries, Australia, Tanzania and Hawaii (Myburg et al., 2003). Fungi in these two subclades are indistinguishable from each other morphologically, although they represent two geographically distinct groups (Gryzenhout et al., 2004). Population-level techniques will most likely be required to determine whether fungi in these two subclades represent distinct species.
A fungus pathogenic to Terminalia ivorensis and Terminalia superba (Combretaceae, Myrtales) in Ecuador has recently been discovered and characterized (Fig. 4). This fungus has been described in the new genus Rostraureum and it is referred to as R. tropicale Gryzenh. & M. J. Wingf. (Gryzenhout et al., 2005a). Rostraureum (Fig. 1h) is characterized by superficial, orange, rostrate conidiomata with long necks, and semi-immersed ascostromata with little stromatic tissue, except for a white sheath of tissue around the perithecial necks. Cryphonectria longirostris also has these characteristics, but can be distinguished from R. tropicale based on conidial size (Gryzenhout et al., 2005a). For these reasons, C. longirostris has been transferred to Rostraureum (Gryzenhout et al., 2005a). In contrast to R. tropicale, there is no evidence to suggest that R. longirostris (Earle) Gryzenh. & M. J. Wingf. is pathogenic (Earle, 1901).
A newly discovered genus related to Cryphonectria and Chrysoporthe
Extensive sampling of fungi similar to Chrysoporthe and Cryphonectria has led to the discovery of a new fungus, Aurapex penicillata Gryzenh. & M. J. Wingf nom. prov., from Colombia. This fungus is associated with canker and die-back symptoms on native Melastomataceae (Tibouchina, Miconia) and Eucalyptus trees (Gryzenhout et al., 2006). Morphologically, this fungus is similar to Chrysoporthella, with black, superficial and pyriform conidiomata (Fig. 1I). It can, however, be distinguished from Chrysoporthella based on the different tissue organizations in the two fungi and because the tips of the conidiomatal necks are orange as opposed to necks that are uniformly black in Chrysoporthella (Fig. 1f). Phylogenetically, the fungus groups alone but close to Cryphonectria and allied genera (Fig. 4). No teleomorph has been found for A. penicillata, and it was thus described as the mitotic genus Aurapex nom. prov. residing in the Diaporthales.
Position of Cryphonectria and allied genera in the Diaporthales
Large subunit DNA sequence comparisons of an extensive collection of genera in the Diaporthales have revealed that species of Cryphonectria, Endothia and Chrysoporthe form a well-supported and distinct clade in the order (Castlebury et al., 2002). This clade has been referred to as the Cryphonectria–Endothia complex. Castlebury (2002) proposed that the unique grouping of these isolates in the Diaporthales is supported by morphological features such as stromatic tissue that is orange at some stage in the life cycle of the fungus, and the tissue that turns purple in 3% KOH and yellow in lactic acid. The unique grouping and morphological characteristics of these genera suggest that Cryphonectria and related genera most likely represent a new family in the Diaporthales.
Recent taxonomic revisions have restricted the name Cryphonectria to only seven species. Other species related to Cryphonectria now reside in the three genera Chrysoporthe, Rostraureum and Amphilogia. The species conclusively shown to belong in Cryphonectria include C. parasitica, which is now also recognized as the type of the genus, C. nitschkei, C. radicalis sensu lato and C. macrospora. In DNA sequence comparisons, these species reside in a distinct clade representing Cryphonectria sensu stricto. Of these Cryphonectria species, only C. parasitica is a serious and primary plant pathogen (Anagnostakis, 1987).
Cryphonectria havanensis, C. coccolobae and C. eucalypti currently retain their position in Cryphonectria. However, the taxonomic relationship of C. havanensis and C. coccolobae with species in the Cryphonectria clade must still be clarified. Likewise, C. eucalypti should reside in a taxon apart from Cryphonectria because of its distinct ascospore morphology (Myburg et al., 2004a) and phylogenetic data presented for the first time in this study.
Cryphonectria spp., as they are now recognized, occur in temperate areas of Asia, Europe and North America (Table 1). Only C. havanensis and C. coccolobae occur in the Caribbean area, while C. eucalypti is known from Australia and South Africa (Table 1). C. parasitica, C. radicalis sensu lato, C. nitschkei and C. macrospora occur on a wide range of woody plants residing in five plant orders (Table 1). These species are all known from Japan, and it seems probable that this part of the world represents a centre of origin for these fungi. Only C. radicalis sensu lato occurs naturally in Europe (Myburg et al., 2004b), and additional species remain to be described in this group. Cryphonectria radicalis is also known in the USA, based on herbarium specimens from 1828 (Myburg et al., 2004b), although its presence in that country needs to be confirmed based on fresh specimens and DNA sequence comparisons.
During the course of the last decade, wide-ranging changes have been made to the taxonomy of Cryphonectria and related genera. These have emerged largely from studies on species such as Chr. cubensis that causes a serious canker disease on Eucalyptus and related plants (Wingfield, 2003). These studies commenced during the time when DNA sequence comparisons and the phylogenetic species concept was emerging as a dominant taxonomic approach, and they would not have been possible without them. Taxonomic studies described in this review provide a framework that should lead to a better understanding of the important tree pathogens residing in this group. They are also likely to lead to the discovery of additional species and to promote a more lucid understanding of the global distribution of invasive or potentially invasive tree pathogens.
We are grateful to Dr Scott Enebak who provided us with photographs of chestnut blight. Financial support for the various studies that made this review possible was provided by the National Research Foundation (NRF), members of the Tree Protection Co-operative Programme (TPCP), the THRIP support programme of the Department of Trade and Industry, and the Department of Science and Technology/NRF Centre of Excellence in Tree Health Biotechnology, South Africa.