Introduction

In the opinion of leading contemporary fungal taxonomists (Moncalvo and Ryvarden 1997; Ryvarden 1985), Ganoderma P. Karst. 1881 has been regarded as an important, but also very difficult genus among the “polypores”, i.e., Agaricomycotina that grow saprobically or parasitically on wood of seed plants and are able to form conspicuous, bracket-like sporocarps. These fungi are now mostly classified in the order Polyporales. Some Ganoderma species have been used in Asia for more than 2,000 years for their medicinal properties and are nowadays cultivated there at an industrial scale (Jong and Birmingham 1992; Wasser and Weis 1997). Therefore, products made from Ganoderma have attained an enormous economic value (Lai et al. 2004). On the other hand, some tropical Ganoderma species also cause great damage in agriculture, e.g., as parasites of oil palms (Paterson 2007). Others may harm ornamental and forest trees and cause root and stem rot (Adaskaveg and Gilbertson 1986).

The present paper is intended to provide an overview of the taxonomy of Ganoderma and related taxa for non-specialists. Some basic knowledge, however, is essential to understand the problems that are discussed here. For instance, the priority of fungal names normally follows stringent rules, which are mainly based on historical aspects. If the same fungus has been described twice, the older name will be given preference and the younger name be treated as a synonym. Exceptions can be made, e.g. in case an older synonym will be discovered for a well-known name that has been in common use for a very long time, in which case it can be retained via a ‘conservation’ proposal. We have provided all species names not only with author citation but also with the year of publication, and hope it will thus be easier to follow our line of reasoning.

Taxonomic and nomenclatural history of Ganoderma and Ganodermataceae

The family Ganodermataceae was introduced by Donk (1948) with Ganoderma (Curtis) P. Karst. 1881 as type and is presently classified in the order Polyporales, with five currently accepted genera (Table 1). Another genus that possibly belongs here is Trachyderma Imazeki 1952, illegitimate because of Trachyderma Norman 1853 (a genus of lichenised Ascomycota), although currently regarded as a synonym of Ganderma pending re-evaluation of the type species.

Table 1 Genera of the family Ganodermataceae with species number and synonyms
Full size table

The genus Ganoderma was created by Karsten (1881) to accommodate a single species, Ganoderma lucidum (Curtis) P. Karst. 1881. Soon, Patouillard (1887) transferred other species to Ganoderma, and Karsten (1889) introduced the genus Elfvingia for non-laccate Ganoderma species, while Patouillard (1889) created section Amauroderma for species with spherical and subspherical basidiospores with uniformly thickened walls. In the 20th century, other mycologists developed different generic and species concepts. Murrill (1905) introduced Tomophagus for Polyporus colossus Fr. 1851, and Lloyd rejected Ganoderma and recombined its species epithets in Polyporus or Fomes. Later Torrend (1920), Donk (1933, 1948) and Imazeki (1939) greatly contributed to our knowledge of the genus and its close relatives. Donk (1948) also segregated the Ganodermataceae as a separate family from the other polypores, his concept based on spore colour and morphology. Other authors, for example Steyaert (1972) and Corner (1983), added additional genera, subgenera or new species. More recent studies employed molecular phylogenetic methods and shed some new light on natural relationships. Nevertheless, the affinities of Ganoderma to other genera of the Polyporales remain to be clarified (Hong and Jung 2004).

As proposed by Imazeki (1952), Ganoderma is presently divided into two subgenera: subgen. Ganoderma, comprising laccate species with a palisade of inflated hyphae at the pileus surface, and subgen. Elfvingia for non-laccate species where the palisade of club-shaped cells is absent in the surface of the basidiomes. The type species of Ganoderma, G. lucidum (for typical basidiomes see Fig. 1), was previously known as Polyporus lucidus (Curtis) Fr. 1821 and the basionym, Boletus lucidus Curtis 1781, dates back to the late 18th century. The British mycologist Curtis originally described material that was collected from hazel (Corylus avellana) in Peckham Common, an area of south London, but the type specimen is apparently lost. Nevertheless, it is important to note that G. lucidum is clearly a taxon of European origin, which has strong implications on the taxonomy of the Asian traditional medicinal mushrooms.

Fig. 1
figure 1

Basidiomes of Ganoderma lucidum. Photo taken by Thomas Læssøe in Denmark, 2006

Full size image

The subgenus Elfvingia, typified by G. applanatum (Pers.) Pat. 1887 (synonym Elfvingia applanata (Pers.) P. Karst. 1889; basionym Boletus applanatus Pers. 1789 (see typical basidiomes in Fig. 2), was eventually thought to have a worldwide distribution. However, the type material originates from Europe and is preserved at the Rijksherbarium Leiden, Netherlands. Species of subgenus Elfvingia often produce rather large and conspicuous basidiomes and are thus much more frequently encountered in the field than those of subgenus Ganoderma, but much less cultivated for medicinal purposes.

Fig. 2
figure 2

Basidiomes of Ganoderma applanatum. Photo taken by Jens H. Petersen in Sweden, 2002

Full size image

The other central species in the “G. applanatum – australe complex” of subgenus Elfvingia is G. australe (Fr.) Pat. 1889, which can be distinguished from G. applanatum by having larger basidiospores (Moncalvo and Ryvarden 1997). Ganoderma australe is common in the tropics and was never recorded from Europe. The type specimen of G. australe no longer exists and the only material deposited in the Royal Botanic Gardens Kew under that name is unlikely to correspond to the original collection (Ryvarden and Gilbertson 1993). Ganoderma lipsiense (Batsch) G.F. Atk. 1908, also belongs to this complex, and has been treated by some authors as the correct name for G. applanatum. Some years ago, a proposal to conserve the basionym of G. applanatum (Boletus applanatus Pers. 1800), against the earlier name, Boletus lipsiensis Batsch 1796 was approved by the Nomenclature Committee for Fungi to settle this problem. Therefore, G. applanatum is the correct name for this taxon.

Morphological characteristics and host affinities

According to Ryvarden (2004), the genus Ganoderma has the following salient characteristics: Basidiomes annual or perennial, stipitate to sessile; pileus surface with a thick, dull cuticle or shiny and laccate with a thin cuticle or cuticle of clavate end cells; context cream coloured to dark purplish brown, soft and spongy to firm-fibrous; pore surface cream coloured, bruising brown, the pores regular, 4–7 per mm; tube layers single or stratified, pale to purplish brown; stipe when present central or lateral; hyphal system dimitic; generative hyphae with clamps; skeletal hyphae hyaline to brown, non-septate, often with long, tapering branches; basidia broadly ellipsoid, tapering abruptly at the base; cystidia absent; basidiospores broadly to narrowly ellipsoid with a truncate apex and apical germ pore, wall two-layered, the endosporium brown and separated from the hyaline exosporium by inter-wall pillars, negative in Melzer's reagent, 7-30 μm long.

The family Ganodermataceae can be distinguished macroscopically from morphologically similar basidiomycete genera with bracket-like basidiomes, such as Fomes and Fomitopsis, by its brown (vs. white) spore deposit. Within the Ganodermataceae the distinction between the genera is sometimes not simple when comparing macromorphological characteristics. In general, the spores are a reliable characteristic for separating the different genera. For instance, Humphreya has peculiar spores with a distinct reticulate alveloid or honeycomb pattern (Ryvarden and Johansen 1980). Amauroderma can be distinguished from Ganoderma by growing on the ground out of buried roots, being apparently parasitic, and the truncate spores of Ganoderma (Ryvarden 2004).

For species segregation in Ganoderma, basidiospore morphology and surface texture of the pileus are regarded to be the most important discriminative features (Steyaert 1980; Zhao 1989; Moncalvo and Ryvarden 1997; Tham 1998). Other micro- and macromorphological characters may be variable and can thus not be used as reliable characters (Hong and Jung 2004; Douanla-Meli and Langer 2009). The presence of chlamydospores is an important characteristic for species identification, as it is also corroborated by molecular phylogenetic studies. Only a few Ganoderma species, such as G. colossus, G. subamboinense and G. weberianum, contain chlamydospores in their basidiomata (Douanla-Meli and Langer 2009; Hong and Jung 2004).

As judged mainly from the chorological data available through meticulous field work in Europe and other parts of the Northern hemisphere, the apparent host affinities can often also be of diagnostic value for species determination. For instance, several Ganoderma species grow on different angiosperm hardwoods, while others (e.g., G. tsugae Murrill 1902 and its relatives) seem to be constantly associated with conifers. Some species even seem to be closely associated with a certain plant genus (e.g. G. carnosum and Abies), while others, including G. lucidum in Europe, may occur on a broader range of “substrates”. The host plant, however, cannot always be determined with certainty, since Ganoderma basidiomes are often found on heavily decayed wood in mixed forests. Needless to say, the host tree species have often not been reported for many species in tropical countries that were encountered in forests where even new species and genera of plants remain to be discovered.

Species diversity and biogeography

Ganoderma is a cosmopolitan genus with a wide distribution in tropical and temperate regions, but its true diversity and the biogeography of its species still remains to be established. Diverse species concepts and inadequate specimens make it difficult to get an overview of the real distribution for many species (Moncalvo and Ryvarden 1997). Due to frequent misidentifications, determinations need to be considered with caution. Despite these facts the diversity of Ganoderma has been studied thoroughly for a few countries. Smith and Sivasithamparam (2000a, b; 2003) studied the genus in Australia, whereas Gottlieb and Wright (1999a, b) dealt with the South American species. These studies contributed significantly to our understanding of the genus and provided a vague perception of the real distribution of many species. Nevertheless, some species could not be studied properly due to the absence of authentic material (Moncalvo et al. 1995a; Smith and Sivasithamparam 2000a) and some parts of the world appear to be insufficiently investigated, e.g. tropical Africa (Ryvarden and Johansen 1980; Douanla-Meli 2007). In general, the tropic regions are problematic with a lack of material and dubious identifications (Corner 1983; Moncalvo et al. 1995a). One reason for the high diversity in the tropics and the consequential complexity of identifications may be that the genus has its origin there. The high phenotypic plasticity and the results of molecular phylogenetic studies both indicate that Ganoderma is a young genus that has only recently originated from the tropics and is not highly specialised. The species of Ganoderma in the temperate regions are still in a spreading phase (Moncalvo et al. 1995a). This was later confirmed by Moncalvo and Buchanan (2008), with a substitution rate calculation of an ITS dataset, leading to an estimated origin of the genus dating back only ca. 30 million years.

While different sources provide contradictory information about species numbers (Table 1), Ryvarden (1995) proposed a rather broad concept and suggested to reduce the number of species, as he found that the same taxon has often been described as a new species in two or more independent studies. A world monograph, relying on comparative studies of all type materials, is still not available. Nevertheless, intensive morphological studies and the use of molecular data has thus far resulted in the discovery of 14 new species in the 21st century (Ryvarden 2000, 2004; Smith and Sivasithamparam 2003; Ipulet and Ryvarden 2005; Torres-Torres et al. 2008; Douanla-Meli and Langer 2009; Welti and Courtecuisse 2010; Kinge and Mih 2011; Cao and Yuan 2013).

The data in Table 1 illustrate the multitude of names that were introduced in the history of the genus. On the other hand, European species names have often been used by those taxonomists who applied broad species concepts (and, as can be seen below, by non-specialists who reported secondary metabolites from these fungi). Ganoderma was, therefore, regarded to be “in a taxonomic chaos”, by Ryvarden (1991). Twenty years later the genus was still regarded to be “in need of a revision” (Smith and Sivasithamparam 2003), even after various new methods had been introduced, leading to substantial improvements in our knowledge of the relationships among Ganoderma and its close relatives. Studies of culture and mating systems (Adaskaveg and Gilbertson 1986, 1988, 1989), isozymes (Hseu 1990; Gottlieb et al. 1995; Smith and Sivasithamparam 2000b), but especially employing molecular phylogenetic methods, have led to a new understanding of the genus (Moncalvo et al. 1995a, b; Smith and Sivasithamparam 2000a; Hong and Jung 2004). Unfortunately, most of these studies did not deal with the same material and are thus not comparable. Some of these techniques require living cultures (or alternatively frozen tissue samples), but unfortunately the strains used in the aforementioned papers (and often, the corresponding basidiome specimens) have not been deposited in the public domain collections. As not even the most important taxa have been lecto-, neo- or epitypified by material that can be cultured, their relevance is limited. Similar problems are now arising in the Ascomycota, where there is also a great need for lecto-, neo- or epitypification before the application of a unified nomenclature following the One fungus-One Name concept will become feasible (cf. Rossman et al. 2013; Stadler et al. 2013)

The high morphological variability and the lack of unified taxonomic criteria led to a confusing taxonomical state. Several poorly defined species (e.g. G. lucidum, G. resinaceum, G. tsugae) have received multiple names (Douanla-Meli and Langer 2009). This is due to application of different species concepts from several taxonomists who had previously treated this genus (Moncalvo and Ryvarden 1997), but also to the high morphological variability of the pileus with the change of geographic and climatic variables (Ryvarden 1995; Kim et al. 2002). According to Gottlieb and Wright (1999b), not even the macromorphology of the pileus and the spore structure and shape are trustworthy because of a lack of unifying criteria. Reliable and important criteria to distinguish species seem to be the host range as well as the geographic distribution (Gottlieb and Wright 1999b; Hong and Jung 2004). To examine Ganoderma species, other methods such as DNA sequencing are helpful to support the morphological results (Hong and Jung 2004).

Molecular phylogenetic studies

With DNA sequencing a new method became available that is now one of the most important ones for phylogenetic studies. rDNA was established as a good marker (White et al. 1990) at a generic and subgeneric level (Lee and Taylor 1992; Wingfield et al. 1994; Cooke and Duncan 1997). Moncalvo et al. (1994, 1995a, b) and Bae et al. (1995) were the first ones who used internal transcribed spacer (ITS) and large subunit (LSU) rDNA sequences to distinguish between species of Ganoderma. The intraspecific variation for ITS was found to be <2 % in Ganoderma (Moncalvo et al. 1994, 1995b). Hong et al. (2002) and Hong and Jung (2004) and Douanla-Meli and Langer (2009) carried out further studies with mitochondrial small subunit (SSU) rDNA sequences. While Douanla-Meli and Langer used SSU only as a phylogenetic marker, Hong et al. (2002) studied also the secondary structure of the SSU domains. The results showed that the structure of variable domains can be used as a valuable marker and that the SSU region has three times more information than ITS among Ganoderma species. Their studies also revealed that the information of the SSU region is concentrated mostly in five variable domains (V1, V4, V5, V6 and V9). To obtain additional genetic information, additional DNA markers should be used (Yang and Feng 2013). Only a few other studies also amplified fragments of RPB2 (Wang et al. 2012), RPB1, and TEF1-α (Cao et al. 2012); the genes for the two largest subunits of the RNA polymerase II (RPB1 and RPB2) and the translation elongation factor 1-α (TEF1- α).

Studies on isozymes in species of Ganoderma could help to distinguish five different species from Australia (Smith and Sivasithamparam 2000b). They recognised genetic variability at the species and population level and were able to show the potential of this method. Other studies used isozymes for species and strain identification before, but without allelic interpretations (Park et al. 1986; Hseu et al. 1989; Lee and Lee 1991; Gottlieb et al. 1995, 1998). However, Gottlieb and Gottlieb and Wright (1999a) could find no correlation between isozymic pattern and morphological features.

Studies that dealt with the phylogeny of Ganoderma mainly covered the status of G. lucidum, but due to these and other molecular studies Ganoderma could be divided into different monophyletic groups or even taxa (Fig. 3). Moncalvo et al. (1995a) already established that the species on conifers (e.g. G. carnosum, G. oregonense, G. tsugae and G. valesiacum) clustered in a monophyletic group. In addition, some light was shed on the geographic separation and distribution of different species, supported by culture characteristics and presence of chlamydospores. Cao and Yuan (2013) worked with the results from Smith and Sivasithamparam (2000a) and additional sequences. Smith and Sivasithamparam (2000a) mostly studied material from Australia, but included data sets published earlier by Moncalvo. The calculations resulted in five terminal clades named “G. australe”, “G. cupreum”, “G. incrassatum”, “G. weberianum” and “Ganoderma sp.” Recently reported ITS sequence data even supported the separation of certain species. In their studies Ganoderma lucidum appears closely related to the only group that is still in need of intensive verification, the G. resinaceum group with species such as G. subamboinense, and G. weberianum. The species G. flexipes, G. curtisii, G. multipileum, G. philippii, G. tropicum, and G. lingzhi/sichuanense were fairly well resolved with a solid bootstrap support (Yang and Feng 2013). Regarding geographic distribution, Moncalvo and Buchanan (2008) working with a major dataset of ITS sequences from species of the G. applanatum-australe complex, could distinguish single geographic clades. Figure 3 includes ITS sequences from different studies and gives an overview on possible relationships of groups and single species. Ganoderma colossus represents the outgroup as a distinct species/genus, with G. resinaceum and its relatives next to it, but with a weak bootstrap support. The subgenus Elfvingia is clearly separated from the other Ganoderma species; among those, the Asian and European groups of G. lucidum appear in distant clades.

Fig. 3
figure 3

Maximum likelihood (ML) tree giving an overview about phylogenetic relationships in the genus Ganoderma. The alignment of 41 ITS sequences was generated with Clustal W (Thompson et al. 1994) and edited by GBlocks (Castresana 2000). The tree was calculated with RAxML 7.2.8. (Stamatakis 2006) and the nucleotide model GTR Gamma I using Geneious 7.0.6 (created by Biomatters). Numbers above nodes represent ML Bootstrap values from 1000 replicates. The origin of the sequence data has been compiled in Table S1 (Supplementary Information)

Full size image

Hong and Jung (2004) divided the genus via mt SSU rDNA sequences in six monophyletic groups that are well supported by host specificity and presence/absence of chlamydospores. The clade they named “G. applanatum group” contains non-laccate species, such as G. applanatum and G. lobatum. Hence, the division of the two subgenera based on phenotype-derived data was reflected very well by molecular phylogeny.

According to a comparison of geographic origins and the secondary structure models of variable domains of the small subunit rDNA (SSU), G. applanatum is polyphyletic, indicating the presence of a species complex, and should be investigated further to clarify the relationships in this group. The tropical strain of “G. applanatum” included is probably G. australe, which is in need of a careful neotypification. In Europe, G. australe was mistaken for the domestic G. adspersum (Smith and Sivasithamparam 2000a). The results of Hong and Jung (2004) indicate that the acceptance of a renamed Tomophagus might be justified. Ganoderma meredithiae in the G. meredithiae group (Hong and Jung 2004) does not produce chlamydospores and grows on conifers similar to the species in the G. tsugae group. Some species in the G. tsugae group do not clearly separate from each other, e.g. G. tsugae from North America and G. valesiacum from Europe seem to represent a single species. Species in the group of G. resinaceum such as G. pfeifferi and G. subamboinense, also produce chlamydospores (Hong and Jung 2004). “Ganoderma lucidum” strains from North America in this group might be conspecific with G. resinaceum from Europe. The worldwide distribution of G. lucidum was negated early (Moncalvo et al. 1995a), strains from Asia separate clearly from strains from North America and Europe (see chapter 1.4.).

With all the confusion about misapplied names and the resulting misperception concerning the distribution, the results from DNA sequencing brought a new insight. It could be shown that there is an extensive convergence or parallelism of morphological characters that has occurred during the evolution of the genus, several morphological differences may have occurred with little divergence time (Moncalvo et al. 1995a). These results were based on morphologically well-studied material, whereas many other published phylogenies were relying on gene sequences of which no voucher material had been deposited in the public domain for verification, and no details on the morphological identification were provided.

Ganoderma lucidum and ´Lingzhi´

The most famous species within the genus has been described in traditional Asian medicine under the Chinese names “Ling-zhi” “Chi-zhi” or “Rui-zhi” (Teng 1963; Liu 1974; Tai 1979; Ying et al. 1987; Zhao and Zhang 2000), which has often been associated with the “European” name, G. lucidum. In 1969 the fungus “Lingzhi” was cultivated successfully for the first time (Yu and Shen 2003) and currently Ganoderma products have an estimated global turnover of more than two billion US$ (Lai et al. 2004; Wachtel-Galor et al. 2004). It is commercialised as a dietary supplement, especially in Asia (Dai et al. 2009), and also used in cosmetics (Hyde et al. 2010).

Ganoderma lucidum sensu lato is circumscribed as a laccate species with basidiomes of varying colour, from orange-red to brownish or black, which are sessile or stipitate; the basidiome surface consists of pilocystidia. The context is plain or duplex, with white to brown colouring. The pores are light coloured, the basidiospores are double-walled with differently expressed spines or apically truncated projections (Moncalvo et al. 1995a). According to the literature, this species seems to be almost cosmopolitan, as it has been reported from Europe (Steyaert 1972; Ryvarden and Gilbertson 1993), Asia (Hong and Izawa 1994; Núñez and Ryvarden 2000; Zhao and Zhang 2000), America (Bazzalo and Wright 1982; Gilbertson and Ryvarden 1986), Africa (Ryvarden and Johansen 1980) and Oceania (McKenzie and Foggo 1989; Quanten 1997). However, these reports are largely based on applications of broad species concepts, e.g. where all laccate specimens samples were regarded as G. lucidum; even several different cultivated Asian species have also been mistaken for this fungus. This fact was also corroborated by molecular phylogenetic studies (Moncalvo et al. 1994, 1995a, b; Pegler and Yao 1996). The actual biogeographic distribution of G. lucidum is still not clear (Moncalvo et al. 1995a; Postnova and Skolotneva 2010). Unfortunately, the holotype collected in Europe was lost and attempts to collect a specimen to serve as epitype from the UK have so far been unsuccessful (Steyaert 1972; Moncalvo and Ryvarden 1997); the name was lectotypified by Steyaert (1961) with an illustration. An epitypification based on fresh material that can be cultured and subjected to DNA sequencing of its genome remains to be carried out to meet the goals of modern fungal taxonomy.

A comparison of molecular phylogenetic data showed that the Asian G. lucidum strains are not conspecific with the European ones (Moncalvo et al. 1994, 1995a, b). As the holotype represents a European taxon, the taxonomy of the medicinal mushrooms, especially those that are cultivated at a commercial scale in Asia needed to be revised, and this process is ongoing. Yang and Feng (2013) determined that G. lucidum from Europe and the Asian “G. lucidum” may both occur in China. However, owing to the fact that their argumentation was largely based on ITS sequences, which have been proven insufficient for segregation of species in many fungal taxa (Kuhnert et al. 2014), this remains to be confirmed by extensive morphological studies on a broad range of specimens and a multi-gene genealogy including gene loci that are different from ribosomal DNA. In a recent evaluation of xylariaceous Ascomycota, ITS sequences also often proved insufficient for segregation of species, whereas data derived from partial beta tubulin sequences provided a much better resolution that was also in agreement with the phenotypes, including morphology as well as secondary metabolite production (cf. Bills et al. 2012; Kuhnert et al. 2014; Stadler et al. 2014). These so-called “housekeeping genes” that can be used for a precise resolution can often only be obtained from cultures or after destruction of the specimens, hence it should be practical to obtain cultures from the respective material as well.

An additional, commercialized species from tropical areas and South China was meanwhile identified as G. multipileum Hou 1950 (Wang et al. 2009). Asian records of “G. lucidum” and G. lucidum from Europe differ morphologically in the presence of melanoid bands that makes the context dirty white in the Asian material. Moreover, the pileus is purplish red to reddish brown with a yellow pore surface and thicker dissepiments at maturity (Cao et al. 2012; Yang and Feng 2013). However, there are additional species that have to be considered, and it seemed practical to reach an agreement on how the most widely cultivated Ganoderma species in Asia should be named. Recently, Chinese mycologist groups have proposed different approaches to this matter: While Wang et al. (2012) proposed to use the name G. sichuanense J.D. Zhao and X.Q. Zhang 1983, Cao et al. (2012) proposed the new name G. lingzhi Sheng H. Wu, Y. Cao & Y.C. Dai 2012. The problem with the typification of G. sichuanense arose as the published DNA sequence of its holotype specimen turned out to be derived from a misidentification or from sample confusion in the laboratory. The DNA sequence showed strong homology to G. weberianum, a species that is morphologically quite different from the Chinese Lingzhi, and for which a specimen was also concurrently under study. Attempts to re-sequence the DNA of the holotype of G. sichuanense have failed (Cao et al. 2012; Wang et al. 2012; Yang and Feng 2013). An epitype for G. sichuanense was designated (Yao et al. 2013). This epitypification procedure was straightforward, thus making the new taxon G. lingzhi taxonomically superfluous because the rules of fungal nomenclature require that the oldest valid name of any given taxon should be given preference. Currently, the latter species is therefore regarded as a later synonym of G. sichuanense in Species Fungorum (http://www.speciesfungorum.org), and the nomenclatural dispute seems to have been solved. This example shows that, despite the undisputed value of PCR-based data for exploration of biodiversity research, special care must be taken, not to overestimate the value of molecular phylogenetic data in fungal taxonomy. Errors and confusions can easily occur, and all important taxonomic matters must be verified by morphological studies, preferably using multiple reference materials. DNA extraction of fungarium material is a destructive procedure and is not permitted by the terms of use of material from many international fungaria. Moreover, the DNA in basidiomes may degenerate over time, and basidiomes often only yield sufficient DNA to amplify the ribosomal DNA genes but cannot be used to obtain additional molecular data from other genes. It appears good practice, therefore, to deposit a living ex-type culture together with the type specimen. Such a culture can be preserved permanently and used for whole genome sequencing and other purposes such as further morphological studies. The fact that Yao et al. (2013) have considered all these aspects, including deposition of an ex-epitype culture, gives rise to some hope that the nomenclature of the Lingzhi has now been stabilised permanently. Images of type and authentic material of G. sichuanense (including a misidentified “paratype”) are depicted in Fig. 4. We propose that a similar procedure should be carried out for Ganoderma lucidum in the UK. Notably, the only alternative to “save” G. lingzhi would be a proposal proposal to the nomenclature committee for fungi (NCF), and chances are very slim that such a proposal would find approval. For good reasons, the introduction of new names for validly described species has in the past mostly been rejected, since this might possibly lead to an unstable situation in fungal nomenclature, which the user communities would certainly not appreciate.

Fig. 4
figure 4

Images of type material and other important specimens of Ganoderma sichuanense, taken at the Fungarium, Institute of Microbiology, Chinese Academy of Sciences (HMAS) with kind assistance of Prof. Y.-J.Yao & co-workers. a-c: Holotype of G. sichuanense (Acc. no. HMAS 42798): a. Basidiomes contained in the specimen. b. Close-up of laccate surface. c. Close up of hymenophor; d. Basidiomes of epitype specimen (Acc. no. HMAS 252081; cf. Yao et al. 2013); e. Basidiomes of the first cultivated strain of Ganoderma in China (Acc. no. 240103); f. basidiomes of another cultivated specimen (originating from Shandong Prov; Acc. no. HMAS 240178), illustrating the morphological plasticity of the basidiomes; g, h. Basidiomes of paratype specimen of G. sichuanense (Acc. no. HMAS 43728), which constitutes a different species, probably corresponding with G. weberianum (cf. Wang et al. 2012)

Full size image

Secondary metabolites of Ganoderma and their potential use in chemotaxonomy

The majority of known secondary metabolites from Ganoderma species (see Figs. 5, 6, 7 for selected chemical structures) belong to the class of lanostane-type triterpenoids, which are derived from lanosterol (1), such as ganoderic acids, lucidenic acids or applanoxidic acids (Kubota et al. 1982; Kikuchi et al. 1985; Ríos et al. 2012; Wang et al. 2007, Wang and Liu 2008; Iwatsuki et al. 2003; Mizushina et al. 1999; Chairul and Hayashi 1994; Tokuyama et al. 1991). Furthermore, ergostane-type steroids (e.g. ergosterol 2) and seco-lanostane-type triterpenoids (e.g. australic acid 3) were isolated frequently (Rösecke and König 2000; Weng et al. 2011; Leon et al. 2003; Yoshikawa et al. 2002), followed by pentacyclic triterpenes (e.g. friedelin 4; Turner and Aldridge 1983). Gradually, many of these compounds have been found in different species of Ganoderma, such as stellasterin in G. lucidum (Yokokawa 1987), G. amboinense (Lin et al. 1993), G. australe (Gerber et al. 2000) and G. lipsiense (syn. for G. applanatum; Rösecke and König 2000). The majority of these reports did not involve an in-depth characterisation of the morphology of the starting material and in many cases, no voucher specimens have been deposited that would allow for retrospective revision of the taxonomy. Therefore, the chemotaxonomic characterisation of a certain species based on metabolites from these structural classes is highly problematic.

Fig. 5
figure 5

Secondary metabolites of Ganoderma sp. with frequently occurring chemical structures

Full size image
Fig. 6
figure 6

Uncommon secondary metabolites from Ganoderma species

Full size image
Fig. 7
figure 7

Uncommon secondary metabolites from Ganoderma species

Full size image

Currently, only a few other natural product scaffolds are known from Ganodermataceae. These include farnesyl hydroquinones such as ganomycin A and B (5), which have been obtained from powdered basidiomes of G. pfeifferi (Mothana et al. 2000). However, even these compounds cannot be regarded as ‘specific’, since ganomycin B has been isolated also from G. colossus (El Dine et al. 2009), which might not even be retained in the genus, according to the molecular phylogeny presented in Fig. 3 . Moreover, other, highly similar derivatives (fornicins A-C 6) have been identified in G. fornicatum (Niu et al. 2006). Sato et al. (2009) reported novel farnesyl hydroquinone-triterpene conjugates. In ganosinensins A-C (7), isolated from G. sinense, ganodermanontriol is connected to the farnesyl hydroquinone unit via an ester bond at C-24 or C-26. An esterification during the extraction process is considered unlikely due to the mild conditions, but may also not be excluded (Sato et al. 2009). In addition to the conjugates, various alkaloids (sinensines A-E 8, Liu et al. 2011) and triterpenoids containing a four-membered ring (ganosinensic acids 9, Wang et al. 2010) were discovered from the species G. sinense. A biosynthetic pathway from lucidenic acid has been proposed for ganosinensic acids A and B (9) (Wang et al. 2010).

Colossolactones C-G, IV and VIII (10) were obtained from different specimens of G. colossus (Kleinwächter et al. 2001; El Dine et al. 2008a, b). Generally, they also belong to the group of triterpenoids, but notable in this case is the incorporated seven-membered lactone. Similar structures have been observed only in Tomophagus cattienensis (Hiena et al. 2013) and some plants of the Schisandraceae (Wang et al. 2006; Xue et al. 2011). Another unusual triterpenoid derivative, including a spiro-lactone moiety, is represented by austrolactone (11) from G. australe (Leon et al. 2003).

For G. tsugae, lanostanoids and triterpene glycosides (e.g. tsugaric acids) are primarily known (Lin et al. 1997; Su et al. 2000), but recently a novel benzofuran has been characterised by La Clair et al. (2011). Ganodone (12) was discovered in basidiomes collected in New York State (USA) and displayed cytostatic activity in different tumor cells in the micro- to nanomolar range (La Clair et al. 2011). Ganofuran B (13), isolated from a fungus named G. lucidum, also contains a benzofuran moiety and could conceivably be derived from ganomycin B. (Adams et al. 2010).

Further compounds can be assigned to the class of sesquiterpenes. Ganomastenols A-D (14), the earliest examples of cadinene derivatives from the genus Ganoderma, have been identified from cultures of G. mastoporum (Hirotani et al. 1995). Cryptoporic Acids H and I (15), drimane-type sesquiterpenes including three carboxylic acid groups and mainly occurring in Cryptoporus species (Hashimoto et al. 1989), were isolated from cultures of G. neojaponicum (a member of the G. lucidum complex, collected at Chiba, Japan (Hirotani et al. 1991)). Campos Ziegenbein et al. (2006) investigated volatile compounds of G. lucidum and found various terpenes such as R-(−)-linalool, S-(+)-carvone, α-bisabolol and the phenylpropanoid trans-anethol as major constituents.

In case this was verified by comprehensive chemotaxonomic work, the species G. applanatum could eventually be distinguished chemotaxonomically by the presence of benzopyranone derivatives such as “Ganoderma aldehyde” (16) and applanatins, which have been isolated from specimens originating from both, North America and Asia. (Ming et al. 2002; Wang et al. 2007). In addition, some indene and isochromanone derivatives, unfortunately named “applanatines A-E” (17), and therefore easily to be confused with the above mentioned applanatins, have been obtained from culture extracts (Fushimi et al. 2010). Another potential chemotaxonomic marker compound is ganodermycin (18) derived from cultures of a fungus named “G. applanatum“from French Guiana (Jung et al. 2011).

Most of the metabolites known from Ganoderma have been isolated from specimens assigned to “Ganoderma lucidum” (cf. Table 2). The secondary metabolism of other species was studied less extensively, and for some species such as G. weberianum, G. subamboinense or G. sichuanense no data seem to be available. In fact, a comprehensive study of type and other well-characterised specimens by means of HPLC profiling of standardised extracts from their basidiomes and cultures might even be helpful to settle the remaining questions in the taxonomy of the genus. A comparison of chemical traits has been rather helpful in many other groups of fungi (Frisvad et al. 2008), and in certain ascomycetes such as the Xylariales, chemotaxonomic data even proved to be phylogenetically informative (Bitzer et al. 2008; Stadler et al. 2010). In Xylariaceae, the characteristic compounds of the stromata remain stable for several decades, and according to our preliminary results, the same seems to hold true for the triterpenes in the pileus of Ganoderma, which allows for examination of a broad range of herbarium specimens. Since the triterpenoids are highly concentrated in the surface of the pileus, chemotaxonomic studies including HPLC-DAD/MS could even be extended to old type specimens. However, fresh material – even in larger quantities must be made available too, in order to serve for preparative extraction and isolation of the chemotaxonomically relevant compounds, so they can serve as standards. The cultures from the fresh specimens should also undergo chemotaxonomic studies and molecular phylogeny.

Table 2 Detailed information on the above mentioned secondary metabolites of Ganoderma species and references
Full size table

In summary, a procedure similar to the above mentioned polyphasic studies on Xylariaceae taxonomy appears feasible and practical, to resolve the taxonomy and phylobiogeography of Ganoderma, and to obtain a clearer picture on the taxonomic relevance of the numerous bioactive triterpenoids that are produced by this highly interesting fungi.