Truffle trouble: what happened to the Tuberales? (original) (raw)

Introduction

Fungi pursuing the truffle strategy by producing underground sporocarps have long been recognized as a polyphyletic group with representatives in the former Zygomycota now Glomeromycota (Endogone, Glomus a.o.), Ascomycota, and Basidiomycota. Those with asci were at one time all placed in the Tuberales (e.g. Tulasne and Tulasne, 1851, Fischer, 1897, Knapp, 1950, Hawker, 1954, Eckblad, 1968, Korf, 1973a). Nannfeldt (1946) wrote: ‘The question is raised whether Tuberineae is monophyletic or whether it is composed of different operculates that have evoluted

±

convergently into hypogeous forms.’ Malençon (1938) also advanced ideas about the evolution of truffles and their transformation from epigeous apothecial species to hypogeous truffles, but, as pointed out by Burdsall (1968), his system relied too heavily on macroscopic features. Korf (1973b) discussed the evolution of convoluted pezizalean forms, both above and below ground, and although he accepted the Tuberales, he indicated that at least some of the taxa were derived along various evolutionary lines within the Pezizales. He considered Tuberales to be a biological unit rather than a phylogenetic one. Trappe (1971) published a similar statement, and finally Trappe (1979), proposed that the order be abandoned, with one major part being moved to the Pezizales and just Elaphomyces to the Elaphomycetales. Fischer (1897) had earlier referred Elaphomyces to the ‘_Plectascineae_’ but alongside the Terfeziaceae. Later (Fischer 1938), Terfeziaceae reappeared within the Tuberales. Trappe (1979) kept some hypogeous lines, as families, within his Pezizales, but other hypogeous taxa were placed alongside epigeous species in various mixed families. Burdsall (1968) had already convincingly merged one tuberalean genus (Geopora) with the pezizalean genus Sepultaria. Eckblad (1968) gave many clear arguments for not accepting the Tuberales but, nevertheless, concluded the opposite. In the first Outline of the Ascomycetes (Eriksson 1982) Tuberales (with Geneaceae, Terfeziaceae, and Tuberaceae) were relegated to synonymy of Pezizales. Ainsworth & Bisby's Dictionary of the Fungi (Hawksworth 1983) likewise abandoned the use of Tuberales and listed the order under Pezizales (and Elaphomycetales). Trappe's hypothesis was tested in a long-lasting study of the ultrastructure of pezizalean taxa guided by Kimbrough and summarized in Kimbrough (1994), that for example, led to the placement of Hydnobolites in the Pezizaceae, based on both cytological and ultrastructural features of asci and ascospores. Also the placement of Barssia in the Helvellaceae followed from these studies. The most important character used was the morphology of the complicated septal pore-apparatus at the base of the asci (Kimbrough 1994). Another prominent feature, the number of nuclei in the mature spores that originated in Berthet's (1963) studies on epigeous Pezizales, was also taken into account when trying to delimit natural groups of truffles (e.g. Berthet, 1982, Donadini, 1986a, Donadini, 1986b). With the onset of the molecular taxonomy era, these early hypotheses have gradually been confirmed and expanded upon, or in some cases, corrected (e.g. O'Donnell et al., 1997, Norman and Egger, 1999, Percudani et al., 1999, Hansen et al., 2005, Perry et al., 2007). In a comprehensive treatment of European (mainly Italian) truffles Montecchi & Sarasini (2000) refer former Tuberales taxa to Elaphomycetales, with just Elaphomyces, and Pezizales with seven families: Pezizaceae (four genera), Pyronemataceae (four genera), Geneaceae (two genera), Helvellaceae (three genera), Balsamiaceae (two genera), Terfeziaceae (four genera) and Tuberaceae with two genera. Although, they cite recent molecular results, they have chosen a conservative approach by following the systems proposed in Trappe (1979) and Pegler et al. (1993). One group of researchers (Parguey-Leduc et al., 1990, Parguey-Leduc et al., 1987b, Janex-Favre and Parguey-Leduc, 2003) proposed to accept Tuberales based mainly on the genera Tuber and Terfezia that were considered closely related, mostly based on a perceived different development of asci and ascospores. van Brummelen (1994) gave a summary of the arguments put forward up to that time. Eriksson (2006b), influenced by data published by e.g. de Hoog et al. (2005), discussed what to do nomenclatorily if Pezizales are restricted to Pezizaceae. Although Tuberales are a possible choice, he proposed to find another name. Currently, however, there is no supported molecular phylogenetic evidence that suggests Pezizaceae are not part of the Pezizales (the Pezizaceae are supported as monophyletic by a BS value of 100

%, but the relationships among the included families in e.g. de Hoog et al. (2005) are without support).

The purpose of this paper is to review morphological and biological traits, and the systematics of the passively dispersed, more or less hypogeous Pezizales. Using all currently available LSU sequences from pezizalean truffles, in analyses with a broad sample of epigeous pezizalean taxa, we will further investigate the phylogenetic relationships and evolution of these truffle fungi. Ascomycetous truffles, which are now considered to be non-pezizalean (Elaphomyces, Eurotiomycetes), are not treated in detail. The taxonomic position of all accepted taxa at and above generic level are given and compared with previous classifications. The accepted classification is based on molecular phylogenetic analyses and morphological characters.

Ascomycete truffles can be defined as producing sporocarps below or at ground level and with a simultaneous loss of active spore dispersal. In several genera, for example Geopora and Helvella, species with intermediate characters can be found. Also Sarcosphaera coronaria is an example of a fungus that has nearly become a truffle. It forms apothecia below ground and often opens by a rather small aperture, but as the spores are actively ejected it can still be classified as a “cup fungus”. The genus Caulocarpa was based on such hypogeous Sarcosphaera ascomata (Trappe 1975c). Although some species tend to produce sporocarps in or on the litter, we still group them with the truffles as long as they have lost active spore dispersal. Glaziella and Paurocotylis are good examples.

The ascomata are typically fleshy but can be quite hard and cartilaginous. An outer rind (peridium) is often present and can be almost woody and sculptured. Even at maturity the spores do not become powdery, except in a few genera (e.g. Carbomyces) that are adapted to extreme xeric conditions. There is a continuous variation from truffles with a single cavity lined with a hymenium, often with a single opening, to truffles with intricate foldings or with pockets of asci in a firm gleba. Weber et al. (1997) defined three different types of hypogeous ascomata within the Pezizales: ptychothecia with persistent, recognizable hymenia and variously folded or even solid ascomata; stereothecia without hymenia and solid ascomata; and exothecia with external hymenia. None of these ascoma types can accommodate Paurocotylis and Glaziella. These genera produce ascomata that are hollow, without paraphyses, and furthermore, are unusual in being fully exposed at maturity. Hansen et al. (2001) reviewed the morphological features of the truffles considered to belong to the Pezizaceae. Those pezizalean species that have been studied in ontogenic detail, such as Tuber and Terfezia species (Janex-Favre & Parguey-Leduc 2003), start out as apothecial before folding occurs. The asci can at one end of the variation resemble those of operculate species being cylindrical with spores in one row or at the other end be completely globose with or without a pedicel and with a variable number of often very large spores. The ascospores vary in colour from hyaline to almost black, and in surface features from smooth and thin-walled to very thick-walled with intricate ornamentation. The ascus walls can be more or less layered and amyloid or inamyloid. The Pezizaceae are characterized by amyloid asci, but this feature appears to have been lost in many pezizaceous truffles (Hansen et al., 2001, Hansen et al., 2005).

Castellano et al. (1989) have published a slightly dated key to the spores of genera found in north temperate forests. An updated key, taking further characters into use, can be found on the Internet (http://natruffling.org/ascokey.htm), and an earlier printed version was published by Trappe & Castellano (1992). Trappe's (1979) synoptical key is still useful. In Europe two main illustrated accounts with keys are current (Montecchi and Sarasini, 2000, Pegler et al., 1993). Other important contributions include Lange, 1956, Lawrynowicz, 1988, and Montecchi & Lazzari (1993).

The names of pezizalean truffles are given sanctioned status if included in Fries (1821–1832) and should be used when available for a given taxon. In practice, however, another tradition has evolved, where Vittadini's (1831) much more accurate work on European truffles has been used as the de facto starting point for especially Tuber nomenclature. As Trappe (2001) has pointed out, it will be necessary to propose these Vittadini names for conservation over the sanctioned Friesian names in order not to disrupt the very long usage of these names for such economically important organisms.

Although, false truffles (hypogeous Basidiomycota) have been collected in extreme arctic environments, the true truffles would appear to have a more limited distribution, with a clear peak in diversity in temperate–subtropical, often rather dry climates. Although a high number of publications are dedicated to truffles, a reasonable picture of the diversity and distribution of the group has still not been achieved. Castellano et al. (2004) from one long Australian study suggest a figure of 600 species (although mainly of false truffles), most of which remain to be described. A part of this project was described in Claridge et al. (2000). Only ten of these species belong to the Ascomycota, and two apparently to undescribed genera. [See also the extensive review of Australian and New Zealand sequestrate fungi by Bougher & Lebel (2001).] Only Europe and parts of North America can be claimed to be reasonably well covered with respect to hypogeous fungi (Castellano et al. 2004). Distributions of European taxa are dealt with in Lawrynowicz (1991). Parts of Asia would seem to be equally rich in truffles. Africa and South America are apparently especially poor in hypogeous ascomycetes but be aware of the likely differences in sampling efforts in various regions. Verbeken & Walleyn (2003) in a checklist of subsaharan sequestrate fungi only reported one pezizalean species, Terfezia decaryi from Madagascar. In addition, three species are known from the southern dry lands of continental Africa, including the Kalahari (Marasas and Trappe, 1973, Ferdman et al., 2005). Two were separated as new genera (Ferdman et al. 2005). The third, Terfezia austroafricana, was listed as a member of Terfezia subgen Mattirolomyces and may require a new combination, as Mattirolomyces has been raised to generic rank. Although too little is known, it is fairly clear that many localized endemics are to be found among pezizalean truffles.

It has been hypothesized that all, or nearly all, truffles are passively dispersed with animal vectors, but there is very little experimental evidence to support this assertion. Various small mammals, including Australian marsupials (e.g. Claridge & May 1994), and voles and chipmunks in North America, collect and often hoard ascomata and by this activity are thought to play an active dispersal role (e.g. Fogel and Trappe, 1978, Maser et al., 1978). The stomach contents of voles and chipmunks have been found to contain over 70

% truffles. So far it has not been shown that pezizalean truffle spores can germinate after gut passage but in all likelihood they can. The volatile compounds truffles exude when ripe clearly substantiate the claim that these mammals are the key dispersal vectors. Also larger mammals such as boar and deer are well known for their ability to locate and digest truffles, and presumably, also act in a beneficial way to the truffles by their dispersal abilities. The volatile compounds may resemble pheromones (Claus et al. 1981) and can also be used in species recognition (e.g. Marin et al., 1984, Pacioni et al., 1990). Trappe (1977) and Trappe et al. (2001) have speculated that the ectomycorrhizal truffle partners migrated along with the rodent dispersers and the truffles themselves, many populations later becoming isolated as a result of continental drift. Many invertebrates (Diptera etc) also actively seek out truffles, but although a more parasitic aspect to this relationship can be postulated, additional dispersal ability cannot be ruled out. Even birds have been claimed to actively seek out truffles and possibly act as dispersal vectors (Alsheikh and Trappe, 1983b, Castellano et al., 2004). One example concerns the desert truffle Phaeangium (or Picoa) lefebvrei, which is believed to be dispersed by various species of desert-adapted larks, but also by cream-coloured courser and hoopoe. Another case deals with Paurocotylis pila, which at maturity has epigeous, orange–red fruit bodies coinciding with the fall of likewise bright-coloured Podocarpus fruits, known to be bird dispersed. Whether birds may also be involved in the dispersal of introduced British populations of Paurocotylis is not known.

Section snippets

Taxon sampling and alignment

To summarize and determine the phylogenetic placement of hypogeous taxa within Pezizales, LSU rDNA sequences from 48 hypogeous species (represented by 55 specimens) and 134 epigeous pezizalean species (represented by 141 specimens) were compiled for analyses (for sequence accession numbers, see online Supplementary Data Table 1). Sequences were selected to represent all sub-lineages within Pezizales based primarily on Hansen et al., 2001, Hansen et al., 2005, O'Donnell et al., 1997, and Perry

Phylogenetic relationships of truffles within lineage A

The LSU dataset of lineage A included 973 characters with 338 being parsimony informative. Parsimony analyses resulted in 1391 equally MPTs (1327 steps, CI

=

0.333, RI

=

0.678). The Pezizaceae are highly supported as monophyletic (PB 99

%, PP 100

%), with Ascobolaceae as the sister group (PB 97

%, PP 100, Fig 2). The strict consensus tree of all MPTs is highly resolved, but the deep level relationships are not well supported. Fourteen fine-scale lineages that correspond to the lineages resolved in

Lineage A

The Ascobolaceae have no confirmed hypogeous representatives but various truffle taxa have at times been placed in the family, e.g. Sphaerosoma and Ruhlandiella (as Muciturbo) (e.g. Castellano et al. 2004). See Fig 2, Fig 5.

Pezizaceae Dumort. 1829 (syn. Terfeziaceae E. Fisch. 1897)

The family Terfeziaceae as defined by Zhang, 1992a, Zhang, 1992b) are included in this family, but was accepted in the latest Dictionary of the Fungi (Kirk et al. 2001). Recent molecular results (e.g. Norman and

Phylogenetic relationships of truffles within Pezizales

Within the last 13

y molecular phylogenetic studies have gradually confirmed and greatly expanded our knowledge on a repeated evolution of ascomycetous truffles across Pezizales. The first study to address the controversial issue of the placement of Tuber was that of Landvik & Eriksson (1994a), who confirmed the placement within Pezizales, as predicted by Trappe (1979) and others. Elaphomyces was erroneously indicated to be nested within Pezizales (Landvik & Eriksson 1994a; but see Landvik &

Conclusion

In conclusion, the trend that started with abandoning the Tuberales, now robustly confirmed, has continued at the family level where ‘pure’ hypogeous monophyla have been reduced to a single taxon, the Tuberaceae. At least 15 independent origins of hypogeous forms within the Pezizales are supported by the LSU rDNA gene trees. Different types of hypogeous ascomata forms, infolded or chambered ptycothecia, solid ptycothecia and stereothecia, appear to have evolved multiple times independently with

Acknowledgements

Thanks to Jens H. Petersen for providing and editing photographs, and to Christian Lange, Jan Vesterholt, Johan Nitare, Juan Santos, and Manuel Tabarés for allowing us to use their pictures. Jim Trappe is thanked for help with references, advice on truffle matters in general and for supplying many of the specimens used by K.H. for molecular work. Rosanne Healy is thanked for providing specimens of Pachyphloeus. Donald H. Pfister is thanked for discussions and support of this work. The

Copyright © 2007 The British Mycological Society. Published by Elsevier Ltd. All rights reserved.