Myco-photobiontal selection in a Mediterranean cryptogam community with Fulgensia fulgida


  • A. Beck,

    Corresponding author
    1. Lehrstuhl für Pflanzensystematik, Universität Bayreuth, Universitätsstraße 31, D-95447 Bayreuth, Germany;
      Author for correspondence: A. Beck Tel: +49 (0)89 17861303 Fax: +49 (0)89 172638 Email:
    Search for more papers by this author
  • T. Kasalicky,

    1. Institut für Systematische Botanik, Ludwig-Maximilians-Universität München, Menzinger Straße 67, D-80638 München, Germany
    Search for more papers by this author
  • G. Rambold

    1. Lehrstuhl für Pflanzensystematik, Universität Bayreuth, Universitätsstraße 31, D-95447 Bayreuth, Germany;
    Search for more papers by this author

Author for correspondence: A. Beck Tel: +49 (0)89 17861303 Fax: +49 (0)89 172638 Email:


  •  To examine the mode of dispersal and photobiont selectivity of Fulgensia fulgida we investigated the photobionts associated with the lichens of a community related to the Toninio–Psoretum decipientis association.
  •  Photobionts and mycobionts were analysed using morphological (light microscopy) and molecular (internal transcribed spacer (ITS) and partial large subunit (LSU) nuclear ribosomal DNA sequencing) techniques.
  •  The thalli of the mycobiont of F. fulgida, which were identical in their ITS and partial LSU-sequence, contained two strains of Trebouxia asymmetrica, one of them also present in a thallus of Toninia sedifolia. Other co-occurring lichens, namely Squamarina lentigera, Catapyrenium michelii and Collema cristatum, were shown to have Asterochloris irregularis, Myrmecia and Nostoc as photobionts.
  •  It is suggested that F. fulgida is selective in its photobiont choice, that dispersal of the lichen has occurred by way of ascospores and subsequent independent relichenization events and that F. fulgida and T. sedifolia share the same photobiont pool. The results also suggest possible mechanisms of relichenization and strategies for future research on the selectivity of lichen bionts.


Lichens are prime examples of symbiotic associations. They were even referred to in the original definition of a symbiosis (De Bary, 1879), but the associations are still poorly understood. In the majority of cases (about 97%) it is unclear which species of algal organisms are involved. However, to answer questions concerning selectivity, coevolution and the mode of relichenization, knowledge of the identity of both partners is a prerequisite.

In sexually reproducing lichens, where fungal components propagate via ascospores, relichenization is suspected to be a necessary event in the life cycle of the system. For the establishment of a functioning, mutualistic symbiosis, germinating hyphae of the mycobionts need to meet and associate with compatible free-living or already lichenized photobiontal cells (Beck et al., 1998). But the likelihood of potential mycobionts meeting a compatible, aposymbiotic alga has been considered to be rather low. However, the chance of contact is considerably increased if a potential mycobiont is capable of forming a temporary aggregate with a less compatible biont and remains in that state until the association with a compatible photobiont is established and thallus formation initiated (Ott, 1987). In this way, a lichen mycobiont can survive and spread on the substrate for an extended period of time.

Where such interim photobionts are not present and relichenization events rare, a high mycobiontal spore input is necessary to assure the establishment of a lichen population. Such high input may be provided by parental thalli in the surrounding environment. In populations with a closely neighboured, genetically homogenous source of spores, the newly developing mycobionts can be expected to be genetically homogenous as well. Should the neighbouring source of spores be already genetically inhomogeneous or recombination between genetically different chromosomes from different germ cells has occurred during the process of meiosis in the parental asci, the offspring are expected to be genetically heterogeneous.

Genetic uniformity of the algal photobiont depends on the photobiont-referring selectivity of the mycobiont as well as on the availability of suitable photobiontal species at a particular stand. In the case of low selectivity and accessibility, more than one algal clone is expected to occur in lichen populations, even those comprising genetically uniform mycobionts. Such heterogeneity of the photobiont may be considered to reflect multiple relichenization events actually having taken place in the population. In studies on lichen populations, morphological, chemical, isoenzymatic and molecular traits of the fungal component have been used for the identification of the lichen individuals (Fahselt, 1996). However, the genetic traits and their variation of both the fungal and the algal components need to be examined to understand the dynamics of the whole systems.

In the present study, the ITS and partial LSU nrDNA polymorphism of fungal and the ITS nrDNA polymorphism of the algal bionts of a whole population of a sexually reproducing lichen species was examined. This was exemplified by a population of Fulgensia fulgida (Nyl.) Szatala (Teloschistaceae) in Southern France, growing in a stand which was covered by a lichen community related to the Toninio–Psoretum decipientis Stodiek association. F. fulgida is a frequently occurring species in the Mediterranean but nearly absent North of the Alps. It grows on calcareous ground in open grassland, usually directly on soil or over mosses. By contrast to other species of the genus, F. fulgida occasionally covers rock surfaces as well. The species is well characterised by its growth form, pyriform and clavate ascospores and the presence of depsidones in addition to anthraquinones.

Materials and Methods

Lichen samples

All investigated lichen thalli were taken from a stand of a Toninio-Psoretum decipientis-related community. The population selected was a rather close assemblage of thalli in a sociologically uniform stand of lichens, about 30 dm2. It was collected in France, Alpes Maritimes, 4 km northeast of Gourdon, on a 45° inclined, west-exposed boulder, which is elevated about 30 cm above the ground. Among the Fulgensia thalli present, five were selected and analysed with respect to their myco- and photobionts (photobiont clones M-99.021 to M-99.025). In addition, the photobiont of co-occurring thalli of Toninia sedifolia (Scop.) Timdal (B-207) and Squamarina lentigera (Weber) Poelt (M-99.109) were analysed in more detail. All material was collected by D. Triebel and G. Rambold (no. 6254 a–g) and has been deposited in the lichen herbarium of the Botanische Staatssammlung München (M).


Photobionts were isolated using a single-cell-manipulator as described in Beck & Koop (2001) and grown as described in Friedl (1989a). Culture strains of the isolated photobionts are maintained at the Lehrstuhl für Pflanzensystematik, University of Bayreuth. The photobionts were examined both in the lichenized and cultured states by standard light microscopic techniques. For identification, the isolated strains were compared with cultures of all known species of Trebouxia (Gärtner, 1985; Ettl & Gärtner, 1995), obtained from culture collections (SAG, Schlösser, 1994, UTEX, Starr & Zeikus, 1993).

DNA extraction, PCR and sequencing


For the DNA sequence analyses one photobiont clone of each lichen thallus was selected from the isolated lichen photobionts (Table 1). This was considered to be sufficient as all clones from one strain were found to be morphologically identical. Additionally four more clones, obtained from the thallus with the differing photobiont clone, were sequenced. DNA was extracted from log phase cultures using a Qiagen DNeasy Plant Mini Kit, following the standard extraction protocol as supplied by the manufacturer. ITS-regions were amplified using PCR with primers and under conditions as described by Friedl (1996). The ITS-region of the photobiont of Toninia sedifolia was amplified from DNA of two whole areoles using the photobiont specific primer AL1500bf (Helms et al., 2001) instead of NS7m. The PCR products were cleaned with Qiaquick (Qiagen GmbH, Hilden, Germany) spin columns, following instructions in the manual and sequenced directly over both strands using a Licor X4 automated sequencer. Sequencing primers were 1800F (Friedl, 1996) and ITS 4 (White et al., 1990).

Table 1.  Photobionts analysed from samples of collection GR 6254, fragments a–f
Clone no.Lichen speciesFragmentPhotobiont speciesNumber of clones analysed by LMSequenced cloneType of sequence
M-99.021Fulgensia fulgidaaTrebouxia asymmetrica13B2a
M-99.022Fulgensia fulgidaaTrebouxia asymmetrica14B4a
M-99.023Fulgensia fulgidabTrebouxia asymmetrica13A3, B2, B4, C2, C3b
M-99.024Fulgensia fulgidacTrebouxia asymmetrica15B3a
M-99.025Fulgensia fulgidafTrebouxia asymmetrica13A5a
M-99.109Squamarina lentigeradAsterochloris irregularis10  
B207Toninia sedifoliabTrebouxia asymmetricasequenced directlya

Data analysis

The sequences obtained were manually aligned with available Trebouxia ITS-sequences (Table 2), representing the major lineages within the genus Trebouxia as reported by Beck et al. (1998) and Helms et al. (2001), using the program GeneDoc ( After regions of ambiguous alignment and the 5.8S nrDNA had been excluded (positions 44–80, 152–185, 300–452 (5.8S nrDNA), 478–495 and 579–592; positions are given in respect to the sequence of T. arboricola SAG 219–1a, GenBank acc. no. Z68705), the alignment with 453 aligned nucleotides was subjected to maximum parsimony analyses using the program-package PAUP*4.0b4a (Swofford, 2000). From the 453 positions used, 277 were constant and 122 variable positions were parsimony informative. The analyses were done using a branch and bound search and the resulting phylogram was midpoint-rooted as no suitable outgroup for Trebouxia species was available. Stability of monophyletic groups was tested using the bootstrap method (Felsenstein, 1985) with 500 replications.

Table 2.  GenBank accession no. of ITS nrDNA sequence and references of Trebouxia strains used for the phylogenetic analyses
Trebouxia speciesStrain numbersource of photobiontsAccession no.
T. arboricolaSAG 219–1aFree-living?Z68705
 M-96.025C1Xanthoria parietinaAJ007387
T. asymmetricaSAG 48.88Diploschistes diacapsisAJ249565
 M-99.023C2Fulgensia fulgidaAF344175
 M-99.024B3Fulgensia fulgidaAF344176
 B207Toninia sedifoliaAF344177
T. gelatinosaUTEX 905Flavoparmelia caperataZ68698
 G-86.108B2Flavoparmelia caperataZ68697
T. giganteaUTEX 2231Caloplaca cerinaAJ249577
T. impressaM-96.027D1Physcia adscendensAJ007383
T. incrustataUTEX 784Lecanora dispersaAJ293795
T. jamesiiM-97.017A2Lecidea silaceaAF128270
T. usneaeG-87.019A1Parmotrema tinctorumZ68702


Total DNA of the mycobionts was extracted using Qiagen DNeasy Plant Mini Kit, following the standard extraction protocol with minor modifications (elution buffer AE was replaced with sterile water, in two elution steps). The undiluted mycobiont DNA extraction was amplified using PCR and Perkin-Elmer Amplitaq Gold polymerase. After an initial denaturation step of 95°C for 10 min, the PCR ran for five cycles (95°C for 1 min, 53°C for 30 s, 72°C for 2 min) and 35 cycles (95°C for 1 min, 50°C for 20 s, 72°C for 2 min) with a final extension step of 72°C for 10 min. The primers nu-SSU-1203–5′ (‘nongreen algal’; Gargas & DePriest, 1996), ITS1F (‘fungal-specific’; Gardes & Bruns, 1993), ITS4 (White et al., 1990), S155–5′ (Döring et al., 2000) LR3, LR5 (, and the recently developed LSU155–5′ (‘mycobiont-specific’; Döring et al., 2000) were used as PCR primers. The PCR products were purified with QIAquick PCR Purification Kit (Qiagen) and eluted with sterile water. Both strands were sequenced using the PCR primers and primer ITS3 (White et al., 1990). The cycle sequencing reaction ran for 28 cycles (95°C for 30 s, 50°C for 15 s, 60°C for 4 min). Sequences were obtained by a Perkin-Elmer ABI 377 automatic sequencer, using the Big Dye Terminator Reaction Kit (Perkin-Elmer Inc., Wellesley, MA, USA). Fragments were assembled with the Sequencher 3.0 package (Gene Code Corporation Inc., Ann Arbor, MI, USA).

Sequence alignment

Overlapping PCR products generated sequences covering the 3′ part of the SSU, the ITS-regions (the spacers ITS1 and ITS2 and the enclosed highly conserved 5.8S rRNA gene) and the 5′ part of the LSU. The sequences were aligned with MegAlign in the LaserGene 1.61 package (DNASTAR Inc., Madison, WI, USA), using the default (gap penalty 10, gap length penalty 10) and manually optimised settings. PCR priming sites were excluded.


The lichen community investigated shares only one species, Toninia sedifolia, with the original stand of the Toninio-Psoretum decipientis (= Thalloedema-caeruleonigricans–Lecidea–decipiens association, Stodiek, 1937), but is similar to the Fulgensietum fulgentisGams 1938. The main differences were the absence of Psora decipiens (Hedwig) Hoffm. and Squamarina cartilaginea (Wirth) P. James and the replacement of F. fulgens (Sw.) Elenkin with F. fulgida. The Fulgensietum fulgentis has been considered a synonym to the first community by Wirth (1980). The species F. fulgens or F. bracteata (Hoffm.) Räsänen, often reported for this association, have not been found in the investigated stand. They are replaced by F. fulgida, which is much more common in the Mediterranean region. Very similar communities, which also comprise F. fulgida and lack F. fulgens, were described from Spain by Llimona (1982), who assigned them to the Toninion caeruleonigricantis Hadac 1948. The most characteristic species of some described Toninio-Psoretum decipientis-related lichen sociological units and the presently studied community are listed in Table 3. Since the inventory of the investigated stand seems to correspond sufficiently with other reports of the Toninio-Psoretum decipientis (one character species is replaced by a congeneric species), the investigated community is considered as belonging to the Toninio-Psoretum decipientis in a wider sense. It may be possible to differentiate between different subgroups within this association as soon as an improved data background has been achieved.

Table 3.  Species inventory of the Toninio-Psoretum decipientis and related lichen communities
 Stodiek (1937)Gams (1938)Llimona (1982)Scholz (1995)Present study
Catapyrenium michelii (A. Massal.) R. Sant.   ++
Catapyrenium squamulosum (Ach.) O. Breuss   + 
Collema fuscovirens (With.) J. R. LaundonRare    
Collema cristatum (L.) Wigg.    +
Diploschistes ocellatus (Vill.) Norman  +  
Endocarpon pusillum Hedw.   + 
Fulgensia bracteata (Hoffm.) Räsänen   + 
Fulgensia fulgens (Sw.) Elenkin + + 
Fulgensia fulgida (Nyl.) Szatala  + +
Psora decipiens (Ehrh.) Hoffm.++++ 
Psora saviczii (Tomin) Follm. & Crespo   + 
Squamarina cartilaginea (With.) P. James in Hawksw. +   
Squamarina lentigera (Weber) Poelt +  +
Toninia sedifolia (Scop.) Timdal+++++


Within the population examined, all mycobiont sequences were identical, lacking any variation. Compared with two samples of F. fulgida from Greece (GenBank acc. nos AF278776, AF 278777) and two samples of the same species from Italy (GenBank acc. nos AF278774, AF 278775), which were genetically identical to each other, the obtained sequences of the population’s mycobionts (GenBank acc. nos AY051357, AY051358, AY051359) showed only the following minor differences: one transition in the ITS1-region (C/T change), an additional nucleotide (A) in the ITS2-region, one transition in the 5′ part of the LSU (C/T change) and one transition in a spliceosomal intron of the LSU (C/T change). All sequences contained a spliceosomal intron starting at position 808 in the LSU with respect to the sequence of Saccharomyces cerevisiae (GenBank acc. no. J01355) (Kasalicky et al., 2000).


All photobiont clones from the Fulgensia fulgida thalli were morphologically identical. The cells possessed a finely lobed chloroplast (‘crenulate’, as defined by Gärtner, 1985) and were often arranged in packages of autospores (mainly tetrads). This feature is characteristic of T. gigantea (Hildreth & Ahmadjian) Gärtner (1985) in the sense of Friedl (1989b). Very few cells, however, had cell wall thickenings at one side of the cell, a feature characteristic of Trebouxia asymmetrica Friedl & Gärtner, 1998, which is closely related to T. gigantea (Friedl, 1989b). To test the relationships of the F. fulgida photobiont the ITS nrDNA was sequenced.

The sequence analyses demonstrated that all the studied photobiont clones did indeed belong to Trebouxia asymmetrica and that they were genetically closely related (Fig. 1). The photobiontal isolates of the Fulgensia thalli examined appeared to be genetically homogenous (Type a sequences in Table 1) except the material derived from one particular thallus (Type b sequences in Table 1). Within that thallus the obtained photobiont strains were homogenous as well. The photobiont strains from four thalli (M-99.021B2, M-99.022B4, M-99.024B3 and M-99.025A5) were identical in their ITS and 5.8 S nrDNA sequence. The five photobiont strains analysed from the fifth thallus (M-99.023A3, B2, B4, C2 and C3) all differed from the strains of the other thalli in only one position, a C/T transition at the base corresponding to position 475 in the sequence of Trebouxia arboricola (SAG 219–1a, GenBank acc. no. Z68705). All these ITS-sequences of photobiont clones from Fulgensia fulgida were identical to the sequence of the type strain of Trebouxia asymmetrica (GenBank acc. no. AJ249565) except for three transitions and two deletions (transition C/T at positions 67 and 551, transition A/G at position 496, deletion of a T at positions 158 and 271/272; numbers are given with respect to the sequence of T. arboricola SAG 219–1a). The genetic differences between all other available Trebouxia ITS-sequences are much larger. None of the photobiont-strains from Fulgensia fulgida had a Group I-intron in position 1512 of the SSU nrDNA (with respect to the Escherichia coli rDNA), which has often been found in other Trebouxia strains, being genetically identical in this respect too.

Figure 1.

Unrooted phylogeny of Trebouxia ITS nrDNA sequences. One tree has been obtained using the maximum parsimony method and a branch and bound search. Bootstrap values were independently calculated for 500 replicates using unweighted maximum parsimony methods. The scale indicates the distance due to 10 evolutionary steps.

The maximum parsimony analyses of Trebouxia strains resulted in a single, most parsimonious tree with a consistency index (CI) of 0.7655 (Fig. 1). The investigated photobiont clones from Fulgensia fulgida clearly belong to Trebouxia asymmetrica, because their monophyly with T. asymmetrica is well supported in bootstrap tests (97%). Therefore it is evident, that the ability to develop cell wall thickenings at one side of the cell is an important phylogenetic trait characteristic of this species, even if only a few cells exhibit this trait. The relationship of this species to other, morphologically quite similar, species mentioned earlier (namely T. gigantea and T. incrustata Ahmadjian ex Gärtner, 1985) was not well resolved (bootstrap values around or below 50%). It will be interesting to test whether the inclusion of further sequences from strains of these species will clarify or blur the boundaries within this species complex. The clones of T. arboricola formed a monophyletic clade, but with only 82% bootstrap support.

Apart from Fulgensia fulgida, the photobiont inventory of some co-occurring lichen species, such as those of Toninia sedifolia and Squamarina lentigera, was investigated as well. Within the community Toninia sedifolia was poorly represented, whereas S. lentigera had a rather high percentage of coverage. The ITS and 5.8 S nrDNA sequence of the photobiont of Toninia sedifolia proved to be identical to the sequence of those Trebouxia asymmetrica strains from the four Fulgensia thalli examined (Type a sequences in Table 1). Thus the early record of T. potteri for Toninia caeruleonigricans (Lightf.) Th.Fr. [=T. sedifolia] by Reháková (1968) could not be verified. The photobiont clones M-99.109, obtained from the thallus of S. lentigera, all belonged to the genus Asterochloris. They corresponded to Asterochloris irregularis (Hildreth & Ahmadjian) Friedl ined. because of the elliptical cells and the broadly lobed chloroplast. Further species occurring at the examined stand were Collema cristatum (L.) Wigg. and Catapyrenium michelii (A. Massal.) R. Sant. The algal partners of these were checked in squash preparations of the thallus and resulted to be Nostoc Vaucher and Myrmecia Printz, respectively. Additionally, traces of indeterminable sterile crustose lichens were present but not considered in this analysis. The dominant moss was identified as belonging to the genus Grimmia Hedw.


This investigation demonstrates that the Fulgensia fulgida-population is highly selective in its photobiont choice, being associated with only one photobiont species, namely T. asymmetrica. As this photobiont is associated with other lichen mycobionts as well (e.g. Toninia sedifolia), the fungal–algal association is not a specific one at morpho-species level. T. asymmetrica is also known from very distantly related lichens as Diploschistes Norman (Graphidales, Thelotremataceae; Friedl & Gärtner, 1988). However it might be possible that different strains within a morpho-species are confined to certain lichen groups. The genetic investigations showed that two different strains of T. asymmetrica, which were almost identical in their ITS-sequence, are acceptable algal partners for F. fulgida at this stand. Only one of these strains has been shown to form lichens with other mycobionts, namely Toninia sedifolia. If the other strain should not be found in other lichens, this association would be a specific one.

Alternatively the occurrence of only one Trebouxia species within this F. fulgida-population could be explained by the absence of other Trebouxia species in this stand rather than to the high photobiont selectivity of the mycobiont. We regard this to be unlikely as there is a generally high input of vegetative diaspores of lichens (Ahmadjian, 1988), most likely containing several species of Trebouxia. As the investigated stand was not protected from this input these diaspores should have been accessible for a possible relichenization. Moreover preliminary results suggest that Fulgensia is associated with T. asymmetrica in Italy as well (A. Beck, unpubl.).

Contrary to the photobionts, the mycobionts have been found to be genetically homogenous. The fact that genetic differences exist between mycobionts of Fulgensia fulgida from different localities, for example France and Italy (Kasalicky et al., 2000), gives significance to the observed homogeneity within the population. Mycobiontal genetic variation within lichen populations has been observed already, for example by Zoller et al. (1999), who showed that the ITS-region – and specifically the ITS1-region – revealed genetic variation within and among different populations of Lobaria pulmonaria (L.) Hoffm. mycobionts. The fact that only one ITS-sequence has been found in apothecia of one and the same thallus could be explained by homothallism, which means a fertilisation of trichogynes with pycnospores (which are bacilliform and could serve as spermatia) from the same thallus or cytogamy. Pycnospores have already been interpreted as spermatia, for example by Poelt (1986), amongst others referring to observations made by Honegger (1984) on fusions of pycnospores and trichogynes in Cladonia furcata (Hudson) Schrader. Pycnidia and apothecia were observed on one and the same thallus.

The finding of the uniformity of ITS-sequences within the apothecia of the mycobionts and heterogeneity of the photobionts suggests that the lichen thalli of this stand derive from propagation by ascospore dispersal (which had identical ITS-sequences) and multiple independent relichenization events rather than by thallus fragmentation. Alternatively, independent relichenization events may have occurred already before the establishment of this population and lichenized propagules have played the major role as secondary dispersal units, despite the presence of apothecia. However, nothing is indicative of such a scenario, as F. fulgida appears to be a purely sexual species, showing sterile thalli only at the juvenile stage and never forming asexual, lichenized propagules. In other Fulgensia species, a special type of vegetative diaspore has been recognized and described by Poelt (1965). These flake-like structures have been called schizidia. They are parallel aligned to the thallus and develop from the upper thalline layers, consisting of the upper cortex and parts of the photobiontal and medullar layers, lacking a lower cortex. These fragile structures can easily split off the thallus and are therefore assumed to serve as lichenized dispersal units. Within the genus Fulgensia, they are known from F. bracteata ssp. deformis (Erichsen) Poelt, F. fulgens, F. delphinensis Poelt, F. desertorum (Tomin) Poelt, F. poeltii Llimona and F. subbracteata (Nyl.) Poelt (Poelt, 1965; Gilbert, 1978; Westberg & Kärnefelt, 1998), and are assumed to be of taxonomic relevance, for example by Poelt (1973). F. fulgida has no schizidia and the strictly monophyllous and placodioid thalli also lack other structures, which might facilitate thallus fragmentation. Its compact thallus is firmly fixed to the substrate. For these reasons, dispersal by ascospores seems the most likely distribution mechanism of this species.

The examination of this lichen community from southern France also suggests that Toninia sedifolia and Fulgensia fulgida share the same photobiont pool, a finding that might support the observations of Ott et al. (1995) on Fulgensia bracteata, which is supposed to take over the photobionts of Toninia sedifolia (= T. caeruleonigricans). However it has not yet been shown that Toninia sedifolia is really a potential photobiont source for germinating or aposymbiotic hyphae of Fulgensia. Toninia sedifolia at least seems to be a suitable substrate providing improved conditions for the thallus establishment of Fulgensia, and the results from this study demonstrate that the two lichens have photobionts with identical ITS-sequences. Co-existence of the two lichens in the same habitat and with a narrow ecological amplitude of the photobiont may have favoured photobiont-sharing between Toninia and Fulgensia. It can be estimated that additional lichen species participate in this pool, without being involved in close interactions. Cyanolichens, as Collema cristatum in the investigated stand, often occur together with F. fulgida and may serve as a substrate during the initial growth of the phycolichen by representing a suitable environment with improved water and nitrogen supply.

Perspectives of studies on myco-photobiont relationships in lichenized systems

For characterising symbiotic interactions Galun & Bubrik (1984) distinguished between the terms ‘selectivity’, meaning ‘preferential interaction between organisms’, and ‘specificity’, meaning ‘cell–cell interactions with absolute exclusivity’. Smith & Douglas (1987) referred to specificity as the degree of taxonomic difference between partners with which an organism associates. Other authors treat the two terms as synonyms. The terminology describing preferential partner selection in symbiotic systems is therefore in need of clarification. In lichenological literature, ‘selectivity’ has hitherto been attributed only to the mycobiont of a lichenized system but has never been applied to the photobiontal partner. However, in a classification that has to cover the whole range of possible kinds of reciprocal interactions, it seems appropriate that this term should be addressed to each biont of a symbiotic system. We suggest using the term ‘selectivity’ for the characterisation of interactions between organisms viewed from the perspective of one biont only. Thus it characterises the range of possible partners that can be selected by this biont – its degree of selectivity. ‘Specificity’ should be used for the symbiotic association as a whole, therefore depending on the degree of selectivity of the partaking bionts. A suitable classification of selectivity and specificity may be best elaborated by considering the whole range of theoretically possible constellations. Five levels of selectivity can be distinguished, covering the whole range from very high – association with members of the same strain only – to very low – association with members of different orders or even classes (Table 4, Fig. 2). It is evident from this definition that the correct systematic placement of the bionts is a prerequisite for determining the degree of selectivity.

Table 4.  Levels of selectivity in symbioses (with minor alterations from Smith & Douglas, 1987)
Levels of selectivityRanges of acceptable partners
Very highWithin the same strain
HighWithin the same species
ModerateWithin the same genus
LowWithin the same order
Very lowWithin groupings of higher taxonomic levels
Figure 2.

Schematic illustration of the definition of selectivity, exemplified for lichen mycobionts.

If a mycobiont forms a lichen with photobionts of different families, it exhibits a low degree of selectivity. An example of this kind of interaction is Chaenotheca chrysocephala (Turner ex Ach.) Th.Fr., which can form lichen thalli with either Trebouxia (ord.: Trebouxiales, class: Trebouxiophyceae) or Stichococcus (ord.: inc. sed., class: Trebouxiophyceae) (Tibell, 1980). As other mycobionts can form lichens with the same photobionts, the association is not exclusive. A mycobiont that is found lichenized with only one photobiont species is considered to exhibit a high selectivity. Phlyctis argena (Sprengel) Flotow is an example of this type of association, where the mycobiont was found to be always associated with Dictyochloropsis splendida Geitler (Tschermak-Woess, 1995). The corresponding photobiont establishes associations also with other mycobionts, for instance Chaenotheca brunneola (Ach.) Müll.Arg. (Tschermak-Woess, 1978), so the interaction as a whole is therefore an unspecific one.

If both (or all) partaking organisms exhibit a high degree of selectivity, the association is specific (Table 5). Actually it is impossible to give an example of a truly specific association, because the identity of the photobiont at the species level is only known for about 3% of all described lichens (Honegger, 1996). For this reason it is premature to assign an exclusive selectivity to a photobiont. However, judging by the available data, most photobiont species seem to have a relatively low degree of selectivity, for instance Trebouxia jamesii, which is associated with several Lecanoralean fungi (Beck, 1999).

Table 5.  Definition of the different degrees of specificity in binary symbiotic associations Thumbnail image of

Tripartite associations, as present in cephalodiate lichens, have to be resolved in three binary ones and the pairs have to be estimated individually.

The study of the kind of selectivity in lichen associations will be an important task for the near future to clarify whether a high degree of mycobiontal and/or photobiontal selectivity can be found in certain lichen groups, which would be indicative that coevolutionary processes may have taken place there. These studies are also necessary to test whether the photobionts are potential indicators of phylogenetic relationships of the mycobionts (Rambold et al., 1998).

Investigations of the selectivity of single mycobiont species from different localities and their algal counterparts can focus on various aspects: focusing on individual lichen thalli; on whole lichen populations; and on the phylogenetic relationships (of the mycobionts). The investigation of a single mycobiont species may be of interest and may be carried out at a local or global scale. On a local scale, it may reveal polymorphism in the mycobiontal populations. However, to test for the degree of selectivity of a mycobiont, the associated photobionts have to be investigated at a global scale, preferentially from the most diverse habitats. Based on morphology, studies of this kind have first been performed by Tschermak-Woess on Phlyctis argena by Tschermak-Woess (1995). Currently various working groups focus on such questions by applying molecular techniques. Recently Kroken & Taylor (2000) demonstrated selectivity of photobiont choice in the genus Letharia, always being associated with algae from the Trebouxia jamesii species complex. Analysing both, the myco- and photobionts of each thallus by molecular techniques they conclude, that cospeciation is not a common rule, but that there is evidence for selectivity of a given fungal species for an algal species. More such investigations will certainly provide better insights on the degree of photobiont-referring selectivity of lichen mycobionts.

Information on the environmental influence of selectivity is provided by studies of whole lichen communities. Different lichen species may share the photobionts when growing side by side, an observation that can only be addressed when all thalli of a lichen community are analysed (Beck, 1999). Studies of this kind are very important in order to make estimations on the relichenization events taking place in saturated communities, where established lichen thalli may act as a photobiont source for arriving mycobiontal spore material.

Population studies like the present one, combined with mycobiontal spore germination capability tests, will allow the investigation of the modes of relichenization in more detail. Furthermore, experiments on lichen resynthesis with the bionts of the original association and lichen resynthesis with one mycobiontal strain and with different algal strains and vice versa will provide further knowledge on the potential range of compatible partners.

The only way of proving specificity of the partner selection is to investigate different lichen species and to determine whether the partners are exclusively associated with each other (even when occurring in different lichen communities). This means there is a huge number of lichens to be investigated, which means that only a very limited number of thalli per species can be investigated. Possible results of such investigations include the verification of perfect coevolution – leading to cospeciation. Should such cospeciation have taken place, it could be visualized by congruent underlying myco- and photobiontal phylogenies (Fig. 3).

Figure 3.

Scheme for supposed coevolution between myco-(MB) and photobionts (PB) in a lichen associations leading to cospeciation.

Up until now not enough data are available to estimate whether such scenarios exist for taxa associated in lichenized systems at all. In the case of trebouxiophyte photobiont Coccomyxa (Friedl, 1998) being found in families of various relationships, as in the Icmadophilaceae, Peltigeraceae, Clavariaceae (Multiclavula) and Tricholomataceae (Omphalina), it is evident that cospeciation did not occur as these unrelated mycobionts are associated with members of the same photobiont genus.

Existing patterns of selectivity in lichens may also be due to environmental factors. It seems possible that certain ancestral, as well as recent, myco-photobiontal-constellations were due to the similar ecological preferences of both bionts (Fig. 4a). In this case, the underlying phylogeny of the myco- and photobionts would not need to be congruent (Fig. 4b). It is an important task for upcoming studies to investigate such preferences in detail.

Figure 4.

(a) Partner selection in lichen symbioses could be due to similar ecological preferences of the corresponding bionts. Thus the phylogenies of the underlying myco- and photobionts do not need to be congruent (b).

As phylogenetically related species might share similar environmental demands, the resulting pattern of photobiont distribution would look similar to one caused by perfect coevolution (but lack congruence of the underlying phylogenies).

As many lichen photobionts are found mainly within lichen thalli (e.g. Trebouxia, the most common lichen photobiont), mycobiontal factors are an important part of their environment. Therefore even distantly related taxa may share the same photobiont taxon through providing a suitable environment for these algae (Fig. 5). Perfect cospeciation between the partaking bionts would not be a precondition for such patterns, only a general ability to associate by lichenization. Recent results from investigations of the Acarosporetum sinopicae Hil. 1924 seem to support this hypothesis (Beck, 1999).

Figure 5.

Partner selection in lichen symbioses might be due to similar environmental factors for the photobiont provided by the mycobionts because of convergent mycobiontal evolution.


We thank C. Stadtkus (Botanische Staatssammlung München), for help with the sequencing work, and Gert Helms (University of Göttingen) for making available the sequence of an unpublished algal-specific PCR-primer. T.K. acknowledges Heidi Döring (Natural History Museum London, now University of Bayreuth) and Mats Wedin (Natural History Museum London, now University Umeå) for advice during molecular analysis of the Fulgensia mycobionts. We extend our thanks to an anonymous referee for his suggestions on an earlier version of this manuscript, A. Davies for improving the English text and to F. Schuhwerk for his advice in lichen sociological questions.