Introduction
Lichens, as found in nature, are the symbiotic phenotype of lichen-forming fungi in association with their photoautotrophic partner. Species names of lichens refer to the fungal partner. Lichen photobionts, mostly green algae or cyanobacteria, very rarely Xanthophyceae or Phaeophyceae (Tschermak-Woess Reference Tschermak-Woess and Galun1988; Peršoh et al. Reference Peršoh, Beck and Rambold2004), have their own names and phylogenies. Traditionally, species of lichen-forming fungi were described on the basis of morphological and chemical characters (morpho- and chemospecies). Morphological criteria also formed the basis of species descriptions in lichen photobionts. In less than 2% of the c. 13 500 species of lichen-forming fungi known to science has the photobiont ever been identified at species level (Honegger Reference Honegger and Nash2008); this estimate is based on Tschermak-Woess (Reference Tschermak-Woess and Galun1988) and on the recent literature.
As lichen-forming fungi do not easily re-lichenize under sterile culturing conditions, the range of compatible photobiont taxa per lichen-forming fungal species cannot be experimentally approached with re-lichenization experiments in the Petri dish. Instead, the photobiont of lichen specimens, as collected in the wild, is investigated. Traditionally, isolation and culturing under defined sterile conditions, followed by light or electron microscopic analysis and comparison with reference strains, were used (Ahmadjian Reference Ahmadjian1958, Reference Ahmadjian1967; Tschermak-Woess Reference Tschermak-Woess and Galun1988). Accordingly, only a few experts worldwide were able to identify the photobionts of lichen-forming fungi at species level. Since the advent of molecular techniques, the time-consuming isolation and culturing has been largely avoided; instead, photobiont-specific molecular markers applied to whole lichen DNA has facilitated photobiont identification at the species level (Kroken & Taylor Reference Kroken and Taylor2000; Dahlkild et al. Reference Dahlkild, Kallersjo, Lohtander and Tehler2001; Helms et al. Reference Helms, Friedl, Rambold and Mayrhofer2001; Piercey-Normore & DePriest Reference Piercey-Normore and DePriest2001; Tibell Reference Tibell2001; Romeike et al. Reference Romeike, Friedl, Helms and Ott2002; Tibell & Beck Reference Tibell and Beck2002; Helms Reference Helms2003; Piercey-Normore Reference Piercey-Normore2004, Reference Piercey-Normore2006; Yahr et al. Reference Yahr, Vilgalys and DePriest2004, Reference Yahr, Vilgalys and DePriest2006; Blaha et al. Reference Blaha, Baloch and Grube2006; Guzow-Krzeminska Reference Guzow-Krzeminska2006; Muggia et al. Reference Muggia, Grube and Tretiach2008, Reference Muggia, Zellnig, Rabensteiner and Grube2010; Francisco De Oliveira et al. Reference Francisco De Oliveira, Timsina and Piercey-Normore2012). Based on increasing numbers of entries in databases, the studies above have gained novel insights into photobiont diversity and phylogenies. Isolation and culturing are, however, still crucial as reference material, for genetic analyses at the subspecific level and for diverse experimental approaches.
The range of compatible photoautotrophic partners per fungal species and their inter- and intraspecific diversity are ideally studied in a large set of samples from a wide geographical range. However, even the analysis of one or a few samples gives valuable first insights into the taxonomic affiliation of compatible photobionts. The majority of morphologically advanced species of lichen-forming fungi are moderately specific to specific with regard to their photobiont selection, that is a fungal species associates with one or few species of green algae or cyanobacteria (Honegger Reference Honegger1993). A lower specificity towards their photobiont was observed in a few of the lichen-forming ascomycetes forming morphologically less advanced crustose thalli (Friedl Reference Friedl1987; Tschermak-Woess Reference Tschermak-Woess and Galun1988; Beck Reference Beck2002; Helms Reference Helms2003; Blaha et al. Reference Blaha, Baloch and Grube2006; Pérez-Ortega et al. Reference Pérez-Ortega, Ortiz-Álvarez, Allan Green and de los Ríos2012; but see Vargas Castillo & Beck Reference Vargas Castillo and Beck2012). Moreover, lichen mycobionts growing in extreme habitats such as Antarctic or alpine ecosystems tend to associate with a wide range of photobiont strains (Romeike et al. Reference Romeike, Friedl, Helms and Ott2002; Wirtz et al. Reference Wirtz, Lumbsch, Green, Türk, Pintado, Sancho and Schroeter2003; Muggia et al. Reference Muggia, Grube and Tretiach2008; Domaschke et al. Reference Domaschke, Fernandez-Mendoza, Garcia, Martin and Printzen2012; Pérez-Ortega et al. Reference Pérez-Ortega, Ortiz-Álvarez, Allan Green and de los Ríos2012). Interestingly, the most common and widespread aerophilic unicellular green algae, often forming conspicuous green layers on bark or rock surfaces, are very rarely acceptable partners of lichen-forming fungi (Tschermak-Woess Reference Tschermak-Woess and Galun1988; Peršoh et al. Reference Peršoh, Beck and Rambold2004).
More than 80% of the lichen-forming fungi studied associate with green algal photobionts, representatives of the genera Trebouxia de Puymaly and Asterochloris (Tscherm.-Woess) T. Friedl ined. (Trebouxiophyceae sensu Friedl Reference Friedl1995) being the most common and widespread partners in all climates (Ahmadjian Reference Ahmadjian1988; Tschermak-Woess Reference Tschermak-Woess and Galun1988; Rambold et al. Reference Rambold, Friedl and Beck1998; Peršoh et al. Reference Peršoh, Beck and Rambold2004; Beck & Peršoh Reference Beck and Peršoh2009). Probably due to their ability to survive desiccation unharmed, Trebouxia spp. are the photobionts of most lichen-forming fungi in climatically extreme habitats such as Antarctic, Arctic, alpine or desert ecosystems, where the whole thallus is continuously subjected to drought and temperature extremes.
Sexually reproducing lichen-forming fungi are assumed to re-lichenize at each reproductive cycle, that is germinating asco- or basidiospores have to find a compatible photobiont. Contradictory views are found in the literature concerning the abundance of free-living Trebouxia cells and their availability for asco- or basidiospore-derived germlings of lichen-forming fungi. According to Ahmadjian (Reference Ahmadjian1988, Reference Ahmadjian2002a , Reference Ahmadjian and Seckbach b ), Trebouxia species do not normally exist outside lichen thalli. Tschermak-Woess (Reference Tschermak-Woess1978) found free-living Trebouxia cells, but pointed out that they are rare in aerophilic algal communities. Bubrick et al. (Reference Bubrick, Galun and Frensdorff1984) found free-living Trebouxia cells near thalli of Xanthoria parietina, and according to Mukhtar et al. (Reference Mukhtar, Garty and Galun1994) Trebouxia arboricola de Puymaly is one of the most common colonizers of bare rock surfaces after fires in Israel. In a series of elegant in situ re-lichenization studies, Sanders (Reference Sanders2005) observed large numbers of free Trebouxia cells on plastic slides which had been exposed in oak trees (Quercus ilex) with lichen cover in Spain, and germ tubes of Xanthoria parietina ascospores in contact with them. In the phycological literature, the aerophilic T. arboricola, type species of the genus, is referred to as abundant and widespread on saxicolous and corticolous substrata in Europe (Ettl & Gärtner Reference Ettl and Gärtner1995; John et al. Reference John, Whitton and Brook2002; Rindi & Guiry Reference Rindi and Guiry2003).
The present study aims to explore the identity, diversity and phylogeny of the photobionts in Teloschistaceae (Teloschistineae, Lecanoromycetes), the focus being on the genera Xanthoria and Xanthomendoza. Teloschistaceae are lichen-forming ascomycetes with a worldwide distribution. They comprise species with a very wide geographical range such as the ubiquitous Xanthoria elegans and the very widespread X. parietina, alongside species with a small area of distribution such as the South African endemics X. capensis, X. flammea and X. karrooensis. Teloschistaceae are associated with trebouxioid green algal photobionts. Best investigated is the widely distributed type species of the genus, Xanthoria parietina, and the closely related European X. calcicola and X. ectaneoides, here referred to as the X. parietina complex (fungal phylograms in Scherrer & Honegger Reference Scherrer and Honegger2003; Eichenberger Reference Eichenberger2007). According to the literature, the X. parietina complex is associated with Trebouxia arboricola, T. crenulata Archibald, T. decolorans (Ahmadjian) Archibald, and T. italiana Archibald [syn. Asterochloris italiana (Archibald) T. Friedl ined.] (Ahmadjian Reference Ahmadjian1960, Reference Ahmadjian and Seckbach2002b ; Gärtner Reference Gärtner1985a , Reference Gärtner b , Reference Gärtner c ; Honegger & Peter Reference Honegger and Peter1994; Beck et al. Reference Beck, Friedl and Rambold1998). Trebouxia decolorans has been determined to be the photobiont of Xanthomendoza hasseana and Xanthoria tenax in southern California (Werth Reference Werth2012), whereas Trebouxia asymmetrica has been reported as the photobiont of Fulgensia fulgida (Beck et al. Reference Beck, Kasalicky and Rambold2002). In contrast, Teloschistes chrysophthalmus associates with T. gelatinosa (Werth Reference Werth2012; Nyati et al. Reference Nyati, Werth and Honegger2013a ). The Trebouxia photobiont of Teloschistes flavicans was found to be related to Trebouxia galapagensis and T. higginsiae (Reis et al. Reference Reis, Iacomini, Gorin, de Souza, Grube, Cordeiro and Sassaki2005). Antarctic photobionts of Caloplaca belong to the genus Trebouxia (Pérez-Ortega et al. Reference Pérez-Ortega, Ortiz-Álvarez, Allan Green and de los Ríos2012). Trebouxia arboricola, T. decolorans, and T. gigantea were found to be the photobionts of Caloplaca spp. from northern Chile (Vargas Castillo & Beck Reference Vargas Castillo and Beck2012). No data are available on the taxonomic affiliation of the photobionts of other Teloschistaceae.
The goals of the present study are: 1) to evaluate photobiont diversity and phylogenies in a range of Xanthoria and Xanthomendoza spp.; 2) to explore the range of compatible photobionts in a large sample of the X. parietina complex from worldwide locations. As X. parietina was most likely introduced to Australia and New Zealand (Galloway Reference Galloway1985; Rogers Reference Rogers1992), we were interested to see whether it associates with different photobionts in these areas than in Europe or North America; 3) to isolate Trebouxia photobionts of Teloschistaceae into sterile culture as reference strains and for diverse future investigations. Most of the corresponding fungal partners were brought into sterile culture (Honegger Reference Honegger2003), their taxonomic affiliation and phylogenies being analyzed in parallel experiments (Eichenberger Reference Eichenberger2007; Itten & Honegger Reference Itten and Honegger2010).
Materials and Methods
Lichen collection and storage
Freshly collected lichens were either immediately processed or stored, in a desiccated state, at −20°C, where they stay viable for prolonged periods of time (Honegger Reference Honegger2003). Voucher specimens were deposited in the herbarium of ETH Zürich (Z+ZT). Collectors and collecting sites are listed in Table 1. A few experiments were carried out with specimens from the lichen herbarium of the University of Graz, Austria. From a set of samples originating from the campus of the University of Zürich (Zürich-Irchel, numbers 319–320), the thalli were left in situ and only small fragments were removed after photographic documentation.
Photobionts isolated from members of the Teloschistaceae used in the present study, their country of origin, collectors and collection numbers and ITS and rbcL GenBank Accession numbers
*Photobiont species were determined based on ITS and rbcL sequence data where available, authorities are mentioned in Table 2; †*P: photobiont isolated, L: whole lichen DNA used for PCR amplification and sequencing where axenic cultures could not be established, followed by voucher number, thallus number and apothecia or lobe number; Sc: single cell isolate, G: lichen specimens obtained from the herbarium of the University of Graz (GZU). 1syn. Gallowayella borealis (R. Sant. & Poelt) S.Y. Kondr. et al.; 2syn. Oxneria fallax (Hepp.) S.Y. Kondr. & Kärnefelt; 3syn. Gallowayella fulva (Hoffm.) S.Y. Kondr. et al.; 4syn. Gallowayella hasseana (Räsänen) S.Y. Kondr. et al.; 5syn. Jesmurraya novozelandica (Hillmann) S.Y. Kondr. et al.; 6syn. Oxneria ulophyllodes (Räsänen) S.Y. Kondr. & Kärnefelt; 7syn. Honeggeria rosmarieae (S.Y. Kondr. & Kärnefelt) S.Y. Kondr. et al.; 8syn. Massjukiella candelaria (L.) S.Y. Kondr. et al.; 9syn. Xanthodactylon capense (Kärnefelt, Arup & L. Lindblom) S.Y. Kondr. et al., 2009; 10syn. Rusavskia elegans (Link) S.Y. Kondr. & Kärnefelt; 11syn. Xanthodactylon flammeum (L. f.) C.W. Dodge; 12 karrooensis; 13syn. Massjukiella polycarpa (Hoffm.) S.Y. Kondr. et al.; 14syn. Rusavskia sorediata (Vain) S.Y. Kondr. & Kärnefelt; 15syn. Xanthodactylon turbinatum (Vain.) C.W. Dodge.
Photobiont isolation and culture
With a sterile platinum needle, photobiont cells were scraped from the thalline margin of apothecia, or alternatively from the algal layer of lobes in samples with no or few fruiting bodies. Photobiont cells were spread on the surface of agarized non-nutrient, mineral medium [Bold's basal medium (BBM) according to Deason & Bold Reference Deason and Bold1960] contained in Petri dishes, with double amount of nitrogen and with 0·005% (w/v) doxycycline (Sigma-Aldrich, MA, USA) as an antibiotic. These plates were maintained at 15±1°C at a 16:8 h light-dark cycle at c. 5 µE m–2 s–1 for 2–3 weeks until cells started to divide. All cultures were screened regularly; fungal contaminants were immediately cut out. Groups of dividing algal cells were either transferred to Trebouxia medium II according to Ahmadjian (Reference Ahmadjian1967), with only ¼ amount of glucose and casamino acids (Honegger Reference Honegger2004), and cultured for 8–12 weeks, or left on BBM 2N. Most cultures are multi-cell isolates, cells originating from a very small area, but a few are single cell isolates. Approximately 300 sterile photobiont cultures from 12 identified and a few unidentified Xanthoria species, seven Xanthomendoza and two Teloschistes spp. were established. All isolates are stored in liquid nitrogen in our laboratory (Honegger Reference Honegger2003). Reference strains obtained from culture collections are listed in Table 2.
List of reference Trebouxia strains and their ITS and rbcL accession numbers
*type strains are indicated with an asterisk; †*UTEX: Algal Culture Collection at University of Texas; IB: Algal Culture Collection at University of Innsbruck; SAG: Algal Culture Collection at University of Göttingen; other isolates are in private culture collections; ‡Type species of the genus Trebouxia de Puymaly; **references applicable only for nrITS sequences already published, all rbcL sequences were generated in the present study; 1ITS sequences generated by Thomas Friedl or Gert Helms, who kindly provided access to their unpublished sequence data for comparison.
DNA extraction
Genomic DNA was isolated and purified using GFX PCR, DNA and Gel Band Purification Kit (Amersham Biosciences, NJ, USA), following the protocol of the manufacturer with slight modifications. Briefly, algal isolates or lichen samples, respectively, were frozen in liquid nitrogen prior to grinding. After addition of 100 µl of capture buffer to the ground material, the samples were incubated at 60°C for 10 min and subsequently centrifuged. The supernatant was transferred to a GFX column, which had been preloaded with 100 µl of capture buffer, incubated for 3 min at room temperature, centrifuged and washed with 500 µl of washing buffer. The DNA was eluted in 50 µl of elution buffer (10 mM Tris-HCL, pH 8·0) and stored at 4°C.
ITS amplification
The complete internal transcribed spacer region of nuclear ribosomal ITS region (ITS1, 5.8S rDNA and ITS2) of algal isolates was amplified in both directions using primer pair ITS5 and ITS4, as described by White et al. (Reference White, Bruns, Lee, Taylor, Innis, Gelfand, Sninsky and White1990). For whole lichen DNA, 1·5 µM of forward primer AL1500bf (Helms et al. Reference Helms, Friedl, Rambold and Mayrhofer2001) and reverse primer LR3 (Friedl & Rokitta Reference Friedl and Rokitta1997) were used in 50 µl reactions containing 1U Taq polymerase (Sigma-Aldrich), 200 µM of each dNTP, 1× PCR buffer containing 1·5 mM MgCl2, 10 mM Tris-HCl pH 8·3, 50 mM KCl, and 0·001% gelatin (final concentrations). Amplifications were run on a PTC 200 DNA engine (MJ Research, Watertown, MA, USA) with the following PCR conditions: initial denaturation at 95°C for 3 min, followed by 32 cycles (94°C for 40 s, 50°C for 40 s, and 72°C for 80 s), with a final extension step at 72°C for 10 min. Internal primers at 5.8S rDNA were newly designed (Table 3). PCR products were purified with GFX PCR, DNA and Gel Band Purification Kit (Amersham Biosciences), following the standard protocol provided by the manufacturers and sequenced directly.
List of primers used in the present study.
* The position of the primers given with respect to a reference sequence (accession number given in brackets)
When direct sequencing did not give satisfactory results, the samples were cloned using pGEM®-T Easy Vector System (Promega Corp., WI, USA) and competent XL10-Gold® Escherichia coli cells (Stratagene, CA, USA). Plasmid DNA was isolated using GFXTM Micro Plasmid Prep Kit according to the manufacturer's protocol (Amersham Biosciences).
rbcL amplification
Six different primers were newly designed (Table 3) for amplification and sequencing of the large subunit (rbcL) of the plastid gene ribulose-1, 5-biphosphate carboxylase/oxygenase. Concentrations of PCR ingredients were the same as in the ITS amplifications. PCR conditions were as follows: initial denaturation at 95°C for 3 min, followed by 30 cycles (95°C for 45 s, 52°C for 60 s, and 72°C for 80 s), with a final extension at 72°C for 10 min.
Agarose gel electrophoresis
PCR fragments were run on 1·2% agarose gel in 1× Tris-acetate-EDTA (TAE) buffer at 80 V, stained with ethidium bromide and visualized by a UV transilluminator at 302 nm wavelength. Cut gel fragments were purified with the GFX PCR, DNA and Gel Band Purification Kit (Amersham Biosciences), following standard protocol provided by the manufacturer.
Sequencing
Purified PCR fragment (10–20 ng DNA) or plasmid (150–300 ng DNA) was used for sequencing in 10 µl reaction mix containing 120 nM primer, 0·8 µl BigDye Terminator Mix V3.1 (Life Technologies, Rotkreuz, Switzerland), and 1× reaction buffer following the protocol of the manufacturer. Amplification conditions were as follows: initial denaturation at 94°C for 2 min, followed by 60 cycles of 96°C for 10 s, 50°C for 5 s, and 60°C for 3 min (0·9°C/s ramp). The products were analyzed on a HITACHI ABI 3730 DNA Analyzer (Life Technologies).
Phylogenetic analysis
Sequences were analyzed with SequencherTM 4.2.2 (Gene Codes Corp., Ann Arbor, MI, USA) and aligned automatically with Clustal X 1.81 (Thompson et al. Reference Thompson, Gibson, Plewniak, Jeanmougin and Higgins1997), with a gap opening penalty of 10·0 and gap extension penalty of 0·20. Aligned sequences were imported in MacClade 4.06 (Maddison & Maddison Reference Maddison and Maddison2002) and aligned manually. Phylogenetic analysis was carried out using PAUP 4.0b10 (Swofford Reference Swofford1998) by Maximum Likelihood (ML), Maximum Parsimony (MP) and Neighbour-joining (NJ) methods on each locus separately and on a combined dataset containing 39 samples. Ambiguous characters were removed from the analysis. A separate analysis was carried out where missing and ambiguous sites were included, which resulted in a similar phylogram (data not shown). Jackknife values for 500 replicates were calculated separately by MP and NJ analyses. ITS analyses were carried out with complete ITS1, ITS2 and 5.8S rDNA sequences. Intron sequences were cut out from the nrITS alignment since these were present in only 25% of newly generated ITS sequences. In ITS analyses, T. simplex sequences were used as outgroup while in rbcL analyses, Asterochloris sequences were used as outgroup. For the combined analysis, neither Asterochloris sp. sequences nor T. simplex were available and, hence, a midpoint rooted neighbour-joining tree is shown. The displayed tree is a neighbour-joining tree, constructed with MEGA version 5.1 (Tamura et al. Reference Tamura, Dudley, Nei and Kumar2007) and annotated with support values from PAUP.
Results
Isolation and culturing
In the course of ongoing projects, c. 300 photobiont isolates were obtained from most of the freshly collected lichen specimens, with or without prior storage at −20°C. On non-nutrient mineral medium (BBM) all isolates grew well, albeit more or less slowly, and kept their green colouration. There was no evidence of bleaching under the light conditions provided, nor of any dependence on external nutrient supply, as suggested by Ahmadjian (Reference Ahmadjian1960, Reference Ahmadjian2002a , Reference Ahmadjian and Seckbach b ). In our laboratory, the type strain of T. decolorans retained its colour with the same intensity after four months culturing on either BBM 2N or Trebouxia ¼ media. Different growth rates were observed among different isolates, partly even among isolates from samples collected next to each other (e.g. among isolates 319 and 320). Only one algal isolate was normally taken per lichen sample. All except one isolate were phenotypically homogenous, and RAPD-PCR analyses of diverse subsamples per isolate turned out to be homogenous (five subsamples each of five isolates tested with three primers, data not shown). However, two phenotypically different isolates (P-121-I-a, either light green, or dark green to brownish) were obtained from the same apothecium of a X. parietina sample. As concluded from ITS phylogenetic analyses, both isolates represented different genotypes of the same algal species (T. decolorans) (see Fig. 1).
ML phylogram of nrITS region (combined ITS1, ITS2 and 5.8S rDNA). Jackknife values calculated separately for 500 replicates by MP (first number) and NJ (second number) analyses and indicated at the nodes. Trebouxia simplex sequences were used as outgroup. Arrowhead points to the type species of the genus Trebouxia de Puymaly. Samples were labelled as follows: example P 270 I a Telochry E Canary Islands e : P, sterile cultured photobiont of Teloschistes chrysophthalmus, voucher number 270, thallus I, apothecium a, collected from Spain (E), Canary Islands, epiphytic (e). Abbreviations used: Xanthoria : Xca: X. candelaria, Xcl: X. calcicola, Xcp: X. capensis, Xec: X. aureola, Xel: X. elegans, Xfl: X. flammea, Xkr: Xanthoria karrooensis; Xli: X. ligulata, Xp: X. parietina, Xpo: X. polycarpa, Xsp: unidentified Xanthoria or Xanthomendoza sp., Xtu: X. turbinata. Xanthomendoza: Xbo: Xanthomendoza borealis, Xfa: Xm. fallax, Xfu: Xm. fulva, Xha: Xm. hasseana, Xnovo: Xm. novozelandica, Xul: Xm. ulophyllodes, Xweb: Xm. weberi; synonyms see Table 1. Teloschistes: Telochry: Teloschistes chrysophthalmus, Telohos: Telo. hosseusianus. Josefpoeltia: Jb: Josefpoeltia parva (syn. J. boliviensis). Letters A, C, I & S indicate Trebouxia clades (A: arboricola; C: corticola; I: impressa; S: simplex) as proposed by Helms (Reference Helms2003). P: photobiont isolated; L: whole lichen DNA used for amplification; e: epiphytic; s: saxicolous; l: lignicolous/ corticolous; +: sequence contained a 1512 intron. Sequences obtained from databases are in bold and indicated with strain number and accession number; *: unpublished sequence provided by G. Helms. Arrowhead points to type species of the genus.
Continued
ITS phylogeny
A total of 101 photobiont nrITS sequences were obtained in this study, originating from 12 Xanthoria species, seven Xanthomendoza species, two Teloschistes species, Josefpoeltia parva and 10 unnamed Xanthoria and Xanthomendoza samples. A total of 781 characters were included in nrITS (ITS1, 5.8S rDNA and ITS2) phylogenetic analyses, 358 of which were constant, 118 were variable but uninformative, and 305 were parsimony informative. Primer binding sites and adjacent flanking regions were omitted from the analyses. Tree topologies for main clades were identical in ML, MP and NJ analyses. Only one most likely tree resulted in ML analyses (Fig. 1). In 30 out of 101 ITS sequences, a longer ITS fragment was found due to a group I intron at position 1512 as described by Bhattacharya et al. (Reference Bhattacharya, Friedl and Damberger1996, Reference Bhattacharya, Friedl and Helms2002) and Helms et al. (Reference Helms, Friedl, Rambold and Mayrhofer2001). Intron sequences were removed from ITS alignments prior to phylogenetic analysis and investigated in a separate study (Nyati et al. Reference Nyati, Bhattacharya, Werth and Honegger2013b ).
The major clades in both phylogenies were in accordance with the Trebouxia clade system as proposed by Beck (Reference Beck2002) and Helms (Reference Helms2003). In their system, clade A includes T. arboricola (including T. aggregata and T. crenulata), T. decolorans, T. asymmetrica, T. showmanii, T. incrustata and T. jamesii. In the present study, clade A was subdivided into an arboricola cluster (subclades Aa and Ab) and a decolorans cluster (subclades Ac and Ad); unnamed Trebouxia species formed subclade Ae. Clade I comprised the impressa (subclade Ia) and gelatinosa (subclade Ib) clusters. Photobionts of all identified and unidentified Xanthoria spp. analyzed in this study belonged either to T. decolorans, T. arboricola or closely related, unnamed Trebouxia sp. within clade A (Fig. 1). The best represented photobionts in our sample set were from associations with X. parietina, with 49 ITS sequences from specimens collected on 4 continents. Photobionts of Xanthomendoza species belonged to either the arboricola (A) or impressa (I) clades, but none of the Xanthomendoza species associated with both.
Photobionts of eight identified and four unnamed Xanthoria species were represented in the arboricola cluster (subclades Aa and Ab), which is characterized by a 28 nucleotides long insert within ITS1 (Helms et al. Reference Helms, Friedl, Rambold and Mayrhofer2001). Subclade Aa has jackknife support of 87% (MP) and 92% (NJ). It includes the type species of the genus, Trebouxia arboricola (strain SAG 219-I-a, arrowhead), T. aggregata (UTEX 180) and photobiont isolates of X. calcicola, X. aureola, X. ligulata, X. parietina and an unnamed Xanthoria species, phenotypically resembling X. parietina (L-337, Canary Islands). Subclade Ab of the arboricola cluster includes T. crenulata (strain CCAP 219/2), photobionts of X. calcicola, X. parietina, and of the South African endemic species X. capensis. Photobiont isolates of X. turbinata, X. calcicola, X. aureola, X. polycarpa and X. parietina formed a cluster which was only weakly supported in MP analysis and not supported by NJ analysis. Photobionts of unnamed Xanthoria species from New Zealand and Australia formed a small cluster with high support (96% in MP, 93% in NJ), although photobionts from other Xanthoria species in this region fall in separate lineages. Photobionts of X. flammea (ZA), X. aureola, X. karrooensis and X. parietina formed unresolved basal lineages. Subclade Ac, being part of the decolorans cluster, has high jackknife support in MP and NJ analyses (91% and 87% respectively). It includes T. decolorans (UTEX 901) and photobiont sequences of X. parietina (CH, NZ, F, RUS, S, USA), X. candelaria (CZ, IS), X. elegans (Nepal), X. polycarpa (Armenia, CH, USA) and Xm. borealis (Greenland). The ITS sequence of the X. parietina photobiont identified as T. arboricola (AJ007387) by Beck et al. (Reference Beck, Friedl and Rambold1998) also falls in this subclade. Subclade Ad, also being part of the decolorans cluster, is very well supported (100%). It comprised photobiont sequences of four unidentified Xanthoria species clustering within the X. parietina complex (GR), along with photobiont sequences of X. parietina s. str. (AUS, F, E, USA & ZA). The phylogenetic position of several photobionts, which cluster outside subclades Ac and Ad, could not be properly resolved. This part of the tree includes the photobionts of Xanthomendoza fulva (Armenia), Xm. hasseana (USA), Xm. borealis (Greenland), X. candelaria (IS), X. parietina (AUS, F, E and USA), X. polycarpa (USA), and unidentified Xanthoria species (AUS, USA). Subclade Ae has a high jackknife support in both ML and NJ analyses (100%) and includes photobiont sequences of X. elegans (CH), X. sorediata (CH), and an unidentified Xanthomendoza sp. (USA). This ITS subclade Ae most likely represents a cryptic Trebouxia sp. The exact phylogenetic positions of photobionts of Xanthomendoza borealis (G 9306; Greenland) and Xanthoria elegans (KS) could not be resolved. Two morphologically different thalli of Xm. borealis, a narrow and a broad-lobed specimen growing side by side, which were collected at the same locality in Greenland, had different T. decolorans genotypes from different subclades (Fig. 1); their fungal partners turned out not to be conspecific (Eichenberger Reference Eichenberger2007).
Subclade Ia has a very high jackknife support (91% in MP, 100% in NJ; Fig. 1). This might be partly due to the small sample size. It includes T. impressa, T. potteri, T. flava and photobiont sequences of Xanthomendoza fallax (CH, USA), Xm. ulophyllodes (USA) and an unidentified Xanthomendoza sp. (USA). Helms (Reference Helms2003) found the authentic strain of T. impressa (UTEX 893) to be very similar to T. potteri (UTEX 900) and most probably conspecific with T. flava (UTEX 181), as inferred from ITS p-distances. Therefore, all our isolates in this cluster are referred to as T. impressa. The well-supported (100%) gelatinosa cluster (subclade Ib) comprises two separate groups, one of them harbouring the type strain of T. gelatinosa with photobiont sequences from Teloschistes chrysophthalmus (E), and Xanthomendoza weberi (F, USA). A sister clade, also with high support, comprised photobiont isolates of Xm. novozelandica (NZ), Teloschistes hosseusianus (Argentina) and Josefpoeltia parva (Argentina). Teloschistes hosseusianus and Josefpoeltia parva grew side by side and were locally overgrowing each other; it is interesting to see that they associate with largely the same photobiont (one nucleotide difference).
rbcL phylogeny
A total of 1155 characters were included in phylogenetic analyses of the rbcL gene, 925 of which were constant, 34 variable but uninformative, and 196 were parsimony informative. ML (Fig. 2), MP and NJ analyses resulted in similar tree topologies. Asterochloris sequences formed an outgroup. The rbcL phylogeny was largely congruent with ITS phylogeny. Trebouxia arboricola (SAG 219-Ia) and T. aggregata (UTEX 903) were part of subclade Aa (bootstrap support 99%) while T. crenulata (CCAP-219-2) was part of subclade Ab, as was also the case in the ITS phylogram. The rbcL clade Ab had very low (52% in NJ) or no jackknife support (MP). Photobiont isolates of X. flammea (ZA), X. aureola (GR) and of an unidentified Xanthoria sp. (NZ) fell outside subclade Ab. Deduced amino acid sequences within subclade Aa were identical and differed from subclade Ab sequences only marginally (data not shown). Subclade Ac with low support (53% MP, 64% NJ) comprised T. decolorans (UTEX 901) along with isolates which were also part of subclade Ac in the ITS phylogeny. Subclade Ad, which is highly supported in the ITS phylogram had low support (<50%) in rbcL phylogeny (dotted line). The photobiont isolates of Xanthomendoza fallax (CH, USA) and Xm. ulophyllodes (USA) clustered with T. impressa, T. potteri and T. flava in subclade Ia, which was very well supported (100%). All rbcL sequences in subclade Ib, including type strains T. gelatinosa and T. anticipata, were nearly identical. Six representatives of the genus Asterochloris formed the outgroup.
ML phylogram of rbcL locus. Jackknife values calculated separately for 500 replicates by MP (first number) and NJ (second number) analyses are given at the nodes. Asterochloris sequences form the outgroup. Abbreviations used: Xanthoria : Xca: X. candelaria, Xec: X. aureola, Xfl: X. flammea, Xli: X. ligulata, Xp: X. parietina, Xpo: X. polycarpa, Xtu: X. turbinata, Xsp: unidentified Xanthoria or Xanthomendoza sp. Xanthomendoza: Xbo: Xanthomendoza borealis, Xfa: Xm. fallax, Xha: Xm. hasseana, Xul: Xm. ulophyllodes, Xweb: Xm. weberi; Teloschistes: Telochry: Teloschistes chrysophthalmus. A, C and I indicate Trebouxia clades as proposed by Helms (Reference Helms2003). P: photobiont isolated; L: whole lichen DNA used for amplification.
Combined phylogeny
A total of 1921 characters were included in the combined phylogenetic analysis of ITS and rbcL. ML and NJ analyses resulted in similar topologies, containing the clades Aa and Ab belonging to T. arboricola and clades Ac and Ad belonging to T. decolorans, as well as clades Ia and Ib (Fig. 3). As in the separate analyses for each locus, Trebouxia arboricola (SAG 219-Ia) was part of subclade Aa whereas T. potteri (IB 332) and T. flava (IB 346) belonged to subclade Ia. The position of T. showmanii (UTEX 2234) in the tree was associated with a high degree of uncertainty, as indicated by lack of support. Clade I, subclade Ad and subclade Ib were well supported (100% NJ, 100% ML), as was subclade Ia (100% NJ, 95% ML). In contrast, subclade Ab received no statistical support in the combined analysis, and clade A had high support in NJ (99%) but low support in ML (58%). The arboricola clade had relatively high support (99% NJ, 76% ML), but the decolorans clade received support only in NJ jacknifing (80% NJ, 0% ML).
NJ tree of the combined phylogenetic analysis of ITS and rbcL loci. A total of 1921 characters were included in the analysis. For the combined analysis, neither Asterochloris sp. sequences nor T. simplex was available and hence, a midpoint rooted neighbour-join tree is shown. ML and NJ analyses resulted in similar topologies.
Discussion
Photobionts of the Teloschistaceae
All foliose (Xanthoria, Xanthomendoza) and fruticose (Teloschistes) Teloschistaceae investigated in the present study associated with Trebouxia species belonging either to ITS clade A or I sensu Helms (Reference Helms2003), or rbcL clades A or I, respectively (Figs 1 & 2). None of the Xanthoria, Xanthomendoza and Teloschistes species associated with photobionts from two clades, and none of the samples had photobionts of ITS clades S or C or of the genus Asterochloris. The present findings refer to a moderate specificity at genus level within foliose and fruticose Teloschistaceae, species of few subclades of the same Trebouxia clade being acceptable partners. It has to be admitted that the sample size was very small in many of the taxa examined. The range of compatible photobionts in representatives of the genera Teloschistes and Josefpoeltia remains unclear, Trebouxia gelatinosa (clade Ib) being the only green-algal partner so far found associated with these taxa. Other studies have found the same algal species in association with Teloschistes (Reis et al. Reference Reis, Iacomini, Gorin, de Souza, Grube, Cordeiro and Sassaki2005; Werth Reference Werth2012). Only one freshly collected specimen per species was available for South African endemic species, hence the range of compatible photobionts remains unclear for these taxa. It would be interesting to investigate photobiont diversity among crustose Teloschistaceae (genera Caloplaca, Ioplaca etc.). Two studies have investigated the photobionts of Caloplaca spp. In northern Chile, Caloplaca associated with three species of Trebouxia (T. arboricola, T. decolorans, and T. gigantea) (Vargas Castillo & Beck Reference Vargas Castillo and Beck2012), and in Antarctica, a Caloplaca sp. was found in association with three ITS haplotypes, but the species was not determined (Pérez-Ortega et al. Reference Pérez-Ortega, Ortiz-Álvarez, Allan Green and de los Ríos2012). Crustose taxa of lichen-forming ascomycetes turned out to be less specific than foliose and fruticose ones at either the generic (Tibell Reference Tibell2001; Tibell & Beck Reference Tibell and Beck2002) or species level (Beck Reference Beck2002; Helms Reference Helms2003; Blaha et al. Reference Blaha, Baloch and Grube2006). Also, the habitat may influence photobiont selectivity: in extreme climates, lichen-forming fungi tend to associate with a wide range of photobionts (Romeike et al. Reference Romeike, Friedl, Helms and Ott2002; Wirtz et al. Reference Wirtz, Lumbsch, Green, Türk, Pintado, Sancho and Schroeter2003; Muggia et al. Reference Muggia, Grube and Tretiach2008). No co-speciation is evident in the present data set. This is not surprising as the photobionts of Teloschistaceae are also partners of numerous other lichen-forming ascomycetes. A similar situation was found in Cladoniaceae (Piercey-Normore & DePriest Reference Piercey-Normore and DePriest2001) and Physciaceae (Helms Reference Helms2003), but Dahlkild et al. (Reference Dahlkild, Kallersjo, Lohtander and Tehler2001) reported co-speciation for Physciaceae.
In studies on green-algal phylogenies, the rbcL locus, which encodes for the large subunit of Rubisco (ribulose-1,5-biphosphatedecarboxylase), was shown to be a highly suitable molecular marker (McCourt et al. Reference McCourt, Karol, Kaplan and Hoshaw1995; Nozaki et al. Reference Nozaki, Ito, Sano, Uchida, Watanabe, Takahashi and Kuroiwa1997, Reference Nozaki, Onishi and Morita2002; Sherwood et al. Reference Sherwood, Garbary and Sheath2000). The same applies for green-algal photobionts of lichen-forming ascomycetes, as shown in the present study. Rubisco was located in the pyrenoid of Trebouxia species with immunocytochemical techniques (Ascaso et al. Reference Ascaso, Valladares and De Los Rios1995). In Trebouxia decolorans photobionts associated with Ramalina menziesii, rbcL turned out to be variable at the population level (Werth & Sork Reference Werth and Sork2010).
Comparison of morphological and molecular data sets
Comparisons of morphological data, as compiled by Friedl (Reference Friedl1989), and molecular data within the genera Trebouxia and Asterochloris (present study) are summarized in Table 4. ITS clade A, comprising most of the photobionts of Teloschistaceae investigated in this study, includes Trebouxia species from several morphological groupings. The morphology of the samples genetically identified in this study will have to be analyzed in future investigations. The ITS sequence data obtained by Helms et al. (Reference Helms, Friedl, Rambold and Mayrhofer2001), Helms (Reference Helms2003) and in the present study indicate that T. crenulata and T. aggregata are conspecific with T. arboricola. Peršoh et al. (Reference Peršoh, Beck and Rambold2004) consider T. arboricola synonymous with T. decolorans. As both are morphologically distinguishable by the shape of their chloroplast (Gärtner Reference Gärtner1985b ; Friedl Reference Friedl1987) and cluster within different ITS and rbcL subclades, both species names were retained in the present investigation. Some authors refer automatically to T. arboricola when ITS sequences fall into clade A sensu Helms (Reference Helms2003). Morphospecies names given to taxa among the genera Trebouxia and Asterochloris need to be revised in future studies, based on additional genetic and morphological data.
Comparison of molecular markers and morphological characters in the genera Trebouxia and Asterochloris (Trebouxiophyceae, Chlorophyta)
†authorities given in Table 2, *chloroplast shape distinctly different in T. crenulata and T. decolorans (Gärtner Reference Gärtner1985b ). **termed G in Helms (Reference Helms2003), now changed into C (“corticola”; G. Helms, pers. comm.)
A wide range of algal genotypes was found in each ITS and rbcL subclade, comparable to the situation among photobionts of the genera Letharia (Kroken & Taylor Reference Kroken and Taylor2000; Altermann Reference Altermann2009), Cladonia (Piercey-Normore Reference Piercey-Normore2004), Evernia (Piercey-Normore Reference Piercey-Normore2006), Ramalina (Werth & Sork Reference Werth and Sork2010; Francisco De Oliveira et al. Reference Francisco De Oliveira, Timsina and Piercey-Normore2012) or Parmotrema (Ohmura et al. Reference Ohmura, Kawachi, Kasai, Watanabe and Takeshita2006). Studies based on microsatellite markers yielded similar results for Lobaria (Dal Grande et al. Reference Dal Grande, Widmer, Wagner and Scheidegger2012; Werth & Scheidegger Reference Werth and Scheidegger2012; Widmer et al. Reference Widmer, Dal Grande, Excoffier, Holderegger, Keller, Mikryukov and Scheidegger2013). It is interesting to see that identical algal ITS genotypes occurred in the same or even in different Xanthoria spp. from geographically different locations; examples are the photobionts of X. parietina from Corsica and of X. capensis from South Africa (subclade Ab; Fig. 1), of X. parietina from Otago (NZ), Zürich (CH) and Burgundy (F) (subclade Ac; Fig. 1), or of X. polycarpa from Zürich, X. parietina from Zürich (CH), Götland (S) and Burgundy (F) (subclade Ac; Fig. 1). On the other hand, X. parietina thalli collected side by side (populations 144 & 145 from Roussillon, SW France, 120 & 121 from Burgundy, France, and 319 & 320 from Zürich, Switzerland) had partly the same, partly different ITS genotypes of the same subclade (Fig. 1). RAPD-PCR analyses of the sterile cultured fungal partners revealed considerable genetic variation within the populations (populations 120 & 121, 144 & 145, 319 & 320 plus 164 from Brittany analyzed; Itten & Honegger Reference Itten and Honegger2010).
Photobionts of the genus Xanthoria
All Xanthoria species investigated in the present study associated with photobionts of ITS clade A sensu Helms (Reference Helms2003) (Fig. 1). No clear geographical pattern can be seen in the present data set, but T. decolorans (subclades Ac and Ad) was almost exclusively found in epiphytic samples (marked with e in Fig. 1), whereas T. arboricola occurred in saxicolous specimens (forming subclade Aa) in the Northern and Southern Hemispheres and in many of the corticolous samples in the Southern Hemisphere (subclade Ab). The T. arboricola photobiont of saxicolous X. parietina, growing under a willow tree in Zürich, was more closely related to the photobiont of a saxicolous X. parietina from Corsica than to the T. decolorans genotypes isolated from corticolous samples on the respective willow tree. Xanthoria candelaria (CZ, IS) associated with T. decolorans, whereas Aoki et al. (Reference Aoki, Nakano, Kanda and Deguchi1998), using microscopy techniques, identified T. incrustata, another representative from clade A sensu Helms (Reference Helms2003), from a sample collected in Antarctica. Fulgensia fulgida was shown to associate with T. asymmetrica (Beck et al. Reference Beck, Kasalicky and Rambold2002), another representative of ITS clade A.
Photobionts of Xanthoria parietina s. lat.
Early investigators had already discovered a range of phenotypically different strains among Trebouxia isolates derived from thalli of X. parietina, which they interpreted as ecotypes (Thomas Reference Thomas1939; Werner Reference Werner1954; Tomaselli Reference Tomaselli1956). Our present findings are in agreement with earlier reports, based on light and electron microscopic as well as molecular investigations, on T. arboricola, T. decolorans and T. crenulata, all members of ITS clade A sensu Helms (Reference Helms2003) and partly conspecific, being photobionts of X. parietina s. lat. (Ahmadjian Reference Ahmadjian1960; Gärtner Reference Gärtner1985b ; Honegger & Peter Reference Honegger and Peter1994; Beck et al. Reference Beck, Friedl and Rambold1998), X. calcicola and X. aureola included (Scherrer & Honegger Reference Scherrer and Honegger2003).
Asterochloris photobionts are found in Cladoniaceae (Rambold et al. Reference Rambold, Friedl and Beck1998; Peršoh et al. Reference Peršoh, Beck and Rambold2004; Yahr et al. Reference Yahr, Vilgalys and DePriest2004, Reference Yahr, Vilgalys and DePriest2006) and in Stereocaulaceae (Peksa & Skaloud Reference Peksa and Skaloud2011), whereas most foliose Lecanorineae and Teloschistineae select Trebouxia spp. as photobiont. Nevertheless, there are some reports, based on microscopic investigations, of Asterochloris photobionts among Parmeliaceae (summarized by Rambold et al. Reference Rambold, Friedl and Beck1998); these deserve re-investigation with molecular tools. Asterochloris species were reported twice as photobionts of X. parietina, which was postulated to reveal low specificity (Ahmadjian Reference Ahmadjian and Seckbach2002b ). Asterochloris italiana was originally isolated from an Italian sample (Tomaselli Reference Tomaselli1956) as Cystococcus Xanthoriae parietinae. No details are given on isolation techniques, nor is a voucher deposited. One out of Tomaselli's three different photobiont isolates, originating from three different X. parietina specimens, was kept in the Cambridge Culture Centre (CCC) as T. decolorans. It became the type strain of A. italiana (sub Trebouxia italiana), the cells of which are mentioned to be multinucleate (Archibald Reference Archibald1975). Peršoh et al. (Reference Peršoh, Beck and Rambold2004) speculate in this particular case on confusion of strains. The fate of this type species cannot be reconstructed. Ahmadjian (Reference Ahmadjian and Seckbach2002b ) mentioned A. irregularis (sub Trebouxia irregularis) as photobiont of X. parietina, without giving any further details. The few Xanthoria parietina specimens investigated, and the phenotypically very similar but phylogenetically different Xanthoria samples from Australia, Tasmania and New Zealand, all corticolous, had photobiont genotypes either from subclade Ab (T. arboricola), Ad (T. decolorans) or from the assembly of T. decolorans genotypes which fall between subclades Ac and Ad. The genetic diversity of some of the corresponding fungal partners was studied with fingerprinting techniques (RAPD-PCR applied to sterile cultured single- or multispore-isolates, Honegger et al. Reference Honegger, Zippler, Scherrer and Dyer2004). These data suggest a relatively high similarity of Australian X. parietina with samples from the Western Mediterranean, including the Balearic and Canary Islands. The photobiont of X. parietina from Port Fairy, Australia (voucher no. 133) falls in subclade Ad, which comprises an interesting assembly of Trebouxia decolorans genotypes isolated from corticolous samples growing in coastal areas from Brittany to Majorca, Canary Islands, Greek Islands, South Africa and South-Eastern Australia. The mycobiont of an unnamed epiphytic Xanthoria species from Canberra (AUS; voucher no. 276), which is morphologically similar to X. parietina, was strongly dissimilar and formed an outgroup in the fingerprinting experiments (Honegger et al. Reference Honegger, Zippler, Scherrer and Dyer2004); its Trebouxia photobiont was found in the unresolved part of the “arboricola cluster” (Fig. 1).
Algal theft by Xanthoria spp. from Physcia species?
Based on the assumption of scarcity of free-living Trebouxia photobionts outside lichen thalli, Ott (Reference Ott1987a , Reference Ott b ; Ott et al. Reference Ott, Schröder and Jahns2000) addressed the question of how germinating ascospores of the always richly fertile X. parietina and X. polycarpa, both with no vegetative symbiotic propagules, acquire a compatible photobiont. She postulated temporary association of Xanthoria germlings with ultimately incompatible green-algal cells and/or invasion by ascospore-derived germ tubes into the thalli of adjacent Physcia spp. (Lecanorineae, Lecanoromycetes), theft of their Trebouxia photobiont and subsequent development of a brightly yellow-coloured thallus on or within the grey Physcia thalli. However, upon careful dissection, presumed chimaerae of X. parietina and Physcia tenella and/or P. adscendens were invariably found to be juvenile thalli of Xanthoria polycarpa which, at a young age, may be as grey as adjacent, small-lobed Physcia adscendens due to very small amounts of anthraquinones in their vegetative thallus, only pycnidial ostioles and apothecial discs being coloured by yellow anthraquinones (Honegger et al. Reference Honegger, Conconi and Kutasi1996). In their inventory of photobiont diversity within crustose and foliose species of the Physcietum adscendentis, X. parietina being part of this community, Beck et al. (Reference Beck, Friedl and Rambold1998) showed with molecular markers that the photobionts of Physcia spp. are not associated with X. parietina. Similar results were found for lichen communities of southern California: Physciaceae associated with different photobiont clades than Teloschistaceae, and no algal sharing was detected among representatives of the two families when thalli growing side by side were examined (Werth Reference Werth2012). Extensive studies on photobiont diversity within the Physciaceae (Dahlkild et al. Reference Dahlkild, Kallersjo, Lohtander and Tehler2001; Helms et al. Reference Helms, Friedl, Rambold and Mayrhofer2001; Helms Reference Helms2003) support this view.
The present findings on photobiont diversity in X. parietina and X. polycarpa indicate that both species associate with photobionts of clade ‘A’, that is with genotypes of T. decolorans (corticolous samples in the Northern Hemisphere) or T. arboricola (saxicolous X. parietina in the Northern Hemisphere, corticolous X. polycarpa in NZ). Thus photobionts of Physcia tenella and P. adscendens (subclade I1 sensu Helms Reference Helms2003) are unlikely acceptable algal partners of either mycobiont. Sorediate structures, as described by Ott et al. (Reference Ott, Schröder and Jahns2000) as evidence for colonization of sorediate Physcia thalli by X. polycarpa, are within the range of phenotypic plasticity of X. polycarpa (Eichenberger Reference Eichenberger2007). However, as already described by Ahmadjian (Reference Ahmadjian1960) with microscopy techniques and confirmed with molecular methods (Beck et al. Reference Beck, Friedl and Rambold1998), Buellia punctata, an inconspicuous crustose species of the Physcietum adscendentis, has the same T. decolorans photobiont as X. parietina and, according to the present findings, as X. polycarpa.
Are Trebouxia spp. free-living?
Ahmadjian (Reference Ahmadjian2002a , Reference Ahmadjian and Seckbach b ) wrote about “lingering lichen myths” such as the belief that Trebouxia spp. occur free-living outside lichen thalli and that they are photoautotrophic. Instead, he suggests Trebouxia spp. are not independent organisms, but heterotrophic ones, “both in the lichen thallus and also growing independently in culture”. Our long-term culturing experiments on agarized non-nutrient mineral media leave no doubt about the ability of Trebouxia species to live as independent, photoautotrophic organisms. Based on diverse microscopic observations on free-living Trebouxia cells in nature (Tschermak-Woess Reference Tschermak-Woess1978; Bubrick et al. Reference Bubrick, Galun and Frensdorff1984; Gärtner Reference Gärtner1985a ; Mukhtar et al. Reference Mukhtar, Garty and Galun1994; Ettl & Gärtner Reference Ettl and Gärtner1995; Schroeter & Sancho Reference Schroeter and Sancho1996; John et al. Reference John, Whitton and Brook2002; Rindi & Guiry Reference Rindi and Guiry2003; Sanders Reference Sanders2005; Handa et al. Reference Handa, Ohmura, Nakano and Nakahara-Tsubota2007; Hedenås et al. Reference Hedenås, Blomberg and Ericson2007), it seems reasonable to assume Trebouxia species are very widespread and distinctly more common than previously hypothesized. The fact that closely related Trebouxia genotypes occur in thalli of different sexually-reproducing Xanthoria spp. with no vegetative propagules on different continents, as shown in the present investigation (Fig. 1), indirectly indicates that these photobionts must be available in nature for re-lichenization events. Molecular probes might be used in future experiments to detect the availability of free-living Trebouxia photobionts of lichen-forming fungi in environmental samples.
Our sincere thanks are due to all friends and colleagues worldwide who kindly collected fresh specimens for this project (see Table 1); to Prof. Georg Gärtner, Innsbruck, for generous gifts of reference strains, to Prof. Dr Thomas Friedl and Dr Gert Helms, Göttingen, for giving us access to unpublished sequence data; to Dr Louise Lindblom, Bergen, for identifying Xanthomendoza weberi; to two unknown referees for constructive comments on this manuscript and to the Swiss National Science Foundation for financial support (grant no. 31-103860 to RH).
