View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Helsingin yliopiston digitaalinen arkisto

Phylogenetic studies of cyanobacterial lichens

Filip Högnabba

Botanical Museum Finnish Museum of Natural History University of Helsinki Finland

Plant Biology Department of Biological and Environmental Sciences University of Helsinki Finland

Academic dissertation

To be presented with the permission of the Faculty of Biosciences of the University of Helsinki for public criticism in Auditorium XII of the University Main Building, Unioninkatu 34, on March 30th 2007 at 12 o'clock noon.

Helsinki 2007 © Filip Högnabba (summary, articles III, IV) © International Association for Plant Taxonomy (article I) © The British Mycological Society (article II) © The Willi Hennig Society (article V)

Cover layout: Patrik Högnabba

Author’s address: Botanical Museum P.O. Box 7 FI-00014 University of Helsinki Finland e-mail: [email protected]

ISBN 978-952-92-1767-0 (paperback) ISBN 978-952-10-3798-6 (PDF) http://ethesis.helsinki.fi

Yliopistopaino Helsinki 2007 Phylogenetic studies of cyanobacterial lichens

Filip Högnabba

This thesis is based on the following articles:

I Myllys, L., Högnabba, F., Lohtander, K., Thell, A., Stenroos, S., Hyvönen, J. 2005. Phylogenetic relationships of Stereocaulaceae based on simultaneous analysis of beta-tubulin, GAPDH and SSU rDNA sequences. Taxon 54: 605- 618.

II Högnabba, F. 2006. Molecular phylogeny of the genus Stereocaulon (Stereo- caulaceae, lichenized ascomycetes). Mycological Research 110: 1080-1092.

III Högnabba, F., Stenroos, S., Thell, A., Myllys, L. Evolution of cyanobacterial symbioses in Ascomycota. Manuscript.

IV Högnabba, F., Stenroos, S., Thell, A. Phylogenetic relationships and evolution of photobiont associations in Lobariaceae (Lecanoromycetes, Ascomycota). Manuscript.

V Stenroos, S., Högnabba, F., Myllys, L., Hyvönen, J., Thell, A. 2006. High selectivity in symbiotic associations of lichenized ascomycetes and cyano- bacteria. Cladistics 22: 230-238.

These are referred to in the text by the corresponding Roman numerals: Contributions

The following table shows the major contributions of authors to the original articles or manuscripts. The initials refer to the authors of the articles in question.

I II III IV V Original idea LM FH SS, FH FH, SS SS Data LM, FH FH SS, FH, AT FH, AT, SS FH, SS Analyses LM FH FH FH FH, SS Manuscript preparation LM, FH FH FH, SS FH SS, FH

Supervised by: Doc. Soili Stenroos, Botanical Museum, Finnish Museum of Natural History, University of Helsinki Prof. Jaakko Hyvönen, Plant Biology, Department of Biological and Environmental Sciences, University of Helsinki

Reviewed by: Dr. Harrie Sipman, Botanical Museum Berlin-Dahlem, Free University of Berlin, Germany Dr. Gunilla Ståhls-Mäkelä, Zoological Museum, Finnish Museum of Natural History, University of Helsinki

Examined by: Prof. Anders Tehler, The Swedish Museum of Natural History, Stockholm, Sweden Summary

Filip Högnabba Botanical Museum, P.O. Box 7, FI-00014 University of Helsinki, Finland.

Introduction undergoes speciation. In modern classifications lichens are therefore treated as lichen-forming Lichens are intriguing mutualistic symbiotic fungi (Gargas et al., 1995; Honegger, 1996; assemblages between fungi (mycobiont) and Tehler, 1996). The lichen-forming life strategy green algae (phycobiont) or cyanobacteria has been adopted independently several times (cyanobiont) (see e.g. Honegger, 2001). during the evolutionary history of fungi (Gargas Associations including all three components in et al., 1995; James et al., 2006). Thus lichen- various combinations are also relatively forming fungi are not a monophyletic group. common (Hawksworth, 1988). Lichens are Lutzoni et al. (2001) show that lichenization therefore not organisms but can be viewed as may have occurred only once within the ecological phenomena or small ecosystems Ascomycota, but do not exclude the possibility (Tehler, 1996). Lichen symbiosis is a successful that this life strategy may have developed life strategy and lichens can be found in almost independently up to three times within the all terrestrial ecosystems, colonizing a wide Ascomycota. More importantly, Lutzoni et al. range of substrates. A few also occur in (2001) show that the symbiotic life strategy has freshwater and some are even found been lost repeatedly and that major fungal submerged in marine environments (Nash III, lineages are derived from lichen-forming 1996). Symbiosis between fungi and ancestors, a hypothesis also supported by autotrophic organisms, of which lichen Lutzoni et al. (2004). symbiosis is an example, has been suggested to At present about 13 500 fungal species be very old (Tehler, 1983; Tehler et al., 2003). have been recognized to be involved in lichen Recent fossil records support this view and it symbioses (Kirk et al., 2001). Of all lichen- has been shown that lichen-like symbioses forming species, 98% belong to the phylum were present 600 million years ago (Yuan et Ascomycota (Honegger, 1996). Of the al., 2005). approximately 32 000 species described in The taxonomy of lichen associations is Ascomycota, roughly 40% are lichen-forming based on the mycobiont, as this is the (Kirk et al., 2001). The majority of lichen component of the lichen that reproduces and symbioses (approximately 90%) are bipartite

5 associations between fungi and green algae, (Galloway, 1988; White & James, 1988). It has while about 10% are bipartite associations with been proposed to treat the cyanobacterial a cyanobacterium as the photobiont photomorphs as forma of the green algal (Tschermak-Woess, 1988). Tripartite species (Laundon, 1995), to give the associations, where green algae are more or cyanobacterial counterparts names without less evenly distributed in the lichen thallus nomenclatural status within quotation marks while cyanobacteria occur in specialized (Jørgensen, 1996), or to append “cyan.” or structures (cephalodia) spatially isolated from “chlor.” after the correct name (Heidmarsson, the green algae, occur in about 3-4% of lichens 1997). However, no consensus has been (Honegger, 1996). reached on this matter. One particular group of lichen-forming The main goal of my thesis was to fungi consists of species that are able to form reconstruct phylogenetic relationships of photomorphs, where the same fungal species defined groups of lichen-forming fungi that form symbioses with either green algae or include species involved in symbioses with cyanobacteria. The photomorphs may be cyanobacteria. Phylogenetic hypotheses morphologically very different, but representing natural relationships between morphologically similar photomorphs has also taxa should form the basis for classification. been demonstrated (James & Henssen, 1976). Based on phylogenetic relationships, the origin These photomorphs may occur separately or and evolution of specific symbioses, particularly form composite thalli (photosymbiodemes). those involving cyanaobacteria, were studied. The ability of lichen fungi to form Genetic diversity and phylogenetic photomorphs has resulted in taxonomical relationships of symbiotic cyanobionts were problems, particularly for those species in also studied in order to examine selectivity of which the photomorphs exist independently cyanobionts and mycobionts as well as possible (Laundon, 1995; Jørgensen, 1996, 1997, 1998; co-evolution between partners involved in Heidmarsson, 1997). It has been difficult to lichen associations. demonstrate that morphologically distinct The aims of study I were to examine the lichens contain the same fungus, and hence delimitation and position of Stereocaulaceae photomorphs of the same fungal species have based on multiple gene loci. In study II my at times been assigned to different genera. aims were to present a more detailed study of Careful morphological studies have, however, the phylogenetic relationships of the genus tied together morphologically different Stereocaulon (Stereocaulaceae) in order to test photomorphs (James & Henssen, 1976). The the existing infrageneric classification, to development of molecular tools has made the investigate the placement of some crustose recognition of photomorphs easier (Armaleo & species described in the genus, and to test the Clerc, 1991; Goffinet & Bayer, 1997; Stenroos suggested transfer of Muhria to Stereocaulon. et al., 2003; Takanashi et al., 2006a). According The aims of study III were to reconstruct the to the code of botanical nomenclature one phylogenetic relationships of Ascomycota and taxon can have only one correct name to examine the evolution of cyanobacterial (McNeill et al., 2006). However, different symbioses within the phylum. In study IV the names have often been assigned to different goals were to investigate the delimitation of photomorphs of the same fungal species the genera currently included in the family

6 Lobariaceae, to examine the relationships of of their simple morphology. Molecular markers previously poorly studied austral species, and can also be readily utilized in studies of co- finally to examine evolution and stability of evolution and specificity between mycobionts various symbiotic associations found within the and photobionts, even without identification of family. In study V genetic variation and the partners. phylogenetic relationships of cyanobionts were The phylogenetic analyses presented here examined and compared to those of free living are based solely on DNA sequence data. In cyanobacteria in order to study selectivity of studies I through IV the genetic diversity and mycobionts and photobionts involved in lichen phylogenetic relationships of the mycobionts symbioses. were studied using different combinations of sequences from the following gene regions: the ribosomal RNA encoding genes of the nuclear Materials and Methods ribosomal DNA cluster, i.e. the nuclear small subunit ribosomal DNA (nSSU, 18S), the DNA sequence data nuclear large subunit ribosomal DNA (nLSU, 28S), and the 5.8S nuclear ribosomal DNA; the Many lichens show great morphological non-coding internal transcribed spacers 1 and plasticity, which often causes problems in 2 (ITS1 and ITS2) of the nuclear ribosomal securely interpreting relationships between DNA cluster; the small subunit ribosomal RNA taxa. Furthermore, in many groups of lichens encoding gene of the mitochondrial ribosomal morphology is very simple and characters DNA (mtSSU); the protein coding nuclear useful for identification are not readily available beta-tubulin gene; the protein coding nuclear (e.g. crustose lichens). The use of DNA glyceraldehyde 3-phosphate dehydrogenase sequence data has thus had a considerable gene (GAPDH). Of these the most widely used impact on the taxonomy as well as on gene region in fungal systematics is the nuclear evolutionary studies of lichen-forming fungi; ribosomal DNA cluster (Lutzoni et al., 2004). the inclusion and position lichens in the fungal As the nuclear ribosomal genes are the most system has been securely assessed based on commonly used they are often chosen in order sequence level data. In addition, as discussed to maximise the potential for combining newly above, the identification of morphologically produced sequences with those available from different photomorphs of the same fungal GenBank. Furthermore, the nuclear ribosomal species can be achieved using DNA sequence genes are relatively easy to amplify due to the data. Recent results based on DNA sequences large number of primers available for this gene have also revealed that the same fungal species region. The ITS1 and ITS2 regions are may adopt very different life strategies (Wedin comparatively variable and therefore assumed et al., 2004). to be suitable for analysis at infrageneric and Photobionts included in lichen symbioses even infraspecific levels. For this reason the show a reduced morphology (Friedl & Büdel, ITS1-5.8S-ITS2 portion of the nuclear 1996). Identification and evolutionary studies ribosomal DNA has often been used to study on photobionts have thus been difficult. DNA closely related species. The nSSU and nLSU are sequence data and phylogenetic analyses have less variable and provide phylogenetic permitted identification of photobionts in spite information at higher taxonomic levels. The

7 mtSSU has also become commonly used in deletions (indels). Parsimony analysis given phylogenetic studies, particularly those at equal weight to all of these types minimize the family or higher taxonomic levels (e.g. Schmitt overall number of hypothesized et al., 2001; Lumbsch et al., 2002, 2004, 2005; transformations, and results in the most Wedin et al., 2005). Recent studies have parsimonious solution(s) (Frost et al., 2001, shown that the beta-tubulin gene provides 2006; Grant, 2003; Grant & Kluge, 2003). phylogenetic information at infrageneric and Assigning different weights to different types of infraspecific levels (Myllys et al., 2001, 2003; transformations in parsimony analyses Articus et al., 2002; Stenroos et al., 2002a; minimizes certain transformations at the Thell et al., 2002; Högnabba & Wedin, 2003; expense of others according to a priori Molína et al., 2004). In particular, the intron specified assumptions about evolution, and regions and the third-codon position in the always leads to less parsimonious solutions exon regions have been shown to contain overall (Frost et al., 2001, 2006; Grant, 2003; phylogenetic information even below the Grant & Kluge, 2003). A priori assumptions of species level (Myllys et al., 2001). Also, the evolutionary process are also included in GAPDH gene has shown to be variable and maximum likelihood and Bayesian analyses useful even in studies of interspecific and where models of character evolution and intergeneric relationships (Myllys et al., 2003; statistical methods are used to estimate Thell et al., 2004). phylogeny. Methods including models of In study V the genetic diversity and evolutionary processes are problematic, mainly phylogenetic relationships of cyanobionts were due to the high numbers of untested a priori studied using sequences from the following assumptions involved. Evolution is an gene regions: the small subunit of the extremely complicated process and it is ribosomal DNA (16S); the ribulose-1.5- therefore unrealistic to try to explain it using bisphosphate carboxylase/oxygenase large simplified models, especially when our goal is subunit (rbcL) and ribulose-1.5-bisphosphate to reconstruct phylogenies of very old lineages. carboxylase/oxygenase small subunit (rbcX) An agnostic approach that minimizes a priori genes; the tRNA-Leu gene. assumptions and “lets the data speak for itself” was therefore applied throughout this thesis. To achieve this, the parsimony optimality Phylogenetic analyses criterion and unweighted analyses were exclusively used to infer phylogenetic Several methods are available for inferring relationships. phylogenetic relationships (Felsenstein, 2004). Phylogenetic relationships are revealed by At present most analyses are conducted using comparing character states expressed by methods based on parsimony, maximum characters that are homologous, i.e. share a likelihood or Bayesian inference, all of which common ancestor. Sequence alignment, i.e. try to explain observed variation by minimizing constructing hypotheses about nucleotide the number of character transformations homology a priori is often problematic. The (Frost et al., 2006). For DNA sequence data, same dataset may be aligned in various ways possible transformations are transitions, depending on the criteria used to derive transversions, and nucleotide insertions/ hypotheses of nucleotide homology and to

8 detect possible indels (Wheeler, 1995). to this, manually created or manually optimized Different alignments may support different alignments are nearly always subjective and not phylogenetic hypotheses, thus poorly chosen repeatable. To minimize the number of hypotheses of homology and indel recognition ambiguous homology assumptions, regions that may introduce errors to the hypotheses about are not readily aligned are often removed from phylogeny. Grant & Kluge (2004) argued that the analyses. However, the decisions on which nucleotide homology must be evaluated with regions to exclude are always arbitrary and reference to a topology. For protein-coding subjective. Furthermore, exclusion of data genes without length variation apparently might also result in the loss of informative unambiguous alignments, i.e. unambiguous characters. It has also been shown that assumptions of nucleotide homology, can be resolution will be lost if enough data are constructed. However, introns in protein- discarded (Cameroon et al., 2004). When using coding genes (such as in beta-tubulin, see study a static alignment procedure the homology II), may show length variation, which if used assumptions will remain unchanged during the may cause alignment problems. Non-coding reconstruction of the phylogenetic hypotheses. regions on the other hand often show To avoid problems with conventional considerable length variation, causing problems alignment programs and manual sequence with homology assumptions and often making alignment, and to assess nucleotide homology unambiguous alignment impossible. Finding the in reference to tree topologies, direct optimal alignment even for a small number of optimization (Wheeler, 1996) was used for the short sequences is impossible, as already for analyses presented in this thesis. Direct five sequences with five nucleotides each there optimization is a heuristic procedure that are 1.05 * 1018 possible alignments (Slowinski, enables optimization of unaligned sequences of 1998). In a more realistic example presented different length directly onto competing tree by Slowinski (1998), where a maximum of topologies. The method operates by three gaps representing indel events can be constructing hypothetical sequences at each included in each of the five sequences, there internal node on a given tree. The hypothetical still remain over 500 million possible sequences are based on sequences at the alignments. closest descendant nodes and are constructed The standard approach in phylogenetic in order to minimize the cost of nucleotide analyses of molecular data is to do (multiple) transformations and indels that must be sequence alignment(s) followed by a tree hypothesized between the descendants. The search. Usually the alignment is created either overall cost of a tree is the sum of the costs manually or with an alignment program such as for all transformations (including indels) at each Clustal X (Jeanmougin et al., 1998). However, internal node. This procedure is performed on Clustal and other straightforward alignment competing topologies and the topology with programs often introduce obvious errors into the lowest cost is the one considered to be the homology assumptions, and even the order the best representation of the observed data, of the included terminals may cause identical i.e. the most parsimonious tree. As the sequences to be aligned in different ways. sequences are optimized directly onto These shortcomings are often addressed by competing topologies, unique base-to-base manual adjustment of the alignment. In addition homology assumptions are made for each tree.

9 Therefore the homology assumptions are said advantage of parallel computing using to be dynamic, in contrast to the static processor clusters at the Center of Scientific homology assumptions made prior to analysis Computing (CSC), Espoo, Finland, and at the with conventional methods. The distinction Finnish Museum of Natural History (FMNH), between static and dynamic is also made in the Helsinki, Finland. At CSC version 3.0.6 of the treatment of indels. In conventional methods, POY program runs on a Unix platform using indel events are represented by the insertion up to 16 1.1 GHz processors of the IBM of gaps in the alignment. In the tree search eServer cluster 1600, while at FMNH version these gaps are treated either as missing data or 3.0.11 of the program runs on a Beowulf as a fifth character state. In direct optimization, cluster with 21 2.4 GHz processors employing indels are seen as transformation events that Scyld Unix and parallel virtual machine (PVM). have occurred between ancestral and Although we were able to use parallel descendent nucleotide sequences and in that computing, the calculation time required was sense they are equivalent to other inferred often the limiting factor. In order to reduce transformations. For further discussion on calculation time sequences were cut into direct optimization see Wheeler (1996, 2000, shorter fragments (studies II, IV), an approach 2001, 2002), Phillips et al. (2000), Frost et al. that can speed up the analyses considerably (2001, 2006), and Schulmeister et al. (2002). and which was essential for their satisfactory Direct optimization is implemented in the completion (Janies & Wheeler, 2002). In order program POY (Wheeler et al., 1996-2003), in to determine where to cut the sequences which sequence optimization on different tree preliminary alignments were made either topologies takes advantage of heuristic tree manually or using Clustal X (Jeanmougin et al., search methods, such as branch swapping, tree 1998) or Dialign (Morgenstern, 1999). This drifting, treefusing, and ratcheting (Goloboff, procedure introduces some subjectivity into 1999; Nixon, 1999) to find the most the analyses. However all the cuts were made parsimonious solution. The problem of finding within highly conserved regions. In this the optimal alignment for a given topology is approach homology is assumed of certain NP-complete (NP = non-deterministic genes, or stretches of sequences, but these polynomial time; Wang & Jiang, 1994), i.e. a assumptions are not fixed a priori to the level problem for which an exact solution cannot be of individual nucleotides. In studies III and IV determined or verified in any practical way calculation times were further reduced by (Frost et al., 2006). Finding optimal trees conducting preliminary analyses using shortcut among competing topologies is in itself a NP- heuristics. These analyses usually find complete problem. Thus, as POY cladograms that are only marginally longer than simultaneously tries to find the optimal those found by more intense search strategies homology assumptions and tree topologies the (Wheeler et al., 2006). These cladograms were program has to deal with two NP-complete used as input topologies in the final analyses problems nested within each other. Therefore where they were further refined by additional the direct optimization procedure is branch-swapping. computationally demanding and requires high Bremer support values (Bremer, 1988, calculation power. For all the analyses 1994) were calculated in order to identify presented in this thesis we were able to take clades that lack support. These clades are

10 more likely to be altered with further sampling other families based on evidence from nSSU of either characters or terminals. More rDNA data. Cladia and Pilophorus have been commonly used branch support metrics in shown to be more closesly related to parsimony analyses are bootstrap (Felsenstein, podetiate genera of Cladoniaceae than to 1985) and jackknife (Farris, 1996) values. When Stereocaulon (Stenroos & DePriest, 1998; calculating bootstrap values artificial matrices Wedin et al., 2000). The pseudopodetiate are generated by resampling of characters, genus Austropeltum has been shown to be while jackknifing is based on partial exclusion closely related to species of Sphaerophoraceae of characters. The calculation of Bremer (Wedin & Döring, 1999; Wedin et al., 2000). support values is more “realistic” in the sense Wedin et al. (2000) and Stenroos et al. (2002b) that the original data is not manipulated. It circumscribed Stereocaulaceae to include should be pointed out, however, that the Stereocaulon only. In an analysis based on calculation of support metrics does not mitochondrial SSU and nuclear LSU data contribute any new data to the analyses. presented by Wiklund & Wedin (2003), Furthermore, well supported hypotheses may Pilophorus and Stereocaulon formed a mono- become less supported when new taxa or phyletic group. Based on ITS and nSSU rDNA characters are added (Sober, 1988; Kluge, data, Ekman & Tønsberg (2002) showed that 1997). The best test of phylogenetic the majority of the species of Lepraria and hypotheses is simply the addition of characters Leproloma form a monophyletic group that is or terminals to later analyses. I agree with sister to Stereocaulon. Thus Lepraria, including Grant & Kluge (2003) when they claim that too Leproloma, is currently classified in Stereo- much attention has been directed toward caulaceae (Eriksson, 2006). these metrics, and also with their quote from The genus Muhria was assigned to Popper (1983): “The cult of impressive Stereocaulaceae by Jørgensen & Jahns (1987). technicalities or the cult of precision may get This monotypic genus is morphologically the better of us, and interfere with our search similar to the crustose species of Stereocaulon for clarity, simplicity, and truth”. but was considered distinct based on apothecia development. Ekman & Tønsberg (2002) indicate a close relationship between Muhria Results and Discussion and Stereocaulon, as Muhria forms the sister taxon to the monophyletic Stereocaulon based Phylogenetic relationships of Stereocaulaceae on ITS sequence data. In a study by Printzen & Kantvilas (2004) Muhria is nested within The family Stereocaulaceae has traditionally Stereocaulon based on ITS sequence data, included species characterized by the indicating its inclusion in Stereocaulon. formation of pseudopodetia, i.e. fruticose However, no final conclusion was reached in secondary structures arising by elongation of this study due to the limited sampling of thalline tissue, and cephalodia. The use of DNA species. sequence data has had a great impact on the Hertelidea is a recently described genus in circumscription of the family. Many of the Stereocaulaceae (Printzen & Kantvilas, 2004). pseudopodetiate genera formerly classified in The inclusion of Hertelidea in Stereocaulaceae Stereocaulaceae have been transferred to is based on similarities of morphology and

11 chemistry, and is further supported by ITS disagreement with this hypothesis. The sequence data. In the Bayesian analyses phylogeny presented in study I would allow the presented by Printzen & Kantvilas (2004), inclusion of the crustose Stereocaulon cumu- Hertelidea botryosa forms the sister taxon to a latum and S. leucophaeopsis in Muhria. This Lepraria + Stereocaulon clade. Taxon sampling in would, however, necessitate further nomen- this study is, however, limited and based on a clatural changes due to the basal position of single DNA region. Broader sampling of taxa as Stereocaulon tornese. To maximize nomen- well as of characters within Stereocaulaceae, clatural stability the inclusion of Muhria in including Hertelidea, would be necessary to Stereocaulon is therefore preferred. The formal reach a final conclusion about the status of this renaming of Muhria is included in study II. species. Based on anatomical, morphological, and chemical characters five crustose species have Phylogeny of Stereocaulon been assigned to Stereocaulon, although the genus traditionally included only fruticose Species traditionally assigned to the genus species (Purvis & James, 1985; Fryday & Stereocaulon are characterized by a crustose Coppins, 1996; Timdal, 2002; Fryday & Glew, primary thallus from which a fruticose 2003). Printzen & Kantvilas (2004) showed that secondary thallus develops by elongation of the crustose Stereocaulon cumulatum is the thalline tissue into stalks, i.e. pseudopodetia, sister taxon of a Muhria + Stereocaulon (fruti- which are in many cases richly branched. These cose) clade based on ITS sequence data. pseudopodetia support phyllocaldia and, in In study I the phylogenetic relationships of most of the species, cephalodia. Fries (1857, Sterecaoulaceae were investigated using beta- 1858) presented the first monographic tubulin, GAPDH and nSSU rDNA sequences. treatment of the genus. Several refinements In our combined analysis, Stereocaulaceae, and alternatives to Fries’ concepts (Nylander including Muhria, Lepraria, and Stereocaulon, 1858-1860; Tuckerman, 1872; Vainio, 1890, forms a monophyletic group that appears as 1915; Riddle, 1910; Magnusson, 1926; Dodge, the sister group to Cladoniaceae (Fig. 4 in 1929; Räsänen, 1943; Lamb, 1951) were study I). Within Stereocaulaceae Lepraria forms published before Lamb (1977) devised the a monophyletic group. Muhria is clearly nested scheme that is currently accepted. A more within Stereocaulon, and thus Stereocaulon is recent infrageneric classification of the genus monophyletic only with the inclusion of Muhria. was published by Dombrovskaya (1996) in her The crustose Stereocaulon tornense is the treatment of Stereocaulon of the former Soviet earliest diverging taxon in the Stereocaulon + Union. Being restricted in its geographical Muhria clade. The other two crustose scope it cannot be accepted as the valid Stereocaulon species included, S. cumulatum and infrageneric classification. S. lecophaeopsis, together with Muhria, form a Monophyly of Stereocaulon, including the sister group to the fruticose species of crustose species and Muhria, was confirmed by Stereocaulon. Distribution of the crustose more extensive species sampling from within growth form in Stereocaulon indicates that it is Stereocaulon (Fig. 1 in study II). Thus, the the primitive state in the genus. The results traditional view of Stereocaulon as only obtained in study II were, however, in including fruticoses species is discarded. The

12 current infrageneric classification of subsection Holostelidium, as all other included Stereocaulon is not supported as such. species of section Holostelidium are classified in The two main subgenera of Stereocaulon, subsection Aciculisporae. Therefore, a more Holostelidium and Stereocaulon, which are thorough sampling of species from section distinguished based on differences in Holostelidium is required before any final pseudopodetial development, are separated in conclusion can be reached. Section the phylogeny. The monospecific subgenus Stereocladium includes the single species Pilophoropsis could unfortunately not be Stereocaulon nanodes. Its nested position within included in the study. The members of group 8 (Fig. 1 in study II), together with subgenus Holostelidium, excluding Stereocaulon species of the sections Stereo-caulon and sorediiferum which appears as the sister species Denudata, challenges the need to place it in a of the rest of the genus, form a monophyletic section of its own, although it is group. The members of the subgenus Stereo- morphologically distinct. caulon excluding Stereocaulon pileatum, S. verru- Of all the subsections, Aciculisporaea is the culigerum, and S. virgatum also form a mono- only one that is monophyletic and can be used phyletic group. Most of the representatives of without alteration in a forthcoming the two subgenera appear separated into two classification. As stated above however, different lineages that can also be distinguished sampling from other groups within the section by morphological characters. These lineages Holostelidium is insufficient. For a detailed should form the basis for any forthcoming discussion of the subsections, see study II. classification. Further studies need to be The nested positions of the crustose undertaken to clarify the position of the species and of Muhria shown in study II species not occurring in any of these main strongly support the conclusions of study I, lineages. that these species should be included in The section Lobophoron is monophyletic Stereocaulon. Therefore a new combination was and could be used at some level in a future made for Muhria urceolata. Contradicting the classification. The sections Stereocaulon, results of study I, however, the crustose Denudata, and Holostelidium are not mono- species are well nested within Stereocaulon. The phyletic. Of these, the sections Stereocaulon crustose growth form was therefore not and Denudata are well represented in the supported as the likely ancestral state in present study and thus it is concluded that they Stereocaulon, as has been suggested (Lamb, should not be recognized as currently 1951; study I). Rather, the nested position of delimited. The non-monophyly of the section the crustose species within the genus indicates Holostelidium is due to the basal placement of that these species represent a lineage in which Stereocaulon sorediiferum, while the rest of the the fruticose growth form has been lost or is Holostelidium is monophyletic. With the not currently expressed. The monophyly of the exclusion of S. sorediiferum, the section crustose species indicates that this group could Holostelidium could also be used in a future be formally treated in an infrageneric classification. The position of S. sorediiferum classification. remains unclear due to missing ITS sequences A new infrageneric classification of for this species in the analyses. Furthermore, S. Stereocaulon is clearly needed in order to sorediiferum is the single representative of reflect phylogenetic relationships within the

13 genus. A new classification is not, however, Based on molecular data the family presented at this stage. The results should Lobariaceae has been shown to be mono- rather be regarded as representing a phyletic. Within Lobariaceae the genus Sticta hypothesis that reflects problems with the appeared monophyletic in studies by Thomas current classification and hopefully stimulates et al. (2002) and Stenroos et al. (2003), while further studies of the group. One of the main the results of Takanashi et al. (2006a, 2006b) reasons for not presenting a new classification suggested that the genus is polyphyletic. Lobaria is insufficient morphological understanding of was found to be monophyletic by Wiklund & the genus. There are no obvious morphological Wedin (2003) and Wedin & Wiklund (2004). characters defining the various groups revealed By contrast, Stenroos et al. (2003), by DNA sequence data, other than the Miadlikowska & Lutzoni (2004), and Takanashi distinction between the two subgenera based et al. (2006a, 2006b) all found Lobaria to be on differences in pseudopodetial development, non-monophyletic. Pseudocyphellaria was found and the crustose growth form, which defines a to be clearly non-monophyletic by Thomas et monophyletic group. The characters used by al. (2002), Stenroos et al. (2003), and Lamb (1977) to separate infrageneric groups Takanashi et al. (2006a, 2006b). The results of are in many cases problematic as most of them Miadlikowska et al. (2002) also suggested non- are very general and allow for much variation. monophyly of Pseudocyphellaria, with two Further taxon sampling for molecular studies species grouping outside of the rest of the as well as studies of morphology and chemistry genus. are obviously required before a conclusive In study IV we present analyses based on hypothesis on the infrageneric classification of the most extensive taxon sampling from the Stereocaulon can be presented. Lobariaceae undertaken so far. Southern Hemisphere species in particular were extensively sampled, as these have clearly been Phylogeny of Lobariaceae underrepresented in previous studies. Lobariaceae were revealed to be monophyletic Lobariaceae include the genera Dendriscocaulon, (Fig. 1 in study IV), thereby confirming the Lobaria, Pseudocyphellaria, and Sticta (Eriksson, results of previous studies. Sticta, including 2006). The status of Dendriscocaulon has Dendriscocaulon, also forms a monophyletic remained unclear, as several species described group. The nested position of Dendriscocaulon within the genus have been shown to be dendroides in Sticta suggests that it could be a cyanomorphs of species of Lobaria and Sticta. photomorph of the latter. Lobaria and Lobariella and Lobarina are recent segregates Pseudocyphellaria are clearly non-monophyletic. from Lobaria and are as yet poorly known Lobaria species occur in two clades, neither of (Yoshimura, 1998, 2002; Stenroos et al., 2003). which is exclusively comprised of species The established genera of Lobariaceae, Lobaria, currently classified in the genus; representa- Pseudocyphellaria and Sticta, have been tives of Lobarina and Lobariella, as well as separated based on characters of their cortical Pseudocyphellaria, are nested within both of tissue, i.e macula-like thin patches in Lobaria, these clades. Pseudocyphellaria species are pseudocyphellae in Pseudocyphellaria and distributed in as many as five clades. Most of cyphellae in Sticta. the species are distributed in the three clades

14 consisting solely of species currently classified were included turned out not to be in Pseudocyphellaria. Two of these groups form monophyletic. Stereocaulon saxatile, for a larger monophyletic entity while all three example, seems to be wildly misunderstood as plus the Sticta clade form a monophyletic it appeared in three of the major clades within group. Finally, a few Pseudocyphellaria species Stereocaulon. On the other hand, samples of are nested in two clades including mainly Stereocaulon tomentosum from distant Lobaria species. populations in Finland and southern South The results presented in study IV clearly America group together, indicating that show that a new classification of Lobariaceae is delimitation of the species may be accurate needed. Although our study represents the (Fig. 1 in study II). most extensive sampling of Lobariaceae to An interesting case is found in group 4 (Fig. date, there are still many species that remain 1 in study II), where the foliose Stereocaulon to be sampled. Thorough studies of foliolosum and the sorediate S. coniophyllum are morphology are also required before a final intermixed. Perhaps these should be regarded proposal for a new classification is made. The as belonging to a “species pair”, differing only characters traditionally used to separate the in the mode of reproduction. Whether the genera, such as the cortical tissue, are clearly morphotypes of such species pairs should be not useful. A characteristic that has been treated as different species or not was discussed in connection with recent discussed by Poelt (1970, 1972) and Tehler phylogenetic studies of Pseudocyphellaria (1982). According to the phylogenetic species (Thomas et al., 2000, 2002) is the distinction concept different species should occur in between species with yellow and white separate clades. If they do not, as in the case of medulla. Our observations, however, do not S. foliolosum and S. coniophyllum, they could be reveal this character to be useful either. Our treated as forms of the same species (Tehler, study is a first step towards a classification that 1982). An example of a possible “species pair” better represents the natural relationships is found also in Lobariaceae, where within the family. Pseudocyphellaria anomala and P. anthraspis (the latter being the fertile counterpart of the sorediate P. anomala; Brodo et al., 2001), group Species delimitation together in clade B (Fig. 1 in study IV). It remains to be tested by the inclusion of Species delimitation of lichens is not always multiple samples whether these species group clear as many species show considerable in separate clades or are intermixed in a morphological variation. The morphological monophyletic group. variation in Stereocaulon is sometimes higher For those species of Lobariaceae for which within than between species (Carlin, 1990). To multiple samples were included, most species test if the species of Stereocaulon and form monophyletic entities. However, Lobariaceae form monophyletic entities, Pseudocyphellaria crocata and P. intricata (two multiple samples of species were included in specimens referred to as P. cf. intricata) seem studies II and IV. The problem of variation is to be misunderstood, as they are both clearly reflected in the phylogeny as 10 of the 21 non-monophyletic. Within the Sticta clade Stereocaulon species for which multiple samples there are also a few non-monophyletic species.

15 However, this clade is highly unresolved and group has evolved late within the Ascomycota. therefore difficult to interpret (Fig. 1 in study The few other studies that include IV). cyanobacterial lichens are mostly restricted to The results presented in studies II and IV particular groups and are therefore based on indicate that further research at the species limited taxon sampling. level is very much needed, particularly within In study III we examined the evolution of Stereocaulon, but also for some species of cyanobacterial symbioses within the Lobariaceae. In order to reach final conclusions Ascomycota. Our results reveal that species about species concepts extensive sampling associated with cyanobacteria are widely within single species and the use of genetic distributed across the Ascomycota, indicating markers providing sufficient information at low repeated gains of the symbiotic association taxonomic levels will be required. Ideally, the with cyanobacteria. Gain of cyanobacterial results based on molecular data should be photobionts among non-lichenized species was linked to morphological characters. indicated in only one ambiguous case, while switches from green algal to cyanobacterial photobionts have taken place repeatedly. Evolution and specificity of cyanobacterial Tripartite associations have been formed symbioses repeatedly through the gain of green algae as well as through the gain of cyanobacteria. No Cyanobacterial symbioses are found in various losses of cyanobacteria resulting in a non- groups within the Ascomycota. Bipartite lichenized life strategy were observed; neither cyanobacterial symbioses are found primarily in was any switch from the cyanobacterial state the class Lichinomycetes and in the order to the green algal state. Cyanobacterial Peltigerales (Lecanoromycetes). The small symbiosis thus appears to be stable in family Arctomiaceae also includes bipartite comparison to green algal symbiosis, which has cyanobacterial lichens (Lumbsch et al., 2005). A been lost repeatedly (study III; Lutzoni, 2001). tripartite symbiotic life strategy is found in Study III also suggests stability of the tripartite Peltigerales, Agyriaceae, Cladoniaceae, symbiosis as only two ambiguous cases were Coccotremataceae, Pilocarpaceae, Porpidi- observed where one of the two photobionts in aceae, and Stereocaulaceae, amongst other a tripartite association might have been lost; groups. this indicates loss of cyanobacteria. However, The evolution of cyanobacterial symbioses this is in contradiction to the pattern found in within Ascomycota remains little studied. study IV, as repeated losses of green algae Cyanobacteria were among the first resulting in bipartite cyanobacterial photobionts available. This was presented as assemblages have taken place within evidence for the hypothesis that the Lobariaceae. Ancestral state reconstruction in peltigeralean group evolved early within the study IV shows that the ancestor of Ascomycota (Hawksworth, 1988). However, Lobariaceae was associated with cyanobacteria. Wiklund & Wedin (2003), Miadlikowska & Green algae have been gained repeatedly in the Lutzoni (2004), and Wedin & Wiklund (2004) family to form tripartite symbioses. The gain of studied the phylogenetic relationships of green algae has not resulted in a complete loss peltigeralean fungi and demonstrated that this of cyanobacteria in any of the Lobariaceae

16 included in our studies. Thus the stability of Knowledge of the evolution of interactions the cyanobacterial symbioses found in study III between cyanobacteria and lichen-forming was confirmed. fungi is still limited. Further studies are Lichen symbioses offer a unique opportunity to therefore necessary to clarify evolutionary study the specificity and co-evolution of the patterns and factors that may affect the included partners. Various levels of specificity formation of these intriguing and successful life have been demonstrated between mycobionts strategies. and their green algal photobionts (e.g. Beck el al., 1998, 2002; Kroken & Taylor, 2000; Dahlkild et al., 2001; Helms et al., 2001; Conclusions Piercey-Normore & DePriest, 2001; Yahr et al., 2004, 2006). Furthermore, Paulsrud & Based on three independent loci the Lindblad (1998), Paulsrud et al. (1998, 2000, circumscription of the family Stereocaulaceae 2001) showed, using molecular data, that to include Stereocaulon and Lepraria was mycobionts are selective towards cyanobionts. supported. The affinity of the genus Hertelidea By contrast, Oksanen et al. (2002), Rikkinen et with Stereocaulaceae remains to be confirmed. al. (2002), Summerfield et al. (2002), Lohtander The genus Muhria was shown to be best placed et al. (2003), Wirtz et al. (2003), O’Brien et al. in Stereocaulon, and thus the necessary new (2005), and Summerfield & Eaton-Rye (2006) combination for Muhria urceolata was made. demonstrate that the fungi and the Stereocaulon urceolatum (syn. Muhria urceolata) cyanobacteria may not require a specific shows close affinities to the other crustose partner. species of Stereocaulon, all of which, based on In study V we examined the selectivity of molecular data, seem to be correctly placed mycobionts and cyanobionts in lichen together with fruticose species in Stereocaulon. associations. High selectivity of mycobionts Phylogenetic relationships of the species of towards cyanobionts was found particularly in Stereocaulon reveal that the current infra- the genera Pseudocyphellaria, Sticta, Leptogium, generic classification, which is based on and Collema. Species assigned to these genera morphology, anatomy, and chemistry, requires seem to form lichen associations with specific major revision. No obvious morphological strains, species, or species groups of characters that would define the observed cyanobacteria. Species of the genera groupings within the genus have been found. Stereocaulon, Nephroma, Peltigera, and Lobaria Therefore, thorough studies of morphology as show lower selectivity and form lichen well as further sampling of taxa not included associations with a broad spectrum of thus far will be necessary before any conclusive cyanobacteria. High selectivity towards the infrageneric scheme for Stereocaulon can be photobiont was, however, found in Peltigera presented. Lobariaceae as currently and Nephroma in a study by Myllys et al. (2007). circumscribed form a monophyletic group. No indication of co-evolution between cyano- Within the family the genus Sticta including and mycobionts was demonstrated, since Dendriscocaulon dendroides is resolved as cyanobionts grouping in the same clades share monophyletic. The genera Pseudocyphellaria and different species, genera and even families of Lobaria are clearly non-monophyletic. Lobarina, mycobionts. recently segregated from Lobaria, shows close

17 affinities with the latter and its status is Acknowledgements therefore questionable. Lobariella, which is also a recent segregate of Lobaria, was The completion of this thesis has occasionally monophyletic. However, the nested position of required hard work. However, most of the Lobariella within Lobaria challenges its identity. time the work has been great fun, which The nested position of Dendriscoculon mostly has been due to the large group of dendroides within Sticta indicates that it should people that have taken part or followed the be treated as a cyanomorph of a Sticta species. project in various ways. Without these people To conclude, a major revision of the family I suppose that this thesis would look Lobariaceae is also called for. The completely different, or more likely not exist at morphological characters currently used to all. My sincerest thanks are due to my separate the genera within the family are supervisors Dr. Soili Stenroos and Professor obviously not useful if the classification is to Jaakko Hyvönen for creating perfect reflect natural relationships. Useful characters opportunities for me to do this job. Financial still remain to be discovered, and therefore support through Soilis projects has enabled me thorough studies of morphology are still to focus on the research. Soili has allowed me required. Species concepts need to be critically to work independently without any extra revised in Stereocaulon as well as in some pressure. This suits me well and is certainly the Lobariaceae species. In Stereocaulon, several of best way to get me to do my best. Still, she has the species for which multiple samples were always been ready to answer my numerous included were shown to be non-monophyletic. questions and patiently read my attempts to Within Lobariacea a few species also turned write manuscripts and kindly pointed out out to be seriously misunderstood. Further mistakes in my, often, sloppy written texts. studies at the species level are therefore Thanks to Jaakko I have been able to use the necessary. All of the groups that were studied most advanced methods and facilities for include species associated with cyanobacteria. completing the phylogenetic analyses. Jaakko The cyanobacterial symbiosis was shown to also always had time and patience to answer have evolved repeatedly in the Ascomycota my questions regarding phylogenetic analyses. and to be more stable than the symbiosis with Professor Teuvo Ahti and Dr. Orvo green algae. Of the genera including Vitikainen have kindly helped me from my first cyanobacterial photobionts, Pseudocyphellaria, day at the Botanical museum. Thanks to their Sticta, Collema, and Leptogium seem to be highly broad knowledge in lichenology they have selective towards the cyanobiont, while always been able to answers to my many Stereocaulon, Nephroma, Peltigera, and Lobaria questions, and their extensive private libraries show less specificity. No evidence of co- that have been freely available have been evolution between the components in invaluable for my work. The group of cyanobacterial lichens could be demonstrated. lichenologists working at the Botanical Museum Many questions concerning the evolution of in Helsinki has grown considerably during the cyanobacterial symbiosis remain open for past two years. After moving to Kaisaniemi Dr. future research. Leena Myllys and Dr. Katileena Lohtander have generously shared their knowledge with me, providing a nice opportunity to get different

18 views on various problems. The two most Finland to projects led by Soili Stenroos. recent members of our group, Saara Velmala Financial support has also been provided by and Satu Laitinen, have further strengthen the Societas pro Fauna et Flora Fennica, Vasa nice feeling of working in a research team nation, Svenska kulturfonden, Svensk which I am sure will lead to even better result Österbotniska samfundet, and Otto A. Malms in the future. Our distant collaborator Dr. donationsfond. Arne Thell have always encouraged me in my Without mentioning any names (the list work and always had time to review would be to long and I would probably forget manuscripts. Jaana Kekkonen has been of great someone) I would like to thank all the friends help by ensuring that the work in the MES-lab at Vasa nation that I have get to know during has been running without interruptions. the years in Helsinki. Spending time with you I wish to thank Aino Juslén for many has provided a meaningful leisure time and valuable discussions during our PhD projects. many pleasant moments, which has given me Aino also taught me, with her example, the energy to concentrate also on my work. important art of finishing the work. Thanks I am not sure if my family really has also to Jässe who often cheered up the lunch understood how lichens can keep someone brakes with political discussions and other busy for many years. Still they have always stories, to Sirkka in the library who always supported what I have been doing. helped me when I couldn’t find what I was Sussi, thanks for everything! looking for, to Laura Kivistö who kindly checked my determinations of Stereocaulon species from Finland. Sincere thanks also to all References the other colleagues at the Botanical museum who have made it a nice place to work at. Armaleo, D., Clerc, P. 1991. Lichen chimeras: I am obliged to Professor Hiroyuki DNA analysis suggests that one fungus forms Kashiwadani and Professor Masakane Inoue for two morpotypes. Experimental Mycology 15: taking good care of me while visiting Japan. 1-10. Professor Kashiwadani also helped me greatly Articus, K., Mattsson, J.-E., Tibell, L., Grube, with determination of Japanese Stereocaulon M., Wedin, M. 2002. Ribosomal DNA and β- species and professor Inoue provided tubulin data do not support the separation of important material for the study. Professor the lichens Usnea florida and U. subfloridana as Stefan Ekman helped me greatly when visiting distinct species. Mycological Research 106: Bergen for collecting a rare species of 412–418. Stereocaulon. Curators at the herbaria B, BG, Beck, A., Friedl, T., Rambold, G. 1998. GZU, HBG, O, TUR, and U kindly provided Selectivity of photobiont choice in a defined material for my study. lichen community: inferences from cultural Dr. Harrie Sipman and Dr. Gunilla Ståhls- and molecular studies. New Phytologist 139: Mäkelä are gratefully acknowledged for 709–720. efficient reviewing of this thesis. Dr. Neil Bell Beck, A., Kasalicky, T., Rambold, G. 2002. kindly corrected the language, sincere thanks. Myco-photobiontal selection in a This work has been supported financially Mediterranean cryptogam community with mainly by research grants from the Academy of Fulgensia fulgida. New Phytologist 153:

19 317–326. Felsenstein, J. 2004. Inferring phylogenies. Bremer, K. 1988. The limitations of amino acid Sinauer Associates Inc., Sunderland, sequences data in angiosperm phylogenetic Massachusetts. reconstruction. Evolution 42: 795–803. Friedl, T., Büdel, B. 1996. Photobionts. In: Nash Bremer, K. 1994. Branch support and tree III, T.H., ed. Lichen biology. Cambridge stability. Cladistics 10: 295–304. University Press, Cambridge, 217-239. Brodo, I.M., Duran Sharnoff, S., Sharnoff, S. Fries, Th.M. 1857. De Stereocaulis et Pilophoris 2001. Lichens of North America. Yale commentatio. Wahlström & Co, Uppsala. Univeristy Press, New Haven and London. Fries, Th. M. 1858. Monographia Stereo- Cameroon, S.L., Miller, K.B., D’Haese, C.A., caulorum et Pilophorum. Nova Acta Regiae Whiting, M.F., Barker, S.C. 2004. Societatis Scientiarum Upsaliensis Seriei III 2: Mitochondrial genome data alone are not 307-380. enough to unambiguously resolve the Frost, D.R., Rodriques, M.T., Grant, T., Titus, relationships of Entognatha, Insecta and T.A. 2001. Phylogenetics of the lizard genus Crustacea sensu lato (Arthropoda). Cladistics Tropidurus ( Squamata: Tropiduridae: 20: 534-557. Tropidurinae): direct optimization, Carlin, G. 1990. Anteckningar om variationen descriptive efficiency, and sensitivity analysis hos några Stereocaulon-arter. Manuscript, 20 of congruence between molecular data and pp. + 14 Figs. morphology. Molecular Phylogenetics and Dahlkild, Å., Källersjö, M., Lohtander, K., Evolution 21: 352-371. Tehler, A. 2001. Photobiont diversity in the Frost, D.R., Grant, T., Faivovich, J., Bain, R.H., Physciaceae (Lecanorales). Bryologist 104: Haas, A., Haddad, C.F.B., De Sá, R.O., 527-536. Channing, A., Wilkinson, M., Donnellan, S.C., Dodge, C.W. 1929. A synopsis of Stereocaulon Taxworthy, C.J., Campbell, J.A., Blotto, B.L., with notes on some exotic species. Annales Moler, P., Drewes, R.C., Nussbaum, R.A., de Cryptogamie Exotique 2: 93-153. Lynch, J.D., Green, D.M., Wheeler, W.C. Dombrovskaya, A.V., 1996. Rod Stereocaulon na 2006. The Amphibian Tree of Life. Bulletin of territorii byvshego SSSR. Mir i Semja–95, Sankt- the American Museum of Natural History Peterburg. 297:1-371. Ekman, S., Tønsberg, T. 2002 Most species of Fryday, A.M., Coppins, B.J. 1996. A new Lepraria and Leproloma form a monophyletic crustose Stereocaulon from the mountains of group closely related to Stereocaulon. Scotland and Wales. Lichenologist 28: 513- Mycological Research 106: 1262-1276. 519. Eriksson, O.E. 2006. Outline of Ascomycota – Fryday, A.M., Glew, K.A. 2003 Stereocaulon 2006. Myconet 12: 1-82. nivale, comb. nov. Yet another crustose Farris, J.S., Albert, V.A., Källersjö, M., species in the genus. Bryologist 106: 565- Lipscomb, D., Kluge, A.G. 1996. Parsimony 568. jackknifing outperforms neighbour-joining. Galloway, D.J. 1988. Studies in Pseudocyphellaria Cladistics 12: 99–124. (lichens) I. The New Zealand species. Felsenstein, J. 1985. Confidence limits on Bulletin of the British Museum (Natural phylogenies: an approach using the History), Botany Series 17: 1-267. bootstrap. Evolution 39: 783-791. Gargas, A., DePriest, P.T., Grube, M., Tehler,

20 A. 1995. Multiple origins of lichen symbioses Springer Verlag, Berlin, Heidelberg, New in fungi suggested by SSU rDNA phylogeny. York, 165-188. Science 268: 1492-1495. Högnabba, F., Wedin, M. 2003. Molecular Goffinet, B., Bayer, R.J. 1997. Characterization phylogeny of the Sphaerophorus globosus of mycobionts of photomorph pairs in the species complex. Cladistics 19: 224–232. Peltigerineae (lichenized ascomycetes) based James, P.W., Henssen, A. 1976. The on internal transcribed spacer sequences of morphological and taxonomic significance of the nuclear ribosomal DNA. Fungal Genetics cephalodia. In: Brown, D.H., Hawksworth, and Biology 21: 228-237. D.L., Bailey, R.H., eds. Lichenology: Progress Goloboff, P.A. 1999. Analyzing large data sets and problems. Academic press, London, 22- in reasonable times: solutions for composite 77. optima. Cladistics 15: 415-428. James, T.Y., Kauff, F., Schoch, C.L., Matheny, Grant, T. 2003 Against sensitivity analysis in B.P, Hofstetter, V., Cox, C.J., Celio, G., phylogenetic systematics. In: Muona, J., ed. Gueidan, C., Fraker, E., Miadlikowska, J., Abstracts of the 21st annual meeting of the Lumbsch, H.T., Rauhut, A., Reeb, V., Arnold, Willi Hennig Society. Cladistics 19: 148-163. A.E., Amtoft, A., Stajich, J.E., Hosaka, K., Grant, T., Kluge, A.G. 2003. Data exploration Sung, G.-H., Johnson, D., O’Rourke, B., in phylogenetic inference: scientific, heuristic, Crockett, M., Binder, M., Curtis, J.M., Slot, or neither. Cladistics 19: 379-418. J.C., Wang, Z., Wilson, A.W, Schüßler, A., Grant, T., Kluge, A.G. 2004. Transformation Longcore, J.E., O’Donell, K., Mozley- series as an ideographic character concept. Stanridge, S., Porter, D., Letcher, P.M., Cladistics 20: 23–31. Powell, M.J., Taylor, J.W., White, M.M., Hawksworth, D.L. 1988. The variety of fungal- Griffith, G.W., Davies, D.R., Humber, R.A., algal symbioses, their evolutionary Morton, J.B., Sugiyama, J., Rossman, A.Y., significance, and the nature of lichens. Rogers, J.D., Pfister, D.H., Hewitt, D., Botanical Journal of the Linnean Society 96: Hansen, K., Hambleton, S., Shoemaker, R.A., 3-20. Kohlmeyer, J., Volkmann-Kohlmeyer, B., Heidmarsson, S., Mattson, J.-E., Moberg, R., Spotts, R.A., Serdani, M., Crous, P.W., Nordin, A., Santesson, R., Tibell, L. 1997. Hughes, K.W., Matsuura, K., Langer, E., Classification of lichen photomorphs. Taxon Langer, G., Untereiner, W.A., Lücking, R., 46: 519-520. Büdel, B., Geiser, D.M., Aptroot, A., Helms, G., Friedl, T., Rambold, G., Mayrhofer, Diederich, P., Schmitt, I., Schultz, M., Yahr, H. 2001. Identification of photobionts from R., Hibett, D.S., Lutzoni, F., McLaughlin, D.J., the lichen family Physciaceae using algal- Spatafora, J.W., Vilgalys, R. 2006. Re- specific ITS rDNA sequencing. Lichenologist constructing the early evolution of Fungi 33: 73–86. using a six-gene phylogeny. Nature 443: 818- Honegger, R. 1996. Mycobionts. In: Nash III, 822. T.H., ed. Lichen biology. Cambridge University Janies, M., Wheeler, W.C. 2002. POY version Press, Cambridge, 24-36. 3.0 documentation and command summary, Honegger, R. 2001. The Symbiotic Phenotype uppdate November 6, 2002. American of Lichen-Forming Ascomycetes. In: Hock, Museum of Natural History, New York, B., ed. The Mycota IX, Fungal Associations. available from ftp.amnh.org/pub/molecular.

21 Jeanmougin, F., Thompson, J.D., Gouy, M., Lumbsch, H.T., del Prado, R., Kantvilas, G. Higgins, D.G., Gibson, T.J. 1998. Multiple 2005. Gregorella, a new genus to accomodate sequence alignment with Clustal X. Trends in Moelleropsis humida and a molecular Biochemical Sciences 23: 403-405. phylogeny of Arctomiaceae. Lichenologist 37: Jørgensen, P.M. 1996. On the nomenclature of 291-302. lichen phototypes. Taxon 45: 663-664. Lumbsch, H.T., Schmitt, I., Palice, Z., Wiklund, Jørgensen, P.M. 1997. Lichen phototypes E., Ekman, S., Wedin, M. 2004. Supraordinal nature’s unmanageable misprints? Taxon 46: phylogenetic relationships of Lecanoro- 721-722. mycetes based on a Bayesian analysis of Jørgensen, P.M. 1998. What shall we do with combined nuclear and mitochondrial the blue-green counterparts? Lichenologist sequences. Molecular Phylogenetics and 30: 351-356. Evolution 31: 822-832. Jørgensen, P.M., Jahns, H.M. 1987. Muhria, a Lumbsch, H.T., Wirtz, N., Lindemuth, R., remarkable new lichen genus from Schmitt, I. 2002. Higher level phylogenetic Scandinavia. Notes from the Royal Botanic relationships of euascomycetes (Pezizo- Garden Edinburgh 44: 581-599. mycotina) inferred from a combined analysis Kirk, P.M., Cannon, P.F., David, J.C., Stalpers, of nuclear and mitochondrial sequence data. J.A., eds. 2001. Ainsworth & Bisby's dictionary of Mycological Progress 1: 57-70. the fungi, ninth ed. CAB International, Lutzoni, F., Kauff, F., Cox, C., McLaughlin, D., Wallingford. Celio, G., Dentinger, B., Padamsee, M., Kluge, A.G. 1997. Testability and the refutation Hibbett, D., James, T.Y., Baloch, E., Grube, and corroboration of cladistic hypotheses. M., Reeb, V., Hofstetter, V., Schoch, C., Cladistics 13: 81-96. Arnold, A.E., Miadlikowska, J., Spatafora, J., Kroken, S., Taylor, J.W. 2000. Phylogenetic Johnson, D., Hambleton, S., Crockett, M., species, reproductive mode, and specificity of Shoemaker, R., Sung, G.-H., Lücking, R., the green alga Trebouxia forming lichens with Lumbsch, H.T., O’Donnell, K., Binder, M., the fungal genus Letharia. Bryologist 103: Diederich, P., Ertz, D., Gueidan, C., Hansen, 645–660. K., Harris, R.C., Hosaka, K., Lim, Y.-W., Lamb, M.I. 1951. On the morphogy, phylogeny, Matheny, B., Nishida, H., Pfister, D., Rogers, and taxonomy of the lichen genus J., Rossman, A., Schmitt, I., Sipman, H., Stone, Stereocaulon. Canadian Journal of Botany 29: J., Sugiyama, J., Yahr, R., Vilgalys, R. 2004. 522-584. Assembling the fungal tree of life: progress, Lamb, M.I. 1977. A conspectus of the lichen classification, and evolution of subcellular genus Stereocaulon (Schreb.) Hoffm. Journal traits. American Journal of Botany 91: 1446- of the Hattori Botanical Laboratory 43: 191- 1480. 355. Lutzoni, F., Pagel, M., Reeb, V. 2001. Major Laundon, J.R. 1995. On the classification of fungal lineages are derived from lichen lichen photomorphs. Taxon 44: 387-389. symbiotic ancestors. Nature 411: 937-940. Lohtander, K., Oksanen, I., Rikkinen, J. 2003. Magnusson, A.H. 1926. Studies on boreal Genetic diversity of green algal and Stereocaula. Göteborgs Kungliga Vetenskaps- cyanobacterial photobionts in Nephroma och Vitterhets-samhälles Handlingar 30: 1-89. (Peltigerales). Lichenologist 35: 325–339. McNeill, J., Barrie, F.R., Burdet, H.M.,

22 Demoulin, V., Hawksworth, D.L., Marhold, Cladistics 15: 407-414. K., Nicolson, D.H., Prado, J., Silva, P.C., Skog, Nylander, W. 1858-1860. Synopsis methodica J.E., Wiersema, J.H., Turland, N.K., eds. 2006. Lichenum omnium hucusque cognitorum International Code of Botanical Nomenclature præmissa introductione lingua gallica tractata. I. (Vienna Code). ARG Gantner Verlag, Ruggell. L. Martinet, Paris. Miadlikowska, J., Lutzoni, F. 2004. Phylogenetic O’Brien, H., Miadlikowska, J., Lutzoni, F. 2005. classification of peltigeralean fungi Assessing host specialization in symbiotic (Peltigerales, Ascomycota) based on cyanobacteria associated with four closely ribosomal RNA small and large subunits. related species of the lichen fungus Peltigera. American Journal of Botany 91: 449-462. European Journal of Phycology 40: 363–378. Miadlikowska, J., McCune, B., Lutzoni, F. 2002. Oksanen, I., Lohtander, K., Paulsrud, P., Pseudocyphellaria perpetua, a new lichen from Rikkinen, J. 2002. A molecular approach to western North America. Bryologist 105: 1- cyanobacterial diversity in a rock-pool 10. community involving gelatinous lichens and Molína, M.C., Crespo, A., Blanco, O., Lumbsch, free-living Nostoc colonies. Annales Botanici H.T., Hawksworth, D.L. 2004. Phylogenetic Fennici 39: 93–99. relationships and species concepts in Parmelia Paulsrud, P., Lindblad, P. 1998. Sequence s. str. (Parmeliaceae) inferred from nuclear variation of the tRNALeu intron as a marker ITS rDNA and β-tubulin sequences. for genetic diversity and specificity of Lichenologist 36: 37-54. symbiotic cyanobacteria in some lichens. Morgenstern, B. 1999. DIALIGN2: improve- Applied and Environmental Microbiology 64: ment of the segment-to-segment approach 310-315. to multiple sequence alignment. Paulsrud, P., Rikkinen, J., Lindblad, P. 1998. Bioinformatics 15: 211-218. Cyanobiont specificity in some Nostoc- Myllys, L., Lohtander, K., Tehler, A. 2001. β- containing lichens and in a Peltigera aphthosa tubulin, ITS and group I intron sequences photosymbiodeme. New Phytologist 139: challenge the species pair concept in Physcia 517–524. aipolia and P. caesia. Mycologia 93: 335–343. Paulsrud, P., Rikkinen, J., Lindblad, P. 2000. Myllys, L., Stenroos, S., Thell, A., Ahti, T. 2003. Spatial patterns of photobiont diversity in Phylogeny of bipolar Cladonia arbuscula and some Nostoc-containing lichens. New Cladonia mitis (Lecanorales, Euascomycetes). Phytologist 146: 291-299. Molecular Phylogenetics and Evolution 27: Paulsrud, P., Rikkinen, J., Lindblad, P. 2001. 58–69. Field investigations on cyanobacterial Myllys, L., Stenroos, S., Thell, A., Kuusinen, M. specificity in Peltigera aphthosa. New 2007. High cyanobiont selectivity of epiphytic Phytologist 152: 117–123. lichens in old growth boreal forests of Phillips, A., Janies, D., Wheeler, W.C. 2000. Finland. New Phytologist 173: 621-629. Multiple sequence alignment in phylogenetic Nash III, T.H. 1996. Introduction. In: Nash III, analysis. Molecular Phylogenetics and T.H., ed. Lichen biology. Cambridge University Evolution 16: 317–330. Press, Cambridge, 1-7. Piercey-Normore, M.D., DePriest, P. T. 2001. Nixon, K.C. 1999. The parsimony ratchet, a Algal-switching among lichen symbioses. new method for rapid parsimony analysis. American Journal of Botany 88: 1490–1498.

23 Poelt, J. 1970. Das Kontzept der Artenpaaren Stenroos, S., Hyvönen, J., Myllys, L., Thell, A., bei den Flechten. Vertrage aus dem Ahti, T. 2002a. Phylogeny of the genus Gesamtgebiet der Botanik, Neue Folge 4: Cladonia s. lat. (Cladoniaceae, Ascomycetes) 187-198. inferred from molecular, morphological, and Poelt, J. 1972. Die Taxonomische Behandlung chemical data. Cladistics 18: 237–278. von Artenpaaren bei den Flechten. Botaniska Stenroos, S., Myllys, L., Thell, A., Hyvönen, J. Notiser 125: 77-81. 2002b. Phylogenetic hypotheses: Popper, K.R. 1983. Realism and the Aim of Cladoniaceae, Stereocaulaceae, Baeo- Science. Routledge, London. mycetaceae, and Icmadophilaceae revisited. Printzen, C., Kantvilas, G. 2004. Hertelidea, Mycological Progress 1: 267–282. genus novum Stereocaulacearum (asco- Stenroos, S., Stocker-Wörgötter, E., Yoshi- mycetes lichensiati). Bibliotheca Licheno- mura, I., Myllys, L., Thell, A., Hyvönen, J. logica 88: 539-553. 2003. Culture experiments and DNA Purvis, O.W., James, P.W. 1985. Lichens of the sequence data confirm the identity of Lobaria Coniston copper mines. Lichenologist 17: photomorphs. Canadian Journal of Botany 221-237. 81: 232-247. Riddle, L.W. 1910. The North American Summerfield, T.C., Eaton-Rye, J.J. 2006. species of Stereocaulon. Botanical Gazette 50: Pseudocyphellaria crocata, P. neglecta and P. 285-304. perpetua from the Northern and Southern Rikkinen, J., Oksanen, I., Lohtander, K. 2002. Hemispheres are a phylogenetic species and Lichen guilds share related cyanobacterial share cyanobionts. New Phytologist 170: symbionts. Science 297: 357. 597-607. Räsänen, V. 1943. Das system der Flechten. Summerfield, T.C., Galloway, D.J., Eaton-Rye, Acta Botanica Fennica 33: 1-82. J.J. 2002. Species of cyanolichens from Schmitt, I., Messuti, M.I., Feige, G.B., Lumbsch, Pseudocyphellaria with indistinguishable ITS H.T. 2001. Molecular data support rejections sequences have different photobionts. New of the generic concept in the Coccotrema- Phytologist 155: 121–129. taceae (Ascomycota). Lichenologist 33: 315- Takanashi, K., Tsubota, H., Deguchi, H. 2006b. 321. Taxonomic revision or the genus Sticta Schulmeister, S., Wheeler, W.C., Carpenter, (Lobariaceae, Lichenized Ascomycota) in J.M. 2002. Simultaneous analysis of the basal East Asia. In: Congress Handbook & Abstracts lineages of Hymenoptera (Insecta) using Book 1, 8th International Mycological Congress, sensitivity analysis. Cladistics 18: 455-484. Cairns 21-25 August 2006. Slowinski, J.B. 1998. The number of multiple Takanashi, K., Wang, L.-S., Tsubota, H., alignments. Molecular Phylogenetics and Deguchi, H. 2006a. Photosymbiodemes Sticta Evolution 10: 264–266. wrightii and Dendriscocaulon sp. (lichenized Sober, E. 1988. Reconstructing the Past: ascomycota) from Yunnan, China. Journal of Parsimony, Evolution and Inference. MIT press, the Hattori Botanical Laboratory 100: 783- Cambridge, Massachusetts. 796. Stenroos, S., DePriest, P.T. 1998. SSU rDNA Tehler, A. 1982. The species pair concept in phylogeny of cladoniiform lichens. American lichenology. Taxon 31: 708-717. Journal of Botany 85: 1548–1559. Tehler, A. 1983. The genera Dirina and

24 Roccellina (Roccellaceae). Opera Botanica 70: Tuckerman, M.A. 1872. Genera lichenum: an 1-86. arrangement of the North American lichens. Tehler, A. 1996. Systematics, phylogeny and Journal Steam Press, Lewiston, Maine. classification. In: Nash III, T.H., ed. Lichen Vainio, E.A. 1890. Étude sur la Classification biology. Cambridge University Press, Naturelle et la Morpologie des Lichens du Cambridge, 217-239. Brésil. Acta Societatis pro Fauna et Flora Tehler, A., Little, D., Farris, J.S. 2003. The full- Fennica 7 (1): i–xxx, 1–247; (2): 1–256. length phylogenetic tree from 1551ribosomal Vainio, E.A. 1915. Additamenta ad sequences of chitinous fungi, Fungi. Lichenographiam Antillarium illustrandam. Mycological Research 107: 901-916. Annales academiae scientiarum fennicae Ser. Thell, A., Feuerer, T., Kärnefelt, I., Myllys, L., A. 6: 1-226. Stenroos, S. 2004. Monophyletic groups Wang, L., Jiang, T. 1994. On the complexity of within the Parmeliaceae identified by ITS multiple sequence alignment. Journal of rDNA, β-tubulin and GAPDH sequences. Computational Biology 1: 337–348. Mycological Progress 3: 297–314. Wedin, M., Döring, H. 1999. The phylogenetic Thell, A., Stenroos, S., Feuerer, T., Kärnefelt, I., relationship of the Sphaerophoraceae, Myllys, L., Hyvönen, J. 2002. Phylogeny of Austropeltum and Neophyllis (lichenized cetrarioid lichens (Parmeliaceae) inferred Ascomycota) inferred by SSU rDNA from ITS and β -tubulin sequences, sequences. Mycological Research 103: morphology, anatomy, and secondary 1131–1137. chemistry. Mycological Progress 1: 335–354. Wedin, M., Wiklund, E., 2004. The phylo- Thomas, M.A., Ryan, D.J., Farnden, K.J.F., genetic relationships of Lecanorales suborder Galloway, D.J. 2002. Observations on Peltigerineae revisited. Symbolae Botanicae phylogenetic relationships within Lobariaceae Upsalienses 34: 469-475. Chevall. (Lecanorales, Ascomycota) in New Wedin, M., Döring, H., Ekman, S. 2000. Zealand, based on ITS-5.8S molecular Molecular phylogeny of the lichen families sequence data. Bibliotheca Lichenologica 82: Cladoniaceae, Sphaerophoraceae and Stereo- 123-138. caulaceae (Lecanorales, Ascomycotina). Thomas, M.A., Ryan, D., Galloway, D.J. 2000. Lichenologist 32: 171–187. The phylogenetic relationship of the New Wedin, M., Döring, H., Gilenstam, G. 2004. Zealand Lobariaceae based on ITS-5.8S Saprotrphy and lichenization as options for moleclar sequences data. In: Book of Abstracts, the same fungal species of different Progress and Problems in Lichenology at the substrara: envionmental plasticity and fungal Turn of the Millenium, The fourth IAL lifestyles in the Stictis-Conotrema complex. symposium, Barcelona 3-8 september 2000. New Phytologist 164: 459-465. Univeristy of Barcelona, 95. Wedin, M., Wiklund, E., Crewe, A., Döring, H., Timdal, E. 2002. Stereocaulon cumulatum comb. Ekman, S., Nyberg, Å., Scmitt, I., Lumbsch, nov., another crustose species in the genus. H.T. 2005. Phylogenetic relationships of Lichenologist 34: 7-11. Lecanoromycetes (Ascomycota) as revealed Tschermak-Woess, E. 1988. The algal partner. by analyses of mtSSU and nLSU rDNA In: Galun, M., ed. Handbook of lichenology, Vol. sequence data. Mycological Research 109: 1. CRC Press, Boca Raton, Florida, 39-92. 159-172.

25 Wheeler, W.C. 1995. Sequence alignment, the Lecanorales suborder Peltigerineae. parameter sensitivity, and the phylogenetic Cladistics 19: 419-431. analysis of molecular data. Systematic Biology Wirtz, N., Lumbsch, H.T., Green, T.G.A., Türk, 44: 321–331. R., Pintado, A., Sancho, L., Schroeter, B. Wheeler, W.C. 1996. Optimization alignment: 2003. Lichen fungi have low cyanobiont the end of multiple sequence alignment in selectivity in maritime Antarctica. New phylogenetics? Cladistics 12: 1-9. Phytologist 160: 177–183. Wheeler, W.C. 2000. Heuristic reconstruction Yahr, R., Vilgalys, R., DePriest, P.T. 2004. of hypothetical-ancestral DNA sequences: Strong fungal specificity and selectivity for sequence alignment versus direct algal symbionts in Florida scrub Cladonia optimization. In: Scotland R., Pennington lichens. Molecular Ecology 13: 3367–3378. R.T., eds. Homology and systematics: coding Yahr, R., Vilgalys, R., DePriest, P.T. 2006 characters for phylogenetic analysis. Systematic Geographic variation in algal partners of Association, Taylor & Francis, London, 106- Cladonia subtenuis (Cladoniaceae) highlights 113. the dynamic nature of a lichen symbiosis. Wheeler, W.C. 2001. Homology and the New Phytologist 171: 847–860 optimization of DNA sequence data. Yoshimura, I. 1998. Vainio and Lobaria, old and Cladistics 17: S3–S11. modern concepts. In: Marceli, M.P., Ahti, T., Wheeler, W.C. 2002. Optimization alignment: eds. Recollecting Edvard August Vainio. down, up, error, and improvements. In: CETESB, Sao Paulo, 85-94. DeSalle, R., Giribet, G., Wheeler, W. C., eds. Yoshimura, I. 2002. Lobariella. In: Nash III, T. Techniques in molecular systematics and evo- H., Ryan, B.D., Gries, C., Bungartz, F., eds. lution. Birkhäuser, Basel, 55–69. Lichen flora of the Great Sonoran Desert Region Wheeler, W.C., Aagesen, L., Arango, C.P., vol. I. Lichens Unlimited, Arizona State Faivovich, J., Grant, T., D’Haese, C., Janies, University, Tempe, Arizona, 207-272. D., Smith, Wm.L., Varón, A., Giribet, G. Yuan, X., Xiao, S., Taylor, T.N. 2005. Lichen- 2006. Dynamic Homology and Phylogenetic Like Symbiosis 600 Million Years Ago. Systematics: A Unified Approach Using POY. Science 308: 1017-1020. American Museum of Natural History, New York. Wheeler, W.C., Gladstein, D.S., De Laet, J. 1996–2003. POY: Phylogeny reconstruction via optimization of DNA data. Computer software distributed by the authors and from the American Museum of Natural History, New York, available from ftp://ftp.amnh.org/ pub/molecular/poy. White, F.J., James, P.W. 1988. Studies in the genus Nephroma II. The southern temperate species. Lichenologis 20: 103-166. Wiklund, E., Wedin, M. 2003. The phylogenetic relationships of the cyanobacterial lichens in

26