Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

I. Introduction

Our understanding of evolution and natural selection is often overshadowed by the popular view of the “struggle for existence”, most often envisioned as stressful competition for resources within the same habitat. This notion distracts from other fundamental evolutionary phenomena, such as the novelty of life forms stemming from symbioses. To the naked eye, a premier example is provided by lichen symbiosis, which gives rise to light-exposed and unique phenotypes, the lichen thalli, which are highly diverse in shape, chemical content and ecological demands (Grube and Hawksworth 2007). With an estimated age of 400–600 mio years, lichens are among the oldest symbioses of fungi (Yuan et al. 2005). Conservative estimates suggest that more than 18,800 fungal species evolved in a lichenized stage (Feuerer and Hawksworth 2007). The number may still be higher, because phenotypically poorly recognized species are hidden in taxonomically described lineages. Meanwhile, new data highlight lichens also as more complex symbioses than previously thought. In this chapter we focus on the diversity of bacteria in lichens and their possible roles.

A widely accepted definition describes lichens as “an ecologically obligate, stable mutualism between an exhabitant fungal partner (the mycobiont) and an inhabitant population of extracellularly located unicellular or filamentous algal or cyanobacterial cells (the photobiont)” (Hawksworth and Honegger 1994). In fact, the symbiosis is obligately required for the formation of the reproductive structures by the fungal partners and the thallus structure in general. The interpretation of lichens as a clear case of mutualism—as put forward in most biology textbooks—is difficult to prove and has been a matter of controversy in the scientific community. The lichen symbiosis has originally even been interpreted as a mild sort of algal “slavery” (a “helotism”, Schwendener 1869): a dominant fungal partner usually shapes the lichen structure to form a thallus that provides shelter for the controlled growth of one or several algal or cyanobacterial organisms as primary partners. The sexuality of the algae is generally suppressed in their lichenized stages but algal cell densities are much higher than in free-living stages.

Advanced types of lichen thalli represent without doubt the most complex structures in the fungal kingdom. Unlike the mycelia of most other fungi, the lichen thalli are exposed to the light and develop important diagnostic characters for determination. Moreover, most lichen symbioses are robust and endure the stresses of a dynamic environment. Lichen thalli may persist for many years, even centuries, although growth and metabolic activity might be restricted to their young and regenerating parts (see also Honegger 2008). It is therefore not surprising that lichens are also preferred habitats of many other microorganisms. Particularly well-known are the lichenicolous fungi, which had been studied even before the symbiotic nature of lichens was discovered in the nineteenth century. Lichenicolous fungi develop diagnostic reproductive structures on their host lichens. The biological impact of such fungi on lichen biology is fairly well-known and ranges from parasitic to commensalic life styles. So far, more than 1,500 species of lichenicolous fungi have been described (Lawrey and Diederich 2003). Other fungi, which are isolated from the surface of lichens, appear to be non-specific and symptomless epibionts (Petrini et al. 1990; Prillinger et al. 1997). There are also diverse inconspicuous fungi which occur internally in lichens (Girlanda et al. 1997; Arnold et al. 2009).

In contrast, the significance of bacterial colonization in lichen thalli has been poorly considered in lichenological research in the past, despite the ubiquity of bacteria in microscopic images (Figs. 17.1, 17.2). Probably the first observations about bacteria were made by Thaxter (1892). He described Chondromyces lichenicola (now placed as Melittangium lichenicola in Myxobacteriaceae; McCurdy 1971) as a parasite on lichens. The infection is characterized by tiny knob-like structures on the host. This species, however, has hardly been collected and rarely seems to form its diagnostic phenotypes, although it can be detected among lichen-associated bacterial communities (unpublished data). Uphof (1925) suggested presence of purple bacteria in the thalli of Herpothallon rubrocinctum (as “Chiodecton sanguineum”), although this was then disapproved and seen as misinterpretation of crystallized compounds (Suessenguth 1926). About at the same time, Cengia-Sambo (1925, 1931) also observed bacteria in lichens and assigned them to Azotobacter on the surface of the (Nostoc-containing) cephalodia of Peltigera aphthosa. He coined the term “polysymbiosis” for such multipartite relationships. However, these works were not much noticed by the lichenologists of that time. Further evidence for the presence of bacteria in lichens was then provided by a series of papers by Russian and Armenian authors. Henckel and Yuzhakova (1936) and Iskina (1938) detected nitrogen-fixing bacteria in lichens by cultivation on nitrogen-free Ashby medium and assigned these strains to Azotobacter. Using phase contrast microscopy, Henckel (1946) discovered large numbers of bacteria (assigned to Azotobacter) on different species of lichens.

Fig. 17.1.
figure 1

Bacterial cells are present on the surface and to a varying degree immersed in the intercellular matrix of the internal podetium surface in Cladonia arbuscula. SEM image by E. Stabentheiner

Full size image
Fig. 17.2.
figure 2

Biofilm-like bacterial localization of the internal surface of the reindeer lichen Cladonia arbuscula. Acridine orange staining; CLSM image by M. Cardinale. Bar 10 μm

Full size image

Sporadic findings of lichen-associated bacteria have been published since then. Krasilnikov (1949) could not confirm the previous findings of lichen-associated Azotobacter in his samples, but found other bacterial groups, such as pseudomonads, Actinobacteria and Myxobacteria. Refusing the term symbionts for lichen-associated bacteria, he assumed the presence of Azotobacter is random and that the composition and abundance of microbial consortia depends on the state of the lichen thallus. At the same time, Navahradak (1949) reported the common presence of cellulose-degraders, including Cellvibrio, in lichens. Scott (1956) detected Azotobacter in lichens and found that a part of measured nitrogen-fixation might be provided also by bacteria other than cyanobacteria. Besides Azotobacter, other genera were repeatedly reported from lichens, such as Bacillus (Henckel and Plotnikova 1973), Beijerinckia (Panosyan and Nikogosyan 1966), Clostridium (Iskina 1938) and Pseudomonas (Henckel and Plotnikova 1973). Nitrogen fixation for lichens was repeatedly considered as a function of bacteria in these and other works (e.g. Lambright and Kapustka 1981), while the presence of actinobacteria in cyanobacterial lichens prompted Zook (1983) to suggest also a defensive role for bacteria in lichens. Lenova and Blum (1983) supposed that up to millions of bacterial cells could be present per gram of a lichen thallus. The names of culturable bacteria of these early works on lichen-associated bacteria reflect the state of taxonomical knowledge of the times when classification relied only on phenotypical and physiological methods. In the light of contemporary methodology, the previous determinations are certainly outdated and incomplete.

Current research with molecular and novel microscopic methods established baseline information about diversity, abundance and localization of bacteria on lichen thalli (Cardinale et al. 2006, 2008; Grube and Berg 2009; Grube et al. 2009; Hodkinson and Lutzoni 2009; Selbmann et al. 2010; Bates et al. 2011; Bjelland et al. 2011; Mushegian et al. 2011; Hodkinson et al. 2012). These assessments revealed that Alphaproteobacteria are predominant in growing parts of lichens (Grube et al. 2009; Mushegian et al. 2011; Cardinale et al. 2012a) but other bacterial groups are also found in significant amounts.

II. Multitrophic Interactions in Lichens

Lichens are widely popularized as a two-tier partnership of fungi and algae, but it was recognized in the early twentieth century that more than one photoautotrophic partners can be present, e.g. when the thalli include cyanobacteria. About 2 % of the lichenized fungal species have cyanobacterial symbionts in addition to green-algal partners. These associations are known as tripartite lichens. The nitrogen-fixing role of the incorporated cyanobacteria is well-established (Cengia-Sambo 1931; Millbank and Kershaw 1969). The cyanobacteria in tripartite lichens are often internalized by the fungal thallus in special organs called cephalodia. Actually, these structures were described long before the symbiotic nature of lichen was discovered (Flörke 1819). In the cephalodia, the cyanobacterial Nostoc strains have a higher relative number of heterocysts, compared with lichens containing Nostoc as a photosynthetic partner alone. This conforms to a functional and spatial separation of primary carbohydrate- and nitrogen-fixation (as nitrogen-fixation is catalyzed in the cyanobacterial heterocysts). The division of work is not complete, since the cyanobacteria are still photosynthetically active. Beside their presence in surface-borne or internal cephalodia, cyanobacteria in tripartite lichens can be present as a layer beneath green-algae (as in the genus Solorina), or more unspecifically as colonies adjacent to the thallus. Such affinity to neighboring cyanobacterial colonies has been recognized by the term cyanotrophy (Poelt and Mayrhofer 1988). Cephalodiate and cyanotrophic lichens are good indicators for the possible re-distribution of metabolic functions to more than two partners in the lichen symbiosis.

The concept of lichens as self-contained ecosystems, originally put forward by Farrar (1976), is corroborated when additionally associated microorganisms are considered. Lichenicolous fungi, some of which digest the host structures might be interpreted as the degraders in a simple system with algal producers and the lichen mycobiont as consumers. However, the finding of commensalic and symptomless fungi, and highly abundant, host-specific and diverse bacterial communities add to the biological complexity in lichen symbioses. The functional roles of these additional players in the lichen ecosystems are still far from settled. Here we want to give an overview of what has been achieved so far.

Earlier approaches to the role of bacteria in lichens relied on functions found in the culturable fraction of the associated bacterial community. Gonzáles et al. (2005) used an enrichment medium to isolate actinobacteria from tropical and polar lichens. Most strains represented Micromonospora and Streptomyces, but other Actinobacteria were present as well. Half of the isolated strains showed antimicrobial activities, with activity against Gram-positive Staphylococcus aureus occurring at highest frequency (23%). This confirms previous assumptions of a defensive role of lichen-associated bacteria. Similarly N-free enrichment cultures were used by Liba et al. (2006) to find nitrogen fixers by acetylene-reducing assays in green-algal lichens from coastal Brazil. The results were confirmed by nifH gene dot blots. The 17 detected strains were from Gammaproteobacteria (such as Acinetobacter, Pantoea, Pseudomonas, Serratia, Stenotrophomonas). Further analysis revealed excretion of amino acids and indole acetic acid (IAA) in 14 strains. Eight strains solubilized phosphate and four released ethylene. Stenotrophomonas strains released both ethylene and IAA. These data would support a role of bacteria in nitrogen fixation in green-algal lichens, with a role in nutritional and hormonal amendment. These functions have not been experimentally tested in lichen thalli, but an artificial co-culture study of lichen mycobionts with nitrogen-fixing bacteria indicated an increased capacity of rock weathering with bacterial symbionts (Seneviratne and Indrasena 2006).

However, the culturable bacteria are only a minor part of the communities. Studies using fluorescence in situ hybridization and confocal laser scanning microscopy (FISH-CLSM) clearly showed that Alphaproteobacteria represent a dominant fraction in the studied lichen parts and form extensive, often biofilm-like colonies (Cardinale et al. 2008; Fig. 17.3). Unfortunately, the most abundant lichen-dwelling Alphaproteobacteria have not been found among cultured isolates so far. Other bacterial groups occur often in smaller numbers, and it seems that culturable bacteria are present in rather low numbers in the growing parts. The presence of endocellular bacteria is still uncertain, although Cardinale et al. (2008) already indicated that bacteria may penetrate hyphal walls.

Fig. 17.3.
figure 3

Extensive colonies of Alphaproteobacteria (darker spots) at the lower part of the soil-inhabiting lichen Arthrorhaphis citrinella. 3-D reconstruction of a FISH-CLSM image by L. Muggia and B. Klug

Full size image

Fingerprinting analysis using SSCP revealed host species specificity of the lichen-associated bacterial communities (Grube et al. 2009), which was subsequently confirmed by pyrosequencing studies (e.g. Bates et al. 2011). In rock-inhabiting crustose lichens, which suffuse the uppermost substrate layers with their entire surface, specific influence is also seen below the lichen thalli (Bjelland et al. 2011). Although a higher bacterial diversity is observed in the underlaying rocks, the abundance of bacteria is higher in the epilithic lichen thalli. These results demonstrated that the lichen–rock interfaces are complex habitats, where the macroscopic lichen structures influence the composition of microbial metacommunities. Bjelland et al. (2011) also showed that Acidobacteria—another group of mostly unculturable bacteria—are a significant component in three of four studied rock-inhabiting lichens. Recent 454-amplicon sequencing data confirmed substantial numbers of Acidobacteria in certain host species (Mushegian et al. 2011; Hodkinson et al. 2012; Pankratov 2012; Grube et al., unpublished data).

A first analysis of putative functions at the protein level in an entire lichen system was provided by Schneider et al. (2011). Environmental proteomics techniques gave insight into the total protein composition of the lung lichen Lobaria pulmonaria. Their data revealed a high number of protein reads from prokaryotes, which was equal to the eukaryotic algal partner in this lichen. Among bacteria the majority of reads were from Alphaproteobacteria, confirming the results of DNA-based analyses, but also Archaea (0.5 %) were detected. As expected, most of the green algal proteins were involved in energy production and conversion. Carbohydrate transport and metabolism played a role in both algal and fungal proteins. Fungal functions were more diverse, and considerable numbers of proteins seem involved in biogenesis. Post-transcriptional modifications, protein turnover and chaperones are also important, but form a higher ratio in the bacterial reads. The involvement of bacterial nitrogen fixation was not apparent from Schneider et al. (2011). Lack of nitrogenase activity was also revealed in preliminary metatranscriptomic data (Hodkinson 2011), even though nifH genes were previously detected in green algal lichens by PCR approaches (Grube et al. 2009).

Some of the protein sequences indicated production of secondary metabolites by the prokaryotic partners. Lichens are well-known for their diversity of bioactive molecules, but so far most are products of fungal origin, whereas only very few cyanobacterial compounds are known. The bacterial production of compounds in lichen thalli certainly requires further attention, because such compounds may contribute to functional roles of bacteria in lichen biology.

The functional assignments of proteins provide a rather general insight in the functional network of the lichen symbiosis but cannot predict substrates and activities of expressed proteins. It is also necessary to analyze the protein levels across various physiological stages (e.g. dry to wet thallus conditions) to understand the extent of proteomic variation. Moreover, a combined -omics approach is clearly required to uncover the diverse bacterial functions in lichens.

So far data from culture-based and culture-independent studies suggest a multitude of bacterial functions in the lichen symbioses, in addition to functions of fungal and algal symbionts (Fig. 17.4; Grube and Berg 2009). One important additional source of information will be the detailed study of the lichen metabolome and to specifically target low-molecular-weight substances which might have signaling functions. Further, the analysis of genomes from lichen symbionts will reveal possible ancient horizontal gene transfers as footprints of a long history of bacterial-fungal interactions (Schmitt and Lumbsch 2009; McDonald et al. 2012).

Fig. 17.4.
figure 4

Extended concept of the lichen symbiosis, including the association with bacteria

Full size image

III. Variation of Bacterial Composition Within and Among Conspecific Thalli

Only few studies so far analyzed the variation of the lichen-associated microbiome within single thalli. Mushegian et al. (2011) studied intrathalline variation of multiple individuals of closely related foliose lichens with T-RFLP and pyrosequencing. They found that the internal part of lichens comprises richer and more consistent microbiota, whereas the margins are species-poor and more disparate in their composition. The authors concluded that the central parts of these lichens had more time for a consolidation of the bacterial communities whereas marginal parts comprise more random communities. This is contrasted by a microscopic study of Cardinale et al. (2012a), who found that the bacterial community of the growing parts of reindeer lichen thalli is rather uniform and composed of abundant Alphaproteobacteria, whereas more variation is found in the ageing parts. The somewhat contrasting results can be explained by the difference in the growth form of the investigated lichens. The central parts of the foliose species analyzed by Mushegian et al. (2011) are composed of a mixture of old and regenerative, young thallus parts. One species produces fruit-bodies, whereas the other species forms isidia as mitotic outgrowths of thalli for joint symbiont dispersal. This mixture of differently aged structures might account for the more complex community found in central thallus parts. In contrast, the shrub-like reindeer lichen studied by Cardinale et al. (2012a) has a clear age gradient from the exposed growing tips to the substrate-near, degrading thallus basis.

Cardinale et al. (2012a) also studied the effects of light exposition as an additional factor for microbiome variation. Exposition was clearly shown to affect the microbial communities in old and young parts of thalli. Old parts in shaded sites displayed a distinct increase of Betaproteobacteria, and of yet unknown Bacteria, whereas Alphaproteobacteria are less prominent.

IV. Biogeography

Many lichen species have extremely wide-ranging geographic distributions, with the same species often found across continents and hemispheres. Another line of lichen microbiome research therefore focuses on the question whether the host-specific composition of bacterial communities follow the distribution of their lichen hosts. Hodkinson et al. (2012) found evidence for geography as an important factor shaping lichen-associated community structures, but their analyses included several distinct species, sampled across the ecologically dissimilar sampling sites.

Cardinale et al. (2012b) focused on Lobaria pulmonaria as a single widespread species for biogeographic comparison of two groups of lichen-associated bacteria. As a flagship species in conservation biology and an indicator of forest continuity, Lobaria pulmonaria has a quite narrow ecological range but is found in different continents. The authors compared samples from sites in Europe (at various distances from each other) and in African Islands. The analysis of SSCP fingerprints indicated a higher correlation of the predominant Alphaproteobacteria with geography than found for Burkholderiales. The latter is regularly detected in the culturable fraction but apparently present at low abundance in lichens. Cardinale et al. (2012b) also suggest that Alphaproteobacteria are readily co-dispersed in joint fungal–algal propagules, called isidia, whereas Burkholderiales (Betaproteobacteria) would more likely be taken up from the local environment. This pattern may also underscore the tight association of lichens with Alphaproteobacteria . Printzen et al. (2012) focuses on the alphaproteobacterial community in another lichen species with extremely wide geographic distribution, Cetraria aculeata. This study revealed a less diverse community in polar habitats (in contrast to soils in polar latitudes; e.g. Chu et al. 2010), but antarctic and arctic communities were more similar to each other than to those of lichen samples of other proveniences. Interestingly, this pattern agrees with what has been reported previously for the lichen photobiont partner. Studies of the algal partners suggested a response of lichens to environmental conditions in form of symbiont switches or alteration of their relative ratios. We argue the bacterial community similarly responds to habitat conditions, which could eventually increase the adaptivity of the holobiont.

V. New Bacterial Species from Lichens

Lichens are a rich source of new bacterial strains and lineages. Especially actinobacterial species were isolated from lichens and described (Li et al. 2007; An et al. 2009; Yamamura et al. 2011). Cardinale et al. (2011) described a new actinobacterial species, Frondihabitans cladoniiphilus, from the reindeer lichen Cladonia arbuscula. This new species has interesting features which might be connected with an adaptation to the lichen habitat. In contrast to close relatives, cells of this species can form autoaggregates, which might be important for host attachment. A new species has also been discovered in Betaproteobacteraceae (Herminiimonas saxobsidens is a new member of Oxalobacteraceae; Lang et al. 2007).

Apart from new culturable species described in Actinobacteria and Betaproteobacteria, new lineages of commonly lichen-associated but still uncultured bacteria have been detected in Alphaproteobacteria. The lichen-associated Rhizobiales 1 clade (LAR1; Hodkinson and Lutzoni 2009) has frequently been found in association with green-algal lichens (all from North America; Bates et al. 2011; Hodkinson et al. 2012), and this lineage was also present in a sequence library derived from Antarctic lichens (de la Torre et al. 2003). LAR1 has been characterized so far only by 16rRNA gene sequences. A study of Antarctic lichens revealed even more yet unclassified culturable bacterial species, representing several lineages, including Actinobacteria, Firmicutes, Proteobacteria and Deinococcus (Selbmann et al. 2010). A phylogenetic analysis revealed that many of the strains cluster in well-supported groups that only comprised bacteria from lichens. All strains grew well at low temperatures but also at 25 °C, and therefore are psychrotolerant rather than psychrophilic.

VI. Compounds Produced by Lichen-Associated Bacteria

Bacteria in lichens are also a valuable source of effective new secondary metabolites. A strain of Streptomyces, isolated from the reindeer lichen Cladonia uncialis is particularly exciting due to the production of several novel compounds. Among these, the enediyne uncialamycin (Davies et al. 2005) is a compound with strong antibacterial activity against the human pathogens Burkholderia cepacia (MIC = 0.001 μg/mL) and Staphylococcus aureus (MIC = 0.0064 μg/mL). Uncialamycin was only recently also synthesized in the laboratory (Nicolaou et al. 2007). The same Streptomyces strain (“S. uncialis” ined.) also produces a series of new bis-indole alkaloids, the cladoniamides (Williams et al. 2008). Cladoniamides are unusual among bis-indole natural products: most bis-indoles have an indolocarbazole structure, whereas the cladoniamides have a rarely observed indenotryptoline structure (Fig. 17.5). Their biosynthetic gene cluster (cla) has now been characterized in more detail (Ryan 2011) and compared with the BE-54017 (abe) gene cluster from environmental DNA. BE-54017 and its derivatives are structurally similar to the cladoniamides. The elucidation of the cla gene cluster and comparison with the abe cluster supports the hypothesis that indenotryptoline cores are biosynthetically derived from the oxidative rearrangement of an indolocarbazole precursor. Another lichen-derived streptomycete strain produces the tetrapeptide lichostatinal (Lavallée 2011), which represents a cathepsin K inhibitor and is thus of interest for combating osteoporosis. Yet another Streptomyces strain isolated from lichens produced a new angucycline (JBIR-88) and a new butenolide (JBIR-89), which also displayed inhibitory effects on certain cell lines and a bacterial strain (Motohashi et al. 2010).

Fig. 17.5.
figure 5

Examples of products from lichen-associated Steptomycetes

Full size image

VII. Biotechnological Applications

The biotechnological potential of lichen-associated bacteria is directly related to ecological, morphological and chemical characteristics of the hosts. For example, the ecology of lichens is often characterized by strongly changing conditions of water availability. Gasser et al. (2012) found a high proportion of bacterial poly-β-hydroxybutyrate (PHB) producers as well as genes involved in PHB synthesis in different lichen species. Synthesis of storage compounds such as PHB is a strategy of Gram-negative bacteria to increase survival in changing environments (Steinbüchel and Valentin 2006). Another example is the use of lichen-associated bacteria as antagonists towards fungal plant pathogens for biological plant protection. In their natural environment, pathogen defense against parasites could be one function of the bacterial community in lichens (Grube et al. 2009): up to 100 % antagonists against the plant pathogens Alternaria alternata and Phytophthora infestans were found in the culturable fraction of different lichen species (unpublished data).

These are only two examples for promising biotechnological applications with lichen-associated bacteria, but many more may be developed in the future. We also consider lichens as sources of novel antibiotics and new signalling molecules, or enzymes working under extreme conditions.